DEVELOPMENT OF NOVEL FAR -RED/NEAR -INFRARED DYE -HCRBPII BASED IMAGING TAGS FOR BACKGROUND -FREE LIVE CELL IMAGING By Wei Sheng A DISSERTATION Submitted to Michigan State University in partial fulfillment of the requirements for the degree of Chemistry Ñ Doctor of Philosophy 2019 ABSTRACT DEVELOPMENT OF NOVEL FAR -RED/NEAR -INFRARED DYE -HCRBPII BASED IMAGING TAGS FOR BACKGROUND -FREE LIVE CELL IMAGING By Wei Sheng Modern fluorescence imaging technologies, including deep-tissue imaging and super -resolution microscop ies, require novel fluorescent labeling tags possessing non-conventional optical features , among which most desired ones are high brightness in the far-red/near-infrared (NIR) region and turn -on/off control in a spatiotemporal manner. Previously, we demonstrated the ability of fine tuning the absorption spectra of a protein -bound natural chromophore over an unprecedented range (474 ~ 664 nm). The goal of this PhD research is to exploit protein -ligand interactions for the develop ment of protein -based pigment s as NIR fluorescent tag s for background -free live cell imaging. In the past half century, tremendous efforts have been invested in the optimization and derivatization of GFP-like fluorescent proteins (FPs). More recently, growing attention on phytochrome -based FPs has even upsized the rep ertoire of available FPs with many enhanced optical features . Giving this advancement, certain pit falls are still limiting their uses in modern fluorescence imaging. In this context, synthetic dyes provide a broad er chemical space for tailoring desired optic features including spectral wavelengths, brightness, stability, and many more photophysical and /or photochemical functionalities. To achieve high contrast imaging with minimal background interference , three different strategies have been applied here . 1) NIR emission is approached by utilizing a dye capable of specific complexation with a target pro tein via imine bond formation. Upon protonation of the imine, the complex experiences a large bathochromic shift as a result of a strong intramolecular charge transfer (ICT) process . A light -triggered imine isomerization is further incorporated to furnish a photoswitchable tag and negate the routine wash steps in live cell experiments. Rational protein engineering affords a faster variant that allows unprecedented spatiotemporal control of this no -wash bright NIR imaging. (2) A rare organic super photobase is identified, exhibiting a 14 -unit change in pKa upon light excitation. Steady -state and ultrafast spectroscopic measurements ascribe this event to an excited -state proton transfer (ESPT) process. This ESPT feature is recapitulated in a protein -ligand mic ro-environment, yielding protein -dye complexes with extremely high fluorescence quantum yields (up to 92%) and large pseudo -Stokes shifts (> 200 nm). Our optimal mutant bound to the dye boasts millisecond binding rate and enables live cell imaging with neg ligible background. (3) A general approach to fluorogenicity , i.e., the ability to turn on fluorescence , is designed by coupling a quenching moiety capable of photoinduced electron transfer (PeT) to our dyes. The fluorescence is negligible before the Micha el addition of engineered cysteine residue (the trigger) with the quencher moiety. A 30 -fold fluorescence enhancement is achieved in vitro with an electronically tuned quencher group. Currently, further modifications are in progress to optimize the quenche d system for in vivo applications. iv Dedicated to Anna, my parents, my grandma, and my dear friends, for your everlasting encouragement and enlightenment v ACKNOWLEDGEMENT S The first and most important person in my PhD research period is my advisor Professor Dr. Babak Borhan. I am exceedingly grateful to him for his inculcation of me with an independent and resilient research attitude, which I deem as the most critical charac ter he incarnated into my mind that is beneficial for not only my research, but also my reflections towards life. Babak is a very knowledgeable and insightful scholar with endless energy and enthusiasm for chemistry research. As spotlighted by his three distinct research areas that span widely from total synthesis and synthetic methodologies to bioorganic chemistry and CD spectroscopies, his well -rounded figure as a Ò chemistÓ (rather than a pure organic chemist ) always inspires me to grow into the same type of multifaceted researcher that solves scientific problem by exploiting every means of technologies from all scientific disciplines. This influence is pivotal in developing my own research interest throughout my PhD program. BabakÕs continuous support and encouragement helped me pursuing the research subjects vide infra that may not be regarded as traditional organic chemistry topics by many others. Nourished by his open -mind edness , I was able to explore and experiment into different chemistry disciplines for the fulfillment of my PhD central theme, and comfortab ly claim myself as a problem -solving scientist without putting any boundary to my research. There are many other group members I need to express gratitude to. Dr. Chrysoula Vasileiou is not just a senior in the succession of PhD researchers of the bio -orie nted projects in the group, more importantly to me she is a listener, a helper, and a life -long vi friend that supports my life as a foreign student in the United States. Dr. Ipek Yapici is the senior member who guided me into the subject of research and also a good friend who paid heartful attention to my well -being inside the group. Her rhapsodic joyfulness and emotion always make the days in Rm.600 extremely vivid. I always enjoyed talking and discussing with Dr. Hadi Gholami for broad synthetic topics and graduate life. Shared more common feelings than other lab folks, the small Chinese community of Dr. Xinliang Ding, Dr. Jun Zhang, and Dr. Yi Yi has made my graduate life in US much smoother and connected, as our shared joyfulness comes from the leisure tim e and is beyond the research laboratory. I am also grateful to my many dear friends for their mutual supports: Hadi Nayebi, Nastaran Salehi Marzijarani, Xiaopeng Yin, Yubai Zhou, Chenchen Yang, Peng Wang, Ruipeng Mu, Tayeb Kakeshpour . The list should also include other current and past lab members and can go miles long. Support and collaborations are also vital to my research. I am very thankful to my committee members Dr. Xuefei Huang, Dr. James Geiger, and Dr. William Wulff, for their help in the graduate course s and useful guidance and comments on my research projects. Especially I am grateful to Dr. Geiger for the intense collaboration with his research group on my research projects. Being more specifically, I am thankful to Dr. Zahra Assar, Dr. Alireza Ganbarpour, and Nona Ehyaei for their constructive suggestions and the protein crystallography work. I obtained many instructions from different people on operations of essential instrumentation that facilitated my research work. Thanks to Dr. Melinda Fram e for the help with fluorescence microscopes, Dr. Tony Schilmiller and Dr. Lijun Chen for the training of high -resolution mass spectrometer, and Dr. Dan Holmes for NMR . For vii collaborations inside group, I am grateful to Dr. Elizabeth Santos for her indefati gable effort in protein engineering which attributes to the better understanding of hCRBPII and Dr. Setare Tahmasebi Nick for her crucial input in the photoswitch project . I greatly enjoyed the collaboration with laser spectroscopists Dr. Muath Nairat, Dr. Marcus Dantos and Dr. Ehud Pines for their invaluable and incredible work on the photobase project. The collaboration with Chenchen Yang, Dr. Margaret Young, Dr. Sofia Lunt and Dr. Richard Lunt on the material engineering project widely expande d my research scope and capability. Research is not the sole part of the graduate life. Chasing a dream of research and becoming a graduate student on the other side of earth can be extremely challenging. The endless love, understanding, and support from m y parents will never be paid back. Even possessing the enthusiasm to chemistry, my faith in pursuing the degree would have worn out if without the dedicated backup and care from my beloved wife Anna. My final sincere acknowledgement is devoted to Mr. Thoma s Almer, Mrs. Andree Almer, and Mrs. Rita Hsie h. Fate has brought them to me and bonded us as family while I am thousands of miles away from home. viii TABLE OF CONTENTS LIST OF TABLES ............................................................................................................ x!LIST OF FIGURES .......................................................................................................... xi!LIST OF SCHEMES .................................................................................................... xxiv!KEY TO SYMBOLS AND A BBREVIATIONS ............................................................. xxvi!CHAPTER I. A BRIEF O VERVIEW: THE EMERGIN G FIELD OF MODERN FLUORESCENCE IMAGING TAGS AND NEAR -INFRARED EMITTERS. .................... 1!I.1!Constitutively fluorescent proteins ................................................................... 1!I.1.1!The origin of FPs, and the innate chromophores ........................................ 3!I.1.2!Classification of FPs ................................................................................... 5!I.2!General applications of fluorescence imaging ............................................... 10!I.2.1!Modern imaging applications of FPs ......................................................... 11!I.2.2!Other trending applications of fluorescence bioimaging ........................... 17!I.2.2.1!Biosensing ........................................................................................ 17!I.2.2.2!Emerging new imaging agents and their applications ...................... 22!I.2.2.3!Multimodality fluorescence imaging and intraoperative imaging guided cancer surgery .................................................................................... 28!I.3!Site -specific labeling methods for in vivo fluorescence imaging .................... 30!I.3.1!Covalent enzymatic labeling ..................................................................... 30!I.3.2!Noncovalent affinity labeling ..................................................................... 33!I.3.3!Bioorthogonal chemical labeling ............................................................... 37!I.3.4!Labeling approaches with short oligomer tags ......................................... 42!I.4!Ongoing pursuit of far -red/near -infrared fluorochromes ................................. 45!I.4.1!Far-red and NIR FPs ................................................................................. 46!I.4.2!NIR organic dyes ...................................................................................... 48!I.4.2.1!Polymethine dyes ............................................................................. 48!I.4.2.2!Donor-Acceptor NIR dyes ................................................................. 54!I.4.2.3!Other NIR dyes ................................................................................. 60!REFERENCES .......................................................................................................... 64!CHAPTER II. DEVELOPI NG DYE -HCRBPII COMPL EXES INTO NO -WASH BACKGROUND-FREE NEAR -INFRARED TAGS VIA P HOTOSWITCHING. .............. 97!II.1!Preliminary work towards the hCRBPII -based fluorescent tag ...................... 98!II.2!The requirements of a successful fluorescence imaging tag ....................... 101!II.3!Early attempts in finding fluore scent ligands for hCRBPII ............................ 103!II.4!Approach to NIR emission with large Stokes shift ....................................... 110!II.5!Discovery of photochromic dye -hCRBPII complexes ................................... 117! ix II.6!Developing CF -1 into a NIR tag for no -wash in vivo imaging ....................... 125!II.7!Engineering ps4 into a truly fluorogenic tag for spatiotemporal imaging ...... 139!II.8!Conclusion and future research directions ................................................... 146!II.8.1!Future structural modification of FR -1V .............................................. 147!II.8.2!FR-1V/hCRBPII as a dual -emission ratiometric pH probe .................. 148!II.9!Experimental section .................................................................................... 151!REFERENCES ........................................................................................................ 183!CHAPTER III. EXCITED STATE PROTON TRANSF ER OF FR PHOTOBASES AND DEVELOP ING DYE -HCRBPII COMP LEXES INTO BRIGHT, L ONG STOKES SHIFT, FAST -FORMING TAGS VI A EXCITED -STATE HYDR OGEN BONDING. .................. 188!III.1!Discovery of intermolecular ESPT in the solution study of FR0 ................... 189!III.2!Time -resolved study of FR0 -SB ESiPT ........................................................ 197!III.3!Recapitulate ESHB in hCRBPII complexes to develop bright, long Stokes Shift, and fast -forming imaging tag ......................................................................... 212!III.3.1!ESHB of photoacids and the applications ........................................... 212!III.3.2!ESHB in dye/hCRBPII complexes ....................................................... 214!III.3.3!A bright ESHB tag with large Stokes shift and fast -forming kinetics ... 218!III.4!Dual-color live cell imaging enabled by a single dye .................................... 222!III.5!Conclusion and significance ......................................................................... 227!III.6!Experimental section .................................................................................... 228!REFERENCES ........................................................................................................ 233!CHAPTER IV. MINIMIZI NG IMAGING BACKGROUN D OF DYE -HCRBPII TAG S VIA PHOTOINDUCED ELECTRO N TRANSFER QUENCHING . ....................................... 238!IV.1 !Current strategies to enable no -wash live cell imaging ................................ 238!IV.1.1 !Fluorogenicity based on physical interaction with environments ........ 239!IV.1.2 !Fluorogenicity based o n chemical modification of fluorophores .......... 241!IV.2 !Designing a PeT -quencher tethered FR0 as a dark stain ............................ 249!IV.3 !Modification of the maleimide quencher and in vitro, in vivo studies ........... 258!IV.4 !Conclusion and futur e directions .................................................................. 272!IV.5 !Experimental section .................................................................................... 274!REFERENCES ........................................................................................................ 285! x LIST OF TABLES Table II-1 Spectral features of selected DiAI-2V/hCRBPII complexes. ....................... 106!Table II-2 Spectral features of FR dyes in different organic solvents. ......................... 113!Table II-3 Spectral features of FR-1V and its SB in different solvents. ........................ 115!Table II-4 Wavelength comparison of FR-1V and its SB, PSB. ................................... 116!Table II-5 Selected FR-1V/hCRBPII mutants showing fluorescent photochromism. ... 122!Table II-6 Feature comparison between ps4 and ps5. ................................................ 141!Table II-7 Comparison of CF-1 and CF-2 with other far -red/NIR FPs. a ....................... 146!Table III-1 Spectral comparison of FR0-SB and FR-1V-SB in organic solvents. ......... 190!Table IV-1 Oligomeric nature of psES235 cysteine mutations. ................................... 265!Table IV-2 QY and fluorescence recovery kinetics of psES325 and psWS51 -54. ...... 268!Table IV-3 HRMS of Q2mFR0/hCRBPII adducts. ....................................................... 269! xi LIST OF FIGURES Figure I -1 Structure of Aequorea victoria GFP. Showing the side (left) and top (right) views of the protein, the intrinsically generated fluorophore, and several key residues for chromophore maturation (PDB ID: 1w7s). ....................................................................... 1!Figure I -2 A representative set of chromophores responsible for fluorescence in constitutively fluorescent metazoan proteins. ................................................................... 2!Figure I -3 Reported fluorescence brightness (extinction coefficient times fluorescence quantum yield) of a number of popular or unique fluorescent proteins as a function of emission wavelength. The dashed line serves as a guide to highlight the genera l trend. The coloring of the tags approximates the appearance of the corresponding emission wavelength to the naked eye. The brightness of some fluorophores has been altered by a small amount for clearer visual presentation. Copied from ref [7]. ................................ 3!Figure I -4 Phylogenetic tree of representative FPs, copied from ref [17]. The branches with a posterior probability of 0.95 are collapsed. Individual sequences are represented by boxes corresponding to their color type (see legend). Representative FPs are shown with their commercial names and/or mutant derivatives in brackets. ............................... 6!Figure I -5 Application of fluorescent proteins. Dark gray and light gray petals show structural and functional applications, respectively. This artwork is copied from ref [17]. ........................................................................................................................................ 10!Figure I -6 Three image sequences show the FLAP image in pseudocolour (left panel) and phase -contrast image (right panel) at the following times after bleaching: 0 s (a, d and g), 50 s (b and e), 100 s (c, f and h) and 200 s (i). In these sequences the i ndividual channels were acquired in multi -tracking mode. Bleach region shown as a circle in each image. Threshold level set to 8% of maximum signal. Bleach times 8.6 s (a) and 13.5 s (d and g). Scale bars 10 ! m. Image copied from ref [84]. .............................................. 12!Figure I -7 Three modes of FRET sensing for biomolecular interactions. Artwork copied from ref [85]. ................................................................................................................... 13!Figure I -8 Schematic representation of the FLINC -AKAR design principle, the domain structure of FLINC ÐAKAR1 and the acquisition "of super -resolution activity images with pcSOFI. Artwork copied from ref [95]. ............................................................................ 15!Figure I -9 In vivo dynamic GSH imaging. (a) Chemical structure of QG3.0 probe. (b) Real -time GSH imaging with QG3.0 (1 ! M) in A549 cells under glucose -deprivation with a perfusion sys tem. Overlay and time -lapse ratio imaging of GSH under high -glucose (25 mM: 25G) or glucose -free (0 mM: 0G) conditions. 1: 0 Ð3 min, 25 mM glucose; 3 Ð12 min, 0 mM glucose; 12 Ð30 min, 25 mM glucose. 2: 0 Ð3 min, 25 mM glucose; 3 Ð30 min, 0 mM xii glucose. scale ba rs, 20 ! m. (c) Time -course of GSH concentration under glucose deprivation (red, condition 1; blue, condition 2). Images and plot copied from ref [121]. ........................................................................................................................................ 19!Figure I -10 Methods of voltage sensing with small molecules. (a) Voltage -dependent traverse of solvatochromic dyes in lipid membranes. (b) Electrochromic voltage -sensing. (c) Voltage -dependent FRET acceptor redistribution. (d) Voltage -sensing based on PeT perturbatio n. Artwork copied from ref [128]. ................................................................... 20!Figure I -11 Spontaneous voltage and epifluorescence imaging with BeRST 1 in GFP -labeled cells of rat hippocampal neurons. (a) GFP and (b) stained with BeRST 1. Scale bar is 20 ! m. (c) Optical traces of spontaneous activity in neurons from panels (a) and (b). Numbers next to traces correspond to indicated cells in panel (b). Optical sampling rate is 500 Hz. Images and plot copied fro m ref [131]. ................................................... 21!Figure I -12 Cell surface tension imaging. (a) Schematic of the FRET -quenching mechanism of tension sensor function, copied from ref [132]. (b) Reflection interference contrast microscopy (RICM) and fluorescence overlay, and quantified heat map of tension for cells cu ltured on the cRGDfK (TAMRA -QSY9) and SHAVSS (Alexa 488 -TAMRA) peptide tension probe surfaces. Scale bar: 20 ! m. Images copied from ref [134]. ........ 22!Figure I -13 Schematic representation of dye -loading NP design for multicolor imaging via sequential FRET cascade. ............................................................................................. 23!Figure I -14 Tracking RGB barcoded cancer cells in zebrafish embryo. A) Six batches of D2A1 cells were labeled with fluorescent NPs generating RGB barcodes. Labeled cells were then mixed and injected intravascularly in 2 d postfertilization zebrafish embryos. BÐD) Tumor cells that arrested in the vasculature are imaged, at 3 h post -injection, in the caudal plexus region of the zebrafish embryo. Clusters and individual cells can be found and distinguished based on their color. Images copied from ref [142]. .......................... 24!Figure I -15 1O2-sensitizer -amplified NIR afterglow. (a) Schematic illustration of the proposed mechanism. (b) Schematic illustration of SPN -NCBS pre -irradiated by an 808 -nm laser for afterglow enhancement versus a 514 -nm laser. (c) After glow luminescence images of SPN -NCBS pre -irradiated at 514 (left) or 808 nm (right). The NP solutions were pre -irradiated by 808 - or 514 -nm laser for 1 min, and then the afterglow images were acquired under bioluminescence mode with an acquisition time of 30 s after removal of the laser source. Images copied from ref [144]. ............................................................. 25!Figure I -16 Schematic representations of nucleic acids -based imaging probes. (a) Spinach sensor, comprising Spinach (black), a transducer (orange), and a target -binding aptamer (blue). Target (purple) binding to the aptamer promotes stabilization of the transducer st em, enabling Spinach to fold and activate DFHBI (green) fluorescence. (b) General structure of molecular beacons. ....................................................................... 27!Figure I-17 Schematic labeling reactions of SNAP/CLIP -tags. ...................................... 31! xiii Figure I -18 Fluorogenic TMP -tag. (a) Cartoon of the trimeric TMP -quencher -fluorophore to be cleaved in a proximity -induced S N2 reaction. (b) Structure of TMP -BHQ1 -Atto520. ........................................................................................................................................ 33!Figure I -19 Design of the PYP -tag mutant PYP3R and its fluorogenic probe, with a focus on electrostatic int eractions and the pK a value of the leaving group. Copied from ref [236]. ........................................................................................................................................ 35!Figure I -20 Y-FAST -tag. (a) Schematic showing the spectral changes during Y -FAST binding: s pectral shift and QY increase. (b) Confocal time lapse showing two cycles of labeling/unlabeling (Ex/Em 488/493 -575 nm) of HeLa cells expressing Y -FAST. Cells were repeatedly incubated with HMBR -containing culture medium for 20 s and HMBR -free culture medi um for 40 s. HMBR concentration was 5 !M. (c) Temporal evolution of the cell fluorescence upon addition (+) and removal ( -) of HMBR. Images and plots copied from ref [238]. ................................................................................................................. 36!Figure I -21 Fluorogenic ligands of FAP: thiazole orange (TO) and malachite green (MG). ........................................................................................................................................ 37!Figure I -22 Two-step labeling scheme of site -specific protein labeling via PRIME and chelation -assisted CuAAC. In the first step, W37VLplA covalently attaches a copper -chelating picolyl azide ( pAz) onto genetically fused LAP. In step two, CuAAC selectively derivatizes pAz with a probe -conjugated terminal alkyne. ............................................. 42!Figure I -23 The ideal NIR window for in viv o imaging, showing minimal light absorption by haemoglobin (Hb, green), oxyhaemoglobin (HbO 2, red) (<650 nm) and water (>900 nm). Copied from ref [306]. ............................................................................................ 45!Figure I -24 Phytochrome -derived BV -binding FP. (a) Scheme showing Z/E isomerization between Pr and Pfr states of BV adduct. (b) Absorption spectra of Pr and Pfr states. Q band is the major absorption. Soret band is the pyrrole absorption. (c) A truncated PAS -GAF NIR FP (PDB ID: 4XTQ). ........................................................................................ 46!Figure I-25 Indocyanine green (ICG). ............................................................................ 47!Figure I -26 Representative non -convergent polymethine cyanines. (a) Cyanines with various rigidified heptamethine backbones and terminal groups. (b) Conformation -restrained pentamethine cyanines. (c) Large Stokes shift heptamethine cyanines. ...... 49!Figure I -27 Chemical structures of PPCy dyes. (a) Standard PPCys. (b) Further -extended PPCys. (c) PPCy aza -BODIPY analogues. .................................................................... 50!Figure I -28 NIR BODIPYs. (a) BODIPY framework. (b ) aza -BODIPYs. (c) "coupled" BODIPYs. (d) "restricted" BODIPYs. .............................................................................. 52! xiv Figure I -29 NIR ring -fused BODIPYs. (a) [a/h] -fused BODIPYs. (b) [b/g] -fused BODIPYs. ........................................................................................................................................ 53!Figure I -30 (a) NIR fluorescent squaraines. (b) NIR -II absorptive croconates. Scheme shows mesoionic oxidation. ............................................................................................ 55!Figure I-31 Restricted and C10 -replaced rhodamines. .................................................. 56!Figure I-32 Representative NIR D -A-D type of dyes. .................................................... 57!Figure I-33 Selected ICT dyes and D -A hybrids. ........................................................... 58!Figure I-34 Selected NIR rylenes. ................................................................................. 59!Figure I-35 Selected NIR porphyrinoids. ....................................................................... 60!Figure I -36 Examples of NIR AIEgen. (a) Red -shifted TPE emission by extended conjugation. (b) Aggregation -induced bathochromic shift in emission. .......................... 61!Figure II -1 Crystal structure of hCRBPII KL mutant complexed with all -trans -retinal. Retinal is shown in red. The active lysine residue Q108K is shown in light green. Other key residues engineered to regulate the absorption wavelength of the bound retinylidene are shown in teal, with oxygen atoms colored red and nitrogen atoms colored blue (PDB ID: 4EXZ). The scheme shows the iminium (PSB) formation between lysine 108 and retinal aldehyde. ............................................................................................................. 99!Figure II -2 Structure of MCRA and its schematic PSB formation with active lysine 132 of CRABPII. The absorption (red) and emission (blue) spectra are of the R132K:R111L (KL) mutant complex with MCRA . ........................................................................................ 100!Figure II -3 Crystal structure of hCRBPII -KL (Q108K:K40L) mutant, highlighting the cavity for ligand binding (meshed cages). PDB ID: 4EXZ. The bound all -trans -retinal is deleted for clarity. ...................................................................................................................... 104!Figure II -4 Live cell imaging with DiAI -2V and MCRA for performance testing. a) Clone map of hCRBPII mutant KLVSWW:L117E in ECFG -GalT vector. b) Comparison of images of DiAI -2V and MCRA complexed with KLVSWW:L117E in HeLa cells. Cells were stained with 250 nM of each dye and incubated for 10 min at 37 ¡C. Cells were washed with DPBS twice before imaging. ECFP channel: "ex = 458 nm, BP 475 -525. MCRA channel: "ex = 594 nm, LP 615. DiAI -2V channel: "ex = 514 nm, LP 560. BR : bright -field. ...................................................................................................................................... 108!Figure II -5 Different resonance forms proposed for push -pull chromophores and related shape of the absorption spectra. The charge distribution alon g the push -pull systems of merocyanines and cyanines are depicted by three states: M 1 (neutral for merocyanine, zwitterionic for cyanine), M 2 (least bond length alteration), and M 3 (bipolar for xv merocyanine, inverted zwitterionic for cyanine). Merocyanines are featured by one electron rich amine group (D, donor) and one more electronegative terminal group (A, acceptor) on either ends of the push -pull chain. Typical cyanines bear one neutral amine (D) and one charged iminium (A) on either ends of the push -pull chain. ..................... 109!Figure II -6 Schematic cross section of the ground state and the lowest excited singlet state potential hypersurfaces along the reaction coordinate corresponding to the ICT state model. ........................................................................................................................... 111!Figure II -7 Normalized spectra of FR-1V in various solvents: (a) absorbance, (b) emission. Normalized spectra of FR-1V-SB ( n-butylamine) in various solvents: (c) absorbance, (d) emission. Tol: toluene; THF: tetrahydrofuran; EA: ethyl acetate; DC M: dichloromethane; DMF: N,N-dimethyl formamide; DMSO: dimethyl sulfone; MeCN: acetonitrile; EtOH: ethanol; MeOH: methanol. Scheme shows the SB formation between FR-1V and n-butylamine. ................................................................................................................... 114!Figure II -8 Comparison of (a) absorbance and (b) emission spectra of FR-1V-SB and FR-1V-PSB in DMSO and ethanol. .................................................................................... 116!Figure II -9 Flexible docking of FR-1V with all -trans -retinal bound hCRBPII KL mutant (PDB ID: 4EXZ). Green stick is the crystal structure of retinal. Orange stick is the pose of FR-1V. Docking was performed by AutoDock Vina. Residues !within 5.5 † distance to FR-1V are highlighted in line form. ..................................................................................... 120!Figure II -10 Cis-trans imine photoisomerization of all -trans -retinal bound CRABPII complexes. a) The iterative color switching of all-trans retinal/CRABPII complex under UV and green light irradiations. (PDB ID: 4YFP) b) Scheme illustrating the photocycle of retinylidene imine isomerization and the acid -base equilibrium of the thermodynamic SB and kinetic PSB. Figure and scheme are r eproduced from ref [43]. ............................ 121!Figure II -11 Normalized absorption and emission spectra of the two photoswitching states of CF-1 (FR-1V/ps4 complex). ÒOFFÓ state: the co lorless thermal equilibrated state after binding. ÒONÓ state: the blue -colored state after UV irradiation. The equilibrium illustrates the reversible switching under different conditions. ..................................... 123!Figure II -12 Binding of ps4 (20 ! M) and FR-1V in PBS buffer (pH 7.3) over time, incubated at 25 ¡C. (a) Stacked traces of absorbance spectra, FR-1V (7 ! M); (b) Time -course absorbance spectrum monitored at 378 nm, FR-1V (6 ! M). Data points were fitted to an exponential rise function. ..................................................................................... 125!Figure II -13 Switching fatigue resistance of CF-1. Stacked traces of (a) absorbance spectra, (b) emission spe ctra, over cycles of alternate UV and yellow light irradiations. For each cycle, excitation at 378 nm lasts 20 s with U -360 UV Bandpass filter (Blue); excitation at 600 nm lasts 15 min with Y -50 (500nm) Longpass filter (Yellow). (c) CF-1 absorbance monitor ed at 600 nm, and (d) CF-1 emission intensities monitored at the xvi maxima of ÒOFFÓ and ÒONÓ states in non -degassed solutions. 0.6% absorption and 1% emission intensities decreased per cycle. All measurements were in PBS (pH 7.3). .. 126!Figure II -14 CF-1Õs emission switching time of the "OFF" and "ON" states. Monitored at the emission maxima under continuous excitation. ...................................................... 127!Figure II -15 Temperature -dependent CF-1 ON-to-OFF thermal switching, absorbance monitored at 600 nm. ................................................................................................... 128!Figure II -16 Thermal ON -to-OFF switching of CF-1. Rate constants plotted against temperature. (a) Arrhenius plot. (b) Eyring plot. ........................................................... 130!Figure II -17 Thermal versus yellow light -assisted (LP 5 00 filter) ON -to-OFF switching of CF-1, monitored at absorption maximum. Data points collected at 3 -min interval. ...... 130!Figure II -18 Working pH range of CF-1. Stacked tr aces of emission spectra of ÒOFFÓ (Ex. = 378 nm) and ÒONÓ (Ex. = 600 nm) states: (a) pH 7 to 9; (b) pH 7 to 6. (c) pH -dependent fluorescence intensities of ÒOFFÓ and ÒONÓ states, showing 80% intensities within pH 6.6 ~ 7.6 range. .................................................................................................................. 131!Figure II -19 MTT assay of FR-1V in (a) HeLa and (b) U2OS cell lines. The absorbance was measured 550 nm, and corrected at 655 nm. Cells were treated with FR-1V for 24 h before MTT staining. Error bars represent the standard deviation of the readouts from ( n = 6) cells. ...................................................................................................................... 132!Figure II -20 Size exclusion chromatography of hCRBPII mutant ps4 expressed and purified from BL21(DE3)pLysS (buffer: 10 mM Tris "HCl, 2 M NaCl, pH 8.1) using a BioLogic DuoFlow Qua dTec 10 system (Bio -Rad) equipped with a Superdex 120 16/600 GL size -exclusion column (GE Healthcare). ................................................................ 133!Figure II -21 Schematic maps of the (a) hCRBPII -ECFP and (b) EGFP -hCRBPII -SP fusion constructs. SP: signaling peptides. SP = 3 #NLS (nuclear localization sequence), NES (nuclear export sequence), and CAAX (prenylation tag). ............................................. 134!Figure II -22 Time serial confocal fluorescence images of U2OS cells expressing construct pFlagCMV2 -EGFP -ps4-3 #NLS, stained with 1 !M FR-1V and incubated at (a) 20 ¡C, and (b) 37 ¡C. Channels: EGFP (Green), "ex = 488 nm, BP 505 -530; CF-1_ON (Red), "switch = 405 nm, "ex = 594 nm, LP 650. Scale bar, 20 !m. ................................ 135!Figure II -23 CF-1 labeling in U2OS and HEK293 cell lines. Left an d middle column: cells expressing C -terminal ECFP, showing whole cell fluorescence. Right column: cells expressing N -terminal EGFP, showing nuclei -localized fluorescence. NLS = nuclear localization sequence. BR: bright -field. Channels: ECFP, "ex = 458 nm, BP 465 -510IR; EGFP, "ex = 488 nm, BP 505 -530; CF-1_ON, "switch = 405 nm, "ex = 594 nm, LP 650; Bright field, "ex = 488 nm. Cells were stained with 1 !M FR-1V and incubated at 37 ¡C for 5 min before imaging. No washing steps required. Scale bar, 20 ! m. ......................... 136! xvii Figure II -24 Compartmentalized CF-1_ON labeling in HeLa cells. (a) Labeling of different CF-1 fusion proteins. NLS = nuclear localization sequence. NES = nuclear export sequence. CAAX = prenylation tag. DIC = differential interference contrast. Cells were stained with 2 !M FR-1V and incubated at 37 ¡C for 5 min before imaging. No washing steps required. Scale bar, 10 !m. ................................................................................. 137!Figure II -25 Labeling specificity of nuclei -localized CF-1 in HeLa cells. Top left: EGFP channel; top right: CF-1_ON channel; bottom left: merged channels + brightfield; bottom right: line profiles of fluorescence intensities in two channels along the blue arrow in the merged image . Scale bars, 20 µm. .............................................................................. 138!Figure II -26 Normalized absorption and emission spectra of the two photoswitching states of CF-2 (FR-1V/ps5 complex). ........................................................................... 139!Figure II -27 Swit ching fatigue resistance of CF-2 monitored at 605 nm absorbance in PBS (pH 7.3). For each cycle, excitation at 378 nm lasts 20 s with U -360 UV Bandpass filter; excitation at 600 nm lasts 15 min with Y -50 (500 nm) Longpass filter. ............... 140!Figure II -28 (a) Time -course absorbance spectrum monitored at 382 nm of ps5 and FR-1V binding. Data points were fitted to an exponential rise function. (b) Time -course spectrum of CF-2 ON-to-OFF thermal switching at 23 ¡C. Absorbance monitored at 605 nm. Data points were fitted to an exponential decay function. ..................................... 141!Figure II -29 Compartmentalized CF-2_ON labeling in HeLa cells. Labeling of different CF-2 fusion proteins. NLS = nuclear localization sequence. NES = nuclear export sequence. CAAX = prenylation tag. DIC = differential interference contrast. Cells were stained with 2 !M FR-1V and incubated at 37 ¡C for 5 min before imaging. No washing steps required. Scale bar, 10 !m. ................................................................................. 142!Figure II -30 Labeling specificity of nuclei -localized CF-2 in HeLa cells. Top left: EGFP channel; top right: CF-2_ON channel; bottom left: merged channels + DIC; bottom right: line profiles of fluorescence intensities in two channels along the blue arrow in the merged image . Scale bars, 20 µm. Circles and number i ndicating mean SBR (n = 440). ........ 143!Figure II -31 ON-to-OFF thermal switching kinetics of (a) CF-1 and (b) CF-2 in HeLa cells. Region of interest was excited continuously with 594 nm laser (1.63 !W). Mean ROI intensity was recorded every 6 sec. ............................................................................. 144!Figure II -32 In cellulo NIR fluorescence (LP650) ÒON/OFFÓ cycling of HeLa cells expressing EGFP -ps5-CAAX, showing 10 switching cycles. Arrows: OFF -to-ON, 405 nm laser 4 sec; ON -to-OFF, r.t. 3 m in. Scale bar: 10 !m. .................................................. 144!Figure II -33 Spatiotemporal confocal imaging of HeLa cells, expressing nuclei -localized EGFP -fused ps5. Cells were stained with 2 !M FR-1V for 5 min at 37 ¡C and imaged without washing. Red channel: CF-2_ON. Following the arrows: before switch -on; 1 st cell xviii switch -on; 2 nd cell switch -on; 3 rd cell switch -on; whole field switch -on. Compare with green channel: colocalized reference EGFP. Scale b ars, 10 !m. .......................................... 145!Figure II -34 The prototype of a dual -emission ratiometric pH probe. a) Stacked traces of the absorption spectra of FR-1V/ph1 (Q108K:K40L:T51V:T53S:V62E) complex upon iterative light irradiation. UV = U -360 UV Bandpass filter. VIS = Y -50 (500 nm) Longpass filter. Stacked traces of b) SB and c) PSB emission under different pH. d) The correlation of the logarithmic fluorescence ra tio and pH. A linear function is presented. ............... 150!Figure III -1 Normalized spectra of FR0 in various solvents: (a) absorbance, (b) emission. Normalized spectra of FR0-SB (n-butylamine) in various solvents: (c) absorbance, (d) emission. Tol: toluene; THF: tetrahydrofuran; EA: ethyl acetate; DCM: dichloromethane; DMF: N,N-dimethyl formamide; DMSO: dimethyl sulfone; ACN: acetonitrile; EtOH: ethanol; MeOH: methanol. Scheme sho ws the SB formation between FR0 and n-butylamine. ................................................................................................................... 189!Figure III -2 Solvatochromism of FR0, FR-1V, and their SBs as evident by their emission wavelengths in different s olvents. Plots illustrate the correlation between wavenumbers and solvent polarities. The short wavelength emission maxima of FR0-SB in ethanol and methanol were selected for the plot. ReichardtÕs E T(30) scale is applied. .................... 191!Figure III-3 QY as a function of solvent polarity of FR0-SB and FR-1V-SB. ............... 192!Figure III -4 (a) Absorption and emission of FR0-SB in ethanol. (b) Excitation spectra of FR0-SB in ethanol at 460 nm and 630 nm, corrected against Rhodamine -B. ............. 193!Figure III -5 (a) Stacked absorption traces of FR0-SB acidification with hydrochloric acid. (b) Absorption and emission spectra of acidified FR0-SB. ........................................... 194!Figure III -6 Absorption and e mission spectra of FR0-SB in different alcoholic solvents. (a) methanol, (b) ethanol, (c) 1 -propanol, (d) 2 -propanol, (e) n-amyl alcohol, (f) ethylene glycol. Traces labeled with Peak -i (1 ~ 4) are fluorescence spectra simulated by second derivative fittin g with a Voigtian amplitude distribution using PeakFit TM. The Envelope traces are the convolution of multiple Peak -i, showing a perfect overlap with the experimental spectra. ................................................................................................... 195!Figure III -7 Absorption and emission spectra of FR0-SB dissolved in acetic acid. (a) Absorption spectrum, showing ammonium peaks at 342 nm and iminium peaks at 472 nm. (b) Emission spectra excited at 472 nm. (c) Emission spectra excited at 342 nm. In (b) and (c), emission traces labeled with Peak -i are simulated by second derivative fitting with a Voigtian amplitude distribution using PeakFit TM. The Envelope traces are the convolution of multiple Peak -i. ...................................................................................... 196!Figure III -8 Steady -state absorption (solid) and emission (shaded) spectra of FR0-SB in (a) acetonitrile (ACN) and (b) ethanol (EtOH). The spectra of FR0-PSB (acid ified by perchloric acid in ethanol) are also overlaid. ................................................................ 199! xix Figure III -9 TCSPC traces with single exponential fits for FR0-SB near the emission maxima when dissolved i n ACN and EtOH. Inset shows that the red emission trace at 650 nm of FR0-SB is identical to the FR0-PSB emission. ........................................... 200!Figure III -10 Isotopic effect of FR0-SB in EtOH a nd ethanol -d6 (EtOD). (a) Steady -state absorption and emission spectra of FR0-SB in EtOH and EtOD. (b) TCSPC decay curves of FR0-SB LW emission along with single exponential decay fits in EtOH and EtOD. An isotope effect of 1.6 is observed in EtOD. .................................................................... 201!Figure III -11 TCSPC decay curves of FR0-SB LW emission with single exponential decay fits in MeOH and methanol -d4 (MeOD). An isotope effect of 2.0 is observed in M eOD. ...................................................................................................................................... 202!Figure III -12 (a) Transient absorption spectra of FR0-SB at various time intervals after excitation in EtOH. Labeled arrows show the steps during the ESPT pr ocess. (b) Energy progression during the proton transfer process. ........................................................... 203!Figure III -13 (a) Transient absorption of FR0-SB in ACN, showing the long -lived excited state abs orption and stimulated emission signals from the non -protonated form with biexponential fits at the frequencies of (b) excited state absorption and (c) stimulated emission signals. .......................................................................................................... 205!Figure III -14 Stimulated emission signal from the transient absorption of FR0-PSB in acidic EtOH. Two decay components are observed, with a fast one corresponding to an intramolecular response from FR0-SB, and a slow one corresponding to the excited state lifetime. ......................................................................................................................... 206!Figure III -15 (a) Evolution associated spectra (EAS) of the four levels used in the sequential global analysis model to describe the proton transfer dynamics of FR0-SB in EtOH. (b) Population kinetics of the four levels used in the global analysis model. ..... 207!Figure III -16 Transient absorption trace showing the energy progression during the excited state FR0-SB proton transfer process in MeOH. ............................................. 208!Figure III -17 Transient absorption traces at (a) 570 nm where stimulated emission from the intermediate formation is observed, and (b) 650 nm where stimulated emission from FR0-PSB can be seen while diss olved in EtOH (black) and EtOD (red). Biexponential decay constants are given in the inset along with !the pre -exponential factors in parentheses. An isotope effect of 1.5 is observed during the formation of the partially proton transferred intermediate, wh ile an isotope effect of 2.0 is observed during the formation of final protonated FR0-PSB. ....................................................................... 209!Figure III -18 Transient absorption traces at (a) 580 nm where stimulate d emission from the intermediate formation is observed, and (b) 660 nm where stimulated emission from FR0-PSB can be seen while dissolved in MeOH (black) and MeOD (red). Biexponential decay constants are given in the inset along with the pre -exponential fa ctors in xx parentheses. An isotope effect of 2 is observed during the formation of the intermediate as well as in the final protonated form formation. ......................................................... 210!Figure III -19 The observed intermolecular ESPT dynamics in EtOH along with the associated time constants for the steps as obtained from global analysis (black) and the TCSPC data (colored). ................................................................................................. 211!Figure III -20 Normalized fluorescence spectra of FR0-SB in the acetonitrile solution of imidazole. The arrow shows the LW emission band (630 -660 nm) emerging with the addition of imidazole (from blue to red), indicating the proton transfer from imidazole to FR0-SB. ........................................................................................................................ 212!Figure III -21 Potential residues facilitating excited state hydrogen bonding on FR-1V imine nitrogen atom. Residues within 5 † to the nitrogen atom are shown in green. Residues within 6 † distance are show n in red and magenta. The pose of FR-1V is simulated with a flexible docking method by using AutoDock Vina. The protein matrics is extracted from hCRBPII mutant KL (PDB ID: 4EXZ). FR-1V:carbon (cyan), nitrogen (blue), hydrogen (grey). Magenta residues are simulated with PyMOL. ................................ 216!Figure III -22 Representative FR/hCRBPII complexes showing excited state hydrogen bonding long wavelength emissions. Mutant Q108K:K40D:T53A: R58L:Q38F:Q4F:V62E is in the dimeric form. ................................................................................................... 217!Figure III -23 Excitation spectrum of FR-1V/psES235 600 nm emission. The red trace is the excitation spectrum corrected against Rhodamine -B. The green trace is the absorption spectrum. The blue trace is the normalized brightness curve, plotting the product of # and $ with each square as a data point. .................................................. 218!Figure III -24 Binding kinetics of FR-1V with psES235 . a) Pseudo -first order binding of FR-1V (10 !M) with psES235 (20 !M). The binding was monitored at the 595 nm ESHB emission maximum. An ex ponential fit gives the binding half -time as 0.18 s. b) Second -order rate constant measurement. Variable concentrations of FR-1V were used to bind 100 nM of psES235 . Observable rate constants were derived from single exponential fits. The second -order rate constant was then derived from the linear regression of observable pseudo first order rate constants as a function of concentration. All measurements were taken in PBS buffer (pH = 7.3). .................................................................................... 219!Figure III -25 Compartmentalized FR-1V/psES235 imaging in live HeLa cells. NLS = nuclear localization sequence. NES = nuclear export sequence. CAAX = prenylation tag. Cells were stained with 500 nM FR-1V and incubated at 37 ¡C for 1 min. Cells were washed 3 times with DPBS before imaging. Scale bar, 10 !m. ................................... 220!Figure III -26 ESHB fluorescence specificity of FR-1V/psES235 . Top row shows the merge d EGFP and LP615 channels in cell nuclei, cytosol, and plasma membrane -localized HeLa cells. Bottom row shows the line profile of colocalized EGFP and ESHB signal along the white arrows in the top images. Scale bar, 10 !m. ............................ 221! xxi Figure III-27 Absorption and emission spectra of FR-1V/psST61 . .............................. 223!Figure III -28 Binding kinetics of FR-1V/psST61 . a) Binding of FR-1V (10 !M) and psST61 (10 !M) monitored at the 630 nm fluorescence maximum. The intensity curve is fit with a single exponenti al function. b) Second -order rate constant measurement. Variable concentrations of FR-1V were used to bind 100 nM of psST61 . Observable rate constants were derived from single exponential fits. The second -order rate constant was then derived from the linea r regression of observable pseudo first order rate constants as a function of concentration. All measurements were taken in PBS buffer (pH = 7.3). ..... 224!Figure III -29 Dual-color live cell imaging with FR-1V in HeLa cells co -transfected with pFlag-CMV2 vectors encoding EGFP -psES235 -NES and EGFP -psST61 -3xNLS. a) ESHB channel: "ex = 405 nm, LP615. b) PSB channel: "ex = 514 nm, LP650. c) Reference EGFP channel: "ex = 488 nm, BP505 -530. d) Merged image of a) and b), showing well -defined edges of cell nuclei and surrounding cytosol. Cells were stained with 500 nM of FR-1V and incubated at 37 ¡C for 1 min, then washed with DPBS for 3 times before imaging. Scale bars, 20 !%& ......................................................................................... 225!Figure III -30 Crystal structure of FR-1V/psST61 complex, showing two trajectories of the 40H residue in a 50:50 population ratio. The closer conformer has a 2.9 † distance between the histidine and the imine nitrogen atom. Protein crystallized by Dr. Alireza Ghanbarpour. ............................................................................................................... 226!Figure III -31 Proposed key residues for mutagen esis to suppress K40H ESHB. Distances are measured from the FR-1V/psST61 crystal structure. ............................................ 226!Figure IV -1 Jablo $ ski diagrams of various energy/electron donor -accepto r (D -A) systems. a) Fırster Resonance Energy Transfer (FRET). ': angle between donor emission and acceptor absorption. b) Dexter Energy Transfer (DET). Artwork copied from ref [10]. 244!Figure IV -2 PeT mechanisms. a) Participation of fluorophore HOMO -LUMO and an empty quencher orbital. b) Participation of fluorophore HOMO -LUMO and a filled quencher orbital. ........................................................................................................... 247!Figure IV-3 Structure of model PeT -quenched FR fluorophore. .................................. 249!Figure IV -4 Orbital diagram of Q1FR-1V and the corresponding n-butylimine ( Q1FR-1V_SB ), n-butyliminiu m ( Q1FR-1V_PSB ), calculated with DFT/B3LYP/6 -31+G in vacuum. The orbital views on the left side are of Q1FR-1V. The color code of the right -side energy diagram is: blue (HOMO of FR-1V/SB/PSB ), green (LUMO of maleimide), and red (LUMO of FR-1V/SB/PSB ). ............................................................................. 250!Figure IV -5 Orbital diagram of the n-butylimine of FR-1V tethered with maleimides bearing different substituents. Left -side orbital views are shown with non -substitute d maleimide. The color code on the right -side diagram is: blue (HOMO of FR-1V_SB ), green (LUMO of maleimides), and red (LUMO of FR-1V_SB ). .................................... 251! xxii Figure IV -6 Orbital diagram of the n-butyliminium of FR-1V tethered with maleimides bearing different substituents. Left -side orbital views are shown with non -substituted maleimide. The color c ode on the right -side diagram is: blue (HOMO of FR-1V_PSB ), green (LUMO of maleimides), and red (LUMO of FR-1V_PSB ). ................................. 252!Figure IV -7 Extinction coefficient, absorptio n and emission spectra of Q1FR0 in ethanol. ...................................................................................................................................... 254!Figure IV -8 Fluorescence recovery of Q1FR0 after addition of n-propane thiol (PSH). ...................................................................................................................................... 255!Figure IV -9 Fluorescence recovery of Q1FR0 tested by sequential addition of n-butylamine (nBA) and n-propane thiol (PSH) in different order. a) Reaction scheme of two sequential addition orders. b) Stacked traces of time -dependent absorption of Q1FR0 after addition of excess nBA. c) Stacked traces of time -dependent absorption of Q1FR0_PSH adduct after addition of excess nBA. b) and c) showed the same timescale of imine condensation of nBA with Q1FR0 and Q1FR0_PSH . d) Emission spectrum of Q1FR0 after the sequential addition of PSH then nBA. e) Emission spectrum of Q1FR0 after the sequential addition of nBA then PSH. d) and e) were recorded with the same concentrations of Q1FR0, PSH, and nBA. ................................................................... 256!Figure IV -10 Low stability of Q1FR0. a) Picture shows the fluorescence of Q1FR0 samples stored in ethanol at 4 ¡C overnight (O/N), freshly prepared in ethanol and DMSO. Samples are excited with UV handset. b) Possible degradation pathway. .................. 257!Figure IV -11 a) Absorption and emission spectra of Q2mFR0 in ethanol. b) Stacked absorption traces of Q2mFR0 added with n-propane thiol (PSH) over time. c) Emission spectra of Q2mFR0 before and after addition of n-pro pane thiol. d) Fluorescence recovery measured peak intensity. e) Fluorescence recovery measured by whole spectra integration. .................................................................................................................... 261!Figure IV -12 Chemoselectivity and flu orescence recovery of Q2mFR0 with sequential addition of n-butylamine (nBA) and n-propane thiol (PSH). a) Stacked absorption traces of Q2mFR0 with excess nBA over time. b) Stacked absorption traces of Q2mFR0_PSH adduct with excess nBA over time. a) and b) were recorded in ethanol with same amount of Q2mFR0, nBA, and PSH. c) Absorption and emission spectra of Q2mFR0_nBA. d) Absorption and emission spectra of the final solution of Q2mFR0_PSH added with nBA. c) and d) were recorded in ethanol with same starting concentration of Q2mFR0. e) Emission spectra of Q2mFR0 in ethanol with sequential addition of nBA then PSH. .. 263!Figure IV -13 Flexible docking results of Q2mFR0 in hCRBPII KL mutant (PDB ID: 4EXZ). Residues within 5 † distance to the maleimide double bond carbon atoms are highlighted in cyan and stick form. Q2mFR0 is covalently attached to the Q108K re sidue. The top 9 poses are stacked in and displayed line form. ............................................................. 264! xxiii Figure IV -14 Absorption and emission spectra of Q2mFR0 and psES235 cysteine mutants. a) Stacked abs orption traces of Q2mFR0 and psWS54 over time. b) Emission spectrum of Q2mFR0/psWS54 with 374 nm excitation. Emission spectra with excitation at corresponding absorption maxima at different time points after Q2mFR0 addition to c) psWS51 monomer, d) psWS52 monomer, e) psWS53 monomer. ............................. 266!Figure IV -15 Fluorescence recovery of Q2mFR0/psWS51 -54. a) Fluorescence recovery kinetics monitored at 540 nm. b) Fluorescence recovery ratio of different cysteine mutants over psES235 monitored at 540 nm at different time points. ....................................... 267!Figure IV -16 Confocal imaging of HeLa cells expressing pFlag -CMV2 construct: EGFP -psWS54 -3xNLS. Kalman averaging 4 was applied. The reference EGFP channel was excited with a 488 nm laser and collected with a BP505 -530 bandpass filter. The psWS54 channel was excited wi th a 405 nm laser and collected with a BP505 -575IR bandpass filter. Same excitation power, gain and offset levels were applied. Cells were stained with 2 !M Q2mFR0 and incubated at 37 ¡C for indicated time. No washing steps were applied before imaging. Sc ale bar: 20 !m. ................................................................................ 270!Figure IV -17 Confocal imaging of HeLa cells expressing pFlag -CMV2 construct: EGFP -psWS54 -3xNLS. The reference EGFP channel was excited with a 488 nm laser and collected with a BP505 -530 bandpass filter. The psWS54 channel was excited with a 405 nm laser and collected with a BP530 -600 instead of a BP505 -575IR bandpass filter. Cells were stained with 2 !M Q2mFR0 at room temperature for 35 min. No wa shing steps were applied before imaging. Scale bar: 20 !m. ................................................................... 271!Figure IV -18 Non-specific fluorescence background of Q2mFR0 at different time points. Q2mFR0 (2 !M) wa s used to stain HeLa cells at room temperature. The field of view was excited with a 405 nm laser and images were acquired with a BP505 -575IR bandpass filter. No washing steps were applied. Scale bar: 20 !m. ............................................. 272! xxiv LIST OF SCHEMES Scheme I -1 Proposed mechanism for the biogenesis of the GFP chromophore (Inset: atom numbering used in the t ext). .................................................................................... 4!Scheme I -2 Proposed maturation mechanisms of representative chromophores from GFP-, DsRed - and Kaede -like naturally occurring fluorescent proteins. .......................... 7!Scheme I -3 Mechanistic representation of HaloTag, showing H272F mutation to trap fluorophore -labeled intermediate as a stable covalent adduct. R represents the labeling fluorophore. .................................................................................................................... 32!Scheme I -4 Pictet %Spengler ligation of an aldehyde with a tryptamine nucleophile. ..... 38!Scheme I -5 Traceless Staudinger ligation. .................................................................... 39!Scheme I -6 [3+2] cycloadditions. (a) SPAAC and representati ve cyclooctyenes. (b) SPANC. R is a labeling fluorophore. .............................................................................. 40!Scheme I -7 (a) Tetrazole -alkene photoclick cycloaddition. (b) IEDDA tetrazine -alkene cycloaddition. .................................................................................................................. 41!Scheme I -8 Labeling approaches based on short peptides. (a) FlAsH and ReAsH with tetracysteine motif. (b) Bis -boronic acid RhoBo with tetraserine motif. (c) Dimaleimide coumarin YC20 with vicinal dicysteine motif. ................................................................. 43!Scheme II -1 Structures of thiacarbocyanines ( ThCCs) and diazaindenes ( DiAIs), with different number of vinylene insertions between the head group and the aldehyde. ... 105!Scheme II -2 FR series of ICT dyes, derivatized from FR0 with different lengths and types of (-spacers. ................................................................................................................. 112!Scheme II -3 Proposed mechanism for the cis !trans isomerization of the chromophore in Dronpa and Padron. Bright and dark states correspond to the cis and trans forms of t he chromophore, respectively. The two forms are determined by the pK a difference in the acid -base equilibrium . In Dronpa, the cis form is anionic and bright. ........................... 119!Scheme II -4 Postulated photoswitching cycle of CF-1. ............................................... 124!Scheme II -5 Potential derivatives of FR-1V for improving optical properties. .............. 147!Scheme II -6 Synthesis of ThCC-1V. ............................................................................ 160!Schem e II-7 Unsuccessful attempts for the nucleophilic elongation of ThCC-1V. ...... 162! xxv Scheme II -8 Unsuccessful attempts for the electrophilic elongation of 57. ................. 162!Scheme II -9 Attempted synthesis of ThCC-2V and its phenylimine. ........................... 163!Scheme II -10 Synthesis of DiAIs. ................................................................................ 165!Scheme II -11 Synthesis of FRs. .................................................................................. 173!Scheme III -1 Schematic transformation of (a) single and (b) double ESiPT of FR0-SB in protic solvents. ............................................................................................................. 197!Sch eme IV-1 Synthesis of Q1FR0. .............................................................................. 253!Scheme IV -2 Synthesis of 3 -methoxy maleimide. ....................................................... 258!Scheme IV -3 First attempt to synthesize Q2mFR0. .................................................... 259!Scheme IV -4 Attempts to synthesize Q2mFR0 with diff erent functionalized N-ethyl maleimides. .................................................................................................................. 259!Scheme IV -5 Testing S N2 reactions of maleimide 83. ................................................. 260!Scheme IV -6 Final synthesis of Q2mFR0. .................................................................. 260!Scheme IV -7 Chemoselectivity and fluorescence recovery of Q2mFR0 with sequential addition of n-butylamine (nBA) and n-propane thiol (PSH). a) PSH then nBA. b) nBA then PSH. ............................................................................................................................. 262!Scheme IV -8 Detectable adducts of Q2mFR0/hCRBPII in HRMS. A sequential step of reductive amination after both imine formation and Michael addition was used to derivatize the complex i nto a more stable form for mass detection. ............................ 268! xxvi KEY TO SYMBOLS AND A BBREVIATIONS A alanine, Ala C cysteine, Cys D aspartic acid, Asp E glutamic acid, Glu F phenylalanine, Phe G glycine, Gly H histidine, His I isoleucine, Ile K lysine, Lys L leucine, Leu M methionine, Met N asparagine, Asn P proline, Pro Q glutamine, Gln R arginine, Arg S serine, Ser T threonine, Thr V valine , Val W tryptophan, Trp Y tyrosine, Tyr xxvii † angstrom, 10 -10 meter cm centimeter µm micrometer nm nanometer Da dalton g gram mg milligram µg microgram M molar µM micromolar nM nanomolar mol mole mmol millimole xs excess h hour min minute sec second t time t1/2 half-life, maturation half -time O/N overnight k kinetic rate constant ) molar extinction coefficient xxviii * quantum yield T temperature ! degree Celsius K kelvin mW milliwatt h PlanckÕs constant R gas constant +,,frequency Hz hertz cm-1 wavenumber, aq. Aqueous Ka ionization constant FL fluorescence UV ultraviolet Vis visible IR infrared NIR near-infrared Abs, "abs absorbance, absorption wavelength maximum Em, "em emission wavelength maximum Ex, "ex excitation wavelength SS Stokes shift LSS long/large Stokes shift xxix SW short -wavelength LW long-wavelength QY quantum yield EQE external quantum efficiency BP bandpass (filter) LP longpass (filter) ESA excited state absorption LE local excitation SE stimulated emission TAS transient absorption spectra EAS evolution associated spectra FP fluorescent protein CP chromoprotein GFP green fluorescent protein EGFP enhanced green fluorescent protein BFP blue fluorescent protein YFP yellow fluorescent protein RFP red fluorescent protein PA-FP photoactivatable fluorescent protein PS-FP photoswitchable fluorescent protein RSFP reversibly photoswitchable fluorescent protein PS-RSFP positively photoswitchable R SFP xxx NS-RSFP negatively photoswitchable RSFP FAP fluorogen -activating protein PYP photoactive yellow protein Y-FAST Yellow Fluorescence -Activating and absorption -Shifting Tag AIE aggregation -induced emission AIEgen aggregation -induced emission luminogen /fluorogen ACQ aggregation -caused quenching ESPT excited state proton transfer ESIPT excited state intramolecular proton transfer ESiPT excited state intermolecular proton transfer ESHB excited state hydrogen bonding GSPT ground state proton transfer FRET Fırster resonance energy transfer ICT intramolecular charge transfer PLICT planarized intramolecular charge transfer TICT twisted intramolecular charge transfer MLCT metal -ligand charge transfer ESCT excited state charge transfer PeT photoinduced electron transfer TBET through -bond energy transfer TADF thermally activated delayed fluorescence IC internal conversion xxxi ISC intersystem crossing RIR restriction of internal conversion qABP quenched activity -based probe BiFC bimolecular fluorescence complementation FCCS fluorescence cross -correlation spectroscopy FLIP fluorescence loss in photobleaching FLINC fluorescence fluctuation increase by contact FLAP fluorescence localization after photobleaching FRAP fluorescence recovery after photobleaching FLIM fluorescence lifetime imaging microscopy FISH fluorescence in situ hybridization RESOLFT reversible saturable optical fluorescence transition PALM photo-activated localization microscopy STORM stochastic optical recon struction microscopy pcSOFI photochromic stochastic optical fluctuation imaging FMI fluorescence molecular imaging hFMT hybrid fluorescence molecular tomography FGS fluorescence -guided surgery CALI chromophore -assisted light inactivation FALI fluorophore -assisted light inactivation PDT photodynamic therapy PTT photothermal therapy xxxii PAM photoacoustic microscopy PAT photoacoustic tomography FACS fluorescence -activated cell sorting TCSPC time -correlated single -photon counting N.D. not determined N.O. not observed SB Schiff base PSB protonated Schiff base HBA H-bonding acceptor HBD H-bonding donor EDG electron donating group EWG electron withdrawing group HOMO highest occupied molecular orbital LUMO lowest unoccupied molecular or bital MS molecular sieve nBA n-butylamine PSH n-propane thiol IPTG isopropyl &-D-1-thiogalactopyranoside LB lysogeny broth PBS phosphate -buffered saline FBS fetal bovine serum DMEM DulbeccoÕs Modified EagleÕs Medium xxxiii p-HBI 4-(p-Hydroxy -benzylidene) -5-imidazolinone hAGT human O 6-alkylguanine -DNA alkyl transferase TMP trimethoprim eDHFR E. coli dihydrofolate reductase ICG indocyanine green MB methylene blue , molecular beacon TO thiazole orange MG malachite green PPCy pyrrol opyrrole BODIPY boron dipyrromethene TPE tetraphenylene NCL native chemical ligation CuAAC copper -catalyzed alkyne -azide cycloaddition SPAAC strain -promoted alkyne -azide cycloaddition SPANC strain -promoted alkyne -nitrone cycloaddition IEDDA Inverse -Elect ron -Demand Diels -Alder cycloaddition DIBAL diisobutylaluminium hydride TBAB tetrabutylammonium bromide TsOH p-toluenesulfonic acid scFv single -chain variable fragment NP nanoparticle SNP semiconductor nanoparticle xxxiv QD quantum dot NES nuclear export sequence NLS nuclear localization sequence CAAX prenylation tag VDP vicinal dithiol peptide DIC differential interference contrast SBR signal -to-background ratio POI protein of interest ROI region of interest PPI protein -protein interaction 1 A BRIEF OVERVIEW: THE EMERG ING FIELD OF MODERN FLUORESCENCE IMAGING TAGS AND NEA R-INFRARED EMITTERS . The great adva ncement of modern biology started with the publication of the Micrographia by Hooke in 1665. 1 The microscope has provide d an ideal and fundamental tool for biologists to study single - and multi -cellular organisms with the use of light. Even in the current era of modern biology, it still remains as the first requirement of many researchers to visually investigate the morphologies and structures of under study object s, or to follow dynamic biological processes . I.1!Constitutively fluorescent proteins In the field of microscopy, event was the discovery of the green fluorescent protein (GFP) cloned from the jellyfish Aequorea victoria in 1992,2 and its subsequent demonstration as an in vivo tag for fluorescent imaging. 3 In the two decades that followed , the biological research community witnessed the huge success of GFP and th e l arge 4.2 nmN-term C-term p-HBI R96T203E222 2.4 nmFigure 0-1 Structure of Aequorea victoria GFP. Showing the side (left) and top (right) views of the protein, the intrinsically generat ed fluor ophore, and several key residues for chromophore maturation (PDB I D: 1w7s). 2 family of its enhanced homologues , with the recognition of 2008 Nobel Prize in Chemistry jointly awarded to Osamu Shimo mura, Martin Cha lfie, and Roger Y. Tsien , Òfor the discovery and development of the green fluorescent protein, GFPÓ . Sharing the sim ilar --barrel structure surrounding the embedded chromophores , which forms 11 &-sheet s that host an internal distorted helix (Figure I-1), fluorescent proteins (FP) and non -fluorescent chromoproteins (CP) other than A. victoria from Hydrozoa species were also found in the bioluminescent Anthozoa species with quite conservative scaffolds 4 (Figure I-2). Largely due to the commercial availability and certain advantages, GFP-like FPs have had an enormous impact on biological research activities . NNONNOHNNEBFP Azurite NNOOHNOTagBFP2 NNONHmTurquoise2 NNOOEGFP NNOONmPapaya NNOONOHOmOrange2 mKate NNOONOSNNOONOOHmCherry NNOONOOPSmOrange Figure 0-2 A representative set of chromophores responsible for fluorescence in constitutively fluorescent metazoan proteins. 3 I.1.1!The origin of FPs, and the innate chromophores Current FPs span almost the entire palette of the visible spectrum, with an emission wavelength rang ing from 442 nm to 645 nm (Figure I-3).5-7 Unlike the bioluminescence found in bacteria, unicellular algae, coelenterat es, beetles, fish, and others, 8 FPs do not require access ory cofactors or external enzyme and substrate to form intrinsic chromophores. The maturation of their embedd ed chromophore, 4 -(p-hydroxy -benzylidene) -5-imidazolinone ( p-HBI), is accomplished through an autocatalytic process. Based on a number of crystallographic observations, 9 the most widely accepted thoughinthegreenandyellowregionofthespectrummanyof thesegainshavebeenmoreevolutionarythanrevolutionary. Forexample,EGFPcontinuestobeanFPofchoiceinmany applicationsduetoitsattractivecombinationoffastand quantitativematuration,brightness,photostability,andtoler- ancetowardfusions,welloveradecadeafteritsdiscovery. 81Thissuggeststhatgreen !uorescentproteins,andpossiblycyan andyellowproteinsaswell,areclosetorealizingtheirfull potential.Figure4plotsthe !uorescencebrightnessofanumberof popularorotherwiseinteresting !uorescentproteinsasa functionoftheirwavelengthofmaximalemission.This "gurehighlightstwomajortrends:greenish !uorescentproteinstend toachievethemostfavorable !uorescencebrightness,with performancelaggingtowardeitherendofthevisiblespectrum, butalsothatthereisafairlylargespreadinbrightnessamong thelabels.Thelatterre !ectsthatmanydi #erentproperties governtheattractivenessofaparticularlabelforagiven experiment,andoptimizationofoneparameter(e.g.,bright- ness)cannegativelya #ecttheperformanceofanother(e.g., photostability).Thereforethebrightest !uorescentproteinis notnecessarilythebestchoiceforaparticularexperiment. Giventhewidespreadavailabilityofwell-performingcyan, green,andyellowFPs,themostpromisingopportunitiesin termsofoptimizationpotentialprobablycomefromthe !uorescentproteinsattheedgeofthevisualspectrum,which arelesswellcharacterized.However,asFigure4shows,itis probablethatthespectroscopicperformanceofthesevariantsis fundamentallylimitedcomparedtothatofthegreenvariants, andthechoiceofemissionwavelengthisanimportantfactor forexperimentsthatrequire,e.g.,highsensitivity. Anongoingchallenge,andprobablyoneofmostprominent drivingforcesinFPdevelopment,isthecreationofvariants thatemitredorfar-red !uorescence,thoughitisanopen questiontowhatextentfundamentallimitationswillultimately limittheperformanceachievablewithinthe !uorescentprotein sca#old.TheavailableredFPstendtodisplaymuchlarger degreesofcellulartoxicityandlowertolerancesforprotein fusion82whilealsosu #eringfromslowerorevenincomplete chromophorematuration. 38Importantly,asigni "cantfraction ofevenwell-knownred !uorescentproteinsfailstoachieveany red!uorescencewhatsoever:Inastudycombining !uorescencecorrelationspectroscopyandpulsed-interleavedexcitationon samplesconsistingoftwoFPsfusedtogether, 83theauthors foundthatessentiallyallEGFPfragmentsbecome !uorescent,butonly40%ofmCherryand22%ofmRFPfragmentsreach thefullymaturedredstate.Obviouslythisisanareainwhich furtherimprovementisurgentlyneeded.Giventhestrong dependenceofredFP !uorescenceontheentireprotein structure,evidentfromthedi $cultyinmonomerizingthese FPswithoutcompromisingspectroscopicperformance, 77balancingassociationtendenciesorcellulartoxicitywith favorablespectroscopicpropertiespresentsaformidable challenge.Thee #ectof !uorescentproteincomplexityisnicelyrevealed inadiscussionoftheirphotostabilityorresistancetothe completelossof !uorescencethroughphotodestruction.The exactmechanismsassociatedwithFPphotodestructionremain poorlyunderstood.Increasingevidencepointstothedirect involvementofintermediatenon !uorescent(blinking)species, formedfromthesingletexcitedstate,asbeingthemajor pathwaysforphotodestruction. 79,84Insomecases,however, permanentlossof !uorescencethroughphotodestructioncan beconfusedwithprocessesthatcauseashiftintheemission spectrumorareversiblelossof !uorescence,suchas decarboxylation,85,86 oxidativereddening, 87orreversiblephoto- chromism. 73,88 Amaincomplicationindevelopingmore photostable !uorescentproteinsisthesensitivityofthis parametertotheenvironmentandobservationconditionsof theprotein.Observingagiven !uorescentproteinindi #erentfusionconstructs,di #erentcellularcompartmentsorenviron- mentsorcellstatescanallaltertheabsolutekineticsandeven relativekineticsofphotodestruction. 5Thephotostabilityalso dependsstronglyonthemannerofobservation:Therelative stabilityofdi #erentFPscanchangedramaticallydependingon whethertheobservationisperformedusingconfocalimaging, wide"eldimaging,oranothertypeofobservation.Anexample isTagRFP-T,whichwasshowntobemorephotostablethanits ancestorTagRFPinwhole-colonyscreening, 89butisactually surpassedbyTagRFPathighe rexcitationintensities. 90Similarly,mRaspberryismorephotostablethanmPlumin confocalmicroscopybutlessstableinwide "eldmicroscopy. 91Thisdependenceofthephotodestructionontheimaging conditionsisnotsurprisinggiventhatthisprocessisthoughtto occurfromtransientlyformedintermediatestates.Asaresult, thebestapproachinselectingaphotostable !uorescentprotein istomakeuseoftheliteratureonlyforgeneralguidanceandto compareanumberofwell-performingvariantsundertheactual experimentalconditions. These"ndingsaresomewhatemblematicforthechallenges facedinFPdevelopment.Whileitisstraightforwardto combineDNAmutagenesiswithhigh-throughputscreeningto achievedirectedevolution,achoicehastobemaderegarding whichparameterwillbeoptimized,beit !uorescencebrightness, 81photostability,89folding/maturation, 27photo-chromism,92,93 oreventheexcited-statelifetime. 94Whilethis willresultinimprovementsofthatspeci "cparameter,other parameterswilllikelydeteriorateifnofurtheractionistaken. Also,theresultsstronglydependonthemannerinwhichthe screeningwasperformed,anddi #erentresultswillbeobtained underdi #erentimagingortargetingconditions.Sincethis complexityandsensitivityisanintrinsicpartofthe !uorescentFigure4. Reported!uorescencebrightness(extinctioncoe $cienttimes!uorescencequantumyield)ofanumberofpopularorunique !uorescentproteinsasafunctionofemissionwavelength(dataas reportedinrefs10and44).Thedashedlineservesasaguideto highlightthegeneraltrend.Thecoloringofthetagsapproximatesthe appearanceofthecorrespondingemissionwavelengthtothenaked eye.Thebrightnessofsome !uorophoreshasbeenalteredbyvery smallamountsforclearervisualpresentation. JournaloftheAmericanChemicalSociety Perspectivedx.doi.org/10.1021/ja309768d|J.Am.Chem.Soc. 2013,135,2387 !24022391Figure 0-3 Reported fluorescence brightness (extinction coefficient times fluorescence quantum yield) of a number of popular or unique fluorescent proteins as a function of emission wavelength. The dashed line serves as a guide to highlight the general trend. The coloring of the tags approximates the appearance of the corresponding emission wavelength to the naked eye. The brightness of some fluorophores has been altered by a small amount for clearer visual pr esentation. Copied from ref [7]. 4 mechanism for the formation of p-HBI is depicted as three sequential steps from the tripeptide Ser65 -Tyr66 -Gly67 ( Scheme I-1, Path A ): (1) internal nucleophilic attack of Gly67 amide nitrogen on the carbonyl carbon of Ser65 , (2) dehydration of the hemi -aminal to form an imidazolin -5-one intermediate, (3) oxidation by molecular oxygen along the C'-C & bond of Tyr66 to conjugate the imidazolone . The exact order of the three st eps, however, is still unclear . The original mechanis m proposed by Tsien orders the three steps as cyclization -dehydration -oxidation .10 This mechanistic proposal is based on the observation of accumulated GFP intermediate in E. coli in anaerobic atmosphere. This pre -oxidation intermediate is supported by the 1 (±4) Da mass loss of Tyr66 &-carbon in their kinetic isotope study. In contrast, the Wachter group suggests a reversed order of oxidation and dehydration based on X -ray crystallographic and spectroscop ic data of a Y66L GFP variant, which suggested the condensation product being trapped by ring oxidation ( Scheme I-1, Path B). 11 In this pathway, the last step involves the elimination of the hydroxyl group in the form of a water molecule. This final transformation is coupled OHNHOHOHNNOOHNHOHOHNNOOHH2OO2H2O2HONNOOHHONNOMature Chromophore Precyclization structure Intermediate 1Intermediate 2AIntermediate 2BPath A Path B H2OO2H2O2RC!66C"##N66C65N67O66O65Scheme 0-1 Proposed mechanism for the biogenesis of the GFP chromophore (Inset: atom numbering used in the text). 5 to a proton -transfer reaction which may proceed through hydrogen -bonded solvent molecular network with Arg9 6 and Glu222, whereby the proton abstraction of the Gly67 amide nitrogen or the Tyr66 '-carbon is facilitated by the carboxylate of Glu222, and the formed '-enolate is stabilized by Arg96 through lowering of the respective p Ka value s as an electrophile .12-14 Despite the ambiguity in the exact o rder of the chromophore -forming process, Tyr66 and Gly67 remain strictly evolutionarily conserved among all naturally occurring FPs. In contrast, the residue at position 65 can be mutated to various amino acids. This mutation gives rise to numerous color v ariants throughout the FP families as shown in Figure I-2. I.1.2!Classification of FPs As mentioned earlier, many natural FPs have been found in various marine species other than Aequorea victoria over the decades. In the view of phylogeny, FP is an ancie nt Metazoan gene. 15 It has currently been identified in four different phyla of multicellular animals, namely, Cnidaria, Ctenopho ra, Arthropoda, and Chordata. 16 Interestingly, these four phyla cover most of the basic partitions of the metazoan tree of life (Figure I-4). The original GFP and its numerous engineered derivatives only constitute a very specific sm all branch in this tree of life, albeit sharing the similar &-barrel enclosing the chromophore. The most divergent FP sequences are found in the class Hydrozoa, nevertheless the colors are mostly green. 17 On the contrary, the class Anthozoa provides the greatest diversity of FP colors. Recent r esearch e fforts were focused on the Anthozoa class to identify internal chromophore with substanti al red -shifted emission beyond 540 nm. 18-20 The most common two are the DsRed - and Kaede -like chromophores. 6 As shown in Scheme I-2, all of the intrinsic fluorescent proteins have chromophores consisting the same p-HBI scaffold with different side chains to extend the indeedspecializeinpreyingonmedusae(142).Other- wise,toreconcilethetopologyoftheFPfamilywith thecurrentlyacceptedgrossphylogenyofmulticellular animals,oneisforcedtoassumethatBilateriaand Anthozoainheritedadifferentparalogouslineageof FPsthanCtenophoraandHydrozoaandthatthesetwo paralogouslineagesseparatedfromeachotherperhaps evenbeforetheoriginofmulticellularanimals.Unlikely asthissounds,giventhepeculiarbirth/deathdynamics ofFPgeneswithinagenome(seebelow),thispossi- FIG.12.PhylogenetictreeofrepresentativeFPs,reconstructedfromtheproteinsequencesusingMrBayes3.1(181)undertheÒmixedÓmodel. Thebrancheswithaposteriorprobability !0.95arecollapsed.Individualsequencesarerepresentedbyboxescorrespondingtotheircolortype(see legend).Onlythenamesoftheproteinsmentionedthroughoutthetextofthisreviewareshown,withtheircommercialnamesand/ormutant derivativesgiveninbrackets.ThesystematicafÞnityofthehostorganismsisindicated.Thecladesdesignated AÐDcorrespondtothegroups originallyreportedinRef.239. 1116CHUDAKOVETAL. PhysiolRev ¥VOL90 ¥JULY2010 ¥www.prv.orgDownloaded from www.physiology.org/journal/physrev by ${individualUser.givenNames} ${individualUser.surname} (035.009.054.165) on December 22, 2017.Copyright © 2010 American Physiological Society. All rights reserved.Figure 0-4 Phylogenetic tree of representative FPs, copied from ref [17 ]. The branches with a posterior probability of 0.95 are collapsed. Individual sequences are represented by boxes corresponding to their color type (see legend). Re presentative FPs are shown with their commercial names and/or mutant derivatives in brackets. 7 conjugation. Starting with the tripeptide, the GFP -like chromophore is formed through the sequential cyclization, dehydration a nd oxidation of the con servative Tyr66 and Gly67, and exist s in an equilibrium between deprotonated and protonated form s, of which the latter one emits the green fluorescence. DsRed -like chromophore formation takes an extra oxidation step. A neutral blue -emitting intermediate from cyclization and dehydration of the tripeptide can be oxidized by molecular oxygen at the C '-N bond of residue 65 HOHNHNOONHROGly67Tyr66 Xaa65 H2OO2H2O2HONNONHROH+H+ONNONHROO2H2O2ONNONROO2H2O2H2OONNOHNROO2H2O2ONNONH2ONNHBackbone cleavage H2OUV light ONNORNH2OBackbone cleavage OGFPKaedeasFP595 DsRedONNONSHOKusabira-Orange ONNONH2ONBackbone cleavage Lys65 Cys65zFP538 R = Lys R = Cys R = Met R = His Scheme 0-2 Proposed maturation mechanisms of representative chromophores from GFP -, DsRed - and Kaede -like naturally occurring fluorescent proteins. 8 (Xaa65 in Scheme I-2), thus extending the conjugation of the five -membered imidazolino ne ring to an acylimine group instead of the phenolic Tyr66. 21, 22 While in some other FPs, DsRed -like chromophore can be oxidized from the neutral form of p-HBI equilibrium. 23 Further structural conversion can hap pen to DsRed -like chromophores on acylimine group through hydrolysis, 24 transimination, 25-27 or nucleophilic cyclization. 28, 29 In contrast to GFP - and DsRed -like FPs, Kaede -like FPs show a remarkably different chromophore formation process. Once formed the equilibrated p-HBI precursor, the maturation process skips the additional oxidation observed in DsRed and stops at the GFP-type chromophore. Ultraviolet (UV) -violet light irradiation then initiates a peptide backbone cleavage between Tyr 66 and residue 65. In all of the Kaede -type FPs, His65 appears to be indispensable . Its imidazole ring results in extended conjugation with p-HBI scaffold from the photolysis and is responsible for the mature red form. 30 This photoactivation of Kaede -like FPs became popular and triggered the wide i nvestigation and application of photoactiva table, photoconvertible, and photoswitchable FPs. Rooted in the naturally occurring FPs, innumerable artificial FPs have also been engineered via mutagenesis. It is notable that the chromophore structure only determines part of the spectral properties, the surrounding amino acids and the highly ordered water molecules together form a wide range of interactions and build a network that encircles the whole &-barrel, significantly in fluencing the spectroscopic behavior, absorption and emission maxima, photostabilit y, blinking, pH stability, etc. For example, a Thr203Tyr mutation induced a direct (-( stacking with the chromophore Tyr66, and resulted in significantly red -shifted excit ation and emission maxima. 31 9 Generally, b ased on excitation and emission wavelengths, FPs can be categorized into the following color regimes: UV FPs ( "ex < 380 nm, "em < 450 nm, e.g., Sirius); 32 Blue FPs ("ex ranges 380 ~ 400 nm, "em ~ 450 nm, e.g., Azurite, EBFP2, TagBFP, etc. );33-35 Cyan FPs ( "ex ~ 430 nm, "em ranges 470 ~ 490 nm, e.g., ECFP, mTurquoise, Cerulean, etc.); 10, 36-38 Green FPs ( "ex ranges 480 ~ 510 nm, "em ranges 500 ~ 520 nm, e.g., EGFP, Emerald, TagGFP 2, TurboGFP, etc.); 35, 39-42 Yellow FPs ( "ex ranges 510 ~ 530 nm, "em ~ 540 nm, e.g., EYFP, Venus, mPapaya1, etc.); 31, 43-45 Orange FPs ( "ex ~ 550 nm, "em ~ 560 nm, e.g., mOrange, mKO, etc.); 28, 46 Red FPs ( "ex ranges 550 ~ 580 nm, "em ranges 580 ~ 610 nm, e.g., TagRFP, mRuby, mCherry, mScarlet, etc.); 46-49 Far-red FPs ( "ex ranges 590 ~ 610 nm, "em ranges 630 ~ 670 nm, e.g., mKate2, mRaspberry, mNeptune, NirFP, etc.); 49-53 and Near-infrared (NIR) FPs ( "ex ranges 680 ~ 700 nm, "em > 680 nm, e.g., iFP2.0, iRFP720, GAF -FP, etc.). 54-56 Alternatively, FPs can be categorized based on their functions and utilities, such as large Stokes shift (LSS) FPs ( SS > 100 nm, e.g., Sapphire, mKeima, LSS -mKate2, LSSmOrange, CyOFP1, Sandercyanin, etc.) ,57-64 photoactivatable FPs ( PAFP, of which the fluorescence can be turned on by an activation beam, e.g., PA-GFP, PamKate, etc.), 65, 66 photoconvertible FPs (of which the fluorescence wavelength can be irreversibly convert ed to another by a conversion beam, e.g., mMaple, mEos2, Dendra2, etc.) ,67-69 and photoswitchable FPs ( PSFP, of which the fluorescence wavelengths can be reversibly switched between two distinct states, e.g., Padron, D ronpa, Dreiklang, Kohinoor, etc.). 70-73 Studies are also trending towards further mod ulations of maturation, 10 pH stability, kindling, blinking, fluorescence lifetime , and oligomeric form , to name a few , resulting in a variety of FP variants for numerous fluorescence imaging applications. 74 I.2!General applications of fluorescence imaging Because of the integrity of their expression in aerobic environments, and their ability to be genetically fused to various proteins of interest (POI), FPs have found tremendous applications in molecular and cell biology in the past decades . Traditionally, FPs are used as passive reporters. 75 They can be labelled to proteins, nucleic acids (i.e., DNAs and RNAs), organelles, cells, and even organis ms (for example, whole body imaging), for structural investigations . More detailed sub -categorization s of applications within each branch is depicted in Figure I-5. Impressively , with a wise choice of FPs, blehomotetramersevenatverylow(nanomolar)concen- trations(26). ThetetramericstructureofFPswasÞrstrevealedby crystallographicstudiesoftheredFPDsRedfrom Disco-somasp .(465,500).EachDsRedmonomercontactsthe twoadjacentproteinmoleculesbytwodistinctinterfaces (Fig.4).Thehydrophobicinterfaceincludesaclusterof closelypackedhydrophobicresiduessurroundedbyaset ofpolarsidechains.Thehydrophilicinterfacecontains manyhydrogenbondsandsaltbridgesbetweenpolar residuesandincludesburiedwatermolecules.Inaddi- tion,itisstabilizedbyanunusualÒclaspÓformedby COOH-terminalresiduesofeachmonomer.Thistighttet- ramericstructurepresentedmajordifÞcultiesforthe earlyexperimentaluseofredFPs(seesect. IV).ThesidechainsofaminoacidsburiedinsidetheFP barrelplayessentialrolesinchromophoreformationand Þne-tuningofspectralproperties.Themostimportant residuesarelocatedinthemiddleofthe !-strands,close tothechromophoregroup.Therefore,eachstrandÒcon- trolsÓachromophorefromaparticulardirectionas showninFigure5.Arg96fromstrand4isthemostcritical catalyticresidue(strictlyconservedamongFPs),whichis indirectcontactwiththechromophoreinmatureFPs (Fig.3 C).Arg96promotesproteinbackbonecyclization duringFPmaturation(244,404,405).Catalyticactivityis alsoassociatedwiththeevolutionarilyconservedGlu222 fromstrand11(244,404,405),althoughmutagenesisstud- iesdemonstratedthatinsomecasesthisresiduecanbe replacedwithalmostnodeleteriouseffectsonthematu- rationoftheFPchromophore(101,134).Innerresidues fromstrands7,8,and10inßuencetheFPspectraina differentway.Sidechainsofresidues148,165,167,and 203areincontactwithTyr66ofthechromophore(Fig. 3D)andaretheprimarydeterminantsofitsprotonation state(anionicorneutral),polarization,spatialconforma- tion( cisortrans),androtationalfreedom(19,49,70,324, 340,349,352,414,459,463).Variationsoftheseside chainscandramaticallyaltertheexcitationandemission spectra,maketheproteinhighlyßuorescentornearly completelynonßuorescent,orfacilitatereversiblephoto- switchingofthechromophorebetweentwodifferent states.Correspondingly,anumberofinterestingand sometimesextremelyusefulmutantshavebeenobtained bymutagenesisatthesepoints.Forexample,widelyused yellowvariantsofGFP(seesect. V)carryakeyThr203Tyr substitutionthatresultsinared-shiftofexcitationand emissionmaxima(324).Substitutingthesameposition forIleorHis,onthecontrary,leadstostabilizationofthe protonatedchromophorewithablue-shiftedabsorption at!400nm,whichisseeninsuchGFPvariantsasSap- phire(435,512)andPA-GFP(334).AbrightredFPDsRed wasconvertedintoannonßuorescentchromoprotein statebysubstitutionsSer148Cys,Ile165Asn,Lys167Met, andSer203Ala(51).SomeFPs,suchasasFP595(256)or Dronpa(139),naturallypossesspronouncedreversible photoswitchingcharacteristicsthatarelargelydeter- minedbypositions148,165,167,and203(70). FIG.2.Mainareasofapplicationsofßuorescent proteins.Darkgrayandlightgraypetalsshowstruc- turalandfunctionalstudies,respectively,although boundariesbetweenthemareoftenquitefuzzy. 1106CHUDAKOVETAL. PhysiolRev ¥VOL90 ¥JULY2010 ¥www.prv.orgDownloaded from www.physiology.org/journal/physrev by ${individualUser.givenNames} ${individualUser.surname} (035.009.054.165) on December 22, 2017.Copyright © 2010 American Physiological Society. All rights reserved.Figure 0-5 Application of fluorescent proteins. Dark gray and light gray petals show structural and functional applications, respectively. This artwork is copied from ref [17]. 11 multicolor labeling can be achieved with up to six different colors on a routine basis performed with standard equipment , negating the requirement of spectral unmixing. 58 Potentially, t he number of imaging channels can be further expanded by the combinatorial use of LSS FPs and PSFPs with spectrally resolved excitation and/or emission wavelengths. I.2.1!Modern imaging application s of FPs Although the selection of optimal FPs for a particular study can be complicated by multiple factors, such as maturation time, turnover rate, environment al sensitivity, blinking, and bleaching, 76 it is actually possible to exploit these photophysical and photochemical complexities to transform FPs from passive re porters into smart labels. Compared to classical labeling experiments, 77 new application fields are emerging, i.e., protein -protein interaction (PPI) , sensor, ROS production, DNA sequencing , and drug screening, to name a few (Figure I-5).78-80 Normally, photobleaching is not a desired property for FP labeling. Nonetheless , this phenomenon can be utilized to study the mobility of POI when fused with FP. Two widely-used techniques are fluorescence recovery after photobleaching (FRAP) and fluorescence loss in photobleaching (FLIP). The basic principle of FRAP is to photobleach a small region of interest (ROI) inside the sample cell, then monitor the fluorescence recovery rate, which is determined by the mobility of unbleached FP fusions migrating from an adj acent non -photobleaching region. 81, 82 In contrast , FLIP monitors the rate of fluorescence decay in an ROI by constantly photobleaching its adjacent region , the principle of which follows the same idea of FP migration. 83 Another slight ly complicated 12 yet fundamentally simple technique to study POI trafficking is fluorescence localization after photobleaching (FLAP) . Different from FRAP and FLIP, FLAP requires two FPs fused together. During the study, one FP is photobleached, while the other FP remains intact and serves as an internal fluorescence reference. The intensity ratio profile of the two emission wavelengths is mapped pre - and post - photobleach. POI can then be tracked based on the change of the intensity ratio. 84, 85 A good example is shown in Figure I-6. T he 2D heat maps in the left panels show the movement of labeled POIs in diffe rent region of the sample cells at various time stamps post photobleaching. One interesting application of FPs based on their turnover rate is the Timer FPs. Timer FPs initially fluoresce at one wavelength. Over time, their emission s convert to a different wavelength by various maturation rates which may span from minutes to hours even days. This feature, simi lar to those of radical clocks, is suitable for the use to study progressive processes over a number of time scales. The first known Timer FP is DsRed - FLUORESCENCE LOCALIZATION AFTER PHOTOBLEACHING 111 © 2002 The Royal Microscopical Society, Journal of Microscopy , 205 , 109Ð112 dispersal of the FLAP signal, although it had usually Þlled the local rufße system immediately after bleaching (Fig. 2d). Dur- ing the subsequent 50 s the signal migrated laterally, spread- ing into more distant rufßes along the leading edge (Fig. 2(e) and (f)) and tended to persist locally longer than in the Þrst type of pattern. A third distinct pattern was associated with denser accumulations of expressed actin found often in the perinuclear region of the cell (Fig. 2(g)Ð(i)). When the same cell was targeted in this region, 10 min after the last sequence, the FLAP signal remained precisely localized on the bleach spot and showed no discernible dispersal and very little decay during the 4 min recorded sequence. Fig. 1. Transformed rat Þbroblast showing simul- taneously acquired CFP (a) and YFP (b) images immediately after photobleaching the lamella in a narrow strip (white rectangle). Intensity proÞles integrated between the grey lines before (c) and after (d) photobleaching show CFP (cyan), YFP (yellow) and FLAP (red) signals. The FLAP images correspond- ing to the proÞles (c) and (d) are shown encoded in pseudocolour in (e) and (f). Bleach time 3.8 s. Scale bar 10 µm.Fig. 2. Three image sequences show the FLAP image in pseudocolour (left panel) and phase-contrast image (right panel) at the following times after bleaching: 0 s (a, d and g), 50 s (b and e), 100 s (c, f and h) and 200 s (i). In these sequences the individual channels were acquired in multi-tracking mode. Bleach region shown as a circle in each image. Threshold level set to 8% of maximum signal. Bleach times 8.6 s (a) and 13.5 s ( d and g). Scale bars 10 µm. JMI_1007.fm Page 111 Friday, January 4, 2002 6:14 PMFigure 0-6 Three image sequences show the FLAP image in pseudocolour (left panel) and phase -contrast image (right panel) at the following times after bleaching: 0 s (a, d and g), 50 s (b and e), 100 s (c, f and h) and 200 s (i). In these sequences the individual chan nels were acquired in multi -tracking mode. Bleach region shown as a circle in each image. Threshold level set to 8% of maximum signal. Bleach times 8.6 s (a) and 13.5 s (d and g). Scale bars 10 ! m. Image copied from ref [84]. 13 E5 reported in 2000 , which forms green GFP -like chromophore at the beginning and converts to a tetrameric form over time to become red. 86 Monomeric and faster Tim er FPs have also been developed to meet the requirement of wider in vivo applications. 87 Perhaps the most well -known technique to study protein -protein interaction is the Fırster resonance energy transfer (FRET). In principle, if two fluorochromes are close in distance l ess than 10 †, then the excited donor fluorochrome can transfer part of the energy to the acceptor fluorochrome via a non -radiative long -range dipole -dipole coupling, which requires the spectral overlap of the acceptor absorption spectrum and the donor emi ssion spectrum, and the relative orientation of the acceptor absorption dipole moment and the donor emission dipole moment .88 Although common FRET couples are not limited to FPs (for example, Cy3 and Cy5, FITC and TRITC, Alexa488 and Alexa610, etc.), FPs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igure 0-7 Three modes of FRET sensing for biomolecular interactions. Artwork copied from re f [85]. 14 are the most widely used FRET pairs due to their genetic coding ability, among which the first truly effective pair is CFP and YFP. 89 There are three ways to utilize FRET in an imaging study ( Figure I-7): 1) the interaction of two separately labeled POIs bring the FRET pair into close proximity and in duce the FRET signal; 2) the FRET pair is initially labeled on the two termin i of POI and the loss of FRET signal is traced when segmentation of POI happens; 3) the FRET pair is intramolecularly labeled on POI and the FRET signal is monitored when the POI undergoes conformational change upon protein interaction and/or ligand binding. The application of FRET has been exploded in the past decades. However, the limitation of the experimental conditions, and possible art ifacts such as signal cross -contamination , labeling efficiency, and variation in instrumental setup all increase the complexity in interpreting the intensity -based FRET measurement. 90 In this context, fluorescence lifetime imaging microscopy (FLIM) is developed to overcome these limitations, by taking advantage of the fact that the lifetime of FPs is not affected by these factors. Fluorescence lifetime ( .) i s the unique photophysical character of fluorochromes including FPs. It does not depend on the fluorescence intensity, the concentration of the FPs, the intensity of excitation light source, the dichroic, or the d etection responsiveness of photomultipliers. Moreover, it is generally less sensitive to moderate photobleaching. 91 FLIM has been successfully used with FRET to achieve FLIM -FRET imaging, which allows the investigation of protei n micro -environment with resolution higher than that of conventiona l FRET. It should be noted that FLIM requires more sophisticated instrument setup and is sensitive to environment changes affecting FP nature. 15 Extending from the exploitation of FP photophysical characteristics such as photobleaching, energy transfer , turnover rate, and lifetime, a n alternative technique to detect protein -protein interaction is based on fluorescence intensity fluctuation, or blinking. When two different FPs are in a limited spatial volume and are excited, the two emission intensities will fluctuate together and give a degree of cross -correlation. Fluorescence cross -correlation spectroscopy (FCCS) records the time -correlated fluctuation and provides a means to estimate the interdependence of diffusion rates of the two FPs, therefore to deduce the interaction between two POIs fused with the FPs. 92, 93 Importantly, FCCS differs from FRET in the way that FCCS is less demanding in the distance and the orientation between two FPs mandated by FRET, thus giving more freedom for study design. Besi des, the correlation signal is usually more sensitive than the FRET signal, allowing fluorochromes with lower brightness to be used. 94 Recently, ZhangÕs group has furth er expanded the utilization of fluorescence correlation spectroscopy to combine with a super -resolution microscopic technique, photochromic stochastic optical fluctuation NATURE METHODS | VOL.14 NO.4 | APRIL 2017 | 429ARTICLESboth at the single-pixel level and averaged across the entire cell. The average normalized pcSOFI value rapidly increased in the first minute after PKA activation in HeLa cells treated with a cocktail of the adenylyl cyclase activator forskolin (Fsk) and phosphodi -esterase inhibitor 3-isobutyl-1-methylxanthine (IBMX), reaching a plateau at a 25Ð39% increase after 7Ð10 min ( Fig. 2bÐ d) and showing a dynamic range of up to a 40% increase. Addition of the PKA-specific inhibitor H-89 (20 M) gradually decreased the average normalized pcSOFI value, thus demonstrating the reversibility. Additional experiments using a negative control bio -sensor that could not be phosphorylated ( Fig. 2c ,d) and the mem -brane-targeted PKA inhibitor PKI ( Fig. 2 c) demonstrated that the response was dependent on phosphorylation of the biosensor and PKA activity, respectively. Time courses with H-89 inhibition and a submaximal dose of Fsk ( Supplementary Fig. 6 ) showed that FLINCÐAKAR1 accurately reported over a range of PKA activity. We initially quantified PKA activity every 5 min ( Fig. 2 c) and, after further development, every 30 s ( Fig. 2 d). The kinetics of PKA stimulation, as monitored by FRET-based or FLINC-based AKAR, showed no significant differences (FRET half-time ( t1/2) = 1.2 min, number of cells ( n) = 7; FLINC t1/2 = 0.73 min, n = 4). Importantly, our analysis of FLINCÐAKAR1 fluctuations gener -ated super-resolution images of PKA activity at each time point throughout the treatment course. Monitoring the same profile line in the normalized pcSOFI images before and after Fsk/IBMX treatment ( Fig. 2 e) indicated that subdiffraction-limit activity features ( Fig. 2 f) that responded to stimulation were resolved. Actin-targeted FLINCÐAKAR1 enabled us to distinguish the stimulated PKA activity with an average Gaussian full width at half maximum of 179 6 nm ( n = 7) and 116 6 nm ( n = 7) (mean s.e.m.), in agreement with the respective use of second- and third-order analyses ( Supplementary Fig. 7a ,b). Furthermore, we were able to resolve the stimulated membrane PKA activity in converging filopodial features separated by 160 nm and 107 nm, by using second- and third-order cumulant analyses, respectively (Supplementary Note and Supplementary Fig. 7c ,d). Under the conditions used here ( Supplementary Note ), biosensor diffusion did not affect the temporal and spatial resolution of pcSOFI 11, and its accurate quantification of FLINC. Thus, as a reporter of kinase activity, FLINCÐAKAR1 not only provides a consistent readout for PKA activity but also allows for monitoring of the dynamic changes of PKA activity with high contrast at super-resolution. PKA activity microdomains The PKA activity maps generated by FLINCÐAKAR1 revealed many minute and highly active punctate features on the basal mem -branes of living cells. These activity puncta, with a mean diameter of 350 nm ( Supplementary Note ), were clearly resolved after Fsk/ IBMX stimulation. Activation of PKA induced a twofold increase in microdomain coverage over the basal membrane, but this induc -tion was not observed with the DpTT control or the nonphospho -rylatable mutant (TA) ( Fig. 2 g). Using direct stochastic optical reconstruction microscopy (dSTORM) in total internal reflec -tion fluorescence (TIRF) conditions, we further verified the pres -ence of these highly active PKA microdomains ( Supplementary Note ). Phospho-PKA substrates (p-PKAsubs) were clearly clus -tered ( Supplementary Fig. 8a ,b), forming distinct microdomains with a mean diameter of approximately 250 nm on the basal membrane ( Supplementary Fig. 8c and Supplementary Note ). PAABDSubstratePPAABDSubstrateFLINCmodulatedCapture RFP!uctuationsat each pixelQuantify activitywith increasedresolutionand contrastat each pixelKinase activityPKA substrateFLINCÐAKAR1domain structureCAAXEV linker10 µmFLINC Fsk/IBMXFLINC beforeFLINC H89H-89Fsk/IBMX0510150.91.01.11.21.31.41.5Time (min)Normalized pcSOFI valueWT (n = 4)H-89Fsk/IBMX0510150.91.01.11.21.31.41.5Time (min)5.511.016.522.027.533.00.80.91.01.11.21.3WT (n = 9)TA (n = 7)WT + PKI (n = 4)Time (min)Normalized pcSOFI valueFsk/IBMXH-89TA (n = 5)DronpaFHA1TagRFP-Tabcd12340255075100125Length (µm)Normalized pcSOFI valueBefore FIAfter FIp1eLength (nm)Ð1000100FWHM:158.2 nmFitp1fÐ10123456Fold changeTA DpTT**RefNSWTgFigure 2 | FLINC resolves PKA activity microdomains on the plasma membrane at super-resolution. ( a) Schematic of the FLINCÐAKAR design principle, the domain structure of FLINCÐAKAR1 and the acquisition of super-resolution activity images with pcSOFI. ( b) FLINCÐAKAR1 super-resolution images clearly resolve the response to Fsk/IBMX stimulation (Fsk 50 M and IBMX 100 M) and inhibition (H-89 20 M), and detailed spatial information on membrane PKA activity emerges. Color scales are identical. ( c) Mean normalized pcSOFI response time course from live HeLa cells expressing WT ( n = 9 cells) FLINCÐAKAR1, nonphosphorylatable mutant (TA, n = 7) and WT coexpressed with PKI (WT + PKI, n = 4), after PKA stimulation and inhibition. ( d) The normalized pcSOFI response time course from live HeLa cells expressing WT ( n = 4) FLINCÐAKAR1 and nonphosphorylatable mutant (TA, n = 5), after PKA stimulation and inhibition, determined with a fast acquisition imaging scheme. ( e) A profile line at the same position across the pcSOFI images before and after Fsk/IBMX (FI) stimulation clearly demonstrates sensing of PKA activity at super-resolution; the profile is marked by a red line in b. ( f) Zoom view of the active PKA feature (p1) in f, showing resolution of the Gaussian fitting and full width at half maximum (FWHM) size of this subdiffraction-limit PKA activity microdomain. ( g) Comparison of the changes in the fraction of membrane area occupied by punctate structures after stimulation across various FLINC constructs, all targeted with the same CAAX motif. In FLINCÐAKAR1 experiments, cells (WT, n = 10; TA, n = 8) were stimulated with Fsk/IBMX; in DpTT experiments, cells ( n = 8) received no drug treatment. Unless indicated otherwise, pairwise t-test results are shown for data compared with the reference (ref) construct. NS, not significant; **P < 0.01. In c and d, center line and whiskers mark the average and s.e.m., respectively. Figure 0-8 Schematic representation of the FLINC -AKAR design principle, the domain structure of FLINC ÐAKAR1 and the acquisition "of super -resolution activity images with pcSOFI. Artwork copied from ref [95]. 16 imaging (pcSOFI), and discovered the phenomenon which they termed fluorescence fluctu ation increase by contact (FLINC). 95 Tagged w ith FLINC -based biosensors, they successfully visualized the activity of protein kinase A (PKA) with clustered A kinase -activity reporter (AKAR) proteins in microdomains on the plas ma membranes of a living migrating '4CHO cell at super -resolution (Figure I-8). In contrast to the use of tandem - or biomolecular - based FRET/FLIM/FCCS, FPs can be spl it into two complementary segments to image protein -protein interaction. A single FP gene is split into two halves and fused to two POIs, respectively. When the two POIs interact, the two split FP fragments are brought to close proximity and self -assembled into a whole FP which fluoresce as normal. 96, 97 Alternatively, this type of bimolecular fluore scence complementation (BiFC) can be used to investigate the topology transformation of a single POI. Besides the conventional applications in protein -protein interaction studies, BiFC has also found use in drug discovery and flow cytometry. 98, 99 However, care is necessary to avoid art ifacts such as diffusion -limited signal loss, or complementation -induced POI interaction to affect the interpretation of the data. While advanced applications of FPs are emerging from exploit ing photophysical features of FPs , it should be noted that other FP can always be extensively engineered to meet various imaging demands, such as, pH-stable FPs for ratiometric pH-sensing, 100-103 PAFPs and reversibly photoswitchable FPs (RSF P) for super -resolution imaging/ single -molecular nanoscopy (i.e., P ALM, STORM , RESOLFT ),73, 104 and allosteric redox sensor .105 17 I.2.2!Other trending applications of fluorescence bioimaging The discus sion has hitherto been focused on FPs. Although there are innumerable FP variants possessing widely different features and showing successful applications in different areas, synthetic fluorochromes normally provide a higher degree of scaffold variety and have more freedom to modulate their photophysical characteristics . In this context, fluorescence bioimaging is tremendously advanced by invention of modern dyes and subsequent applications. I.2.2.1!Biosensing An ongoing hot field is fluorescence -based sensing of important molecular spe cies involved in interesting biological and/or pathological processes .106 The interactions between host and guest molecules were used to develop analyte -specific sensors at the time the 1987 Nobel Prize was awarded for supramolecular chemistry .107-109 This approach, howeve r, is often limited by the non -quantitative nature of noncovalent interaction s. Inheriting the concepts of substrate -indicator recognition, f luorochromes have been chemically modulated to reaction -based analyte -specific molecular indicators that respond to various cationic, anionic, redox, metal, and bio macro molecular species, including biologically relevant H +, F -, CN -, H 2O2, RSH, NO, superoxides, nucleic acids, and enzymes .110-112 Over the years, a vast number of reaction -based indicator systems have emerged from a variety of reactions (e.g., boron -oxygen displacement, spirocyclization, aza -cope rearrangement, etc.), and have been heavily reviewed. 113-115 These approaches are transplanted to chemoselective in vivo imaging to map the analyte 18 activities whereof either site -specific, 116, 117 or pan-cellular .118-121 Efforts have also been made to fabricate polymer -based systems to increase signal intensity .122-125 More recently, design principles based on different photophysical and/or photochemical mechanisms have been applied to achieve ratiometric reaction -triggered sensing systems. 126 These principles include 1) intra molecular charge transfer (ICT): these probes have electron donor and acceptor parts which can interact with analyte and result in internal charge redistribution, leading to either absorp tion or emission wavelength shifts ; 2) excited -state intramo lecular proton transfer (ESIPT): these probes have signature large Stokes shift emission bands generated from fast intramolecular proton transfer between their proton donor and acceptor moieties , which can be used as ratiometric fluorescence readout as they are sensitive towards analyte binding; 3) FRET and TBET (denoting through -bond energy transfer): probes of these two types of energy transfer (through space or wave function, respectively) are c omposed of covalently linked donor emitter and acceptor emitter moieties, the latter of which is activated upon reaction with analytes and emits fluorescence signals; 4) monomer -excimer interchange: these probes form excimers at the dormant state and dissociate upon binding of analytes, resulting in wavelength shifting. A noteworthy example is the real -time intracellular glutathione (GSH) dynamic quantification probe (QuicGSH) recently achieved by UranoÕs group. 121 This is the fir st reported reversible reaction -based GSH sensing system that demonstrates high sensitive ness and temporal resolution (t1/2 = 620 ms at [GSH] = 1 mM). This work showed 19 the optimal combination of a tetramethyl rhodamine (TMR) and sil icon rhodamine (SiR) FRET pair and a fast reversible thiol nucleophilic addition on SiR 9 -position (Figure I-9). A renewed research field is the direct noninvasive optical measurement of cell membr ane voltage, or voltage -sensing, although for years patch clamp methods serve d as the gold standard for measuring ion currents and voltages related to neuron physiology. 127 Voltage sensing with fluorescent indicator has a long history . Early probes were only available for Ca 2+ sensing, commonly based on genetically encodable voltage -sensitive phosphatases or rhodopsins. New small molecular design strategies have CO2MeONNHNONHOSiNHNH2MeTMR (donor) SiR (acceptor) CO2MeONNHNONHOSiNHNH2MeGS+ GSH - GSH observedwithinminutesoftheswitchtoglucose-freemedium(Fig.5 andSupplementaryMovie5).Reperfusionofnormalmedium containing25mMglucoseresultedinagradualrecoveryofGSH concentration,whilecontinuedperfusionofglucose-freemedium suppressedtherecoveryofGSHconcentration.SincenopHchange wasobservedduringglucosedepriv ation(SupplementaryFig.17), the!uorescencesignalof QG3.0isconsideredtore !ectintracellular GSHlevel.Thisnovel "ndingthatchangesinGSHleveloccurona fasttimescaleduringglucosest arvationcouldnothavebeenmade withconventional !uorescentprobes. Conclusions Wehaverationallydesignedandsynthesizedreversible !uorescent probes(QuicGSH)thatenablereal-timemonitoringofGSHcon- centrationsinlivingcells,basedonour "ndingthatsomerhoda- minesshowextraordinarilyfastreactionkineticswiththiols.We demonstratedthevalidityoftheseprobesforquantifyingintracellu- larGSHconcentrationbymeasuringGSHlevelsinvariouscelllines andbysuccessfullymonitoringthesecond-orderdynamicsofGSH inlivingcellsforthe "rsttime.Previouslyreported !uorescent protein-basedreversiblesensorsformeasuringGSH/GSSGredox potential42havesubstantiallimitations.Forexample,theyarenot compatiblewithcellsthatexhibitlowtransfectionef "ciency,since theyrequiregeneticexpression.Further,overexpressionofthese sensorproteinswouldpotentiallydisturbphysiologicalredox cyclesinlivingcells,sincethesesensorscontainaglutaredoxin (Grx)moietythatcanmediateconversionofGSHtoGSSG 43.In contrast,ournewlydevelopedsmall-molecule-based,reversible probesdonotrequiregeneticexpressionandaremuchlesslikely toperturbcellularfunctionsduringlive-cellGSHimaging.The resultspresentedheresupportourviewthatQuicGSHprobesare revolutionarytoolsforinvestigatinghowglutathionedynamicsare regulatedinaphysiologicalcontext,duetotheircapabilityfor real-timequanti "cationofGSHwithhightemporalresolution. MethodsSynthesismethodsforallcompounds,characterizationdataanddetailsofthe imagingexperimentsareprovidedintheSupplementaryInformation. RatiometricimagingofGSH. Unlessotherwisestated, !uorescenceimagesof QG3.0werecapturedwithaLeicaSP5underthefollowinginstrumentalconditions: laser,550nm;Ch1,560 Ð605nm(gainofPMT,800V);Ch2,620 Ð700nm(gainof PMT,900V(conditionA)or950V(conditionB)).Ratioimageswereconstructed byMetaMorphsoftware.BinarizationoftheCh1images(TMR)bymeansof automaticthresholding(usingthecommand ÔAutoThresholdforLightObjects Õ)resultedinextractionofthesignalscorrespondingtomitochondria,anddivisionof Ch1byCh2(SiR)affordedratioimages.Regionsofinterest(ROIs)werede "nedas thewholemitochondriaineachcell. StainingofcellswithQuicGSHprobes. Cells(A549,HeLa,SHIN3,SKOV3,H226) wereseededon35mmglass-bottomeddishesorµ-slides(ibidi)andincubatedfor 5mininDMEMorRPMIcontaining QG3.0(0.5Ð5µM,0.5%DMSO)andPluronic F-127(0.01 Ð0.04%,w/v)at37¡Cina5%:95%CO 2:airincubator.Thecellswerewashed, placedinDMEMorRPMI(withoutPluronicF-127),andfurtherincubatedfor 15minbeforeimaging.HBSSwasusedastheimagingmedium.DMEMwasusedfor A549,HeLa,H226cells,andRPMIwasusedforSHIN3,SKOV3cells.ForHUVEC, incubationwasperformedinEBM-2 ªmediumcontaining QG3.0 (5µM,0.1% DMSO)andPluronicF-127(0.04%,w/v)for10min.Thecellswerewashed,placedin EBM-2 ª(withoutPluronicF-127),andfurtherincubatedfor15min. Quanti !cationofGSHinvariouscells. Cellsofeachline(A549,H226,SHIN3, SKOV3,HeLaandHUVEC)wereseededon8-wellµ-slides,ibiTreat,andstainedas describedintheSupplementaryInformation.Ineachexperiment,around50 Ð300cellswereselectedasROIs,andaveragedGSHconcentrationwascalculated (n=approximately50 Ð300)usingthecalibrationcurvesdescribedinthe SupplementaryInformation.Thesameexperimentswereperformedondifferent days( N=5Ð15)inordertoevaluatetheeffectofthecultureconditionsofcells onGSHconcentration.AllthedataareshowninSupplementaryFig.8;thedata showninFig.4bwereextractedfromthewholedatasetbystatisticalprocessingas follows.Randomselection( n=20forA549, n=30forothers,respectively)was performedfromeachpopulation( N),andthewholerandomlyselecteddataset (n=20 Ð30,N=5Ð15,dependingonthecellline)wasstatisticallyanalysedbymeans ofbox Ðwhiskerplots.OutlierswererejectedbyapplyingtheSmirnov ÐGrubbsoutliertest(3 !,P<0.01). Real-timelive-cellimagingofGSHuponH 2O2treatmentwithaperfusionsystem. Time-lapseimagingofGSH !uctuationwasperformedwithaLeicaSP5equipped withaperfusionsystem:aWarnerimagingchamber(RC-20),asix-channel perfusionvalvecontrolsystem(VC-6),amulti-lineinlinesolutionheater (SHM-828)andatemperaturecontroller(TC-324B).Perfusionofthechamberwas performedundergravity-feedata !owrateof0.5mlmin Ð1.Fluorescenceimages (512!512pixels)wereobtainedevery5swithascanrateof400Hz.Cells(A549or HUVEC)wereseededon15mmglassslides(coatedwithpoly- L-lysine),and QG3.0 (1µM)wasloadedasdescribedabove.HBSS(solutionA)and50µMor500µM H2O2inHBSS(solutionB)werepreparedinsyringesAandB,respectively, andthesolutionwasswitchedfromAtoB(andBtoA)asrequired.Theinhibition experimentofglutathionereductase(GR)with1,3-bis(2-chloroethyl)-1- nitrosourea 44(BCNU)wasperformedbypassingHBSSbuffercontaining300µM BCNU(solutionC)insteadofsolutionA. Real-timelive-cellimagingofglucose-deprivation/recoverywithaperfusion system. Thesameperfusionsystemdescribedabovewasused.Perfusionofthe chamberwasperformedundergravity-feedata !owrateof0.5mlmin Ð1.Fluorescenceimages(512 !512pixels)wereobtainedevery30swithascanrateof 400Hz.A549cellswereseededon15mmglassslides(coatedwithpoly- L-lysine), 25G0G25G124681012OverlayabRatio images0612182430 Time (min) GSH (mM)10 min10 min30 min 20 min10 min30 min GSH (mM)012 Figure5| Real-timeGSHimagingwith QG3.0 (1!M)inA549cellsunderglucose-deprivationwithaperfusionsystem. a,Left:overlayimages.Right: time-lapseratioimagingofGSHunderhigh-glucose(25mM:25G)orglucose-free(0mM:0G)conditions.1:0 Ð3min,25mMglucose;3 Ð12min,0mM glucose;12 Ð30min,25mMglucose.2:0 Ð3min,25mMglucose;3 Ð30min,0mMglucose.Imagesaretakenfromtheoriginalmovies(Supplementary Movie5),scalebars,20 µm.b,Time-courseofGSHconcentrationunderglucosedeprivation(red,condition1;blue,condition2).AveragevaluesofGSH concentrationinindividualcells( n=10,for1; n=8for2)areplottedandtheerrorbarsrepresents.d. NATURECHEMISTRY DOI:10.1038/NCHEM.2648 ARTICLESNATURECHEMISTRY |VOL9|MARCH2017| www.nature.com/naturechemistry 285©2017MacmillanPublishersLimited,partofSpringerNature.Allrightsreserved. observedwithinminutesoftheswitchtoglucose-freemedium(Fig.5 andSupplementaryMovie5).Reperfusionofnormalmedium containing25mMglucoseresultedinagradualrecoveryofGSH concentration,whilecontinuedperfusionofglucose-freemedium suppressedtherecoveryofGSHconcentration.SincenopHchange wasobservedduringglucosedepriv ation(SupplementaryFig.17), the!uorescencesignalof QG3.0isconsideredtore !ectintracellular GSHlevel.Thisnovel "ndingthatchangesinGSHleveloccurona fasttimescaleduringglucosest arvationcouldnothavebeenmade withconventional !uorescentprobes. Conclusions Wehaverationallydesignedandsynthesizedreversible !uorescent probes(QuicGSH)thatenablereal-timemonitoringofGSHcon- centrationsinlivingcells,basedonour "ndingthatsomerhoda- minesshowextraordinarilyfastreactionkineticswiththiols.We demonstratedthevalidityoftheseprobesforquantifyingintracellu- larGSHconcentrationbymeasuringGSHlevelsinvariouscelllines andbysuccessfullymonitoringthesecond-orderdynamicsofGSH inlivingcellsforthe "rsttime.Previouslyreported !uorescent protein-basedreversiblesensorsformeasuringGSH/GSSGredox potential42havesubstantiallimitations.Forexample,theyarenot compatiblewithcellsthatexhibitlowtransfectionef "ciency,since theyrequiregeneticexpression.Further,overexpressionofthese sensorproteinswouldpotentiallydisturbphysiologicalredox cyclesinlivingcells,sincethesesensorscontainaglutaredoxin (Grx)moietythatcanmediateconversionofGSHtoGSSG 43.In contrast,ournewlydevelopedsmall-molecule-based,reversible probesdonotrequiregeneticexpressionandaremuchlesslikely toperturbcellularfunctionsduringlive-cellGSHimaging.The resultspresentedheresupportourviewthatQuicGSHprobesare revolutionarytoolsforinvestigatinghowglutathionedynamicsare regulatedinaphysiologicalcontext,duetotheircapabilityfor real-timequanti "cationofGSHwithhightemporalresolution. MethodsSynthesismethodsforallcompounds,characterizationdataanddetailsofthe imagingexperimentsareprovidedintheSupplementaryInformation. RatiometricimagingofGSH. Unlessotherwisestated, !uorescenceimagesof QG3.0werecapturedwithaLeicaSP5underthefollowinginstrumentalconditions: laser,550nm;Ch1,560 Ð605nm(gainofPMT,800V);Ch2,620 Ð700nm(gainof PMT,900V(conditionA)or950V(conditionB)).Ratioimageswereconstructed byMetaMorphsoftware.BinarizationoftheCh1images(TMR)bymeansof automaticthresholding(usingthecommand ÔAutoThresholdforLightObjects Õ)resultedinextractionofthesignalscorrespondingtomitochondria,anddivisionof Ch1byCh2(SiR)affordedratioimages.Regionsofinterest(ROIs)werede "nedas thewholemitochondriaineachcell. StainingofcellswithQuicGSHprobes. Cells(A549,HeLa,SHIN3,SKOV3,H226) wereseededon35mmglass-bottomeddishesorµ-slides(ibidi)andincubatedfor 5mininDMEMorRPMIcontaining QG3.0(0.5Ð5µM,0.5%DMSO)andPluronic F-127(0.01 Ð0.04%,w/v)at37¡Cina5%:95%CO 2:airincubator.Thecellswerewashed, placedinDMEMorRPMI(withoutPluronicF-127),andfurtherincubatedfor 15minbeforeimaging.HBSSwasusedastheimagingmedium.DMEMwasusedfor A549,HeLa,H226cells,andRPMIwasusedforSHIN3,SKOV3cells.ForHUVEC, incubationwasperformedinEBM-2 ªmediumcontaining QG3.0 (5µM,0.1% DMSO)andPluronicF-127(0.04%,w/v)for10min.Thecellswerewashed,placedin EBM-2 ª(withoutPluronicF-127),andfurtherincubatedfor15min. Quanti !cationofGSHinvariouscells. Cellsofeachline(A549,H226,SHIN3, SKOV3,HeLaandHUVEC)wereseededon8-wellµ-slides,ibiTreat,andstainedas describedintheSupplementaryInformation.Ineachexperiment,around50 Ð300cellswereselectedasROIs,andaveragedGSHconcentrationwascalculated (n=approximately50 Ð300)usingthecalibrationcurvesdescribedinthe SupplementaryInformation.Thesameexperimentswereperformedondifferent days( N=5Ð15)inordertoevaluatetheeffectofthecultureconditionsofcells onGSHconcentration.AllthedataareshowninSupplementaryFig.8;thedata showninFig.4bwereextractedfromthewholedatasetbystatisticalprocessingas follows.Randomselection( n=20forA549, n=30forothers,respectively)was performedfromeachpopulation( N),andthewholerandomlyselecteddataset (n=20 Ð30,N=5Ð15,dependingonthecellline)wasstatisticallyanalysedbymeans ofbox Ðwhiskerplots.OutlierswererejectedbyapplyingtheSmirnov ÐGrubbsoutliertest(3 !,P<0.01). Real-timelive-cellimagingofGSHuponH 2O2treatmentwithaperfusionsystem. Time-lapseimagingofGSH !uctuationwasperformedwithaLeicaSP5equipped withaperfusionsystem:aWarnerimagingchamber(RC-20),asix-channel perfusionvalvecontrolsystem(VC-6),amulti-lineinlinesolutionheater (SHM-828)andatemperaturecontroller(TC-324B).Perfusionofthechamberwas performedundergravity-feedata !owrateof0.5mlmin Ð1.Fluorescenceimages (512!512pixels)wereobtainedevery5swithascanrateof400Hz.Cells(A549or HUVEC)wereseededon15mmglassslides(coatedwithpoly- L-lysine),and QG3.0 (1µM)wasloadedasdescribedabove.HBSS(solutionA)and50µMor500µM H2O2inHBSS(solutionB)werepreparedinsyringesAandB,respectively, andthesolutionwasswitchedfromAtoB(andBtoA)asrequired.Theinhibition experimentofglutathionereductase(GR)with1,3-bis(2-chloroethyl)-1- nitrosourea 44(BCNU)wasperformedbypassingHBSSbuffercontaining300µM BCNU(solutionC)insteadofsolutionA. Real-timelive-cellimagingofglucose-deprivation/recoverywithaperfusion system. Thesameperfusionsystemdescribedabovewasused.Perfusionofthe chamberwasperformedundergravity-feedata !owrateof0.5mlmin Ð1.Fluorescenceimages(512 !512pixels)wereobtainedevery30swithascanrateof 400Hz.A549cellswereseededon15mmglassslides(coatedwithpoly- L-lysine), 25G0G25G124681012OverlayabRatio images0612182430 Time (min) GSH (mM)10 min10 min30 min 20 min10 min30 min GSH (mM)012 Figure5| Real-timeGSHimagingwith QG3.0 (1!M)inA549cellsunderglucose-deprivationwithaperfusionsystem. a,Left:overlayimages.Right: time-lapseratioimagingofGSHunderhigh-glucose(25mM:25G)orglucose-free(0mM:0G)conditions.1:0 Ð3min,25mMglucose;3 Ð12min,0mM glucose;12 Ð30min,25mMglucose.2:0 Ð3min,25mMglucose;3 Ð30min,0mMglucose.Imagesaretakenfromtheoriginalmovies(Supplementary Movie5),scalebars,20 µm.b,Time-courseofGSHconcentrationunderglucosedeprivation(red,condition1;blue,condition2).AveragevaluesofGSH concentrationinindividualcells( n=10,for1; n=8for2)areplottedandtheerrorbarsrepresents.d. NATURECHEMISTRY DOI:10.1038/NCHEM.2648 ARTICLESNATURECHEMISTRY |VOL9|MARCH2017| www.nature.com/naturechemistry 285©2017MacmillanPublishersLimited,partofSpringerNature.Allrightsreserved. a b c Figure 0-9 In vivo dynamic GSH imaging. (a) Chemical structure of QG3.0 probe. (b) Real -time GSH imaging with QG3.0 (1 ! M) in A549 cells under glucose -deprivation with a perfusion system. Overlay and time -lapse ratio imaging of GSH und er high -glucose (25 mM: 25G) or glucose -free (0 mM: 0G) conditions. 1: 0 Ð3 min, 25 mM glucose; 3 Ð12 min, 0 mM glucose; 12 Ð30 min, 25 mM glucose. 2: 0 Ð3 min, 25 mM glucose; 3 Ð30 min, 0 mM glucose. scale bars, 20 ! m. (c) Time -course of GSH concentration unde r glucose deprivation (red, condition 1; blue, condition 2) . Images and plot copied from ref [121]. 20 drawn interest in recent years. Briefly, amphipathic synthetic probes can span across the plasma membrane. The sensin g output generally relies on the following mechanisms (Figure I-10): 1) solvatochromism, whereby the membrane voltage change perturbs the dipole moment of solvatochro mic dyes and indu ces emission wavelength shifts ; 2) electrochromism, whereby a molecular Stark effect counts for the fast shifts (fs to ps) in both absorption and emission wavelengths; 3) voltage -dependent redistribution of FRET pairs, of which the lipophi lic acceptor emitter will redistribute upon polarizations and fractionally affect FRET signal; 4) photoinduced ele ctron transfer (PeT) quenching of fluorescent molecular wires, the quenching sensitiv ity of which is affected by the external electric field and results in fluorescence readout turn -on/off.128 An example is given below ßuorophore to the donor via a phenylenevinylene molec-ular wire, thereby decreasing the distance-dependence ofeT [35].Our Þrst generation of PeT-based voltage sensors, Volta- geFluors, or VF dyes (Figure 1), feature a dichlorosulfo- ßuorescein reporter, phenylenevinylene molecular wire,and a dimethylaniline donor [36,37!]. As predicted fromthe model proposed above, VF dyes show a fast ßuores-cence turn-on in response to depolarizations, with asensitivity of 27% DF/F per 100 mV that is linear overthe physiological range of "100 mV. Additionally, VFdyes exhibit no measurable capacitive loading, and cantrack action potential spikes in single trials in mammalianneurons and in ex vivo leech preparations [36]. Consistent with a PeT-based mechanism, modulation of the relativeelectron afÞnities of the donor aniline and acceptor ßuor- ophore alters the apparent voltage sensitivity, improving a2nd generation of VFs to sensitivities of approximately 49% DF/F per 100 mV, while maintaining the speed andlinearity of the Þrst VF dye [37!].Although VF dyes are fast, sensitive, and non-disruptive, their pan-membrane staining limits spatial resolution inheterogeneous samples. Cell-type speciÞc labeling withvoltage-sensitive PeT probes would enable coupling ofthe speed and sensitivity of a small molecule approach with the speciÞcity achieved through genetic means. Toaddress this, we developed a small-molecule, photoacti- vatable optical sensor of transmembrane potential, orSPOT [38!]. In this approach, the ßuorescence ofVF2.1.Cl is quenched by formation of a nitrobenzyl etherSmall molecule fluorescent voltage indicators Miller 77Figure 2(a)(c)(d)(b)fluorescentdepolarizationhyperpolarizationFRET no FRET PeT quenching fluorescenteÐeÐdepolarizationhyperpolarizationdepolarizationhyperpolarizationdepolarizationhyperpolarizationSolvatochromic response Electrochromic optical shiftsVoltage-dependent redistribution Photoinduced electron transfer (PeT)Current Opinion in Chemical BiologyMethods of voltage sensing with small molecules. (a) Some voltage-sensitive dyes display voltage-dependent accumulation in cell membranes. Typically this is associated with a solvatochromic change, where the dye becomes more fluorescent upon association with the relatively non-polar lipid bilayer. An example of this is the slow response component of merocyanine 540. (b) Electrochromic, or ÔfastÕ voltage-sensitive dyes, interactdirectly with the electric field across a cell membrane, causing a change in the molecular orbital energetics of the dye. This results in smallwavelength changes depending on the relative orientation of the fluorophore dipole with the membrane-associated electric field. For the case ofdyes such as di-4-ANEPPS, this results in a bathochromic shift upon hyperpolarization and a hypsochromic shift upon depolarization. (c)Lipophilic anions traverse the membrane in a potential-dependent fashion. Several configurations are possible, however, all include a stationaryfluorophore, pictured here on the outer leaflet of the membrane (green), and a mobile fluorophore or quenching group (red). Changes in membrane potential cause a redistribution of the mobile anion, which changes the efficiency of energy transfer or quenching. (d) Voltage sensing viaphotoinduced electron transfer (PeT) through a molecular wire takes advantage of the sensitivity of eT to externally applied electric fields. At rest,the hyperpolarized, negative membrane potential enhances PeT from the electron-rich donor into the fluorophore (purple), quenching anddiminishing fluorescence. Upon depolarization, the positive membrane potential slow the rate of PeT, resulting in a fluorescence brightening.www.sciencedirect.com Current Opinion in Chemical Biology 2016, 33:74Ð80Figure 0-10 Methods of voltage sensing with small molecules. (a) Voltage -dependent traverse of solvatochromic dyes in lipid membranes. (b) Electrochromic voltage -sensing. (c) Voltage -dependent FRET acceptor redistribution. (d) Voltag e-sensing based on PeT perturbation. Artwork copied from ref [128]. 21 showing the VoltageFluor dyes developed in TsienÕs lab which adopts the PeT approach. 129, 130 Recently, by substituting the fluorescein w ith silicon rhodamine in the molecular wire, MillerÕs group successfully developed BeRST 1 as the NIR version of VoltageFluor, and achieved spontaneous optical voltage -sensing with epifluorescence imaging in rat hippocampal neurons (Figure I-11).131 An interesting optical sensing area is the mechanical tension across cell surface. Transmembrane adhesion receptors are important in cell biology. They mediate cell -cell and cell -extracellular matrix adhesions, and transmit the mechanical forces to the chemo -mechanical coupling cycles controlling cell fate. Salaita Õs group designed a FRET -based emitter -quencher pair to study the interaction between membrane -anchored EGFR and silica chip immobilized EGF ( Figure I-12a).132 Later on, this strategy was expanded to ofCa 2+sensorsareoneofthemostwidelyusedfunctional probesforsystemsandcellularneuroscience.However,because theexcitationandemissionspectrumofGCaMPliessquarely withintheregimeofpreviousgenerationsofVoltageFluors 13,23(aswellasotherfunctionalprobes),thishasprecludedtheuse ofGCaMPswithPeT-basedvoltagesensitivedyes.Neurons expressingthegeneticallyencodedCa 2+sensor,GCaMP6s, 29whichprovidesthelargest !uorescenceresponsetoCa 2+,were stainedwithBeRST1( Figure5 aandb).GCaMP6s-positive neurons( Figure5 b)displayedbrightgreencytosolic !uorescence,whilemembraneswereclearlylabeledwith BeRST1( Figure5 a).Weevokedactionpotentialsvia "eldstimulationandobtainedopticalrecordings, "rstinthegreen channelforGCaMP6sandtheninthefar-redtoNIRchannel forBeRST1,fromasinglecellexpressingGCaMP6sand stainedwithBeRST1( Figure5 aandb).Insequentialoptical traces,boththerisein Vmemandrisein[Ca 2+]iareclearly visible( Figure5 c).DualBeRST1andGCaMP6simaging enablesdiscriminationofvoltageandCa 2+transients,withthe voltagetransient(approximately15% !F/F)precedingthe measuredGCaMP6ssignal(approximately5% !F/F)fora singleevokedactionpotential( Figure5 c).Oneofthe advantagesofvoltageimagingistheabilitytodeconvolvefast spikingthatCa 2+imaging,withitsslowerintrinsicsignalsand slowprobeunbindingkinetics,oftencannotresolve.When neuronsexpressingbothGCaMP6sandstainedwithBeRST1 werestimulatedtoevoketrainsofactionpotentialsat5,10,and 20Hz,discrete Vmemresponsescorrespondingtoindividually evokedactionpotentialswerevisibleintheBeRST1channelat allfrequencies( Figure5 d),whileCa 2+responseasmeasuredby GCaMP6swereonlydiscernibleat5Hz,asmeasuredbysmall changesinontherisingedgeofGCaMP6s !uorescence( Figure5e).DuetotheintrinsicresponsespeedofBeRST1,two-color functionalimagingshouldbegeneralizabletomanyothertypes ofCa 2+indicators.Theseexperimentshighlighttheabilityof BeRST1toparticipateinfunctionalimagingexperimentsin whichmultiplesignalsneedtobeinterpretedinthesame system.ThenonoverlappingsignalsemanatingfromBeRST1and GFP-based!uorophoresprovideauniqueopportunityto performtwo-colorvoltageimaging.Weexpressedthe geneticallyencodedvoltagesensor,ASAP1, 19incultured neuronsandstainedtheseneurons,asbefore,withBeRST1 (Figure6 a!d).Inresponsetoanevokedtrainofaction potentials( Figure6 e),ASAP1providesdecreasesinrelative !uorescenceofabout6% ±1.4%(S.E.M.,withasignal-to-noise ratio,SNR,of8:1, n=4cells),whichisconsistentwiththe reportedvalueof !4.8%!F/F.19BeRST1,inthesamecell, underidenticalstimulationparameters( Figure6 e),givesa15% ±2.1%increase(S.E.M.,SNRof41:1, n=4cells),highlighting theutilityofBeRST1formeasurementoffastspikingevents andfortwo-colorvoltageimagingwithcomplementary geneticallyencodedvoltagesensorsbasedonGFP. 18OpticalElectrophysiologywithBeRST1andChR2. Finally,wehopedtointerfaceBeRST1withoptogenetictools likeChannelRhodopsin2(ChR2)inordertodemonstratethe Figure4. SpontaneousvoltageimaginginGFP-labeledcellswithBeRST1.Epi !uorescenceimagesofrathippocampalneuronsexpressing(a)GFP and(b)stainedwithBeRST1.Scalebaris20 !m.(c)Opticaltracesofspontaneousactivityinneuronsfrompanels(a)and(b).Numbersnextto tracescorrespondtoindicatedcellsinpanel(b).Opticalsamplingrateis500Hz. Figure5. DualvoltageandCa 2+imagingwithBeRST1and GCaMP6s.Epi !uorescenceimagesofarathippocampalneuron stained(a)withBeRST1andexpressing(b)GCaMP6s.Scalebaris 20!m.(c)SequentialCa 2+andvoltageimagingduring "eld stimulationoftheneuroninpanels(a)and(b).Voltage-(upper) andCa 2+-(lower)inducedchangesin !uorescence.(d)Voltage-(left) and(e)Ca 2+-(right)induced !uorescenceresponsetotrainsofaction potentialsattheindicatedfrequency.Opticalsamplerateis500Hzfor BeRST1and40HzforGCaMP6s. JournaloftheAmericanChemicalSociety ArticleDOI:10.1021/jacs.5b06644 J.Am.Chem.Soc. 2015,137,10767 !1077610771ofCa 2+sensorsareoneofthemostwidelyusedfunctional probesforsystemsandcellularneuroscience.However,because theexcitationandemissionspectrumofGCaMPliessquarely withintheregimeofpreviousgenerationsofVoltageFluors 13,23(aswellasotherfunctionalprobes),thishasprecludedtheuse ofGCaMPswithPeT-basedvoltagesensitivedyes.Neurons expressingthegeneticallyencodedCa 2+sensor,GCaMP6s, 29whichprovidesthelargest !uorescenceresponsetoCa 2+,were stainedwithBeRST1( Figure5 aandb).GCaMP6s-positive neurons( Figure5 b)displayedbrightgreencytosolic !uorescence,whilemembraneswereclearlylabeledwith BeRST1( Figure5 a).Weevokedactionpotentialsvia "eldstimulationandobtainedopticalrecordings, "rstinthegreen channelforGCaMP6sandtheninthefar-redtoNIRchannel forBeRST1,fromasinglecellexpressingGCaMP6sand stainedwithBeRST1( Figure5 aandb).Insequentialoptical traces,boththerisein Vmemandrisein[Ca 2+]iareclearly visible( Figure5 c).DualBeRST1andGCaMP6simaging enablesdiscriminationofvoltageandCa 2+transients,withthe voltagetransient(approximately15% !F/F)precedingthe measuredGCaMP6ssignal(approximately5% !F/F)fora singleevokedactionpotential( Figure5 c).Oneofthe advantagesofvoltageimagingistheabilitytodeconvolvefast spikingthatCa 2+imaging,withitsslowerintrinsicsignalsand slowprobeunbindingkinetics,oftencannotresolve.When neuronsexpressingbothGCaMP6sandstainedwithBeRST1 werestimulatedtoevoketrainsofactionpotentialsat5,10,and 20Hz,discrete Vmemresponsescorrespondingtoindividually evokedactionpotentialswerevisibleintheBeRST1channelat allfrequencies( Figure5 d),whileCa 2+responseasmeasuredby GCaMP6swereonlydiscernibleat5Hz,asmeasuredbysmall changesinontherisingedgeofGCaMP6s !uorescence( Figure5e).DuetotheintrinsicresponsespeedofBeRST1,two-color functionalimagingshouldbegeneralizabletomanyothertypes ofCa 2+indicators.Theseexperimentshighlighttheabilityof BeRST1toparticipateinfunctionalimagingexperimentsin whichmultiplesignalsneedtobeinterpretedinthesame system.ThenonoverlappingsignalsemanatingfromBeRST1and GFP-based!uorophoresprovideauniqueopportunityto performtwo-colorvoltageimaging.Weexpressedthe geneticallyencodedvoltagesensor,ASAP1, 19incultured neuronsandstainedtheseneurons,asbefore,withBeRST1 (Figure6 a!d).Inresponsetoanevokedtrainofaction potentials( Figure6 e),ASAP1providesdecreasesinrelative !uorescenceofabout6% ±1.4%(S.E.M.,withasignal-to-noise ratio,SNR,of8:1, n=4cells),whichisconsistentwiththe reportedvalueof !4.8%!F/F.19BeRST1,inthesamecell, underidenticalstimulationparameters( Figure6 e),givesa15% ±2.1%increase(S.E.M.,SNRof41:1, n=4cells),highlighting theutilityofBeRST1formeasurementoffastspikingevents andfortwo-colorvoltageimagingwithcomplementary geneticallyencodedvoltagesensorsbasedonGFP. 18OpticalElectrophysiologywithBeRST1andChR2. Finally,wehopedtointerfaceBeRST1withoptogenetictools likeChannelRhodopsin2(ChR2)inordertodemonstratethe Figure4. SpontaneousvoltageimaginginGFP-labeledcellswithBeRST1.Epi !uorescenceimagesofrathippocampalneuronsexpressing(a)GFP and(b)stainedwithBeRST1.Scalebaris20 !m.(c)Opticaltracesofspontaneousactivityinneuronsfrompanels(a)and(b).Numbersnextto tracescorrespondtoindicatedcellsinpanel(b).Opticalsamplingrateis500Hz. Figure5. DualvoltageandCa 2+imagingwithBeRST1and GCaMP6s.Epi !uorescenceimagesofarathippocampalneuron stained(a)withBeRST1andexpressing(b)GCaMP6s.Scalebaris 20!m.(c)SequentialCa 2+andvoltageimagingduring "eld stimulationoftheneuroninpanels(a)and(b).Voltage-(upper) andCa 2+-(lower)inducedchangesin !uorescence.(d)Voltage-(left) and(e)Ca 2+-(right)induced !uorescenceresponsetotrainsofaction potentialsattheindicatedfrequency.Opticalsamplerateis500Hzfor BeRST1and40HzforGCaMP6s. JournaloftheAmericanChemicalSociety ArticleDOI:10.1021/jacs.5b06644 J.Am.Chem.Soc. 2015,137,10767 !1077610771NMeMeMeO O3SNMeMeNMeMeSiMeMeBeRST 1 Figure 0-11 Spontaneous voltage and epifluorescence imaging with BeRST 1 in GFP -lab eled cells of rat hippocampal neurons. (a) GFP and (b) stained with BeRST 1. Scale bar is 20 ! m. (c) Optical traces of spontaneous activity in neurons from panels (a) and (b). Number s next to traces correspond to indicated cells in panel (b). Optical sampl ing rate is 500 Hz. Images and plot copied from ref [131]. 22 nucleic acids by ChenÕs group. 133 The scope of this method was further extended by the combination of native chemical ligation (NCL) and solid -phase peptide synthesis (SPPS), which enabled wider choices of bio-orthogonal immobilized tension probes and extended imaging timescale ( ~ 3 days) to give quantified optical tension measurement (Figure I-12b).134 I.2.2.2!Emerging new imaging agents and their applications Brightness is the soul of fluorescence imaging agents. The widely -used FPs and small molecular imaging probes, albeit with high fluorescence quantum yields (QY, */01 a) b) Figure 0-12 Cell surface tension imaging. (a) Schematic of the FRET -quenching mechanism of tension sensor function , copied from ref [132]. (b) Reflection interference contrast microscopy (RICM) and fluorescence overlay, and quantified heat map of tension for cells cultured on the cRGDfK (TAMRA -QSY9) and SHAVSS (Alexa 488 -TAMRA) peptide tension probe sur faces. Scale bar: 20 ! m. Images copied from ref [134]. 23 are normally limited by their extinction coefficients ( )), which determine the brightness as the product with QY (brightness = */0,2,)). In view of this, fluorescent polymer nanoparticles (NPs) have recently emerged to meet the demand of superior brightness. 135 Fluorescent polymer NPs can encapsulate large quantities of dyes. The brightness of the NPs is then enlarged by the dye -loading number, N (brightness = */0,2,),2,N). They usually can be > 100 -fold brighter than their single molecular parents , and are also brighter than conjugate polymer NPs and inorganic quantum dots (QDs) .136 Because of their NP forms, they present a number of critically different properties. The sizes of dye -loading NPs a re big, which is not ideal in some intracellular site -specific labeling application s, but is more suitable for t issue and whole animal labeling, 137 coating, and sensing. 138 The phot ostability of encapsulated dyes is arguably believed to be increased due to a more rigid polymer matrix that exclude s oxygen and other chemicals relevant to photo-degradation s. Nevertheless, high loading of dye molecules may cause aggregations in ground and/or excited states, leading to undesired spectra l changes and photochemistry. 139 Compared to single dyes, of which the blinking is normally caused by intersystem -crossing to triplet states and electron transfer, dye -loading NPs can average XNNXnH2CCH2CH3H3C1717X = O, n = 1, DiO X = CMe 2, n = 1, Dil X = CMe 2, n = 2, DiD X = CMe 2, n = 3, DiR Figure 0-13 Schematic representation of dye -loading NP design for multicolor imaging via sequential FRET cascade. 24 out these effects and show continuous emission. In fact, aggregation -induced emission (AIE) based polymer NPs have been developed and shown no blinking. These AIE fluorogen (AIEgen) are also used to overcome the aggregation -caused quenching (ACQ), which becomes a major drawback when dye -loading is high. 140 Another advantage of dye -loading NPs is the broad choices of dye s. This renders the flexibility to tune the colors and absorption/emission bandwidths . Wagh et al. prepared a dye -loading NP enclosing four different cyanine dyes that can d eploy a sequential FRET Figure 0-14 Tracking RGB barcoded cancer cells in zebrafish embryo. A) Six batches of D2A1 cells were labeled with fluorescent NPs generating RGB barcodes. Labeled cells were then mixed and injected intravascularly in 2 d postfertiliza tion zebrafish embryos. BÐD) Tumor cells that arrested in the vasculature are imaged, at 3 h post -injection , in the caudal plexus region of the zebrafish embryo. Clusters and individual cells can be found and distinguished based on their color. Images copi ed from ref [142]. 25 cascade. This design enabled tricolor imaging with excitation of a single light source (Figure I-13).141 KlymchenkoÕs group took a different approach to establish a more sophisticated in vivo cell barcoding system for long -term multicellular developmental imaging tracking. 142 In their study, DiO, Dil, and DiD were also selected for dye loading. Other than FRET cascade, three types of NPs were prepared, and each contained only one cyanine dye. Very bulky perfluorinated tetraarylborates were used as c ounterions to prevent ACQ and achieve high dye loading. These NPs were then injected to D2A1 cancer cell line and 6 cell populations were generate d from permuta tion of FRET -pairing for RGB cell ba rcoding . The prelabeled mixture of tumor cells were intravascularly 1106 VOLUME 35 NUMBER 11 NOVEMBER 2017 NATURE BIOTECHNOLOGY ARTICLES map the lymph nodes with high contrast and no need for real-time excitation during imaging. The afterglow luminescence of SPN-NCBS5 was also tested and compared with NIR fluorescence for passively targeted imaging of tumors in living mice. After tail vein injection of SPN-NCBS5, afterglow and NIR fluorescence signals were acquired in real time. Both signals gradually increased over time, but the SBR of afterglow images was higher than that of NIR fluorescence at all time points (Fig. 5d Ðf). Due to the low background of afterglow, the tumor was visible at t = 1 h post-injection and clearly visualized at t = 2 h post-injection for afterglow imaging ( Fig. 5 e); by contrast, the tumor could be visualized only at t = 8 h post-injection with NIR fluores -cence imaging. At t = 2 h, the SBR of afterglow images was 149.7 9.0Ñ23.3-fold higher than the NIR fluorescence images (6.4 0.9) (Fig. 5 f). Both afterglow and fluorescence signals plateaued at t = 36 h post-injection, indicating the efficient accumulation of SPN-NCBS5 in the tumor. The ex vivo data further illustrated that SPN-NCBS5 had the highest uptake in liver followed by tumor, lung, and other major organs ( Supplementary Fig. 27 ). Thus, the afterglow of SPN-NCBS5 permitted faster and higher contrast imaging of tumors in living mice versus NIR fluorescence imaging. Afterglow imaging of drug-induced hepatoxicity Drug-induced hepatotoxicity is a long-standing concern of modern medicine 35. It is one of the most common reasons that the US Food and Drug Administration withholds drug approval 36. Evaluation of potential hepatotoxicity in advance of regulatory approval is challeng -ing because currently studies performed in vitro often have low pre -dictive power 37. Oxidative stress and the consumption of antioxidants in the liver are concurrent early events in hepatotoxicity 38. Biothiols including cysteine (Cys), homocysteine (Hcy) and glutathione (GSH) constitute a major portion of the total body antioxidants that defend against oxidative stress. Thus, real-time in situ imaging of biothiol levels can be a feasible way to evaluate drug-induced hepatotoxicity. To develop activatable afterglow probes for biothiol imaging, we synthesized an amphiphilic oligomer conjugated with an electron- withdrawing quencher (2,4-dinitrophenylsulfonyl, DNBS) and co- precipitated it with NCBS and MEHPPV ( Fig. 6 a). The resulting activatable nanoprobe (SPN-thiol) had a similar size and morphology as other SPNs ( Supplementary Fig. 28 ). Due to the efficient electron transfer from the core to the quencher, the afterglow of the SPN-thiol was substantially quenched at its initial afterglow ÒoffÓ state ( Fig. 6 a). However, in the presence of biothiols including GSH, Cys and Hcy, MEHPPV +R =OSiC6H13C6H13C6H13Pre-irradiationAfterglowLowPre-irradiation:00.80%5%Flu. (!1010 p/s/cm2/sr)Afterglow (!106 p/s/cm2/sr)Afterglow (!107 p/s/cm2/sr)Flu. (!1011 p/s/cm2/sr)0.60.40.20.0550600650700750800 5500.00.00.00.51.01.52.02.53.0012.5510 Afterglow808 nm514 nm1.02.03.04.05.06.07.00.20.40.60.81.0600650700750800 Wavelength (nm)Wavelength (nm)12.5 510012.5 5NCBS (w/w%)Doping amount (w/w%)101.02.03.04.0Em. 580 nmEm. 780 nmAfterglow (!105 p/s/cm2/sr)HighNanoprecipitationAmpli!edafterglow SPNNIR laserirradiationO21O2NIR renewable and ampli!ed afterglow at 780 nmMEHPPV580 nmhvhvEnergy transfer780 nmIncreased generation of 1O2PEG-b-PPG-b-PEGNCBS808 nm514 nmNCBS514 nm808 nmNCBSabdefc0%5%Flu.Figure 3 1O2-sensitizer-amplified NIR afterglow. ( a) Schematic illustration of the proposed mechanism for 1O2-sensitizer-amplified NIR afterglow. ( b) Schematic illustration of SPN-NCBS pre-irradiated by an 808-nm laser for afterglow enhancement versus a 514-nm laser. ( c) Afterglow luminescence images of 12.5 g/mL SPN-NCBS (based on the mass of MEHPPV) pre-irradiated at 514 (left) or 808 nm (right). The nanoparticle solutions were pre- irradiated by 808- or 514-nm laser (1 W/cm 2) for 1 min, and then the afterglow images were acquired under bioluminescence mode with an acquisition time of 30 s after removal of the laser source. ( d,e) Fluorescence ( d) and NIR-induced afterglow luminescence spectra ( e) of SPN-MEHPPV and SPN-NCBS5 in 1 ! PBS buffer (pH = 7.4). ( f) Quantification of the absolute fluorescence and afterglow luminescence intensities of SPN-MEHPPV at different doping amounts of NCBS. The error bars represent the s.d. ( n = 3). Figure 0-15 1O2-sensitizer -amplified NIR afterglow. (a) Schematic illustration of the proposed mechanism. (b) Schematic illustration of SPN -NCBS pre -irradiated by an 808 -nm laser for afterglow enhancement versus a 514 -nm laser. (c) Afterglow l uminescence images of SPN-NCBS pre -irradiated at 514 (left) or 808 nm (right). The NP solutions were pre -irradiated by 808 - or 514 -nm laser for 1 min, and then the afterglow images were acquir ed under bioluminescence mode with an acquisition time of 30 s after removal of the laser source. Images copied from ref [144]. 26 injected into zebrafish embryos , demonstrat ing a powerful tool for tracking cell migration (Figure I-14). This system was further used for tracking embryogenesis. An interesting application of NP -based fluoresc ence is afterglow imaging. Different from conventional fluorescence imaging, thermal stimulation of the energy storage units in afterglow NPs causes slow release of photons, and results in luminesce nce after the end of light excitation .143 As shown in Figure I-15, PuÕs group used a substituted phenylenevinylene ( MEHPPV)-based semiconduct or NP ( SNP ) doped with a 1O2 sensitizer, silicon 2,3 -naphthalocyanine bis(trihexylsilyloxide) (NCBS), to demonstrate prolonged NIR -afterglow. An 808 -nm laser irradiation sensitizes NCBS to release singlet oxygen, followed by vinyl trapping to form a PPV -dioxetane intermediate. Thermal stimulation breaks the dioxetane and releases 580 -nm photons, of which the energy is transferred back to NCBS and emits 780 -nm NIR light. This technique has been applied to in vivo afterglow imaging of lymph nodes and tumor in mice. 144 In consideration of different forms of imaging agents, the essential core is mostly an organic fluorophore (while inorganic QDs, NPs, and Lanthanides are beyond the scope of this discussion), no matter what form of the carrier is (i.e., pure small molecules, proteins, polymers). In the past decade, considerable effort has expand ed the scope to include nucleic acids, including molecular beacons (MBs ) and RNA aptamers. 145 The most notable fluorescence RNA aptamer is Spinach developed in JaffreyÕs lab. 146 Spinach is an RNA mimic of GFP. It binds to a p-HBI d erivative, 3,5 -difluoro -4-hydroxybenzylidene imidazolinone (DFHBI) and fluoresce with high quantum yield (0.72). Being smaller, Spinach has found a range of applications similar to , but wider than FPs. The innate 27 aptamer feature allows it to interact with nucleic acids, ions, and small molecules as well (Figure I-16a).147-151 Slightly different, MBs are single -stranded DNAs that consist of a stem -loop structure doubly labeled with a fluorophore and a quencher group on each end , which remain in a dark state (Figure I-16b). Target binding induces conformational changes and opens up the hairpin folding, resulting in fluorescence turn -on. MBs are more flexible for rational engineering on different parts , and have many uses including real -time PCR .152 Notably, nucleic acids also found tremendous usage in fluorescence in situ hybridization (FISH) .153 It is also noteworthy to mention that, there is a growing interest in fluorescent antibody labeling, that achieves site -specific labeling based on the highly specific antigen-antibody recognition. 154 Different approaches have been taken t o reduce the size of labeled fluorochromes . Examples include : single -chain variable fragment (scFv) -fused FPs, or fluorobodies ;155 and FabLEM (F ab live endogenous modification), which uses monovalent antigen -binding fragment (F ab) domain to conjugate fluorochromes .156, 157 scFvs have been further developed into fluorogen activating proteins (FAPs) via direct evolution, which bind to a nonfluorescent dye and increase fluorescence. a b Figure 0-16 Schematic representations of nucleic acids -based imaging probes. (a) Spinach sensor, comprising Spinach (black), a transducer (orange), and a target -binding aptamer (blue). Target (purple) binding to the aptamer promotes stabilization of the transducer stem, enabling Spinach to fold and activate DFHBI (green) fluorescence . (b) General structure of molecular beacons. 28 I.2.2.3!Multimodality fluorescence i maging and intra operative imaging guided cancer surgery Fluorescen ce molecular imaging (FMI) is expanding its new roles from microscopic imaging into mesoscopic and macroscopic imaging. Typical imaging depth of conventional microscopy is usually about 10 !m. Because biological samples are not completely transparent, sca ttering of photos happens and limits the microscopic studies of tissues and whole -body animals. The imaging depth is determined by a physical parameter called mean free path (MFP) of a photon, which is the average distance a photon travel s between two cons ecutive scattering events. 158 Confocal and two - or multi photon microscopic techniques , by confining the average angle of scattered photons, have been developed to reach longer transport mean free path (TMFP) and further imaging depth (~ 1 mm) with different temporal resolutions .159 Because of the weaker scattering in NIR region than in the visible and U V regions, r ecent research efforts have been put to develop NIR emitters (see Section I.4). Nevertheless, compared to other optic imaging technique s (e.g., SPECT, PET, CT, MRI) , the reachable distance of FMI is still limited. Hybrid fluorescenc e molecular tomography (hFMT) has been developed to image the fluorescence distribution in tissues and small animals. It is a diffuse optical tomography that uses multi -angle optical illumination to collect fluorescence and excitation photo ns from different projections. hFMT allows three -dimen sional and quantitative imaging with depth up to 1 cm. 160 Another trend is to develop imaging agents that couple multiple optical imaging systems .161-163 For example, ChenÕs group has 29 designed an iron oxide NP doped with Cy5.5 and 64Cu-DOTA to realize PET/MR/NIRF triple -modality imaging of U87MG xenograft mouse tumor model. 164 A new cutting -edge field of multimodal fluorescence imaging application s is intraoperative fluorescence -guided surgery (FGS) . The molecular -level of resolution of FMI is superior to all existing portable optical imaging systems .159 It provides FMI as an ideal tool for indicating the remaining tumor cells in the surgical margin (the surrounding areas of the resection site) during a tumor rem oval in clinical operations. This is utterly important for the patientÕs survival and prognosis , given that a follow -up salvage surgery or adjuvant chemo -/radio -therapy usually cannot mitigate the tumor recurrence. 165 This fluorescence guide for surgery is termed by Tsien as Òlive molecular navigationÓ. 166 Currently there are only two synthetic dyes approved by both FDA and EMA for FGS: indocyanine green (ICG) and fluorescein sodium. Methylene blue (MB) is approved by FDA but not EMA. 5-Aminolevulinic acid (5 -ALA) is the precursor of the fluorescent protoporphyrin IX in certain tumor cells and is also approved by both. 167 A growing number of studies have focus ed on NIR-FGS to identify new fluoresce nt reagents , since NIR light is inherently safe in terms of laser illumination power and is invisible to human eyes , and thus does not alter the appearance of the surgical area. 167-174 A few FGS instrumentation s are alre ady approved by regulation authorities or under clinical trials. Interested readers are directed to the excellent perspective by Nguyen and Tsien. 166 It is important to note , new agents that incorporate more sophisticated functionalities are emerging in the rapid growing area of multimodality fluorescence bioimaging, such as photosensitizing -based chromophore/fluorophore -assisted light 30 inactivation (CALI/FALI) and photodynamic /photothermal therapy (PDT /PTT ),175-178 PPI-based optogenetics, 179-182 and photoacoustic microscopy/ tomography (PAM/PAT) .183-185 Nevertheless, the detailed discussion is beyond the scope of this dissertation. I.3!Site -specific labeling methods for in vivo fluorescence imaging It is pivotal to label fluorescent probes to a specific cohort of targets of interest on a molecular level, to enable the imaging, sensing, and more sophisticated functions. Inarguably, the most commonly used site -specific labeling method is based on genetically addressable FPs. With recent advances in genome engineering (e.g., ZFN, CRISPR and TALEN) , FP fusions can be expressed under the target proteinÕs native promoter. 186, 187 However, facilitated by combinatorial synthesis and modern synthetic methodologies, synthetic dyes are much more structurally diverse .188, 189 They have shown greater flexibility in modulating absorption/emission wavelengths, extinction coefficients, fluorescence quantum yields, photostability, and other physical characteristics such as fluorescence lifetime, singlet -triplet interconversion , to achieve more profound application s.190-194 Thus, various posttranslational site -specific labeling methodologies have been developed to accommodate the exogenous dyes for in vivo imaging. 195-198 I.3.1!Covalent enzymatic labeling The idea to label POIs by enzymatic reactions is spu rred by the fact that the cell utilizes protein posttranslational modification to diversify the structures and functions of biomolecules. 199 In this context, a number of posttranslational modification enzymes have been developed for protein labeling. These include transglutaminase (TGase) ,200-203 sortase, 204-207 farnesyltransferase, 208, 209 biotin ligase (BL) ,210-212 N-myristoyl transferase 31 (NMT) ,213-215 formylglycine -generating enzyme (FGE) ,216-218 phosphopantetheinyl transferase (PPTase) ,219, 220 to name a few. However, these approaches generally require over -expression of non -bioorth ogonal enzymes and sequential two -substrate enzymatic reactions, limiting their labeling specificity and kinetics. Bioorthogonal self -labeling enzymes have been developed in this context. They are engineered to stop the enzymatic cleavage of substrate before releasing, thus increasing the enzymatic kinetics and labeling specificity. Representative self -labeling enzymatic tags are SNAP/CLIP -tag, and HaloTag. SNAP -tag is engineered human O 6-alkylguanine -DNA alkyl transferase (hAGT) . hGAT is a 21 kDa DNA repair protein that catalyzes the irreversible alkyl group transfer of O 6-alkylguanine - or O 6-benzylguanine -DNA to its active site Cys residue .221 hGAT mutant s have been screened from direct evolution and yielded SNAP, with a 50 -fold enhanced enzymatic activity over O6-benzylguanine (BG). 222 This feature has been exploited for site -specific labeling of POI by transferring dye -functionalized BG derivative NNNNHNH2OSFluorophore SNAP POI HNNHNNHNH2OSSNAP POI Fluorophore NNNH2OSFluorophore HHNNNH2OSFluorophore CLIP POI CLIP POI O6-benzylguanine (BG) O2-benzylcytosine (BC) Figure 0-17 Schematic labeling reactions of SNAP/CLIP -tags. 32 to its genetic SNAP fusion. Slightly different, CLIP -tag is also engineered from hAGT via a combination of phage display and yeast surface display. Substrate specificity of SNAP -tag and CLIP -tag is orthogonal (Figure I-17). SNAP -tag only reacts with BG while CLIP -tag specifically reacts with O2-benzylcytosine (BC). This orthogonality has been shown useful in simultaneous protein labeling and two -color im aging.223 SNAP /CLIP-tags have been become an indispensa ble tool in the field of bioimaging and have realized super -resolution STORM imaging. 224, 225 HaloTag is engineered from bacterial haloalkane dehalogenase (DhaA) , and is a suicida l covalent ligand -binding protein similar to SNAP/CLIP -tags. As shown in Scheme I-3, the terminal chloride is removed by nucleophilic displacement with Asp106 of DhaA , and followed by an etherificat ion between the hydrocarbon chain and the hydroxyl that form s a covalent intermediate. In the wild -type DhaA, the ester is further hydrolyzed by the base catalysis promoted by His272. A Phe272 mutation stops the hydrolysis an d traps the intermediate as a stable covalent adduct .226 HaloTag also has fast labeling kinetics and can be used simultaneously with SNAP/CLIP -tags. 227 NTrp 107 HClRNHAsn 41 OOAsp 106 Wt DhaA H272FNTrp 107 HClNHAsn 41 OOAsp 106 RHOHPhe 272 OOGlu 130HaloTag NTrp 107 HClNHAsn 41 OOAsp 106 RHOHPhe 272 OOGlu 130NNHScheme 0-3 Mechanistic representation of HaloTag, showing H272F mutation to trap fluorophore -labeled intermediate as a stable covalent adduct. R represents the labeling fluorophore. 33 I.3.2!Noncovalent affinity labeling There is no clear boundary between covalently self -modified enzymatic tags and tags that are based on ligand -protein affinity. The ligands for the latter system can be designed to have covalent bond formation with a residue in proximity. Nevertheless, for the latter type, the affinity -based ligand -protein interactions remain as the major driving force for the sit e-specific recognition. As an example of proximity -induced affinity tag , Cornis h and co -workers have developed TMP -tag based on the nanomolar noncovalent binding affinity of trimethoprim (TMP) with E. coli dihydrofolate reductase ( eDHFR).228 DHFR was fused to POI and labeled with a cell -permeable TMP heterodimer, which consists of TMP, a fluorophore, and a covalently linked acrylamide group. Upon binding, TMP brought the acrylamid e close to an active c ysteine residue just outside the pocket . A sequential Michael addition of the cysteine with this 34--system covalently anchored Particularly,wefoundthequenchedligand-directedtosyl(Q- LDT)chemistrytobepromisingbecauseithasbeenreportedto haverapidproximity-induced invitro reactivitywithminimal nonspeci!chydrolysis. 27,28Furthermore,weanticipatedwe Figure2. FluorogenicTMP-tag.(a)Cartoonofthe "uorogenicTMP-tag,whichcentersatrimericTMP !quencher(Q) !"uorophore(F)moleculeto becleavedinaproximity-inducedS N2reaction.Speci !cally,weselectedtheCyssidechainasthenucleophileandatosylatelinkerastheelectrophile. WhenTMPbindstoeDHFR,theuniqueCysnearbindingpocketwouldreplacethetosylateleavinggroupattachedtothequencher,andthustheprobe isswitchedon.(b)StructureofTMP-BHQ1-Atto520(TMP-Q-Atto520),the !rst"uorogenicTMP-tag. Scheme1.Retro-syntheticDesignofTMP-Q-Atto520 ACSChemicalBiology Articlesdx.doi.org/10.1021/cb300657r|ACSChem.Biol. 2013,8,1704 !17121706a NNNH2H2NOMe OMe ONHONNNOSOONHONNHOONNNNHNHTMP BHQ1 Atto520 MeO NNO2Figure 0-18 Fluorogenic TMP -tag. (a) Cartoon of the trimeric TMP -quencher -fluorophore to be cleaved in a proximity -induced S N2 reaction. (b) Structure of TMP -BHQ1 -Atto520. b 34 the fluorophore to eDHFR. The labeling kinetics was increased to an 8-min half -time by rationally engineering the optimal distance between the acrylamide and the active Cys .229 FRET-quenching was further introduced to the TMP ligand and made the system fluorogenic (Figure I-18).230-232 SNAP/CLIP -, Halo -, and TMP -tags are able to incorporate a variety of fluorescent dyes. Nonetheless, they normally require extensive wash steps to remove excess free fluorochromes to distinguish the bound and unbound probes. A more r ecent trend is to develop fluorogenic probes that remain in a dark state until site -specific binding happen s. As a generic methodology , a quencher moiety can be conjugated to the fluorescent probe and quench the fluorescence either through FRET or PeT. The se quencher groups can be removed by nucleophilic substitution coupled with the covalent labeling in a similar pattern to the third -generation of TMP -tag described before. 233 A more thorough discussion is presented in Section I.20. Kikuchi and co -workers have developed photoactive yellow protein ( PYP)-tag based on this principle. PYP is a 14 -kDa, water -soluble protein derived from purple bacteria , which does not exist in animal body . It naturally binds to the CoA thioester of 4 -hydroxycinnamic acid through transthioesterification with Cys69. Impo rtantly, this natural ligand can be modified to a structural analog, 4 -hydroxycoumarin , with the high binding affinity. A cova lently-linked fluorophore to t he 4 -hydroxycoumarin derivative can associate with it under polar environment and have FRET -quenching. The fluorophore emission will be turned on after the transthioesterification that breaks the association . Alternatively, the PYP ligand can simultaneously be linked to a quencher and an emitter 35 to realize the same function. 233, 234 A more fluorescent 4 -dimethylamino coumarin has been developed as a dual -function ligand for both binding and fluorescence, so that the coumarin fluorescence can be turned on directly after tra nsthioesterification, negating the need to couple an exogenous fluorophore. 235 pKa modulation of the thioester leaving group together with t he charge optimization surrounding the binding cavity has further improved the transthioesterification kinetics (Figure I-19).236 The PYP ligand has also been modified to accommodate more enzymatic substrate to demonstrate improved fluorogenicity .237 It is noteworthy to mention that , recently Gautier and co -workers have combined direct evolution using yeast display and fluorescence -activated cell sorting (FACS) to engineer PYP into a Yellow Fluorescence -Activating and absorption -Shifting Tag (Y -FAST). Instead of 4 -hydroxycinnamic acid or coumarin, Y-FAST binds to 4 -hydroxybenzylidene -rhodanine (HBR) or 4 -hydroxy -3-methylbenzylidene -rhodanine (HMBR) in a reversible manner. Y -FAST distinguishes from PYP in two ways after ligand Figure 0-19 Design of the PYP -tag mutant PYP3R and its fluorogenic probe, with a focus on electrostatic interactions and the pK a value of the leaving group. Copied from ref [236]. 36 binding: 1) a significant increase in QY of H(M)BR; 2) a large red -shift in a bsorption due to phenol deprotonation (Figure I-20a).238 This negates the need of covalently coupled quencher or fluo rophore moieties of the ligands. The residence time of HBR and HMBR are 60 ms and 160 ms, respectively. This enables the Y -FAST labeling as instantaneous, highly dynamic , and fully reversible . Shown in Figure I-20b, a rapid labeling on -off cycle was demonstrated by exchanging ligand -free medium of the cells labeled with Y -FAST . A follow-up study reports the second generation of Y -FAST -tag that extends the emission into far-red/NIR region by elongating the (-conjugation of HBR analogues. 239 Adopting similar strategies, a number of protein a ffinity-based labeling tags have been engineered, f or instance, TEM -1, a class A --lactamase , and CCD, a cephalosporin -based probe. 240 It is worth noting that the scope of protein hosts is not limited by substrate -(38Ð41):Afirsttargetislabeledwithafirststainandimaged;the labelisremovedbyphysicalorchemicalmeans,afterwhicha secondorthogonalstaincanbeappliedtolabelasecondtarget, andsoon.Eventhoughthesemultiplexingapproachesopen veryinterestingprospectsforimagingseveraltensoftargets inasinglecell,theyremainlimitedtothestudyoffixed,per- meabilizedcellsinwhichtargetsarelabeledwitholigonucleo- tide-orantibody-basedprobes.Acollectionofsystemssuchas Y-FASTcouldextendthisstrategytolivecells,particularlywhen combinedwithrecentlydevelopedtargetedgenomeediting techniques(e.g.,CRISPR-Cas9system)(42). Thepossibilitytocontrolthefluorescenceondemandshould alsofacilitatetheimplementationofFRETmeasurements.Esti- mateofFRETsignalrequiresextensivecontrolstodeterminethe extentofcrosstalkbetweendonorandacceptor(43).Theabilityto performmultipleexperimentsonthesamesampleintheabsence orpresenceofthefluorogen(andthereforewithandwithoutthe contributionofY-FAST)couldimproveFRETimagingprotocols. Y-FASTcouldplaytheacceptorinapairwithCFPorthedonorin apairwithmCherry(see SIAppendix ,TextS5 andFig.S19 forFRETcharacterizationbetweenY-FASTandmCherry).Theuse ofY-FASTasacceptorcould,inparticular,permiteasyde- terminationofFRETefficiencybymeasuringthequenchingofthe donorfluorescenceuponadditionofthefluorogen,or,conversely, thedecreaseinfluorescencequenchingbyrapidwashingofthe fluorogen,thuscompetingwithdonorrecoveryafteracceptor photobleachingtechniques,butwiththeadditionaladvantageof reversibility.ThesmallsizeofY-FASTisalsoanadvantagefor FRET,asitenhancestheenergytransferefficiencybyenablinga priorishorterFırsterdistancesthanGFP-likefluorescentproteins. FRETdetectioncouldfurtherbenefitfromtheabilitytocontrol thelabelingdensityofY-FASTindependentlyofitsexpression leveltosetthedonor:acceptorstoichiometrywithintherangeof 1:10Ð10:1toensuredetectableFRETsignals. Finally,thefastexchangedynamicsofY-FASTcouldbead- vantageouslyexploitedforsuperresolutionimaginginlivecells.As Y-FASTinterconvertsspontaneouslyandrapidlyatthesingle- moleculelevelbetweenadark(unbound)stateandabright (bound)state,itshouldbehaveasablinkingfluorophore(44). Fine-tuningoftheexchangedynamicscouldgiveaccesstoblinking ratesadequateforSingle-moleculeLocalizationMicroscopies(41, 45,46)orSuperresolutionOpticalFluctuationImaging(44,47). Inconclusion,thestrategydevelopedinthisworkisgenericand mayopennewroutesforthedesignofsmartprobesandbiosensors. Inparticular,HMBRbelongstoaseriesofconjugateddonor !acceptorcompoundsexhibitingvario usphotophysical/photochemical behaviors(48)thatcouldfacilitatethedesignofacollectionof FASTscoveringthewholevisiblespectrumforvariousapplicationsin multiplexedbioimagingandbiosensing. MaterialsandMethods MammalianCellCulture. HEK293andHeLacellswereculturedinDMEM supplementedwithphenolred,GlutamaxI,10%(vol/vol)fetalcalfserum (FCS),and1%penicillin !streptomycinat37¡Cwithina5%CO 2atmosphere. Formicroscopicimaging,cellswereseededin !DishIBIDI(Biovalley)coated withpoly- L-lysine.CellsweretransientlytransfectedusingGenejuice(Merck) accordingtothemanufacturer Õsprotocol.Beforeimaging,cellswere washedwithPBSandincubatedinDMEM(withoutphenolred)com- plementedwithHBRorHMBRattheindicatedconcentration. NeuronCultures. Culturesofdissociatedspinalcordneuronswereprepared fromSprague !Dawleyrats(atembryonicday14)asdescribedpreviously (49).NeuronsweremaintainedinneurobasalmediumcontainingB27,2mM glutamax,5U/mLpenicillin,and5 !g/mLstreptomycinat36¡Cand 5%CO 2,cotransfectedatdayinvitro(DIV)15withY-FAST !Gephyrinand CEABDFig.3. On/offfluorescenceswitchingbyiterativelabeling/unlabeling.( AandB)HeLacellsexpressingmCherry-Y-FASTweregrowninamicrofluidicchannel andrepeatedlyincubatedwithHMBR-containingculturemediumfor20sandHMBR-freeculturemediumfor40s.Amultifunctionalfluidiccontrolleren -abledseveralcyclesoflabeling/unlabeling.HMBRconcentrationwas5 !M.( A)Confocaltimelapseshowingtwocyclesoflabeling/unlabeling(Ex/Em488/493 Ð575nm). MovieS2 shows10cyclesoflabeling/unlabeling.( B)Temporalevolutionofthecellfluorescenceuponaddition( +)andremoval( Ð)ofHMBR. (C)ConfocaltimelapseshowingthelabelingkineticsinazebrafishembryoexpressingY-FASTandmCherry(HMBRchannel:Ex/Em491/525 Ð539nm;mCherry channel:Ex/Em561/605 Ð664nm).HMBRconcentrationwas10 !M.Seealso MovieS3 .(D)AzebrafishembryoexpressingY-FASTandmCherrywasimaged beforeadditionofHMBR( ÐHMBR),20minafterincubationwith10 !MHMBR( +HMBR),aftertwowashingsof20min(Wash1and2),andafterreincubation with10 !MHMBR( +HMBR).( E)ConfocalmicrographsofliveHeLacellsexpressingDronpa !NLS(nucleus)andlyn11 !Y-FAST(membrane)showingsequential imagingofnuclearDronpaandmembrane-anchoredY-FASTthroughsequentialon/offlabelingofY-FASTintercalatedwithon/offphotoswitchingofDr onpa (Ex/Em488/493 Ð797nm).HMBRconcentrationwas5 !M.(Scalebars,10 !m.)Plamontetal. PNAS|January19,2016 |vol.113 |no.3 |501CHEMISTRY CELLBIOLOGY andinmulticellularorganisms.Y-FASTisanengineeredvariant ofthemonomeric14-kDaPhotoactiveYellowProtein(PYP)[a blue-lightphotoreceptorfrom Halorhodospirahalophila (23Ð25)]thatweevolvedtoreversiblybind4 -hydroxybenzylidene-rhodanine(HBR)or4-hydroxy-3-methylbenzylidene-rhodanine(HMBR), twofluorogensidentifiedinthecourseofthisstudy(Fig.1 A).HBRandHMBRarenonfluorescentbythemselves,butthey fluoresceyellowlightuponblue-lightexcitationwhenbound toY-FAST.Y-FASTdistinguishesitselffromexistinglabeling systemsbecausethebindingisnotonlyspecificandinstanta- neousbutalsohighlydynamicandfullyreversible.Fluorescence canthusberapidlyswitchedonandoffsimplybyadditionor withdrawalofthefluorogen,providinganadditionaldegree ofcontrol.Thefastbindingdynamicsmightmoreoverdecrease theapparentphotobleachingratebycontinuousrenewalof thefluorogen,assuggestedinpreviousreports(26).Designing afluorogen-basedreportercharacterizedbyareversiblebinding thatisbothspecificandhighlydynamicwas,however,chal- lenging,asthehighoff-ratenecessaryforafastexchangedynam- icstendstodecreasetheaffinityrequiredforhighselectivity. Thus,tomaintainhighselectivity,wereliedontwospectroscopic changesforfluorogenactivation:first,bindingofH(M)BR toY-FASTresultsinasignificantincreaseoffluorescence quantumyield,and,second,itinduceslargeabsorptionredshift. BecauseY-FASTistheonlyspeciespromotingthesetwospectro- scopicchanges,freeornonspeci ficallyboundfluorogendoesnot contributetothefluorescencesignal,ensuringhighimagingcontrast. ResultsFluorogenDesign. HBRiseasilyobtainedinonestepbyin-water condensationoftherhodaninetotheparahydroxybenzaldehyde. Itiscomposedofanelectron-donatingphenolconjugatedto anelectron-withdrawingrhodanine(Fig.1 A).Thispush !pullstructure isanalogoustotheGFPchromophore4-hydroxybenzylidene-5- imidazolinone(5),knowntodeexcitenonradiativelyinsolution buttorelaxtothegroundstateradiativelyintherigidbarrelof GFP(27).HBRdrewourattentionasputativefluorogenforthe designofY-FASTforseveralreasons.First,HBRisalmostfully protonatedatphysiologicalpH(itsp KAis8.4 ±0.1),andit undergoesa50-nmabsorptionredshiftupondeprotonationasa resultofthestrongerelectrondonationofthephenolate(Table 1and SIAppendix ,Fig.S1 AandB).Wethereforeanticipated thataproteintagstabilizingdeprotonatedHBRwouldexhibita red-shiftedabsorptionwithres pecttofreeHBRinpH7.4solu- tions,enablingdiscriminationofthefreeandboundstatesby theirabsorptionproperties.Sec ondly,HBRfluorescenceishighly environment-sensitive.Inwater,theprotonatedanddeproto- natedstatesofHBRemitat470n mand545nmwithfluorescence quantumyieldsof0.02%and0.3%(Table1and SIAppendix ,Fig. S1C),whereas,inviscoussolutions(containing40%glycerol),they exhibitsixfoldandthreefoldhigherbrightness,respectively.Taken togetherthesespectroscopicpr opertiesallowedustoanticipate thatbindingofHBRtoawell-designedproteintagcouldprovidea uniquefluorogeniceffectbasedo ntwospectroscopicchanges:an absorptionredshiftthroughbinding-induceddeprotonationanda fluorescencequantumyieldincreaseviafluorogenimmobilization. Y-FASTIsaVariantofPYPEngineeredbyDirectedEvolution. PYP waschosenasscaffoldforthedesignofY-FASTforseveralrea- sons.First,itsparahydroxycinnamoyl(HC)chromophore ÑcovalentlytetheredtoCys69andresponsibleforitsblue-light photosensingproperties(23 Ð25)Ñsharesstructuralfeatureswith HBR,suggestingthatthebindingsiteofPYPcouldbeengineered tobindHBRselectivelyandreversibly.Moreover,thebinding pocketofPYPaccommodatesHCinitsphenolatedeprotonated form(25),providingaplatformfordesigningvariantsabletosta- bilizedeprotonatedHBRandthusobtainabsorptionredshift uponbinding.Finally,wild-typePYPhasaprovenabilityasa recombinantproteintag(16,28)andisasmallprotein(14kDa) comparedwithGFP-likefl uorescentproteins(26 !30kDa). ToengineerthebindingcavityofPYP,werandomizedloops andresiduesincloseproximitywiththechromophorepocketby saturationmutagenesis( SIAppendix ,Fig.S2 ).Byscreeningyeast surface-displayedlibrariesbyfluorescence-activatingcellsorting (29)inthepresenceofHBR,wesuccessfullyidentified47clones specificallyactivatingHBRfluorescence( SIAppendix ,Fig.S3 ).Theselectedclonesallbelongedtothelibraryconstructedby randomizingtheloop94-101thatgatestheentranceofthebinding pocket.TheemergenceoftheconsensussequenceWxIPTxxx confirmedconvergenceoftheselectionprocess. PhysicochemicalCharacterization. Themostpromisingvariants wereexpressedin Escherichiacoli andpurifiedbyaffinity ABCFig.1. Y-FASTenablesspecificlabelingoffusionproteinsinlivingcells. (A)Y-FASTbindsthefluorogenicHBRandHMBRandactivatestheirfluores- cence(POI,proteinofinterest).Bindinginducestwospectroscopicchanges:an increaseofthefluorescencequantumyieldandanabsorptionredshift(dueto ionization).( B)ConfocalmicrographsofliveHeLacellsexpressingvarious Y-FASTfusionslabeledwith5 !MHMBR(Ex/Em488/493 Ð797nm).Cytoplasm: Y-FAST;Nucleus:H2B-Y-FAST;Cellmembrane:Lyn11-Y-FAST;Mitochondria: Mito-FAST(Mito =MitochondrialtargetingsequencefromsubunitVIIIof humancytochrome coxidase);Golgi:Golgi-Y-FAST(Golgi =N-terminal81 aminoacidsofthehumanbeta1,4-galactosyltransferase);Microtubules: Ensconsin-Y-FAST.( C)Epifluorescencemicrographsofadendriticsegmentofa spinalcordneuroncotransfectedwithmCerulean !Gephyrinthataccumulates atinhibitorysynapses(Ex/Em427/472 ±15nm; Left )andaY-FAST-tagged Gephyrinconstruct(Ex/Em504/542 ±14nm; Center ).After10sofincubation with10 !MHMBR,thefluorescenceofY-FASTwasdetectedintheyellow emissionrange(Ex/Em504/542 ±14nm; Right ).(Scalebars,10 !m.) 498|www.pnas.org/cgi/doi/10.1073/pnas.1513094113 Plamontetal. (38Ð41):Afirsttargetislabeledwithafirststainandimaged;the labelisremovedbyphysicalorchemicalmeans,afterwhicha secondorthogonalstaincanbeappliedtolabelasecondtarget, andsoon.Eventhoughthesemultiplexingapproachesopen veryinterestingprospectsforimagingseveraltensoftargets inasinglecell,theyremainlimitedtothestudyoffixed,per- meabilizedcellsinwhichtargetsarelabeledwitholigonucleo- tide-orantibody-basedprobes.Acollectionofsystemssuchas Y-FASTcouldextendthisstrategytolivecells,particularlywhen combinedwithrecentlydevelopedtargetedgenomeediting techniques(e.g.,CRISPR-Cas9system)(42). Thepossibilitytocontrolthefluorescenceondemandshould alsofacilitatetheimplementationofFRETmeasurements.Esti- mateofFRETsignalrequiresextensivecontrolstodeterminethe extentofcrosstalkbetweendonorandacceptor(43).Theabilityto performmultipleexperimentsonthesamesampleintheabsence orpresenceofthefluorogen(andthereforewithandwithoutthe contributionofY-FAST)couldimproveFRETimagingprotocols. Y-FASTcouldplaytheacceptorinapairwithCFPorthedonorin apairwithmCherry(see SIAppendix ,TextS5 andFig.S19 forFRETcharacterizationbetweenY-FASTandmCherry).Theuse ofY-FASTasacceptorcould,inparticular,permiteasyde- terminationofFRETefficiencybymeasuringthequenchingofthe donorfluorescenceuponadditionofthefluorogen,or,conversely, thedecreaseinfluorescencequenchingbyrapidwashingofthe fluorogen,thuscompetingwithdonorrecoveryafteracceptor photobleachingtechniques,butwiththeadditionaladvantageof reversibility.ThesmallsizeofY-FASTisalsoanadvantagefor FRET,asitenhancestheenergytransferefficiencybyenablinga priorishorterFırsterdistancesthanGFP-likefluorescentproteins. FRETdetectioncouldfurtherbenefitfromtheabilitytocontrol thelabelingdensityofY-FASTindependentlyofitsexpression leveltosetthedonor:acceptorstoichiometrywithintherangeof 1:10Ð10:1toensuredetectableFRETsignals. Finally,thefastexchangedynamicsofY-FASTcouldbead- vantageouslyexploitedforsuperresolutionimaginginlivecells.As Y-FASTinterconvertsspontaneouslyandrapidlyatthesingle- moleculelevelbetweenadark(unbound)stateandabright (bound)state,itshouldbehaveasablinkingfluorophore(44). Fine-tuningoftheexchangedynamicscouldgiveaccesstoblinking ratesadequateforSingle-moleculeLocalizationMicroscopies(41, 45,46)orSuperresolutionOpticalFluctuationImaging(44,47). Inconclusion,thestrategydevelopedinthisworkisgenericand mayopennewroutesforthedesignofsmartprobesandbiosensors. Inparticular,HMBRbelongstoaseriesofconjugateddonor !acceptorcompoundsexhibitingvario usphotophysical/photochemical behaviors(48)thatcouldfacilitatethedesignofacollectionof FASTscoveringthewholevisiblespectrumforvariousapplicationsin multiplexedbioimagingandbiosensing. MaterialsandMethods MammalianCellCulture. HEK293andHeLacellswereculturedinDMEM supplementedwithphenolred,GlutamaxI,10%(vol/vol)fetalcalfserum (FCS),and1%penicillin !streptomycinat37¡Cwithina5%CO 2atmosphere. Formicroscopicimaging,cellswereseededin !DishIBIDI(Biovalley)coated withpoly- L-lysine.CellsweretransientlytransfectedusingGenejuice(Merck) accordingtothemanufacturer Õsprotocol.Beforeimaging,cellswere washedwithPBSandincubatedinDMEM(withoutphenolred)com- plementedwithHBRorHMBRattheindicatedconcentration. NeuronCultures. Culturesofdissociatedspinalcordneuronswereprepared fromSprague !Dawleyrats(atembryonicday14)asdescribedpreviously (49).NeuronsweremaintainedinneurobasalmediumcontainingB27,2mM glutamax,5U/mLpenicillin,and5 !g/mLstreptomycinat36¡Cand 5%CO 2,cotransfectedatdayinvitro(DIV)15withY-FAST !Gephyrinand CEABDFig.3. On/offfluorescenceswitchingbyiterativelabeling/unlabeling.( AandB)HeLacellsexpressingmCherry-Y-FASTweregrowninamicrofluidicchannel andrepeatedlyincubatedwithHMBR-containingculturemediumfor20sandHMBR-freeculturemediumfor40s.Amultifunctionalfluidiccontrolleren -abledseveralcyclesoflabeling/unlabeling.HMBRconcentrationwas5 !M.( A)Confocaltimelapseshowingtwocyclesoflabeling/unlabeling(Ex/Em488/493 Ð575nm). MovieS2 shows10cyclesoflabeling/unlabeling.( B)Temporalevolutionofthecellfluorescenceuponaddition( +)andremoval( Ð)ofHMBR. (C)ConfocaltimelapseshowingthelabelingkineticsinazebrafishembryoexpressingY-FASTandmCherry(HMBRchannel:Ex/Em491/525 Ð539nm;mCherry channel:Ex/Em561/605 Ð664nm).HMBRconcentrationwas10 !M.Seealso MovieS3 .(D)AzebrafishembryoexpressingY-FASTandmCherrywasimaged beforeadditionofHMBR( ÐHMBR),20minafterincubationwith10 !MHMBR( +HMBR),aftertwowashingsof20min(Wash1and2),andafterreincubation with10 !MHMBR( +HMBR).( E)ConfocalmicrographsofliveHeLacellsexpressingDronpa !NLS(nucleus)andlyn11 !Y-FAST(membrane)showingsequential imagingofnuclearDronpaandmembrane-anchoredY-FASTthroughsequentialon/offlabelingofY-FASTintercalatedwithon/offphotoswitchingofDr onpa (Ex/Em488/493 Ð797nm).HMBRconcentrationwas5 !M.(Scalebars,10 !m.)Plamontetal. PNAS|January19,2016 |vol.113 |no.3 |501CHEMISTRY CELLBIOLOGY a b c Figure 0-20 Y-FAST -tag. (a) Schematic showing the spectral changes during Y -FAST binding: spectral shift and QY increase. (b) Confocal time lapse showing two cycles of labeling/unlabeling (Ex/Em 488/493 -575 nm) of HeLa cells expressing Y -FAST . Cells were r epeatedly incubated with HMBR -containing culture medium for 20 s and HMBR -free culture medium for 40 s. HMBR concentration was 5 !M. (c) Temporal evolution of the cell fluorescence upon addition (+) and removal ( -) of HMBR. Images and plots copied from ref [238]. 37 specific enzymes and photoreceptor (i.e., PYP). As mentioned in Section I.2.2.2, scFv has also been engineered to bind a diverse set of fluorogenic dyes .241, 242 These fluorogen -activating proteins (FAPs) bind their ligand s with nanomolar affinity. Conceptually, t he ligands are designed to be ionic and do not show fluorescence in polar environment. Shown in Figure I-21 are representative e xamples: thiazole orange (TO) and malachite green (MG). The high affinity binding results not only in the increase of QY, but also in the red -shift of spectra and increase of extinction coefficient s.243 With multi -functionalized MG ester and corresponding scFvs, these no -wash FAP -tags have been applied in far-red ima ging, super -resolution imaging, and multicolor imaging. 244-248 I.3.3!Bioortho gonal chemical labeling The site -specific labeling tags described in Section I.3.1 and I.3.2 are all protein -based, albeit indispensable, their relative large size (u p to ~ 200 amino acids) may elicit concerns of disturbing the POIs. In contrast, bioorthogonal chemistry relies on the ÒclickÓ reactions fusing the probe and a specialized functional group on POI, and can decrease the size perturbation. The field of bioort hogonal chemistry has witness ed an explosive development since the term was coined by Bertozzi in 2003. 249, 250 SNNHNO3SONMeMeNMeMeTOMGFigure 0-21 Fluorogenic ligands of FAP: thiazole orange (TO) and malachite green (MG). 38 A number of modification methods were established based on the nucleophilic nature of lysine, cysteine, tyrosine, and tryptophan. Reagent scope encloses aldehyde, ketone, olefin, sulfonamid e, phenol, aniline, diazomethine, etc., via either acidic/basic catalysis, or transition metal catalyzed coupling. 251 Among a variety of N -terminus modification s, a well -recognized method is the native chemical ligation (NCL), which is a rapid irreversible intramolecular S -N acyl transfer on thioesters by N -terminal amine. 252 One early example of click reaction is hydrazide/ oxime -ligation, which utilizes an oxyamine or hydrazine group to condense with an aldehyde or ketone group under either acidic or basic physiological conditions. 253, 254 However, the application is limited by the difficulty of introducing aldehyde groups to POIs. A recent example is a variant of Pictet -Spengler ligation. 255 The oxyiminium formed between the aldehyde and an oxyamine undergoes an intramolecular C -C bond formation with the indole nucleophilic C -,and formed a stable oxacarboline product (Scheme I-4). Being small, kinetically stable under physiological cond itions, and essentially absent from biological systems, the azide group has found many uses in biorthogonal click reactions. A classic example is the Staudinger ligation. Taking the prototype of Staudinger reduction, which starts from the phosphine nucleop hilic attack of the azide , followed by N 2 release to yield an aza -ylide, Staudinger ligation is designed with an ortho -NHOHONRONHMeNHONHOMeNRNHOONNMeRScheme 0-4 Pictet % Spengler ligation of an aldehyde with a tryptamine nucleophile. 39 ester group on the phosphine reagent to trap the reactive ylide and form a stable amide bond through intramolecular cyclization. 256 A ÒtracelessÓ version is also developed by inverting the ester to release the phosphine oxide (Scheme I-5).257 Azide -based Huisgen [3+2] cycloaddition has also been modified to a copper(I) catalyzed variant that proceeds without the requirement of high temperature and pressure. 258 Against the common thought that the cyclization is triggered by copper actylide via a six -membered copper(III) metallacycle, t his copper -catalyzed alkyne -azide cycloaddition (CuAAC) mechanistically proceeds with a stepwise C -N bond formation catalyzed by a dicopper complex while the monomeric copper actylide is unreactive. 259, 260 Recently, CalFluor, a universal m otif utilizing PeT quenching of azide was reported for CuAAC to cover across the visible spectrum. 261 In terms of in vivo labeling, CuAAC is not ideal due to its high cellular toxicity and the low performance of the Cu(I) catalyst. A copper -free [3+2] strain -promoted alkyne -azide cycloaddition (SPAAC) was developed , driven by the ring strain -release of cyclooctyne. 262 Shown in Scheme I-6a, more electronically and/or sterically modified cyclooctyne derivatives were synthesized to optimize the SPAAC kinetics. 263-266 The triple bond in DIBO can be further modified into a cyclop ropenone and unmasked by photo-activation. 267 Another closely related [3+2] cycloaddition is the strain -promoted alkyne -nitrone cycloadditio n (SPANC , Scheme I-6b). The more reactive nitron e is used and N3OPh2POR+- N 2NPh2POROH2ONHPh2PHOOOR+Scheme 0-5 Traceless Staudinger ligation. 40 shows 3 .9 M -1"s-1 second -order rate constant in model reactions , almost 3 -fold faster than the azide counterpart . However, the nitrone introduction to POI is not easy to implement. 268 Strained alkenes have also been applied in strain -promoted [3+2] cycloaddition, such as oxa norbo rnadiene derivatives. They are relatively easy to synthesize, but usually suffer from slow reaction kinetics and low specificity. 269 An intriguing photoinducible [3+2] cycloaddition of alkene is the alkene -tetrazole photoclick cycloaddition (Scheme I-7a). Diarytetrazole can undergo a UV -activated reverse [3+2] reaction to release N 2 and form a nitrile -imine in situ , consequently furnish ing a [3+2] product with labeled exogenous alkenes. 270 Comparatively, this photoclick cycloaddition is fast, and has intrinsic fluorogenicity since tetrazole itself is a n FFRRNROSDIFODIBO BARAC TMTH NNN+SPAAC XXNNNRRNOR1R2R+SPANC NOR1R2Ra b Scheme 0-6 [3+2] cycloadditions. (a) SPAAC and representative cyclooctyenes. (b) SPANC. R i s a labeling fluorophore. 41 efficient PeT quencher. 271 A potential concern is the cross -reacti vity with endogenous alkenes. Recently the I nvers e-Electron -Demand Diels -Alder (IEDDA) cycloaddition has drawn much attention (Scheme I-7b).272, 273 Strained trans -cyclooctene is paired with tetrazene in this IEDDA and is 9 orders of magnitude faster in rate than internal cis -olefins. 274 The rate constant highly depends on the electronics of the dienophile. 275 To date, electron -rich norbornene s, cyclopropenes, bicyclo[6.1.0]nonynes, and even unstrained styrenes have all successfully made use of this labeling method. 276-278 Similar to tetrazole, tetrazene also serves as a PeT quencher for the purpose of fluorophore labeling. Other click reactions are still emerging, e.g., redox -activated chemical tagging (ReACT), 279 photochemical thiol -ene reaction, 209 chemoselective rapid a zo-coupling reaction (CRACR), 280 and the phospha -Michael addition .281 Recently, one-pot triazole click reaction has also been realized on the DNA level. 282 Even Pd -catalyzed C -H NNNNAr1Ar2h!NAr1NAr2R1R2NNAr1Ar2R2R1NNNN[4+2] NNNND-A [4+2] retro D-A - N 2NNNHNH+- N 2strained alkene tetrazine a b Scheme 0-7 (a) Tetrazole -alkene photoclick cycloaddition. (b) IEDDA tetrazine -alkene cycloaddition. 42 activation and Ru -catalyzed olefin metathesis have seen growing use in bioorthogonal labeling. Readers are referred to these extensive review articles .283-285 I.3.4!Labeling approaches with short oligomer tags Despite the versatility of systems to incorporate various fluorochromes into targeting biomolecules, bioorthogonal labeling strategies are often complicated by the introduction of uncanonical amino acids (UAAs) , which bear the synthetic functional groups needed by the chemical transformations. Expressing UAAs on POIs is not trivial. It requires the implementation of artificial tRNAs and c orresponding recombinant enzyme cassette to exploit cell machinery, 286 which can in turn elicit the concern of functional perturbation of the target ed cells. To circumvent t his, several strategies have been developed to use small oligomeric peptides and/or nucleic acids as labeling tags (for example, the RNA aptamer Spinach discussed in Section I.2.2.2). Recently, TingÕs lab has engineered the E. coli enzyme lipoic acid ligase (LplA) capable of covalently ligating 7 -hydroxycoumarin to a 13 -residue LplA acceptor peptide (LAP). 287 This PRIME (PRobe Incorporation Mediated by Enzymes) method has found ©201 3 Nature America, Inc. All rights reserved. PROTOCOL NATURE PROTOCOLS | VOL.8 NO.8 | 2013 | 1621Both PRIME and CuAAC labeling steps are efficient. PRIME ligation at the cell surface with pAz reaches ~80% completion after 20 min using purified W37VLplA1. For chelation-assisted CuAAC performed in vitro , we achieved complete conversion to product in <5 min in the presence of 10Ð40 M copper and a CuI-stabilizing ligand Tris-(hydroxypropyltriazolylmethyl)amine (THPTA) 10, in contrast to a <40% conversion with conventional CuAAC under the same conditions. On cells, chelation-assisted CuAAC increased labeling yields by 2.7- to 25-fold compared with conventional CuAAC 1. We estimate the overall two-step labeling yields for both cell-surface proteins and purified pro -teins to be >70% using 30Ð60-min labeling protocols. As much less cytotoxic copper (10Ð100 M, compared with low-millimolar amounts in traditional CuAAC) can be used to achieve similar or better reaction rates, chelation-assisted CuAAC is inherently less toxic than conventional CuAAC 1.Applications of the method Two-step PRIME-CuAAC labeling has been applied to fluoro -phore tagging of a variety of cell-surface and purified proteins, Figure 1 | Site-specific protein labeling via PRIME and chelation-assisted CuAAC 1. ( a) The two-step labeling scheme. In the first step, an engineered PRIME ligase (Trp37 Val mutant of lipoic acid ligase or LplA) covalently attaches a copper-chelating picolyl azide derivative (pAz) onto LplAÕs acceptor peptide (LAP), which is genetically fused to a cell-surface or purified protein of interest (POI). In the second step, pAz-modified proteins are chemoselectively derivatized with a terminal alkyne-probe conjugate (red circle) by chelation-assisted CuAAC. Cu I is generated in-situ from 10Ð100 M Cu IISO4 and 2.5 mM sodium ascorbate. Ligand (L) represents Cu I-stabilizing ligands, such as THPTA 10, BTTAA 27 or TBTA 28. The LAP sequence is GFEID KVWYDLDA 24 (lysine labeling site underlined). ( b) Four different configurations for PRIME-CuAAC labeling. PRIME ligation of pAz can be performed at the cell surface (left), with application of exogenous LplA enzyme to the cell medium. Alternatively, pAz ligation can be performed in the cellÕs secretory pathway (right), using ligase expressed in the ER. Thereafter, CuAAC derivatization of pAz-modified proteins can be performed on live cells or after cell fixation. Key features of each labeling configuration are listed in the table below. Figure 0-22 Two -step labeling scheme of site -specific protein labeling via PRIME and chelation -assisted CuAAC. In the first step, W37VLplA covalently attaches a copper -chelating picolyl azide (pAz) onto genetically fused LAP. In step two, CuAAC selectively derivati zes pAz with a probe -conjugated terminal alkyne. 43 use in single molecular nanoscopy and PPI studies. 288, 289 It is worth noting that , LplA can be engineered to recognize azidoalkanoic acid instead of lipoic acid. This opens an avenue to couple CuAAC with PRIME to avoid potential steric collisions between large fluorochromes and enzyme active site (Figure I-22).289-292 OOHOCO2HAsAsSSSSONOHOAsAsSSSSOOHOCO2HAsAsSSSSONOHOAsAsSSSSONHNCO2HB(OH) 2(HO) 2BONHNCO2HBBOOOOOONHONNNMeO OOOOOMe OONHONNNOOOOSOMe OMe SSHSHdC10!FlAsH-EDT 2ReAsH-EDT 2RhoBoYC20 a b c Scheme 0-8 Labeling approaches based on short peptides. (a) FlAsH and ReAsH with tetracysteine motif. (b) Bis -boronic acid RhoBo with tetraserine motif. (c) Dimaleimide coumarin YC20 with vicinal dicysteine motif. 44 To date, the smallest labeling tag is the tetracysteine motif of biarseni cal FlAsH/ReAsH tags (Scheme I-8a). First developed in TsienÕs lab, four cysteines were placed at the i, i + 1, i + 4, and i + 5 positions of an 3-helix to facilitate the binding with two trivalent arsenics of 4Õ,5Õ-bis(1,3,2 -dithioarsolan -2-yl)fluorescein bis-ethanedithiol adduct (FlAsH -EDT 2). The QY of nonfluorescent FlAsH -EDT 2 can be increased by > 50,000-fold after dithiol transposition with the tetracysteine motif. 293 Soon after its debut, the biarsenical label was modified extensively, both its chemical structure and the peptide motif, to optimize the affinity of the biarsenics with tetracysteine and increase the contrast between specific and nonspecific staining. 294-297 Although these biarsenics are limited by their reductive nature, they have found widespread applications across various fields. 298-300 Most recently, ab initio calculatio ns and excited -state lifetime measurement were conducted to elucidate the origin of rotamer -restricted fluorogenicity of these biarsenical labels. 301 Similar strate gies were adopted to develop other short peptide motifs for site -specific labeling. Schepartz and co -workers reported a bis -boronic acid probe that shows high affinity to a tetraserine motif and negates the use of cytotoxic arsenic (Scheme I-8b).302 By coupling a dimale imide PeT quencher to a coumarin, Keillor and co -workers have developed a fluorogen that can have turn -on fluorescence after the Michael addition with a vicinal dithiol peptide ( Scheme I-8c).303, 304 It could be envisioned that aided by directed evolution, new vicinity -induced multi -transposition tags will emerge in the near future. 45 I.4!Ongoing pursuit of far -red/near-infrared flu orochromes For modern biomedic al fluorescence imaging, especially in vivo imaging, the two pillars are the imaging resolution and the imaging depth. The former has witnessed the third revolutionary wave in microscopy, which was recognized by the 2014 Nobel Prize in Chemistry jointly aw arded to Eric Betzig, Stefan W. Hell, and William E. Moerner, for Òthe development of super -resolved fluorescence microscopyÓ. 305 The latter o ne, however, is embracing the continuous advancement of NIR emitters. Deeper tissue is transparent to NIR light for two reasons: 1) physically the longer the wavelength, the less scattering; 2) physiologically, h aemoglobin s and water are the major absorbers in visible and infrared regions, respectively, while having the lowest absorption coefficients in the NIR -I region (650 Ð 900 nm, Figure I-23).306 Fluorophores suitable for NIR -I imaging have drawn much attention during the past decade. Recently, a number of inorganic and nanomaterials have also been explored in use of NIR -II (1,000 Ð 1,700 nm) imaging. 307 Figure 0-23 The ideal NIR window for in vivo imaging, showing minimal light absorption by h aemoglobin (Hb, green), oxyhaemoglobin (HbO 2, red) (<650 nm) and water (>900 nm). Copied from ref [306 ]. 46 I.4.1!Far-red and NIR FPs Discussed in Section I.1.2, DsRed -like chromophores have extended conjugation through acylimines at residue 65. Actually, DsRed has been used as the parent template for a variety of autoca talytic far-red FPs .18, 19 Representative examples are mKate/2, eqFP650/670, TagRFP657/675, mNeptune, mPlum, and mCardinal, to name a few. 47, 53, 308-310 Their emission maxima are all close to , or beyond 650 nm. The red -shift is based on several strategies, including extended (-conjugation, chromophore isomerization, increased chromophore planarity, hydrogen -bonding to the acylimine oxygen, ground -PAS GAFBVNHHNNHO2CCO2OSHNOABCD1516Cys NHHNO2CCO2BCNHHNOOSCys AD1516Z: Pr state Z: Pfr state far-red NIRautoisomerizestoPVBaftercovalentattachmenttotheCys residue.33,34Slightlydi !erentStokesshiftsobservedinNIRFPsare possiblydeterminedbydi !erentpropertiesofthechromophore electron-conjugatedsystem,a !ectedbychromophoreplanarity andpresenceofthehydrogen-bondchromophoreconformers, andbychromophoreinteractionswiththeimmediateprotein environment,includingrearrangementofthehydrogenbonds aroundthechromophoreandanexcited-stateprotontrans- fer.35,36Theoligomericstateofphytochromeshasastrongin "uenceontheirspectralproperties.Typically,oligomericanddimeric proteinsincorporatechromophoremoree #cientlybecauseofa betterfoldingofdimericproteinsthantheirmonomeric versions.Thiscausesapositivecooperativee !ectof chromophorebindingtooneprotomeronanotheronein thedimer. 30Therefore,amolarextinctioncoe #cientthat dependsontheamountofproteinmoleculeswithbound chromophore(holoform)ishigherfordimericandoligomeric phytochromes.2.3.Light-InducedStructuralChanges Cyanobacterialandalgalphy tochromescansenselight throughoutthewholevisiblespectrum. 37,38However,bacterial phytochromesspeci $callysensefar-redandNIRlightand thereforearegoodtemplatestodevelopopticalprobesand toolsforuseinmammaliantissues.Typically,BphPsexistin oneoftwointerconvertiblestates,eitherthePrstateorthePfr state( Figure3 A).ThePrstateabsorbslightat660 !700nm, whereasthePfrstateabsorbslightat740 !770nm.Inaddition tothemainabsorptionpeak,knownastheQ-band,all phytochromesabsorbat380 !420nm,knownastheSoret band,whichcorrespondstoindividualpyrrolerings. AZ/Eisomerizationofthebilinchromophorearoundits15/ 16doublebondinthemethinebridgebetweentheCandD pyrroleringsoccursuponillumination( Figure3 B).The resultingrotationoftheD-ringinducesconformationalchanges intheproteinbackbone,whicharetransferredfromthePCM totheOM. Severalmechanisticschemesforsignalpropagationtothe OMwereformulatedonthebasisofspectroscopicand crystallographicanalyses.StudiesofcrystallizedPCMsrevealed thatthe !-helicalspine,whichisinvolvedintheformationof phytochromedimers,alsoservesasafunctionallinkerbetween thePCMandtheOM. 8Light-drivenconformationalchangesin theBVchromophorein DrBphP,abacterialphytochromefrom Deinococcusradiodurans ,generaterearrangementoftheGAF domainandC-terminal !-helices,thuspropagatingalight signaltotheOMandmodulatingitsactivity. 39Analysisofthe structureofPhyBfrom Arabidopsisthaliana suggestedasimilar photoconversionmechanismforplantphytochromes. 29Analysesofcrystalstructure s,single-particleelectron microscopy(SPEM),andaproteasesensitivityassayof DrBphPallowedfurthercharacterizationoflight-induced conformationalchanges. 40!43Inthe DrBphPPCM,arotation oftheD-ringcausesareorganizationofhydrogenbondsinthe BVchromophorebindingpocket.Consequently,changesinthe weakinteractionsofproteinsidechainsleadtoexpansionofa conformationalspaceoftheprotein,followedbyrepositioning ofthePHYdomainsinascissor-likemanner( Figure3 C).At thesametime,thePHY-tongueundergoesrefolding.InthePr state,themainpartofthetongueconsistsoftwoantiparallel "-strands,whereasinthePfrstateitconvertsintoan !-helix.40,41Mostlikely,infull-length DrBphP,rotationalmovementin PAS!GAFdomainsistransducedasrotationoftheOMviathe PHYdomain.TheX-raysolutionscatteringdatarevealedthata rotationalmotionofHisKdomainswasthemajorlight-driven structuralrearrangementobservedinfull-length DrBphP.42Figure3. Light-drivenchangesinstructureandspectralpropertiesofphytochromes.(A)AbsorptionspectraofbacterialphytochromesinPrandPfr states.(B)Reversible Z/E(Pr!Pfr)isomerizationoftheC15/C16doublebondinboundBVchromophoreunderilluminationwithfar-redandNIR light.(C)Light-drivenstructuralrearrangementsinbacterialphytochrome.RepositioningofthePHY-tongueandthePHYdomainsaremarkedwith blackarrows. !-Helixand "-sheetsecondarystructuresofthePHY-tongueinPrandPfrstates,respectively,areindicated.Crystalstructureswith PDBIDs4O0Pand4O01wereusedtovisualizethestructure. 43(D)Domainstructureoftranscriptionalrepressor RsPpsRfrom Rhodobactersphaeroides,whichistheclosehomologueofthephytochromebindingpartner RpPpsR2from R.palustris .MultiplePASdomainsareshownin shadesofgreen.Thelong !-helicallinkers,whicharelikelyresponsibleforinteractionwiththephotoreceptorbindingpartners,areshowninyellow. ThecrystalstructurePDBID4HH2wasusedtovisualizethestructure. 47ChemicalReviews ReviewDOI:10.1021/acs.chemrev.6b00700Chem.Rev. 2017,117,6423 !64466426a c b Figure 0-24 Phytochrome -derived BV -binding FP. (a) Scheme showing Z/E isomerization between Pr and Pfr states of BV adduct. (b) Absorption spectra of Pr and Pfr states. Q band is the major absorption. Soret band is the pyrrole absorption. (c) A truncated PAS -GAF NI R FP (PDB ID: 4XTQ). 47 state destabilization and excited -state stabilization, and increased environment polarity. 20, 310, 311 Phytochromes from plant, fungi, and bacteria provide another source of NIR FPs. Their natural function as photoreceptors are to absorb far -red and NIR light and regulate light response. Taking bacteria phytochrome s (BphP) as an example, they consist of the PAS, GAF, and PHY domains which are topologically conservative throughout species. 312 BphP binds biliverdin IX 3,(BV) in the GAF domain via a thioetherification between the active Cys residue and the terminal alkene on the A ring of B VÕs tetrapyrrole. 313 The embedded chromophore can undergo a light -triggered Z/E isomerization at the C 15-C16 double bond, namely, the Pr and Pfr states. The Pr states absorbs light at 660 Ð 700 nm, whereas the Pfr state ab sorbs light at 740 Ð 770 nm (Figure I-24). Substantial mutagenesis has halted this Z/E isomerization to yield permanently fluorescent NIR FPs. Because of the ubiquitous existence of BV in mammalian cells, t he current developed NIR F Ps are all from BphPs. 54, 314, 315 Numerous structural and activity studies have elucidated the essential functions of each domain. 316 With these understandings, novel NIR FPs are engineered based on truncating BphPs to either single GAF domain for small monomeric FPs (e.g., GAF -FP), 56 or to PAS -GAF domains for dimeric FPs (e.g., IFPs NSO3NO3SNaIndocyanine green (ICG) !abs = 788 nm !em = 813 nm Figure 0-25 Indocyanine green (ICG). 48 from DrBphP, and iRFPs from RpBphP). 54, 55 By engineering the tethered PHY domain, photoactivatable and photoswitchable NIR FPs are also developed and are use d in SRM. Assis ted with FACS and random mutagenes is, blue - and red -shifted NIR FPs have emerged across the NIR region. 317 A recent report even shows Sandercyanin , a BV -binding protein from walleye skin mucus, capable of being excited at UV -blue region and emitting in the far -red/NIR region. 64 I.4.2!NIR organic dyes For a long time, NIR organic dyes have been sought for various applications including biomedical imaging and biosensing .318, 319 The classical approach of synthetic dyes to reach NIR window is by extending the (-conju gation. One representative is the ICG approved by FDA for intraoperative fluorescence -guided surgery ( Figure I-25, see Section I.2.2.3). Recent progress in developing NIR dye have been focused on extending conjugation (i.e., either by fusing (hetero)aromatics , or by introduing mo re electron -diffused heteroatom), affec ting bond length alternation, modulating intramolecular charge transfer , and other effects including intramolecular hydrogen bonding and molecular stacking. I.4.2.1!Polymethine dyes Classical cyanine dyes are a family of organic dyes having odd number of carbons in a conjugated polymethine chain with two ends of nitrogen -containing heterocycles. A characteristic polymethine framework does not have significant bond length alternation between each carbon unit. Cyanines commonly feature sharp absorption bands and high extinction coefficients. Their polymethine chain lengths and nitrogen -containing terminal 49 groups can vary. 320 According to Brooker, 321 symmetrical cyanines are non-convergent series , which means the absorption and emission wavelength shifts with the increasing number of vinylene group (termed Òvinylene shiftÓ) are essentially equal ( ~ 100 nm) . While asymmetrical cyanines are convergent series and their vinylene shifts become smaller with the increasing length of methine conjugation. Cyanine dyes are well develo ped throughout the decades. To approach the NIR window, heptamethine cyanines with different terminal groups and rigidified polymethine backbones are synthesized (Figure NNMeMeMeMeMeMe!abs = 638 nm !em = 657 nm """#F = 0.15 Cy5,NNMeMeMeMeORRHHH3, R1 = R 2 = H !abs = 662 nm !em = 677 nm """#F = 0.69 !abs = 670 nm !em = 683 nm """#F = 0.55 4, R1 = SO 3-, R2 = CO 2HNMeMeClMeNMeMeMeI!abs = 780 nm !em = 798 nm IR-780 ,NMeMeMeClNMeMeMe!abs = 797 nm !em = 823 nm IR-797 ,ClNNMeMeClCl!abs = 1051 nm IR-1051 ,BF4NMeMeNHMeNMeMeMeHO3SSO3CO2HHCDN,!abs = 625 nm !em = 755 nm NMeMeNHNMeMeHO3SSO3R5, R =6, R =!abs = 602 nm !em = 757 nm """#F = 0.47 !abs = 617 nm !em = 757 nm """#F = 0.38 a b c Figure 0-26 Representative non -convergent polymethine cyanines. (a) Cyanines with various rigidified heptamethine backbones and terminal groups. (b) Conformation -restrained pentamethine cyanines. (c) Large Stokes shift heptamethine cyanines. 50 I-26a).322-324 However, the quantum yield is usually correlated to a dramatic loss with the increasingly red -shifted absorption. Recent ef forts have conformation ally restrain ed cyanines for increased stability and quantum yield , however the restrained heptamethine has not been synthesized as of ye t (Figure I-26b).325 Interestingly, it was found an amine subs titution on position 4 of the polymethine leads to a significant blue -shift in absorption, NNNBNBXXXXCNCNC8H17OOC8H17YYNNNNBNNBFFFFC8H17OOC8H17YYNNNBNBXXXXNCCNOC12H25OC12H25C12H25OOC12H25SNSBNC12H25OOC12H25CNXXNNBNCXXC12H25OOC12H25OC12H25C12H25OOC12H25OC12H25YY34: X = F, Y = Ar 635: X = F, Y = Ar 1136: X = F, Y = Ar 1537: X = Ph, Y = Ar 638: X = Ph, Y = Ar 1139: X = Ph, Y = Ar 15!"!"!"NOt-BuRAr1Ar2Ar3: R = H Ar4: R = Me RSAr5: R = H Ar6: R = t-BuAr7: R = OMe Ar8: R = SMe Ar9: R = CF 3R2R1Ar10: R1 = R 2 = H Ar11: R1 = H, R 2 = t-BuAr12: R1 = H, R 2 = Br Ar13: R1 = Ph, R 2 = Cl Ar14: R1 = H, R 2 =PhNAr15OAr16 7: X = F, Y = Ar 1 8: X = F, Y = Ar 2 9: X = F, Y = Ar 410: X = F, Y = Ar 612: X = F, Y = Ar 713: X = F, Y = Ar 814: X = F, Y = Ar 915: X = F, Y = Ar 1116: X = F, Y = Ar 1217: X = F, Y = Ar 1318: X = F, Y = Ar 1419: X = F, Y = Ar 1520: X = F, Y = Ar 1621: X = Ph, Y = Ar 122: X = Ph, Y = Ar 223: X = Ph, Y = Ar 424: X = Ph, Y = Ar 625: X = Ph, Y = Ar 726: X = Ph, Y = Ar 827: X = Ph, Y = Ar 928: X = Ph, Y = Ar 1129: X = Ph, Y = Ar 1230: X = Ph, Y = Ar 1331: X = Ph, Y = Ar 1432: X = Ph, Y = Ar 1533: X = Ph, Y = Ar 16#abs (nm) 708712730749763773738773775781791805720#em (nm) 684690707732745753719754761767778789699#abs (nm) 749762776804817827785831837843855881771#em (nm) 737747763790803808776819827832843864759#abs (nm) 827838867912924966#em (nm) 815823849894905941#abs (nm) 661676692#em (nm) 63865567140: Y = Ar 341: Y = Ar 542: Y = Ar 10!!!!!!!"""""""a b c Figure 0-27 Chemical structures of PPCy dyes. (a) Standard PPCys. (b) Further -extended PPCys. (c) PPCy aza -BODIPY analogues. 51 without affecting emission wavelength , which is advantageous for imaging purpose due to the large Stokes shift ( Figure I-26c).326-328 Polymethine dyes are not restricted to traditional cyanines. Essentially, they are characterized with methine chains of degenerated bond length alternation. Developed by DaltrozzoÕs and Zumb uschÕs group s, pyrrolopyrrole (PPCy) is a new class of polymethine dyes. Their pyrrolopyrrole core is an effective cyanine -type mimicry , and is almost exclusively synthesized via the condensation of diketopyrrolopyrroles with 3-substituted acetonitriles. 329 PPCys often show sharp emission bands in the NIR region with high extinction coefficients and exceptional fluorescence quantum yields (Figure I-27a).329-333 Their boron chelation between the nitrogen atoms of the pyrrolopyrrole and terminal aryl groups rigidifies the conformation and renders them with excellent light and thermal stability. Further (-extended PPCys fused with benzobisthiazoles have been synthesized and show extremel y large extinction coefficients and close to NIR-II emission (Figure I-27b).334 As the boron chelate is structurally similar to BODIPY dyes ( vide infra ), PPCy aza -BODIPY analogues have also been synt hesized and exhibit close to NIR-I fluorescence (Figure I-27c).335 Boron dipyrromethene ( BODIPY ), or 4,4 -difluoro -4-bora -3a,4a-diaza -s-indacene, is one of the most frequently investigated family of fluorescent compounds . With many commerci ally available BODIPYs, new structu ral entities are still emerging for both NIR emitting and functional sensing. 336-338 As shown in Figure I-28a, BODIPYs can be regarded as a rigid and cross -conjugated cyanines. Being similar in many aspects, they have sharp absorption and emission spectra and usually small Stokes shifts, with a 52 supe rior advantage in stability. Although common BODIPYs emit in the visible range, their potential in easy structural modifica tion have stimulated numerous derivatizations at all seven carbons. 339 It is known that the replacement of C8 with a nitrogen atom can effectively red -shift the spectra, possibly because of the higher electron density of NBNFF12345678(meso )abghBODIPY NBNNFFR1R1R3R3R2R2NBNNFFArArSSAr = Ph, 2-thienyl R1 = Me, aryl R2 = H, OMe, NMe 2R3 = H, Br 11 compounds, !abs = 643 ~ 799 nm, !em = 672 ~ 823 nm !abs = 710 ~ 733 nm, !em = 732 ~ 757 nm aza-BODIPY NBNFFR1R1R3R3RR2R2R1 = Me, substituted-styryl R2 = OMe, NMe 2, substituted-styryl R3 = H, Ar11 compounds, !abs = 645 ~ 802 nm, !em = 660 ~ 837 nm NBNFFR1MeO 2CR1 = C 2H4CO2Me, Me R2 = 2-thienyl, 4-PhOMe R2R23 compounds, !abs = 649 ~ 674 nm, !em = 681 ~ 686 nm ÒcoupledÓ BODIPY NBNXR1R1R3R3R2R2OOX = C-Ar, N R1 = H, Ar R2 = H, OMe R3 = H, Br 4 compounds, !abs = 728 ~ 765 nm, !em = 742 ~ 782 nm NBNFFOONBNNFFXXR1R1R2R3NBNNFFMeO R2R1R1 = H R2 = aryl, het-aryl orR1, R2 = 1,2-naphthyl 8 compounds, !abs = 668 ~ 774 nm, !em = 692 ~ 815 nm R1 = H, OMe R2 = H, OMe R3 = H, OMe X = CH 2, S4 compounds, !abs = 706 ~ 740 nm, !em = 730 ~ 752 nm ÒrestrictedÓ BODIPY 43,!abs = 732 nm, !em = 747 nm a b c d Figure 0-28 NIR BODIPYs. (a) BODIPY framework. (b) aza -BODIPYs. (c) "coupled" BODIPYs. (d) "restricted" BODIPYs. 53 nitrogen at the meso position that f acilitates stronger resonance. Recent synthetic advances have allowed further development of these Òaza -BODIPYsÓ with moderate to high QY in the NIR -I region (Figure I-28b).340-343 Another common strategy is to extend the conjugation by coupling multiple aryl groups on the dipyrromethen e framework. It has been shown that 3 -/5- disubstituti on is more impactful than 2 -/6- or 1 -/7- diarylation (Figure I-28c).344-346 However, these 3,5 -diaryl -BODIPYs usually have diminished NBNXFFR4R3R1R2X = CH, N; R 1 = H, OMe; R 2 = H, OMe R3 = R 4 = Ph, Nph, thienyl, PhNMe 2, PhOMe, PhF, PhMe 16 compounds, !abs = 644 ~ 793 nm, !em = 671 ~ 841 nm NBNFFR4R3NHR1R2RR1 = H, Me R2 = Me, H R3 = Me, Et, H R4 = Ar, H R = subst. pyrrole 4 compounds, !abs = 608 ~ 680 nm, !em = 687 ~ 733 nm NBNFFR1R2R1PhPhR1 =PhPhR2 = Me, Ph 4 compounds, !abs = 658 ~ 765 nm, !em = 695 ~ 783 nm [a/h]-fused BODIPY NBNFFXXR1R1 = H, CF 3R2 = H, Br R3 = 4-OMe, 2,5-OMe X = O, S 7 compounds, !abs = 690 ~ 766 nm, !em = 701 ~ 820 nm RRR2R2NBNFFNHHN44, !abs = 727 nm !abs = 744 nm " = 327000 M-1 cm -1#F = 0.63 [b/g]-fused BODIPY a b Figure 0-29 NIR ring -fused BODIPYs. (a) [a/h] -fused BODIPYs. (b) [b/g] -fused BODIPYs . 54 extinction coefficients and quantum yields, probably due to the free rotation of the aryl groups. 347, 348 The facile solution is to increase molecular rigidity by fusing the ÒcoupledÓ aryl groups to the pyrrole framework (Figure I-28d). A bathochromic shift together with slight increase in extinction coefficient was observed wh en 3,5 -diaryl were locked either through c halcogen (e.g., O, S) or ethylene unit. 349 A good example is shown by 3,4,4 a-trihydroxanthene -fused BODIPY (43), which shows over 100 nm bathochromic shift compare to the styryl BODIPY analogue. 350 Difluorides on boron have also been substituted by ortho -phenolates on the coupling aryls to form boronate. 351 The similar Òrestrict ionÓ approach has been applied to aza -BODIPY. 352 Besides the (-extension through position 1 -/7- and 3 -/5- via rotatable bonds , a prominent and more challenging direction is to build conjugation through [ a/h] or [ b/g] fused aromatic rings. 353-355 The idea is to have (-extension and at the same time to take advantage of BODIPYÕs high rigidity for high QY. However, the optic al properties are usually difficult to estimate due to the unpredictable ring fusion effects. It has been shown that among a vast number of a/h- or b/g-fused BODIPYs, those with conjug ation extended through heterocyclics as furano, pyrrolo, and thieno -fused BODIPYs are the most effective red -shifted ones ( Figure I-29).340, 356-362 It is noteworthy to mention that a recently published naphtopyrrole -fused BODIPY (44) has shown extraordinary optic al properties. 363 I.4.2.2!Donor-Acceptor NIR dyes Squaraines are a family of dye transitioning from polymethine type to donor -acceptor type. They have an electron -withdrawing cyclobutanone in the center and two electron -donating groups on the edge, while their resonance -stabiliz ed zwitterionic 55 structure still presents traits of polymethine. 364 Squaraines have sharp spectra, extremely large extinction coefficient, and high stability. 365 They have been widely used in imaging, nonlinear optics, photovoltaics, and photodynamic therapy. 366 Their inherent strong intramolecular charge transfer character (donor -acceptor -donor sandwich, D -A-D) makes them quite polarity -sensitive. By adding strong electron -donors or extending t he (-conjugation , the band gap can be effectively lowered and results in bathochromi c shift to NNNNOOR3R2R1R1R2R3OONNNNR1R2R3R3R3R3R1R2R1 = H, OMe, Ph R2 = H, Ph R3 = H, Ph 3 compounds !abs = 711 ~ 732 nm !em = 774 ~ 823 nmR1 = Me, n-C6H13R2 = n-C6H13, -(CH 2)5CO2HR3 = H, Me, -(CH 2)3SO3-3 compounds !abs = 737 ~ 802 nm !em = 751 ~ 817 nmONCCNNNXXC12H25C12H2545, X = H 46, X = Cl 47, X = Br 48, X = I !abs (nm) !em (nm) OOR2R1NR3R3NR4R4R1 = R 2 = H, R3 = Me, -(CH 2)2OHR4 = Me, n-Bu3 compounds !abs = 704 ~ 789 nm !em = 735 ~ 811 nm870885891900890913916922Squaraines OOONR1R2OHOHHOHONR1R2oxidation ONR1R2OHHONR1R2OOOO6 compounds !abs = 880 nm R1 = alkyl, R 2 = alkyl 6 compounds !abs = 1100 nm R1 = alkyl, R 2 = alkyl croconates a b Figure 0-30 (a) NIR fluorescent squaraines. (b) NIR -II absorptive croconates. Scheme shows mesoionic oxidation. 56 NIR region (Figure I-30a).367 Recent advances in squaraine synthesis have enabled the assembly of various heterocyclic groups on the squaric ac id core, including symmetric and asymmetric squaraines. 368-373 An interesting expansion of the zwitterionic core is the croconic acid, which provides an even stronger electron acceptor and hence l owers the band gap more efficiently. Alkyl croconates , albeit not fluorescent, can absorb into NIR -I and are easily red -shifted to 1.1 !m with a mesoio nic form after quick oxidation ( Figure I-30b).374, 375 Xanthene is another important type of dye framework with one monomethine bridging two aromatic rings. Commonly seen examples include fluoresceins, pyronins, rhodamines, and the meso -nitrogen derivative resorufins. Ex cept for fluoresceins, these dyes easily reach the far -red region of visible spectrum (e.g. Atto dyes and some of the AlexaFluor s).376 However, alt hough bearing the methine unit and an electron donor -acceptor pair on two sides, the majority of red xanthene s (i.e., rhodamines and pyronines ) do not display ICT -type environment sensitivity , with asymmetric resorufins (e.g., Nile red and Nile blue) as th e only exceptions . Typically, the spectra of rhodamines are not sensi tive towards the aryl group coupling as discussed in the BODIPY case. Currently XNRNX = SiMe 2, Te=O, CMe 2R = Ar, H ZNR1N3 compounds !abs = 642 ~ 669 nm !em = 660 ~ 686 nmR2R3XYR1 = H, Ar R2 = Et, methoxylethyl R3 = Et, methoxylethyl X = pyrrolidino, piperidino Y = pyrrolidino, piperidino X = Y = indolo Z = O, C(Me) 2R2/Y = julolideno 13 compounds !abs = 633 ~ 700 nm !em = 657 ~ 730 nmSiNR2R1NR2XXR1 = Ar, PhR2 = Me, Et X = N-Me, N-Et, pyrrolidino, piperidino 5 compounds !abs = 645 ~ 721 nm !em = 661 ~ 740 nmFigure 0-31 Restricted and C10 -replaced rhodamines. 57 applied strategies to induce bathochromic shifts in rhodamines i nclude conformation al restrictio ns on both rings and amine lone pairs , and meso -replacement with the first three rows of group 14 , 15 and 16 elements. 377-379 Proper alkylation on the nitrogen atoms will increase the rotational barrier and suppress the twisted intramolecular charge transfer process (TICT) in the excited states, and result in spectral bathochromic sh ifts. 380 Mod ifications include fused aliphatic (e.g., pyrrolidine , julolidene) and aromatic rings, nonetheless, with limited improvement in bathochromic shifts. A recent trend in rhodamine and pyronin is to replace C10 with more diffused elements (Si, Ge, Se, Te) and couple with extended (-conjugation to realize NIR emission (Figure I-31).381-387 It i s noteworthy to mention here, though mostly found use in organic optic materials, highly fused poly -heteroaromatics with properly embedded electron donor and acceptor NXNNNNNSSSSSSTTF,!abs = 535 nm !em = 820 nmNYN""ArArX = S, Se, Y = S, Se, C(R)-C(R) "#=SAr = NNOC8H17OC8H17SC8H17C8H17NHNSN""ArArNNNNNSN""ArArBBTD, TQ18 compounds !abs = 531 ~ 1177 nm !em = 698 ~ 1360 nmSOOQPP 2 compounds !abs = 746 ~ 844 nm !em = 1035 ~ 1100 nmNNNNNNNSNNSNNSN""""""ArArArArArArHAT 3 compounds !abs = 530 ~ 1003 nm !em = 700 ~ 1290 nmFigure 0-32 Representative NIR D -A-D type of dyes. 58 moieties have been successfully developed int o D -(-A or D -(-A-(-D NIR -I/II dyes. Examples include tetrathiafulvalenes (TTF) ,388 benzothiad iazoles (BTD) ,389 benzobisthiadiazoles (BBTD) and Se homologues, 390-393 phenothiazines ,394 thiadiazoloquinoxal ines (TQ) ,395, 396 hexaazatriphenylene (HAT), 397, 398 and quinoxalino[2Õ,3Õ,9,10]phenanthro [4,5 -abc]phenazines (QPP) .399 These core frameworks are usually coupled to electron -rich pyrrole, fluorene, thio phene, or diphenylaniline spacers to obtain lengthy resonating system hence low band gaps (Figure I-32). Although these dyes are typically large and highly hydrophobic, most recently Dai and coworkers ONCCNNDCM!abs = 481 nm !em = 644 nm NOCNCNHONXNCCNX = CH 2, O, null3 compounds !abs = 474 ~ 501 nm !em = 680 ~ 713 nm SO2NNO2nn = 2, 3 !abs = 436 ~ 446 nm !em = 667 ~ 684 nm HBO derivative !abs = 360 nm !em = 522, 780 nm OR3NR2R1CO2HR4R1 = Et; R 2 = Et; R3 = H, Me; R 4 = H, R1/R3 = R 2/R4 = -(CH 2)3-6 compounds !abs = 688 ~ 736 nm !em = 721 ~ 773 nm NR1R2R3NNR4ONXCO2HNCS-NIR XNR4OR1R2R3X = O, NH R1 = H, Me, Et R2 = H, Cl, SO 3-R3 = H, SO 3-R4 = Me, Et 6 compounds !abs = 654 ~ 702 nm !em = 677 ~ 725 nm HD-NIR R1 = H, Me, Ph R2 = H, Me, Ac, Bz R3 = H, OMe R4 = OH, NMe 212 compounds !abs = 570 ~ 718 nm !em = 654 ~ 749 nm X = CC-Fluor ONNXR3R2R1X = O, NH R1 = Et; R 2 = H, Et R3 = H, Me R1/R4 = R 2/R3 = -(CH 2)3-5 compounds !abs = 710 ~ 724 nm !em = 732 ~ 747 nm XC-NIR R4OHOONSO3nSO3HSO3Hn = 1, 2 5 compounds !abs = 491 ~ 643 nm !em = 648 ~ 722 nm Hemicyanine-Coumarin Figure 0-33 Selected ICT dyes and D -A hybrids. 59 have PEGylated a BBTD dye ( CH1055) and successfully achieved micrometer -scale resolution of NIR -II imaging in live mouse. 400 Donor-acceptor type of interactions are not limited to these gigantic polyhetero cyclic dyes, essentially any molecule facilitating ICT process has this band gap lowering feature. 239, 401 Small molecules such as donor -acceptor conjugated fluorenes and oligothiophene s have shown great bathochromic shift to NIR -I edge in responding to environment polarity change. 402, 403 Recently , the old laser dye dicyanomethylene -4H-pyran (DCM) has drawn much attention for NIR applications. 79, 404-406 Another widely applied design is to c ouple two red dyes together to install D -A interaction . For instance, 2-(2Õ -hydroxyphenyl) -benzoxazole (HBO) has been coupled to a dicyanovinylene fragment to realize two well -separated emission maxima, with one into the NIR region that features a characte ristic proton transfer event of the phenol in the excited state .407 NNOROOROnbay position bay position NNOOOOnR1R1R2R2PEG3PEG3n = 0, 1 R1 = H, pyrrolidinyl, 4- t-butylphenoxy R2 = H, 4- t-butylphenoxy 3 compounds !abs = 593 ~ 741 nm!em = 632 ~ 803 nm NOONOONNC10H21C10H21C10H21C10H2150,!abs = 760 nm !em = 775 nm "F = 0.55 NNOOOONNNNNNPhPhOOOOC8H17C8H17C8H17C8H1749,!abs = 778 nm !em = 938 nm Figure 0-34 Selected NIR rylenes. 60 Other examples i nclude indocyanine -coupled xanthene and coumarin dyes (Figure I-33).408-412 I.4.2.3!Other NIR dyes Polyenes can also be pushed into NIR region simply by extending conjugation, though the wavelength shift is not as significant as polymethine dyes (se e I.4.2.1) with increasing number of (-spacers . Rylenes are a family of rigid polymeric naphthalene s coupled through 3-carbons. The diim ides at the terminals consist of a pull -pull system. NIR-I emis sion can be easily approached with the installment of multiple naphthalene units ( n 2). 413, 414 Another strategy to effectively lower the band gap is by insert ing donor groups at the bay position without lengthy extension s (Figure I-34).415 It has been shown recently that an asymmetric donor instal led at the long axis yields a quarterrylene (50) with large extinction coefficient (259,540 M -1 cm-1) and excellent quantum yield (0.55 at 775 nm). 416 Structured as conjugate close -ring polypyrroles, porphyrinoids are receiving attention as a new type NIR emitter. Common core structures of porphyrinoids include porphyrin , phthalocyanine , and expanded ones such as hexaphyrin and rubyrin (Figure I-35).417 These polycyclics usually have very bathochromic spectra ex tending into the NIR XYNNNNMR1R2M = Pd, Pt X = Y = C, R 1 = R 2 = Ph; X = C, R 1 = Ph, Y = N; X = Y = N; 6 compounds !abs = 614 ~ 642 nm!em = 770 ~ 875 nm "F = 0.05 ~ 0.51 NNNHNNHNC6H5C6H5C6H5C6H5C6H5C6H5Hexaphyrin 51,!ex = 514 nm !em = 1050 nm NFSSFNFSSFrubyrin 52,!ex = 532 nm !em = 1118 nm Figure 0-35 Selected NIR porphyrinoids. 61 region. One certain advantage of porphyrinoids is that they can be excited at their short -wavele ngth Soret bands ( "abs = 500 ~ 600 nm) and emit into NIR -II region. Their fluorescence intensity is usually affected by metal -ligand charge transfer (MLCT) when they complex with transition metals. This feature has been extensively exploited in designing NIR metal sensors. 418, 419 The most comm on metals are palladium and platinum. 420 Not limited to these two, othe r close -shell and open -shell metals can also form complexes and more often result in NIR phosphorescence. A more detailed discussion is not fitting the scope here ; the reader is directed to ArmaroliÕs recent review. 421 As a closing remark, it is noteworthy to point out that the approache s to NIR emission sh ould not be limited to intramolecular (-electron dispersion. A rising field is the aggregation -induced emission (AIE). 422 Using the iconic AIEgen tetraphenylene (TPE) as an example, the aggregation stops the free rotations of the pheny l groups and strong emission occurs due to the suppressed nonradiative decay. TPE can be easily modified to couple with a dicyanovinylene acceptor through a thiophene spacer, which results in a significant bathochromic shift in TPE emission ( 53, Figure I-36a).423 Alternatively , red fluorophores can be engineered into AIEgens such that the band gap is further lowered CNNSN54,!abs = 479 nm !em, sol = 553 nm !em, agg = 659 nm MeO OMe SCNNCHOOH53,!abs = 480 nm !em, sol = null !em, agg = 650 nm a) b) Figure 0-36 Examples of NIR AIEgen. (a) Red -shifted TPE emission by extended conjugation. (b) Aggregation -induced bathochromic shift in emission. 62 by excimer -induced intermolecular (-electron dispersion , hence realizing NIR emission (54, Figure I-36b).424 63 REFERENCE S 64 REFERENCE S 1. Hooke, R. Micrographia: or some physiological descriptions of minute bodies made by magnifying glasses with observations and inquiries thereupon. (Royal Society; London; 1665). 2. Prasher, D.C., Eckenrode, V.K., Ward, W.W., Prendergast, F.G. & Cormier, M.J. Primary structure of the Aequorea victoria green -fluorescent protein. Gene 1992, 111, 229-233. 3. Chalfie, M., Tu, Y., Euskirchen, G., Ward, W. & Prasher, D. Green fluorescent protein as a marker for gene expression. Science 1994, 263, 802-805. 4. Matz, M.V. et al. Fluorescent proteins from nonbioluminescent Anthoz oa species. Nat. Biotechnol. 1999, 17, 969-973. 5. Newman, R.H., Fosbrink, M.D. & Zhang, J. Genetically encodable fluorescent biosensors for tracking signaling dynamics in living cells. Chem. Rev. 2011, 111, 3614-3666. 6. Subach, O.M., Cranfill, P.J., Davi dson, M.W. & Verkhusha, V.V. An enhanced monomeric blue fluorescent protein with the high chemical stability of the chromophore. PLoS One 2011, 6, e28674. 7. Dedecker, P., De Schryver, F.C. & Hofkens, J. Fluorescent proteins: shine on, you crazy diamond. J. Am. Chem. Soc. 2013, 135, 2387-2402. 8. Wilson, T. & Hastings, J.W. Bioluminescence. Annu. Rev. Cell Dev. Biol. 1998, 14, 197-230. 9. Zimmer, M. Green Fluorescent Protein (GFP): Applications, Structure, and Related Photophysical Behavior. Chem. Rev. 2002, 102, 759-782. 10. Heim, R., Prasher, D.C. & Tsien, R.Y. Wavelength mutations and posttranslational autoxidation of green fluorescent protein. Proc. Natl. Acad. Sci. U. S. A. 1994, 91, 12501-12504. 11. Zhang, L., Patel, H.N., Lappe, J.W. & Wachter, R.M. Reaction progress of chromophore biogenesis in green fluorescent protein. J. Am. Chem. Soc. 2006, 128, 4766-4772. 12. Barondeau, D.P., Putnam, C.D., Kassmann, C.J., Tainer, J.A. & Getzoff, E.D. Mechanism and energetics of green fluorescent protein chromoph ore synthesis 65 revealed by trapped intermediate structures. Proc. Natl. Acad. Sci. U. S. A. 2003, 100, 12111-12116. 13. Sniegowski, J.A., Lappe, J.W., Patel, H.N., Huffman, H.A. & Wachter, R.M. Base catalysis of chromophore formation in Arg96 and Glu222 variants of green fluorescent protein. J. Biol. Chem. 2005, 280, 26248-26255. 14. Sniegowski, J.A., Phail, M.E. & Wachter, R.M. Maturation efficiency, trypsin sensitivity, and optical properties of Arg96, Glu222, and Gly67 variants of green fluorescent protein. Biochem. Biophys. Res. Commun. 2005, 332, 657-663. 15. Shagin, D.A. et al. G FP-like proteins as ubiquitous metazoan superfamily: evolution of functional features and structural complexity. Mol. Biol. Evol. 2004, 21, 841-850. 16. Labas, Y.A. et al. Diversity and evolution of the green fluorescent protein family. Proc. Natl. Acad. S ci. U. S. A. 2002, 99, 4256-4261. 17. Chudakov, D.M., Matz, M.V., Lukyanov, S. & Lukyanov, K.A. Fluorescent proteins and their applications in imaging living cells and tissues. Physiol. Rev. 2010, 90, 1103-1163. 18. Verkhusha, V.V. & Lukyanov, K.A. The mol ecular properties and applications of Anthozoa fluorescent proteins and chromoproteins. Nat. Biotechnol. 2004, 22, 289-296. 19. Subach, F.V., Piatkevich, K.D. & Verkhusha, V.V. Directed molecular evolution to design advanced red fluorescent proteins. Nat. Methods 2011, 8, 1019-1026. 20. Subach, F.V. & Verkhusha, V.V. Chromophore transformations in red fluorescent proteins. Chem. Rev. 2012, 112, 4308-4327. 21. Strack, R.L., Strongin, D.E., Mets, L., Glick, B.S. & Keenan, R.J. Chromophore formation in DsRed o ccurs by a branched pathway. J. Am. Chem. Soc. 2010, 132, 8496-8505. 22. Subach, O.M. et al. Structural characterization of acylimine -containing blue and red chromophores in mTagBFP and TagRFP fluorescent proteins. Chem. Biol. 2010, 17, 333-341. 23. Pakhom ov, A.A. & Martynov, V.I. Chromophore aspartate oxidation -decarboxylation in the green -to-red conversion of a fluorescent protein from Zoanthus sp. 2. Biochemistry 2007, 46, 11528-11535. 24. Quillin, M.L. et al. Kindling fluorescent protein from Anemonia s ulcata: dark -state structure at 1.38 A resolution. Biochemistry 2005, 44, 5774-5787. 66 25. Gurskaya, N.G., Savitsky, A.P., Yanushevich, Y.G., Lukyanov, S.A. & Lukyanov, K.A. Color transitions in coral's fluorescent proteins by site -directed mutagenesis. BMC Biochem. 2001, 2, 6. 26. Remington, S.J. et al. zFP538, a yellow -fluorescent protein from Zoanthus, contains a novel three -ring chromophore. Biochemistry 2005, 44, 202-212. 27. Pletneva, N.V. et al. Three -dimensional structure of yellow fluorescent protein zYFP538 from Zoanthus sp. at the resolution 1.8 †. Russ. J. Bioorg. Chem. 2007, 33, 390-398. 28. Karasawa, S., Araki, T., Nagai, T., Mizuno, H. & Miyawaki, A. Cyan -emitting and orange -emitting fluorescent proteins as a donor/acceptor pair for fluorescence resonance energy transfer. Biochem. J 2004, 381, 307-312. 29. Kikuchi, A. et al. Structural chara cterization of a thiazoline -containing chromophore in an orange fluorescent protein, monomeric Kusabira Orange. Biochemistry 2008, 47, 11573-11580. 30. Mizuno, H. et al. Photo -induced peptide cleavage in the green -to-red conversion of a fluorescent protein . Mol. Cell. 2003, 12, 1051-1058. 31. Ormo, M. et al. Crystal structure of the Aequorea victoria green fluorescent protein. Science 1996, 273, 1392-1395. 32. Tomosugi, W. et al. An ultramarine fluorescent protein with increased photostability and pH insens itivity. Nat. Methods 2009, 6, 351-353. 33. Mena, M.A., Treynor, T.P., Mayo, S.L. & Daugherty, P.S. Blue fluorescent proteins with enhanced brightness and photostability from a structurally targeted library. Nat. Biotechnol. 2006, 24, 1569-1571. 34. Ai, H. W., Shaner, N.C., Cheng, Z., Tsien, R.Y. & Campbell, R.E. Exploration of new chromophore structures leads to the identification of improved blue fluorescent proteins. Biochemistry 2007, 46, 5904-5910. 35. Subach, O.M. et al. Conversion of red fluorescent p rotein into a bright blue probe. Chem. Biol. 2008, 15, 1116-1124. 36. Rizzo, M.A., Springer, G.H., Granada, B. & Piston, D.W. An improved cyan fluorescent protein variant useful for FRET. Nat. Biotechnol. 2004, 22, 445-449. 37. Goedhart, J. et al. Bright c yan fluorescent protein variants identified by fluorescence lifetime screening. Nat. Methods 2010, 7, 137-139. 38. Markwardt, M.L. et al. An improved cerulean fluorescent protein with enhanced brightness and reduced reversible photoswitching. PLoS One 2011, 6, e17896. 67 39. Yang, T.T., Cheng, L. & Kain, S.R. Optimized codon usage and chromophore mutations provide enhanced sensitivity with the green fluorescent protein. Nucleic Acids Res. 1996, 24, 4592-4593. 40. Cubitt, A.B., Woollenweber, L.A. & Heim, R. in Chapter 2: Understanding Structure ÑFunction Relationships in the Aequorea victoria Green Fluorescent Protein. Green Fluorescent Proteins . Methods in Cell Biology. Vol. 58. (eds. K.F. Sullivan & S.A. Kay), 19 -30 (Academic Press; 1998). 41. Evdokimov, A.G. et al. Structural basis for the fast maturation of Arthropoda green fluorescent protein. EMBO Rep. 2006, 7, 1006-1012. 42. Sarkisyan, K.S. et al. Green fluorescent protein with anionic tryptophan -based chromophore and long fluorescence lifetime. Biophys. J. 2015, 109, 380-389. 43. Nagai, T. et al. A variant of yellow fluorescent protein with fast and efficient maturation for c ell-biological applications. Nat. Biotechnol. 2002, 20, 87-90. 44. Kremers, G.J., Goedhart, J., van Munster, E.B. & Gadella, T.W., Jr. Cyan and yellow super fluorescent proteins with improved brightness, protein folding, and FRET Forster radius. Biochemist ry 2006, 45, 6570-6580. 45. Hoi, H. et al. An engineered monomeric Zoanthus sp. yellow fluorescent protein. Chem. Biol. 2013, 20, 1296-1304. 46. Shaner, N.C. et al. Improved monomeric red, orange and yellow fluorescent proteins derived from Discosoma sp. r ed fluorescent protein. Nat. Biotechnol. 2004, 22, 1567-1572. 47. Merzlyak, E.M. et al. Bright monomeric red fluorescent protein with an extended fluorescence lifetime. Nat. Methods 2007, 4, 555-557. 48. Kredel, S. et al. mRuby, a bright monomeric red fluo rescent protein for labeling of subcellular structures. PLoS One 2009, 4, e4391. 49. Bindels, D.S. et al. mScarlet: a bright monomeric red fluorescent protein for cellular imaging. Nat. Methods 2017, 14, 53-56. 50. Wang, L., Jackson, W.C., Steinbach, P.A. & Tsien, R.Y. Evolution of new nonantibody proteins via iterative somatic hypermutation. Proc. Natl. Acad. Sci. U. S. A. 2004, 101, 16745-16749. 51. Lin, M.Z. et al. Autofluorescent proteins with excitation in the optical window for intravital imaging in m ammals. Chem. Biol. 2009, 16, 1169-1179. 52. Shcherbo, D. et al. Far -red fluorescent tags for protein imaging in living tissues. Biochem. J 2009, 418, 567-574. 68 53. Shcherbo, D. et al. Near -infrared fluorescent proteins. Nat. Methods 2010, 7, 827 -829. 54. Shcherbakova, D.M. & Verkhusha, V.V. Near -infrared fluorescent proteins for multicolor in vivo imaging. Nat. Methods 2013, 10, 751-754. 55. Yu, D. et al. An improved monomeric infrared fluorescent protein for neuronal and tumour brain imaging. Nat. Commun. 2014, 5, 3626. 56. Rumyantsev, K.A. et al. Minimal domain of bacterial phytochrome required for chromophore binding and fluorescence. Sci. Rep. 2015, 5, 18348. 57. Zapata-Hommer, O. & Griesbeck, O. Efficiently folding and circularly permuted variants of th e Sapphire mutant of GFP. BMC Biotechnol. 2003, 3, 5. 58. Kogure, T. et al. A fluorescent variant of a protein from the stony coral Montipora facilitates dual -color single -laser fluorescence cross -correlation spectroscopy. Nat. Biotechnol. 2006, 24, 577-581. 59. Piatkevich, K.D. et al. Monomeric red fluorescent proteins with a large Stokes shift. Proc. Natl. Acad. Sci. U. S. A. 2010, 107, 5369-5374. 60. Piatkevich, K.D., Malashkevich, V.N., Almo, S.C. & Verkhusha, V.V. Engineering ESPT pathways based on str uctural analysis of LSSmKate red fluorescent proteins with large Stokes shift. J. Am. Chem. Soc. 2010, 132, 10762-10770. 61. Shcherbakova, D.M., Hink, M.A., Joosen, L., Gadella, T.W. & Verkhusha, V.V. An orange fluorescent protein with a large Stokes shift for single -excitation multicolor FCCS and FRET imaging. J. Am. Chem. Soc. 2012, 134, 7913-7923. 62. Yang, J. et al. mBeRFP, an improved large stokes shift red fluorescent protein. PLoS One 2013, 8, e64849. 63. Chu, J. et al. A bright cyan -excitable orange fluorescent protein facilitates dual -emission microscopy and enhances bioluminescence imaging in vivo. Nat. Biotechnol. 2016, 34, 760-767. 64. Ghosh, S. et al. Blue protein with red fluorescence. Proc. Natl. Acad. Sci. U. S. A. 2016, 113, 11513-11518. 65. Patterson, G.H. & Lippincott -Schwartz, J. A photoactivatable GFP for selective photolabeling of proteins and cells. Science 2002, 297, 1873-1877. 66. Grotjohann, T. et al. Diffraction -unlimited all -optical imaging and writing with a photochromic GFP. Nature 2011, 478, 204-208. 69 67. Gurskaya, N.G. et al. Engineering of a monomeric green -to-red photoactivatable fluorescent protein induced by blue light. Nat. Biotechnol. 2006, 24, 461-465. 68. McKinney, S.A., Murphy, C.S., Hazelwood, K.L., Davidson, M.W. & Looger, L.L. A bright and photostable photoconvertible fluorescent protein. Nat. Methods 2009, 6, 131-133. 69. McEvoy, A.L. et al. mMaple: a photoconvertible fluorescent protein for use in mul tiple imaging modalities. PLoS One 2012, 7, e51314. 70. Ando, R., Mizuno, H. & Miyawaki, A. Regulated fast nucleocytoplasmic shuttling observed by reversible protein highlighting. Science 2004, 306, 1370-1373. 71. Andresen, M. et al. Photoswitchable fluorescent proteins enable monochromatic multilabel imaging a nd dual color fluorescence nanoscopy. Nat. Biotechnol. 2008, 26, 1035-1040. 72. Brakemann, T. et al. A reversibly photoswitchable GFP -like protein with fluorescence excitation decoupled from switching. Nat. Biotechnol. 2011, 29, 942 -947. 73. Tiwari, D.K. e t al. A fast - and positively photoswitchable fluorescent protein for ultralow -laser -power RESOLFT nanoscopy. Nat. Methods 2015, 12, 515-518. 74. Acharya, A. et al. Photoinduced Chemistry in Fluorescent Proteins: Curse or Blessing? Chem. Rev. 2017, 117, 758-795. 75. Zhang, J., Campbell, R.E., Ting, A.Y. & Tsien, R.Y. Creating new fluorescent probes for cell biology. Nat. Rev. Mol. Cell Biol. 2002, 3, 906-918. 76. Shaner, N.C., Steinbach, P.A. & Tsien, R.Y. A guide to choosing fluorescent proteins. Nat. Metho ds 2005, 2, 905-909. 77. Megason, S.G. & Fraser, S.E. Imaging in systems biology. Cell 2007, 130, 784 -795. 78. Balla, T. Green light to illuminate signal transduction events. Trends Cell Biol. 2009, 19, 575-586. 79. Wu, X. et al. In vivo and in situ tracki ng cancer chemotherapy by highly photostable NIR fluorescent theranostic prodrug. J. Am. Chem. Soc. 2014, 136, 3579-3588. 80. Chen, Z. et al. Highly accurate fluorogenic DNA sequencing with information theory -based error correction. Nat. Biotechnol. 2017, 35, 1170-1178. 70 81. Houtsmuller, A.B. in Fluorescence Recovery after Photobleaching: Application to Nuclear Proteins. Microscopy Techniques . (ed. J. Rietdorf), 177 -199 (Springer Berlin Heidelberg; Berlin, Heidelberg; 2005). 82. Fritzsche, M. & Charras, G. D issecting protein reaction dynamics in living cells by fluorescence recovery after photobleaching. Nat. Protoc. 2015, 10, 660-680. 83. Lippincott -Schwartz, J., Altan -Bonnet, N. & Patterson, G.H. Photobleaching and photoactivation: following protein dynamic s in living cells. Nat. Cell Biol. 2003, Suppl, S7 -14. 84. Dunn, G.A., Dobbie, I.M., Monypenny, J., Holt, M.R. & Zicha, D. Fluorescence localization after photobleaching (FLAP): a new method for studying protein dynamics in living cells. J. Microsc. 2002, 205, 109-112. 85. Ishikawa -Ankerhold, H.C., Ankerhold, R. & Drummen, G.P.C. Advanced Fluorescence Microscopy Techniques ÑFRAP, FLIP, FLAP, FRET and FLIM. Molecules 2012, 17, 4047-4132. 86. Terskikh, A. "Fluorescent Timer": Protein That Changes Color with Time. Science 2000, 290, 1585-1588. 87. Subach, F.V. et al. M onomeric fluorescent timers that change color from blue to red report on cellular trafficking. Nat. Chem. Biol. 2009, 5, 118-126. 88. Fırster, T. Zwischenmolekulare Energiewanderung und Fluoreszenz. Ann. Phys. 1948, 437, 55-75. 89. Miyawaki, A. et al. Fluo rescent indicators for Ca2+ based on green fluorescent proteins and calmodulin. Nature 1997, 388, 882-887. 90. Piston, D.W. & Kremers, G.J. Fluorescent protein FRET: the good, the bad and the ugly. Trends Biochem. Sci 2007, 32, 407-414. 91. Van Munster, E. B. & Gadella, T.W., Jr. ( FLIM: a new method to avoid aliasing in frequency -domain fluorescence lifetime imaging microscopy. J. Microsc. 2004, 213, 29-38. 92. Kim, S.A., Heinze, K.G., Waxham, M.N. & Schwille, P. Intracellular calmodulin availability accesse d with two -photon cross -correlation. Proc. Natl. Acad. Sci. U. S. A. 2004, 101, 105-110. 93. Baudendistel, N., Muller, G., Waldeck, W., Angel, P. & Langowski, J. Two -hybrid fluorescence cross -correlation spectroscopy detects protein -protein interactions in vivo. ChemPhysChem 2005, 6, 984-990. 71 94. Bacia, K., Kim, S.A. & Schwille, P. Fluorescence cross -correlation spectroscopy in living cells. Nat. Methods 2006, 3, 83-89. 95. Mo, G.C.H. et al. Genetically encoded biosensors for visualizing live -cell biochemical activity at super -resolution. Nat. Methods 2017, 14, 427-434. 96. Kerppola, T.K. Visualization of molecular interactions by fluorescence complementation. Nat. Rev. Mol. Cell Biol. 2006, 7, 449-456. 97. Kerppola, T.K. Design and implementation of bimolecular fluorescence complementation (BiFC) assays for the visualization of protein interactions in living cells. Nat. Protoc. 2006, 1, 1278-1286. 98. Michnick, S.W., Ear, P.H., Manderson, E.N., Remy, I. & Stefan, E. Universal strategies in research and drug discovery based on protein -fragment complementation assays. Nat. Rev. Drug Discovery 2007, 6, 569-582. 99. Morell, M., Espargaro, A., Aviles, F.X. & Ventura, S. Study and selection of in vivo protein interactions by coupling bimolecular fluorescence complementation and flow cytometry. Nat. Protoc. 2008, 3, 22-33. 100. Hanson, G.T. et al. Green Fluorescent Protein Variants as Ratiometric Dual Emission pH Sensors. 1. Structural Characterization and Preliminary Application€. Biochemistry 2002, 41, 15477-15488. 101. McAnaney, T.B., Park, E.S., Hanson, G.T., Remington, S.J. & Boxer, S.G. Green Fluorescent Protein Variants as Ratiometric Dual Emission pH Sen sors. 2. Excited -State Dynamics. Biochemistry 2002, 41, 15489-15494. 102. Bizzarri, R. et al. Development of a novel GFP -based ratiometric excitation and emission pH indicator for intracellular studies. Biophys. J. 2006, 90, 3300-3314. 103. Bagar, T., Alte nbach, K., Read, N.D. & Bencina, M. Live -Cell imaging and measurement of intracellular pH in filamentous fungi using a genetically encoded ratiometric probe. Eukaryot. Cell 2009, 8, 703-712. 104. Zhang, X. et al. Highly photostable, reversibly photoswitcha ble fluorescent protein with high contrast ratio for live -cell superresolution microscopy. Proc. Natl. Acad. Sci. U. S. A. 2016, 113, 10364-10369. 105. Tao, R. et al. Genetically encoded fluorescent sensors reveal dynamic regulation of NADPH metabolism. Nat. Methods 2017, 14, 720-728. 106. Chan, J., Dodani, S.C. & Chang, C.J. Reaction -based small -molecule fluorescent probes for chemoselective bioimaging. Nat. Chem. 2012, 4, 973-984. 72 107. Mohr, G.J. Chromo - and fluororeactands: indicators for detection of ne utral analytes by using reversible covalent -bond chemistry. Chem. Eur. J. 2004, 10, 1082-1090. 108. Wu, J. et al. Chromogenic/Fluorogenic Ensemble Chemosensing Systems. Chem. Rev. 2015, 115, 7893-7943. 109. You, L., Zha, D. & Anslyn, E.V. Recent Advances i n Supramolecular Analytical Chemistry Using Optical Sensing. Chem. Rev. 2015, 115, 7840-7892. 110. Hettie, K.S., Klockow, J.L. & Glass, T.E. Three -input logic gates with potential applications for neuronal imaging. J. Am. Chem. Soc. 2014, 136, 4877-4880. 111. Hammers, M.D. et al. A Bright Fluorescent Probe for H2S Enables Analyte -Responsive, 3D Imaging in Live Zebrafish Using Light Sheet Fluorescence Microscopy. J. Am. Chem. Soc. 2015, 137, 10216-10223. 112. Hu, J.J. et al. Fluorescent Probe HKSOX -1 for Ima ging and Detection of Endogenous Superoxide in Live Cells and In Vivo. J. Am. Chem. Soc. 2015, 137, 6837-6843. 113. Cho, D.G. & Sessler, J.L. Modern reaction -based indicator systems. Chem. Soc. Rev. 2009, 38, 1647-1662. 114. Li, X., Gao, X., Shi, W. & Ma, H. Design strategies for water -soluble small molecular chromogenic and fluorogenic probes. Chem. Rev. 2014, 114, 590-659. 115. Tang, Y. et al. Development of fluorescent probes based on protection -deprotection of the key func tional groups for biological imaging. Chem. Soc. Rev. 2015, 44, 5003-5015. 116. Madsen, J. et al. Nile Blue -based nanosized pH sensors for simultaneous far -red and near-infrared live bioimaging. J. Am. Chem. Soc. 2013, 135, 14863-14870. 117. Wu, L., Li, X. , Huang, C. & Jia, N. Dual -Modal Colorimetric/Fluorescence Molecular Probe for Ratiometric Sensing of pH and Its Application. Anal. Chem. 2016, 88, 8332-8338. 118. Sun, Y.Q. et al. A mitochondria -targetable fluorescent probe for dual -channel NO imaging ass isted by intracellular cysteine and glutathione. J. Am. Chem. Soc. 2014, 136, 12520-12523. 119. Brewer, T.F. & Chang, C.J. An Aza -Cope Reactivity -Based Fluorescent Probe for Imaging Formaldehyde in Living Cells. J. Am. Chem. Soc. 2015, 137, 10886 -10889. 73 120. Roth, A., Li, H., Anorma, C. & Chan, J. A Reaction -Based Fluorescent Probe for Imaging of Formaldehyde in Living Cells. J. Am. Chem. Soc. 2015, 137, 10890 -10893. 121. Umezawa, K., Yoshida, M., Kamiya, M., Yamasoba, T. & Urano, Y. Rational design of reve rsible fluorescent probes for live -cell imaging and quantification of fast glutathione dynamics. Nat. Chem. 2017, 9, 279-286. 122. Thomas, S.W., 3rd, Joly, G.D. & Swager, T.M. Chemical sensors based on amplifying fluorescent conjugated polymers. Chem. Rev. 2007, 107, 1339-1386. 123. Feng, X., Liu, L., Wang, S. & Zhu, D. Water -soluble fluorescent conjugated polymers and their interacti ons with biomacromolecules for sensitive biosensors. Chem. Soc. Rev. 2010, 39, 2411-2419. 124. Kim, H.N., Guo, Z., Zhu, W., Yoon, J. & Tian, H. Recent progress on polymer -based fluorescent and colorimetric chemosensors. Chem. Soc. Rev. 2011, 40, 79 -93. 125. Liu, H.Y. et al. Quinoxaline -Based Polymer Dots with Ultrabright Red to Near -Infrared Fluorescence for In Vivo Biological Imaging. J. Am. Chem. Soc. 2015, 137, 10420-10429. 126. Lee, M.H., Kim, J.S. & Sessler, J.L. Small molecule -based ratiometric fluore scence probes for cations, anions, and biomolecules. Chem. Soc. Rev. 2015, 44, 4185 -4191. 127. Peterka, D.S., Takahashi, H. & Yuste, R. Imaging voltage in neurons. Neuron 2011, 69, 9-21. 128. Miller, E.W. Small molecule fluorescent voltage indicators for studying membrane potential. Curr. Opin. Chem. Biol. 2016, 33, 74-80. 129. Miller, E.W. et al. Optically monitoring voltage in neurons by photo -induced electron transfer through molecular wires . Proc. Natl. Acad. Sci. U. S. A. 2012, 109, 2114-2119. 130. Woodford, C.R. et al. Improved PeT molecules for optically sensing voltage in neurons. J. Am. Chem. Soc. 2015, 137, 1817-1824. 131. Huang, Y.L., Walker, A.S. & Miller, E.W. A Photostable Silicon Rhodamine Platform for Optical Voltage Sensing. J. Am. Chem. Soc. 2015, 137, 10767 -10776. 132. Stabley, D.R., Jurchenko, C., Marshall, S.S. & Salaita, K.S. Visualizing mechanical tension across membrane receptors with a fluorescent sensor. Nat. Methods 2011, 9, 64-67. 74 133. Blakely, B.L. et al. A DNA -based molecular probe for optically reporting cellular traction forces. Nat. Methods 2014, 11, 1229-1232. 134. Chang, Y. et al. A General Approach for Generating Fluorescent Probes to Visualize Piconewton Forces at the Cell Surface. J. Am. Chem. Soc. 2016, 138, 2901-2904. 135. Reisch, A. & Klymchenko, A.S. Fluorescent Polymer Nanoparticles Based on Dyes: Seeking Brighter Tools for Bioimaging. Small 2016, 12, 1968-1992. 136. Wolfbeis, O.S. An overview of nanoparti cles commonly used in fluorescent bioimaging. Chem. Soc. Rev. 2015, 44, 4743-4768. 137. Peng, H.S. & Chiu, D.T. Soft fluorescent nanomaterials for biological and biomedical imaging. Chem. Soc. Rev. 2015, 44, 4699-4722. 138. Qu, J.B. et al. Inherently fluor escent polystyrene microspheres for coating, sensing and cellular imaging. Colloids Surf., B 2017, 152, 475-481. 139. Trofymchuk, K., Reisch, A., Shulov, I., Mely, Y. & Klymchenko, A.S. Tuning the color and photostability of perylene diimides inside polyme r nanoparticles: towards biodegradable substitutes of quantum dots. Nanoscale 2014, 6, 12934-12942. 140. Ding, D., Li, K., Liu, B. & Tang, B.Z. Bioprobes based on AIE fluorogens. Acc. Chem. Res. 2013, 46, 2441-2453. 141. Wagh, A. et al. Polymeric nanoparticles with sequential and multiple FRET cascade mechanisms for multicolor and multiplexed imaging. Small 2013, 9, 2129 -2139. 142. Andreiuk, B. et al. Fluorescent Polymer Nanoparticles for Cell Barcoding In Vitro and In Viv o. Small 2017, 13, 1701582. 143. le Masne de Chermont, Q. et al. Nanoprobes with near -infrared persistent luminescence for in vivo imaging. Proc. Natl. Acad. Sci. U. S. A. 2007, 104, 9266 -9271. 144. Miao, Q. et al. Molecular afterglow imaging with bright, biodegradable polymer nanoparticles. Nat. Biotechnol. 2017, 35, 1102-1110. 145. Zheng, J. et al. Rationally designed molecular beacons for bioanalytical and biomedical applications. Chem. Soc. Rev. 2015, 44, 3036-3055. 146. Paige, J.S., Wu, K.Y. & Jaffrey, S.R. RNA mimics of green fluorescent protein. Science 2011, 333, 642-646. 75 147. Paige, J.S., Nguyen -Duc, T., Song, W. & Jaffrey, S.R. Fluorescence imaging of cellular metabolites with RNA. Science 2012, 335, 1194. 148. Song, W., Strack, R.L. & Jaffrey, S.R . Imaging bacterial protein expression using genetically encoded RNA sensors. Nat. Methods 2013, 10, 873. 149. Strack, R.L., Disney, M.D. & Jaffrey, S.R. A superfolding Spinach2 reveals the dynamic nature of trinucleotide repeat -containing RNA. Nat. Methods 2013, 10, 1219-1224. 150. Song, W., Strack, R.L., Svensen, N. & Jaffrey, S.R. Plug -and-play fluorophores exte nd the spectral properties of Spinach. J. Am. Chem. Soc. 2014, 136, 1198 -1201. 151. Strack, R.L., Song, W. & Jaffrey, S.R. Using Spinach -based sensors for fluorescence imaging of intracellular metabolites and proteins in living bacteria. Nat. Protoc. 2014, 9, 146-155. 152. Ding, C. 'Other' applications of single nucleotide polymorphisms. Trends Biotechnol. 2007, 25, 279-283. 153. Mellis, I.A., Gupte, R., Raj, A. & Rouhanifard, S.H. Visualizing adenosine -to-inosine RNA editing in single mammalian cells. Nat. Methods 2017, 14, 801-804. 154. Specht, E.A., Braselmann, E. & Palmer, A.E. A Critical and Comparative Review of Fluorescent Tools for Live -Cell Imaging. Annu. Rev. Physiol. 2017, 79, 93-117. 155. Kaiser, P.D., Maier, J., Traenkle, B., Emele, F. & Rothbauer, U. Recent progress in generating intracellular functional antibody fragments to target and trace cellular components in living cells. Biochim. Biophys. Acta 2014, 1844, 1933-1942. 156. Sato, Y. et al. Genetically encoded system to track histone modification in vivo. Sci. Rep. 2013, 3, 2436. 157. Stasevich, T.J. et al. Regulation of RNA polymerase II activation by histone acetylation in single living cells. Nature 2014, 516, 272-275. 158. Ntziachr istos, V. Going deeper than microscopy: the optical imaging frontier in biology. Nat. Methods 2010, 7, 603-614. 159. Fernandez -Suarez, M. & Ting, A.Y. Fluorescent probes for super -resolution imaging in living cells. Nat. Rev. Mol. Cell Biol. 2008, 9, 929-943. 160. Ntziachristos, V., Tung, C.H., Bremer, C. & Weissleder, R. Fluorescence molecular tomography resolves protease activity in vivo. Nat. Med. 2002, 8, 757-760. 76 161. Olson, E.S. et al. Activatable cell penetrating peptides linked to nanoparticles as dual probes for in vivo fluorescence and MR imaging of proteases. Proc. Natl. Acad. Sci. U. S. A. 2010, 107, 4311-4316. 162. Chi, C. et al. Intraoperative imaging -guided cancer surgery: from current fluorescence molecular imaging methods to future multi -mod ality imaging technology. Theranostics 2014, 4, 1072-1084. 163. Ni, D. et al. Dual -targeting upconversion nanoprobes across the blood -brain barrier for magnetic resonance/fluorescence imaging of intracranial glioblastoma. ACS Nano 2014, 8, 1231-1242. 164. Xie, J. et al. PET/NIRF/MRI triple functional iron oxide nanoparticles. Biomaterials 2010, 31, 3016-3022. 165. Nguyen, Q.T. et al. Surgery with molecular fluorescence imaging using activatable cell -penetrating peptides decreases residual cancer and improve s survival. Proc. Natl. Acad. Sci. U. S. A. 2010, 107, 4317-4322. 166. Nguyen, Q.T. & Tsien, R.Y. Fluorescence -guided surgery with live molecular navigation --a new cutting edge. Nat. Rev. Cancer 2013, 13, 653-662. 167. Vahrmeijer, A.L., Hutteman, M., van d er Vorst, J.R., van de Velde, C.J. & Frangioni, J.V. Image -guided cancer surgery using near -infrared fluorescence. Nat. Rev. Clin. Oncol. 2013, 10, 507-518. 168. Perlitz, C. et al. Comparison of two tricarbocyanine -based dyes for fluorescence optical imagi ng. J. Fluoresc. 2005, 15, 443-454. 169. van de Ven, S. et al. A novel fluorescent imaging agent for diffuse optical tomography of the breast: first clinical experience in patients. Mol. Imaging Biol. 2010, 12, 343-348. 170. Schaafsma, B.E. et al. The clinical use of indocyanine green as a near -infrared fluorescent contrast agent for image -guided oncologic surgery. J. Surg. Oncol. 2011, 104, 323-332. 171. van Dam, G.M. et al. Intraoperative tumor -specific fluorescence imaging in ovarian cancer by folate receptor -alpha targeting: first in -human res ults. Nat. Med. 2011, 17, 1315-1319. 172. Jafari, M.D. et al. The use of indocyanine green fluorescence to assess anastomotic perfusion during robotic assisted laparoscopic rectal surgery. Surg. Endosc. 2013, 27, 3003-3008. 77 173. van der Vorst, J.R. et al. Near-infrared fluorescence -guided resection of colorectal liver metastases. Cancer 2013, 119, 3411-3418. 174. Jewell, E.L. et al. Detection of sentinel lymph nodes in minimally invasive surgery using indocyanine green and near -infrared fluorescence imaging for uterine and cervical malignancies. Gynecol. Oncol. 2014, 133, 274-277. 175. Bulina, M.E. et al. Chromophore -assisted light inactivation (CALI) using the phototoxic fluorescent protein KillerRed. Nat. Protoc. 2006, 1, 947-953. 176. Hoffman oÀKim, D., Diefenbach, T.J., Eustace, B.K. & Jay, D.G. in Chromophore oÀAssisted Laser Inactivation. Methods in Cell Biology . Vol. 82, 335 -354 (Academic Press; 2007). 177. He, J. et al. A genetically targetable near -infrared photosensitizer. Nat. Methods 2016, 13, 263-268. 178. Jung, H.S. et al. A Mitochondria -Targeted Cryptocyanine -Based Photothermogenic Photosensitizer. J. Am. Chem. Soc. 2017, 139, 9972-9978. 179. Glantz, S.T. et al. Functional and topological diversity of LOV domain photoreceptors. Proc. Natl. Acad. Sci. U. S. A. 2016, 113, E1442 -1451. 180. Kaberniuk, A.A., Shemetov, A.A. & Verkhusha, V.V. A bacterial phytochrome -based optogenetic system controllable with near -infrared light. Nat. Methods 2016, 13, 591-597. 181. Repina, N.A., Rosenbloom, A., Mukherjee, A., Schaffer, D.V. & Kane, R.S. At Light Speed: Advances in Optogenetic Systems for Regulating Cell Signaling and Behavior. Annu. Rev. Chem. Biomol. Eng. 2017, 8, 13-39. 182. Rost, B.R., Schneider -Warme, F., Schmitz , D. & Hegemann, P. Optogenetic Tools for Subcellular Applications in Neuroscience. Neuron 2017, 96, 572-603. 183. Frenette, M. et al. Shining light on the dark side of imaging: excited state absorption enhancement of a bis -styryl BODIPY photoacoustic cont rast agent. J. Am. Chem. Soc. 2014, 136, 15853-15856. 184. Yao, J. et al. Multiscale photoacoustic tomography using reversibly switchable bacterial phytochrome as a near -infrared photochromic probe. Nat. Methods 2015, 13, 67-73. 185. Reinhardt, C.J., Zhou, E.Y., Jorgensen, M.D., Partipilo, G. & Chan, J. A Ratiometric Acoustogenic Probe for in Vivo Imaging of Endogenous Nitric Oxide. J. Am. Chem. Soc. 2018, 140, 1011-1018. 78 186. Gaj, T., Gersbach, C.A. & Barbas, C.F., 3rd ZFN, TALEN, and CRISPR/Cas -based meth ods for genome engineering. Trends Biotechnol. 2013, 31, 397-405. 187. Pennisi, E. The CRISPR craze. Science 2013, 341, 833-836. 188. Vendrell, M., Zhai, D., Er, J.C. & Chang, Y.T. Combinatorial strategies in fluorescent probe development. Chem. Rev. 2012, 112, 4391-4420. 189. de Moliner, F., Kielland, N., Lavilla, R. & Vendrell, M. Modern Synthetic Avenues for the Preparation of Functional Fluorophores. Angew. Chem. Int. Ed. 2017, 56, 3758-3769. 190. Xiong, X. et al. Thermally activated delayed fluorescenc e of fluorescein derivative for time -resolved and confocal fluorescence imaging. J. Am. Chem. Soc. 2014, 136, 9590-9597. 191. Grimm, J.B. et al. A general method to improve fluorophores for live -cell and single -molecule microscopy. Nat. Methods 2015, 12, 244-250. 192. Kawasumi, K. et al. Thermally Activated Delayed Fluorescence Materials Based on Homoconjugation Effect of Donor -Acceptor Triptycenes. J. Am. Chem. Soc. 2015, 137, 11908-11911. 193. Cheng, Y. et al. Unparalleled Ease of Access to a Library of B iheteroaryl Fluorophores via Oxidative Cross -Coupling Reactions: Discovery of Photostable NIR Probe for Mitochondria. J. Am. Chem. Soc. 2016, 138, 4730-4738. 194. Grimm, J.B. et al. A general method to fine -tune fluorophores for live -cell and in vivo imagi ng. Nat. Methods 2017, 14, 987-994. 195. Marks, K.M. & Nolan, G.P. Chemical labeling strategies for cell biology. Nat. Methods 2006, 3, 591-596. 196. Sunbul, M. & Yin, J. Site specific protein labeling by enzymatic posttranslational modification. Org. Biomol. Chem. 2009, 7, 3361-3371. 197. Crivat, G. & Taraska, J.W. Imaging proteins inside cells with fluorescent tags. Trends Biotechnol. 2012, 30, 8-16. 198. Dean, K.M. & Palmer, A.E. Advances in fluorescence labeling strategies for dynamic cellular imaging. Nat. Chem. Biol. 2014, 10, 512-523. 199. Rashidian, M., Dozier, J.K. & Distefano, M.D. Enzymatic labeling of proteins: techniques and approaches. Bioc onjugate Chem. 2013, 24, 1277-1294. 200. Griffin, M., Casadio, R. & Bergamini, C.M. Transglutaminases: nature's biological glues. Biochem. J 2002, 368, 377-396. 79 201. Lin, C.W. & Ting, A.Y. Transglutaminase -catalyzed site -specific conjugation of small -molec ule probes to proteins in vitro and on the surface of living cells. J. Am. Chem. Soc. 2006, 128, 4542-4543. 202. Tominaga, J. et al. An enzymatic method for site -specific labeling of recombinant proteins with oligonucleotides. Chem. Commun. 2007, 401-403. 203. Oteng-Pabi, S.K., Pardin, C., Stoica, M. & Keillor, J.W. Site -specific protein labelling and immobilization mediated by microbial transglutaminase. Chem. Commun. 2014, 50, 6604-6606. 204. Popp, M.W., Antos, J.M., Grotenbreg, G.M., Spooner, E. & Ploegh , H.L. Sortagging: a versatile method for protein labeling. Nat. Chem. Biol. 2007, 3, 707 -708. 205. Tanaka, T., Yamamoto, T., Tsukiji, S. & Nagamune, T. Site -specific protein modification on living cells catalyzed by Sortase. ChemBioChem 2008, 9, 802 -807. 206. Antos, J.M. et al. Site -specific N - and C -terminal labeling of a single polypeptide using sortases of different specificity. J. Am. Chem. Soc. 2009, 131, 10800 -10801. 207. Guimaraes, C.P. et al. Site -specific C -terminal and internal loop labeling of p roteins using sortase -mediated reactions. Nat. Protoc. 2013, 8, 1787-1799. 208. Duckworth, B.P., Zhang, Z., Hosokawa, A. & Distefano, M.D. Selective labeling of proteins by using protein farnesyltransferase. ChemBioChem 2007, 8, 98-105. 209. Chen, Y.X., Tr iola, G. & Waldmann, H. Bioorthogonal chemistry for site -specific labeling and surface immobilization of proteins. Acc. Chem. Res. 2011, 44, 762 -773. 210. Chen, I., Howarth, M., Lin, W. & Ting, A.Y. Site -specific labeling of cell surface proteins with biophysical probes using biotin ligase. Nat. Methods 2005, 2, 99. 211. Howarth, M., Takao, K., Hayashi, Y. & Ting, A.Y. Targeting quantum dots to surface prote ins in living cells with biotin ligase. Proc. Natl. Acad. Sci. U. S. A. 2005, 102, 7583-7588. 212. Howarth, M. & Ting, A.Y. Imaging proteins in live mammalian cells with biotin ligase and monovalent streptavidin. Nat. Protoc. 2008, 3, 534-545. 213. Heal, W .P. et al. Site -specific N -terminal labelling of proteins in vitro and in vivo using N -myristoyl transferase and bioorthogonal ligation chemistry. Chem. Commun. 2008, 480-482. 80 214. Heal, W.P., Wickramasinghe, S.R., Leatherbarrow, R.J. & Tate, E.W. N -Myrist oyl transferase -mediated protein labelling in vivo. Org. Biomol. Chem. 2008, 6, 2308 -2315. 215. Heal, W.P., Wright, M.H., Thinon, E. & Tate, E.W. Multifunctional protein labeling via enzymatic N -terminal tagging and elaboration by click chemistry. Nat. Pro toc. 2011, 7, 105-117. 216. Carrico, I.S., Carlson, B.L. & Bertozzi, C.R. Introducing genetically encoded aldehydes into proteins. Nat. Chem. Biol. 2007, 3, 321-322. 217. Rush, J.S. & Bertozzi, C.R. New aldehyde tag sequences identified by screening formyl glycine generating enzymes in vitro and in vivo. J. Am. Chem. Soc. 2008, 130, 12240-12241. 218. Liu, J. et al. An Efficient Site -Specific Method for Irreversible Covalent Labeling of Proteins with a Fluorophore. Sci. Rep. 2015, 5, 16883. 219. Yin, J., Liu, F., Li, X. & Walsh, C.T. Labeling proteins with small molecules by site -specific posttranslational modification. J. Am. Chem. Soc. 2004, 126, 7754-7755. 220. Yin, J., Lin, A.J., Golan, D.E. & Walsh, C.T. Site -specific protein labeling by Sfp phosphopantet heinyl transferase. Nat. Protoc. 2006, 1, 280-285. 221. Keppler, A. et al. A general method for the covalent labeling of fusion proteins with small molecules in vivo. Nat. Biotechnol. 2003, 21, 86-89. 222. Juillerat, A. et al. Directed evolution of O6 -alky lguanine-DNA alkyltransferase for efficient labeling of fusion proteins with small molecules in vivo. Chem. Biol. 2003, 10, 313-317. 223. Gautier, A. et al. An engineered protein tag for multiprotein labeling in living cells. Chem. Biol. 2008, 15, 128-136. 224. Maurel, D. et al. Cell -surface protein -protein interaction analysis with time -resolved FRET and snap -tag technologies: application to GPCR oligomerization. Nat. Methods 2008, 5, 561-567. 225. Jones, S.A., Shim, S. -H., He, J. & Zhuang, X. Fast , three -dimensional super -resolution imaging of live cells. Nat. Methods 2011, 8, 499. 226. Los, G.V. et al. HaloTag: a novel protein labeling technology for cell imaging and protein analysis. ACS Chem. Biol. 2008, 3, 373-382. 227. Erhart, D. et al. Chemic al development of intracellular protein heterodimerizers. Chem. Biol. 2013, 20, 549-557. 81 228. Gallagher, S.S., Sable, J.E., Sheetz, M.P. & Cornish, V.W. An in vivo covalent TMP -tag based on proximity -induced reactivity. ACS Chem. Biol. 2009, 4, 547 -556. 229. Chen, Z., Jing, C., Gallagher, S.S., Sheetz, M.P. & Cornish, V.W. Second -generation covalent TMP -tag for live cell imaging. J. Am. Chem. Soc. 2012, 134, 13692-13699. 230. Wombacher, R. et al. Live -cell super -resolution imaging with trimethoprim conjugates. Nat. Methods 2010, 7, 717-719. 231. Jing, C. & Cornish, V.W. A fluorogenic TMP -tag for high signal -to-background intracellular live cell imaging. ACS Chem. Biol. 2013, 8, 1704-1712. 232. Liu, W. et al. A rapid and fluorogenic TMP -AcBOPDIPY probe for covalent labeling of proteins in live cells. J. Am. Chem. Soc. 2014, 136, 4468-4471. 233. Hori, Y. & Kikuchi, K. Protein labeling with fluorogenic probes for no -wash live -cell imaging of proteins. Curr. Opin. Chem. Biol. 2013, 17, 644-650. 234. Hori, Y., Ueno, H., Mizukami, S. & Kikuchi, K. Photoactive yellow protein -based protein labeling system with turn -on fluorescence intensity. J. Am. Chem. Soc. 2009, 131, 16610-16611. 235. Hori, Y. et al. Development of fluorogenic probes for quick no -wash live -cell imaging of intracellular proteins. J. Am. Chem. Soc. 2013, 135, 12360-12365. 236. Hori, Y., Hirayama, S., Sato, M. & Kikuchi, K. Redesign of a Fluorogenic Labeling System To Imp rove Surface Charge, Brightness, and Binding Kinetics for Imaging the Functional Localization of Bromodomains. Angew. Chem. Int. Ed. 2015, 54, 14368-14371. 237. Kamikawa, Y. et al. Design of a protein tag and fluorogenic probe with modular structure for li ve-cell imaging of intracellular proteins. Chem. Sci. 2016, 7, 308 -314. 238. Plamont, M.A. et al. Small fluorescence -activating and absorption -shifting tag for tunable protein imaging in vivo. Proc. Natl. Acad. Sci. U. S. A. 2016, 113, 497 -502. 239. Li, C. et al. Design and characterization of red fluorogenic push -pull chromophores holding great potential for bioimaging and biosensing. Org. Biomol. Chem. 2016, 14, 9253-9261. 240. Mizukami, S., Hori, Y. & Kikuchi, K. Small -molecule -based protein -labeling technology in live cell studies: probe -design concepts and applications. Acc. Ch em. Res. 2014, 47, 247-256. 82 241. Ozhalici -Unal, H. et al. A rainbow of fluoromodules: a promiscuous scFv protein binds to and activates a diverse set of fluorogenic cyanine dyes. J. Am. Chem. Soc. 2008, 130, 12620-12621. 242. Szent -Gyorgyi, C. et al. Fluor ogen-activating single -chain antibodies for imaging cell surface proteins. Nat. Biotechnol. 2008, 26, 235-240. 243. Bruchez, M.P. Dark dyes -bright complexes: fluorogenic protein labeling. Curr. Opin. Chem. Biol. 2015, 27, 18-23. 244. Yan, Q. et al. Localiz ation microscopy using noncovalent fluorogen activation by genetically encoded fluorogen -activating proteins. ChemPhysChem 2014, 15, 687-695. 245. Telmer, C.A. et al. Rapid, specific, no -wash, far -red fluorogen activation in subcellular compartments by tar geted fluorogen activating proteins. ACS Chem. Biol. 2015, 10, 1239-1246. 246. Larsen, M.B., Hu, J., Frizzell, R.A. & Watkins, S.C. Simple image -based no -wash method for quantitative detection of surface expressed CFTR. Methods 2016, 96, 40-45. 247. Naganbabu, M. et al. Multiexcitation Fluorogenic Labeling of Surface, Intracellular, and Total Protein Pools in Living Cells. Bioconjugate Chem. 2016, 27, 1525 -1531. 248. Saurabh, S., Perez, A.M., Comerci, C.J., Shapiro, L. & Moerner, W.E. Super -resolution Imagi ng of Live Bacteria Cells Using a Genetically Directed, Highly Photostable Fluoromodule. J. Am. Chem. Soc. 2016, 138, 10398-10401. 249. Sletten, E.M. & Bertozzi, C.R. From mechanism to mouse: a tale of two bioorthogonal reactions. Acc. Chem. Res. 2011, 44, 666-676. 250. Niki ), I., Kang, J.H., Girona, G.E., Aramburu, I.V. & Lemke, E.A. Labeling proteins on live mammalian cells using click chemistry. Nat. Protoc. 2015, 10, 780. 251. Sletten, E.M. & Bertozzi, C.R. Bioorthogonal chemistry: fishing for selectivi ty in a sea of functionality. Angew. Chem. Int. Ed. 2009, 48, 6974-6998. 252. Dawson, P.E., Muir, T.W., Clark -Lewis, I. & Kent, S.B. Synthesis of proteins by native chemical ligation. Science 1994, 266, 776-779. 253. Zeng, Y., Ramya, T.N., Dirksen, A., Dawson, P.E. & Paulson, J.C. High -efficiency labeling of sialylated glycoproteins on living cells. Nat. Methods 2009, 6, 207 -209. 254. Yi, L. et al. A Highly Efficient Strategy for Modification of Proteins at the C Terminus. Angew. Chem. Int. Ed. 2010, 49, 9417-9421. 83 255. Agarwal, P., van der Weijden, J., Sletten, E.M., Rabuka, D. & Be rtozzi, C.R. A Pictet -Spengler ligation for protein chemical modification. Proc. Natl. Acad. Sci. U. S. A. 2013, 110, 46-51. 256. Saxon, E. & Bertozzi, C.R. Cell surface engineering by a modified Staudinger reaction. Science 2000, 287, 2007-2010. 257. Saxon, E., Armstrong, J.I. & Bertozzi, C.R. A "traceless" Staudinger ligation for the chemoselective synthesis of amide bonds. Org. Lett. 2000, 2, 2141-2143. 258. Haldon, E., Nicasio, M.C. & Perez, P.J. Copper -catalysed azide -alkyne cycloadditions (CuAAC): an update. Org. Biomol. Chem. 2015, 13, 9528-9550. 259. Worrell, B.T., Malik, J.A. & Fokin, V.V. Direct evidence of a dinuclear copper intermediate in Cu(I) -catalyzed azide -alkyne cycloadditions. Science 2013, 340, 457-460. 260. Ziegler, M.S., Lakshmi, K. V. & Tilley, T.D. Dicopper Cu(I)Cu(I) and Cu(I)Cu(II) Complexes in Copper -Catalyzed Azide -Alkyne Cycloaddition. J. Am. Chem. Soc. 2017, 139, 5378-5386. 261. Shieh, P. et al. CalFluors: A Universal Motif for Fluorogenic Azide Probes across the Visible Spect rum. J. Am. Chem. Soc. 2015, 137, 7145-7151. 262. Agard, N.J., Prescher, J.A. & Bertozzi, C.R. A strain -promoted [3 + 2] azide -alkyne cycloaddition for covalent modification of biomolecules in living systems. J. Am. Chem. Soc. 2004, 126, 15046-15047. 263. Baskin, J.M. et al. Copper -free click chemistry for dynamic in vivo imaging. Proc. Natl. Acad. Sci. U. S. A. 2007, 104, 16793-16797. 264. Laughlin, S.T., Baskin, J.M., Amacher, S.L. & Bertozzi, C.R. In vivo imaging of membrane -associated glycans in develop ing zebrafish. Science 2008, 320, 664 -667. 265. Yao, J.Z. et al. Fluorophore targeting to cellular proteins via enzyme -mediated azide ligation and strain -promoted cycloaddition. J. Am. Chem. Soc. 2012, 134, 3720-3728. 266. Gong, Y. & Pan, L. Recent advances in bioorthogonal reactions for site -specific protein labeling and engineering. Tetrahedron Lett. 2015, 56, 2123-2132. 267. Poloukhtine, A.A., Mbua, N.E., Wolfert, M.A., Boons, G.J. & Popik, V.V. Selective labeling of livi ng cells by a photo -triggered click reaction. J. Am. Chem. Soc. 2009, 131, 15769-15776. 84 268. Ning, X. et al. Protein modification by strain -promoted alkyne -nitrone cycloaddition. Angew. Chem. Int. Ed. 2010, 49, 3065-3068. 269. van Berkel, S.S. et al. Appli cation of metal -free triazole formation in the synthesis of cyclic RGD -DTPA conjugates. ChemBioChem 2008, 9, 1805-1815. 270. Wang, Y., Vera, C.I. & Lin, Q. Convenient synthesis of highly functionalized pyrazolines via mild, photoactivated 1,3 -dipolar cyclo addition. Org. Lett. 2007, 9, 4155-4158. 271. Song, W., Wang, Y., Qu, J., Madden, M.M. & Lin, Q. A photoinducible 1,3 -dipolar cycloaddition reaction for rapid, selective modification of tetrazole -containing proteins. Angew. Chem. Int. Ed. 2008, 47, 2832-2835. 272. Blackman, M.L., Royzen, M. & Fox, J.M. Tetrazine ligation: fast bioconjugation based on inverse -electron -demand Diels -Alder reactivity. J. Am. Chem. Soc. 2008, 130, 13518-13519. 273. Devaraj, N.K. & Weissleder, R. Biomedical applications of tetraz ine cycloadditions. Acc. Chem. Res. 2011, 44, 816-827. 274. Thalhammer, F., Wallfahrer, U. & Sauer, J. Reaktivit−t einfacher offenkettiger und cyclischer dienophile bei Diels -Alder -reaktionen mit inversem elektronenbedarf. Tetrahedron Lett. 1990, 31, 6851-6854. 275. Seitchik, J.L. et al. Genetically encoded tetrazine amino acid directs rapid site -specific in vivo bioorthogonal ligation with trans -cyclooctenes. J. Am. Chem. Soc. 2012, 134, 2898-2901. 276. Lang, K. et al. Genetically encoded norbornene direct s site -specific cellular protein labelling via a rapid bioorthogonal reaction. Nat. Chem. 2012, 4, 298-304. 277. Peng, T. & Hang, H.C. Site -Specific Bioorthogonal Labeling for Fluorescence Imaging of Intracellular Proteins in Living Cells. J. Am. Chem. Soc . 2016, 138, 14423-14433. 278. Shang, X. et al. Fluorogenic protein labeling using a genetically encoded unstrained alkene. Chem. Sci. 2017, 8, 1141-1145. 279. Lin, S. et al. Redox -based reagents for chemoselective methionine bioconjugation. Science 2017, 355, 597-602. 280. Addy, P.S., Erickson, S.B., Italia, J.S. & Chatterjee, A. A Chemoselective Rapid Azo-Coupling Reaction (CRACR) for Unclickable Bioconjugation. J. Am. Chem. Soc. 2017, 139, 11670-11673. 85 281. Lee, Y.J., Kurra, Y. & Liu, W.R. Phospha -Michael Addition as a New Click Reaction for Protein Functionalization. ChemBioChem 2016, 17, 456-461. 282. Kukwikila, M., Gale, N., El -Sagheer, A.H., Brown, T. & Tavassoli, A. Assembly of a biocompatible triazole -linked gene by one -pot click -DNA ligation. Nat. Chem. 2017, 9, 1089-1098. 283. Lim, R.K. & Lin, Q. Bioorthogonal chemistry: recent progress and future directions. Chem. Commun. 2010, 46, 1589-1600. 284. Lang, K. & Chin, J.W. Cellular incorporation of unnatural amino acids and bioorthogonal labeling of proteins. Chem. Rev. 2014, 114, 4764-4806. 285. Patterson, D.M., Nazarova, L.A. & Prescher, J.A. Finding the right (bioorthogonal) chemistry. ACS Chem. Biol . 2014, 9, 592-605. 286. Thyer, R., Robotham, S.A., Brodbelt, J.S. & Ellington, A.D. Evolving tRNA(Sec) for efficient canonical incorporation of selenocysteine. J. Am. Chem. Soc. 2015, 137, 46-49. 287. Uttamapinant, C. et al. A fluorophore ligase for site -specific protein labeling inside living cells. Proc. Natl. Acad. Sci. U. S. A. 2010, 107, 10914-10919. 288. Slavoff, S.A., Liu, D.S., Cohen, J.D. & Ting, A.Y. Imaging protein -protein interactions inside living cells via interaction -dependent fluorophore li gation. J. Am. Chem. Soc. 2011, 133, 19769-19776. 289. Liu, D.S., Phipps, W.S., Loh, K.H., Howarth, M. & Ting, A.Y. Quantum dot targeting with lipoic acid ligase and HaloTag for single -molecule imaging on living cells. ACS Nano 2012, 6, 11080-11087. 290. Uttamapinant, C. et al. Site -specific protein labeling using PRIME and chelation -assisted click chemistry. Nat. Protoc. 2013, 8, 1620. 291. Hauke, S. et al. Two -step protein labeling utilizing lipoic acid ligase and Sonogashira cross -coupling. Bioconjugate Chem. 2014, 25, 1632-1637. 292. Plaks, J.G., Falatach, R., Kastantin, M., Berberich, J.A. & Kaar, J.L. Multisite clickable modification of proteins using lipoic acid ligase. Bioconjugate Chem. 2015, 26, 1104-1112. 293. Griffin, B.A., Adams, S.R. & Tsien, R .Y. Specific covalent labeling of recombinant protein molecules inside live cells. Science 1998, 281, 269-272. 294. Adams, S.R. et al. New Biarsenical Ligands and Tetracysteine Motifs for Protein Labeling in Vitro and in Vivo: Synthesis and Biological App lications. J. Am. Chem. Soc. 2002, 124, 6063-6076. 86 295. Martin, B.R., Giepmans, B.N., Adams, S.R. & Tsien, R.Y. Mammalian cell -based optimization of the biarsenical -binding tetracysteine motif for improved fluorescence and affinity. Nat. Biotechnol. 2005, 23, 1308-1314. 296. Cao, H., Chen, B., Squier, T.C. & Ma yer, M.U. CrAsH: a biarsenical multi -use affinity probe with low non -specific fluorescence. Chem. Commun. 2006, 2601 -2603. 297. Spagnuolo, C.C., Vermeij, R.J. & Jares -Erijman, E.A. Improved photostable FRET -competent biarsenical -tetracysteine probes based on fluorinated fluoresceins. J. Am. Chem. Soc. 2006, 128, 12040-12041. 298. Hoffmann, C. et al. A FlAsH -based FRET approach to determine G protein -coupled receptor activation in living cells. Nat. Methods 2005, 2, 171-176. 299. Adams, S.R. & Tsien, R.Y. Pr eparation of the membrane -permeant biarsenicals FlAsH -EDT2 and ReAsH -EDT2 for fluorescent labeling of tetracysteine -tagged proteins. Nat. Protoc. 2008, 3, 1527-1534. 300. Hoffmann, C. et al. Fluorescent labeling of tetracysteine -tagged proteins in intact cells. Nat. Protoc. 2010, 5, 1666-1677. 301. Walker, A.S., Rablen, P.R. & Schepartz, A. Rotamer -Restricted Fluorogenicity of the Bis -Arsenical ReAsH. J. Am. Chem. Soc. 2016, 138, 7143-7150. 302. Halo, T.L., Appelbaum, J., Hobert, E.M., Balkin, D.M. & Schepa rtz, A. Selective recognition of protein tetraserine motifs with a cell -permeable, pro -fluorescent bis -boronic acid. J. Am. Chem. Soc. 2009, 131, 438-439. 303. Chen, Y. et al. Coumarin -based fluorogenic probes for no -wash protein labeling. Angew. Chem. Int . Ed. 2014, 53, 13785-13788. 304. Wang, Y. et al. Development of a red fluorescent light -up probe for highly selective and sensitive detection of vicinal dithiol -containing proteins in living cells. Chem. Sci. 2016, 7, 518-524. 305. Stelzer, E.H. Better im aging through chemistry. Cell 2014, 159, 1243-1246. 306. Weissleder, R. A clearer vision for in vivo imaging. Nat. Biotechnol. 2001, 19, 316 -317. 307. Hong, G., Antaris, A.L. & Dai, H. Near -infrared fluorophores for biomedical imaging. Nat. Biomed. Eng. 2017, 1, 0010. 308. Shcherbo, D. et al. Bright far -red fluorescent protein for whole -body imaging. Nat. Methods 2007, 4, 741-746. 87 309. Subach, F.V. et al. Red fluorescent protein with reversibly photoswitchable absorbance for photochromic FRET. Chem. Biol. 2010, 17, 745-755. 310. Hense, A., Nienhaus, K. & Nienhaus, G.U. Exploring color tuning strategies in red fluorescent proteins. Photoc hem. Photobiol. Sci. 2015, 14, 200-212. 311. Ng, H.L. & Lin, M.Z. Structure -guided wavelength tuning in far -red fluorescent proteins. Curr. Opin. Struct. Biol. 2016, 39, 124-133. 312. Rockwell, N.C. & Lagarias, J.C. A brief history of phytochromes. ChemPhysChem 2010, 11, 1172-1180. 313. Shcherbakova, D.M. et al. Molecular Basis of Spectral Diversity in Near -Infrared Phytochrome -Based Fluorescent Proteins. Chem. Biol. 2015, 22, 1540-1551. 314. Filonov, G.S. et al. Bright and stable near -infrared fluorescent protein for in vivo imaging. Nat. Biotechnol. 2011, 29, 757-761. 315. Shcherbakova, D.M. et al. Bright monomeric near -infrared fluorescent proteins as tags and biosensors for multiscal e imaging. Nat. Commun. 2016, 7, 12405. 316. Chernov, K.G., Redchuk, T.A., Omelina, E.S. & Verkhusha, V.V. Near -Infrared Fluorescent Proteins, Biosensors, and Optogenetic Tools Engineered from Phytochromes. Chem. Rev. 2017, 117, 6423-6446. 317. Rodriguez, E.A. et al. A far -red fluorescent protein evolved from a cyanobacterial phycobiliprotein. Nat. Methods 2016, 13, 763-769. 318. Yuan, L., Lin, W., Zheng, K., He, L. & Huang, W. Far -red to near infrared analyte -responsive fluorescent probes based on organic fluorophore platforms for fluorescence imaging. Chem. Soc. Rev. 2013, 42, 622-661. 319. Guo, Z., Park, S., Yoon, J. & Shin, I. Recent progress in the development of near -infrared fluorescent probes for bioimaging applications. Chem. Soc. Rev. 2014, 43, 16-29. 320. Bricks, J.L., Kachkovskii, A.D., Slominskii, Y.L., Gerasov, A.O. & Popov, S.V. Molecular design of near infrared polymethine dyes: A review. Dyes Pigm. 2015, 121, 238-255. 321. Brooker, L.G.S. Absorption and Resonance in Dyes. Rev. Mod. Phys. 1942, 14, 275-293. 322. Descalzo, A.B. & Rurack, K. On the signalling pathways and Cu(II) -mediated anion indication of N -meso -substituted heptamethine cyanine dyes. Chem. Eur. J. 2009, 15, 3173-3185. 88 323. Kiyose, K. et al. Molecular design strategies for near -infrared ratiometric fluorescent probes based on the unique spectral properties of aminocyanines. Chem. Eur. J. 2009, 15, 9191-9200. 324. Redy-Keisar, O., Kisin -Finfer, E., Ferber, S., Satchi -Fainaro, R. & Shabat, D. Synthesis and use of QCy7 -derived modular probes for the detection and imaging of biologically relevant analytes. Nat. Protoc. 2013, 9, 27. 325. Michie, M.S. et al. Cyanine Conformational Restraint in the Far -Red Range. J. Am. Chem. Soc. 2017, 139, 12406-12409. 326. Peng, X. et al. Heptamethine cyanine dyes with a large stokes shift and strong fluorescence: a paradigm for excited -state intramolecular charge transfer. J. Am. Chem. Soc. 2005, 127, 4170-4171. 327. Bertolino, C.A., Caputo, G., B arolo, C., Viscardi, G. & Coluccia, S. Novel heptamethine cyanine dyes with large Stoke's shift for biological applications in the near infrared. J. Fluoresc. 2006, 16, 221-225. 328. Pascal, S. et al. Expanding the polymethine paradigm: evidence for the contribution of a bis -dipolar electronic structure. J. Phys. Chem. A 2014, 118, 4038-4047. 329. Kaur, M. & Choi, D.H. Diketopyrrolopyrrole: brilliant red pigment dye -based fluorescent probes and their applications. Chem. Soc. Rev. 2015, 44, 58-77. 330. Fisch er, G.M., Ehlers, A.P., Zumbusch, A. & Daltrozzo, E. Near -infrared dyes and fluorophores based on diketopyrrolopyrroles. Angew. Chem. Int. Ed. 2007, 46, 3750-3753. 331. Fischer, G.M., Isomaki -Krondahl, M., Gottker -Schnetmann, I., Daltrozzo, E. & Zumbusch, A. Pyrrolopyrrole cyanine dyes: a new class of near -infrared dyes and fluorophores. Chem. Eur. J. 2009, 15, 4857-4864. 332. Fischer, G.M. et al. Asymmetric PPCys: strongly fluorescing NIR labels. Chem. Commun. 2010, 46, 5289-5291. 333. Fischer, G.M., Klein , M.K., Daltrozzo, E. & Zumbusch, A. Pyrrolopyrrole Cyanines: Effect of Substituents on Optical Properties. Eur. J. Org. Chem. 2011, 2011, 3421 -3429. 334. Fischer, G.M., Daltrozzo, E. & Zumbusch, A. Selective NIR chromophores: bis(pyrrolopyrrole) cyanines. Angew. Chem. Int. Ed. 2011, 50, 1406-1409. 89 335. Shimizu, S., Iino, T., Araki, Y. & Kobayashi, N. Pyrrolopyrrole aza -BODIPY analogues: a facile synthesis and intense fluorescence. Chem. Commun. 2013, 49, 1621-1623. 336. Ulrich, G., Goeb, S., De Nicola, A., Retailleau, P. & Ziessel, R. Chemistry at boron: synthesis and properties of red to near -IR fluorescent dyes based on boron -substituted diisoindolomethene frameworks. J. Org. Chem. 2011, 76, 4489-4505. 337. Ni, Y. & Wu, J. Far -red and near infrared BODIPY dyes: synthesis and applications for fluorescent pH probes and bio -imaging. Org. Biomol. Chem. 2014, 12, 3774 -3791. 338. Chen, H. et al. A General Strategy Toward Highly Fluorogenic Bioprobes Emitting across the Visible Spectrum. J. Am. Chem. Soc. 2017, 139, 10157-10163. 339. Lu, H., Mack, J., Yang, Y. & Shen, Z. Structural modification strategies for the rational design of red/NIR region BODIPYs. Chem. Soc. Rev. 2014, 43, 4778 -4823. 340. Gresser, R., Hummert, M., Hartmann, H., Leo, K. & Riede, M. Synthesis and characterization of near -infrared absorbing benzannulated aza -BODIPY dyes. Chem. Eur. J. 2011, 17, 2939-2947. 341. Grossi, M. et al. Mechanistic insight into the formation of tetraarylazadipyrromethenes. J. Org. Chem. 2012, 77, 9304-9312. 342. Zhang, X., Yu, H. & Xiao, Y. Replacing phenyl ring with thiophene: an approach to longer wavelength aza -dipyrromethene boron dif luoride (Aza -BODIPY) dyes. J. Org. Chem. 2012, 77, 669-673. 343. Wu, D. & O'Shea, D.F. Synthesis and properties of BF2 -3,3'-dimethyldiarylazadipyrromethene near -infrared fluorophores. Org. Lett. 2013, 15, 3392-3395. 344. Burghart, A. et al. 3,5 -Diaryl -4,4-difluoro -4-bora -3a,4a-diaza -s-indacene (BODIPY) Dyes: Synthesis, Spectroscopic, Electrochemical, and Structural Properties. J. Org. Chem. 1999, 64, 7813-7819. 345. Buyukcakir, O., Bozdemir, O.A., Kolemen, S., Erbas, S. & Akkaya, E.U. Tetrastyryl -Bodipy dy es: convenient synthesis and characterization of elusive near IR fluorophores. Org. Lett. 2009, 11, 4644-4647. 346. Leen, V. et al. 1,7 -Disubstituted boron dipyrromethene (BODIPY) dyes: synthesis and spectroscopic properties. J. Org. Chem. 2011, 76, 8168-8176. 347. Dost, Z., Atilgan, S. & Akkaya, E.U. Distyryl -boradiazaindacenes: facile synthesis of novel near IR emitting fluorophores. Tetrahedron 2006, 62, 8484-8488. 90 348. Jiao, L. et al. Long wavelength red fluorescent dyes from 3,5 -diiodo-BODIPYs. Org. Biomol. Chem. 2010, 8, 2517-2519. 349. Chen, J., Burghart, A., Derecskei -Kovacs, A. & Burgess, K. 4,4 -Difluoro -4-bora -3a,4a-diaza -s-indacene (BODIPY) dyes modified for extended conjugation and restricted bond rotations. J. Org. Chem. 2000, 65, 2900-2906. 350. Jiang, X.D., Gao, R., Yue, Y., Sun, G.T. & Zhao, W. A NIR BODIPY dye bearing 3,4,4a-trihydroxanthene moieties. Org. Biomol. Chem. 2012, 10, 6861-6865. 351. Loudet, A. et al. B,O -chelated azadipyrromethenes as near -IR probes. Org. Lett. 2008, 10, 4771-4774. 352. Zhao, W. & Carreira, E.M. Conformationally Restricted Aza -BODIPY: Highly Fluorescent, Stable Near -Infrared Absorbing Dyes. Chem. Eur. J. 2006, 12, 725 4-7263. 353. Loudet, A. & Burgess, K. BODIPY dyes and their derivatives: syntheses and spectroscopic properties. Chem. Rev. 2007, 107, 4891-4932. 354. Okujima, T. et al. Synthesis of *-expanded BODIPYs and their fluorescent properties in the visible ÐnearÐinfrared region. Tetrahedron 2010, 66, 6895-6900. 355. Umezawa, K., Citterio, D. & Suzuki, K. New trends in near -infrared fluorophores for bioimaging. Anal. Sci. 2014, 30, 327-349. 356. Goeb, S. & Ziessel, R. Convenient synthesis of green diisoindolodithien ylpyrromethene -dialkynyl borane dyes. Org. Lett. 2007, 9, 737 -740. 357. Donyagina, V.F., Shimizu, S., Kobayashi, N. & Luk yanets, E.A. Synthesis of N,N -difluoroboryl complexes of 3,3 qõ-diarylazadiisoindolylmethenes. Tetrahedron Lett. 2008, 49, 6152-6154. 358. Umezawa, K., Nakamura, Y., Makino, H., Citterio, D. & Suzuki, K. Bright, color -tunable fluorescent dyes in the visible -near-infrared region. J. Am. Chem. Soc. 2008, 130, 1550-1551. 359. Umezawa, K., Matsui, A., Nakamura, Y., Citterio, D. & Suzuki, K. Bri ght, color -tunable fluorescent dyes in the Vis/NIR region: establishment of new "tailor -made" multicolor fluorophores based on borondipyrromethene. Chem. Eur. J. 2009, 15, 1096-1106. 360. Awuah, S.G., Polreis, J., Biradar, V. & You, Y. Singlet oxygen gener ation by novel NIR BODIPY dyes. Org. Lett. 2011, 13, 3884-3887. 91 361. Lu, H., Shimizu, S., Mack, J., Shen, Z. & Kobayashi, N. Synthesis and spectroscopic properties of fused -ring -expanded aza -boradiazaindacenes. Chem. Asian J. 2011, 6, 1026-1037. 362. Yu, C . et al. Isoindole -BODIPY dyes as red to near -infrared fluorophores. Chem. Eur. J. 2012, 18, 6437-6442. 363. Sarma, T., Panda, P.K. & Setsune, J. Bis -naphthobipyrrolylmethene derived BODIPY complex: an intense near -infrared fluorescent dye. Chem. Commun. 2013, 49, 9806-9808. 364. Qian, G. & Wang, Z.Y. Near -infrared organic compounds and emerging applications. Chem. Asian J. 2010, 5, 1006-1029. 365. Karpenko, I.A. et al. Squaraine as a bright, stable and environment -sensitive far -red label for receptor -specific cellular imaging. Chem. Commun. 2015, 51, 2960 -2963. 366. Sreejith, S., Carol, P., Chithra, P. & Ajayaghosh, A. Squaraine dyes: a mine of mole cular materials. J. Mater. Chem. 2008, 18, 264-274. 367. Ajayaghosh, A. Chemistry of squaraine -derived materials: near -IR dyes, low band gap systems, and cation sensors. Acc. Chem. Res. 2005, 38, 449-459. 368. Basheer, M.C. et al. Design and synthesis of s quaraine based near infrared fluorescent probes. Tetrahedron 2007, 63, 1617-1623. 369. Umezawa, K., Citterio, D. & Suzuki, K. A Squaraine -based Near -infrared Dye with Bright Fluorescence and Solvatochromic Property. Chem. Lett. 2007, 36, 1424 -1425. 370. Umezawa, K., Citterio, D. & Suzuki, K. Water -soluble NIR Fluorescent Probes Based on Squaraine and Their Application for Protein Labeling. Anal. Sci. 2008, 24, 213-217. 371. Chandrasekaran, Y., Dutta, G.K., Kanth, R.B. & Patil, S. Tetrahydroquinoxaline based squaraines: Synthesis and photophysical properties. Dyes Pigm. 2009, 83, 162-167. 372. Mayerhoffer, U., Fimmel, B. & Wurthner, F. Bright near -infrared fluorophores based on squaraines by unexpected halogen effects. Angew. Chem. Int. Ed. 2012, 51, 164-167. 373. Hu, L., Yan, Z. & Xu, H. Advances in synthesis and application of near -infrared absorbing squaraine dyes. RSC Adv. 2013, 3, 7667-7676. 92 374. Langhals, H. An unexpectedly simple NIR dye for 1.1 microm with a central mesoionic structure. Angew. Chem. In t. Ed. 2003, 42, 4286-4288. 375. Tian, M. et al. Discovery of novel dyes with absorption maxima at 1.1 microm. J. Am. Chem. Soc. 2003, 125, 348-349. 376. Kolmakov, K. et al. Far -Red Emitting Fluorescent Dyes for Optical Nanoscopy: Fluorinated Silicon -Rhodamines (SiRF Dyes) and Phosphorylated Oxazines. Chem. Eur. J. 2015, 21, 13344-13356. 377. Grimm, J.B. et al. Bright photoactivatable fluorophores for single -molecule imaging. Nat. Methods 2016, 13, 985-988. 378. Grimm, J.B., Brown, T.A., Tkachuk, A.N. & Lavis, L.D. General Synthetic Method for Si -Fluoresceins and Si -Rhodamines. ACS Cent. Sci. 2017, 3, 975-985. 379. Ikeno, T., Nagano, T. & Hanaoka, K. Silicon -substituted Xanthene Dyes and Their Unique Photophysical Properties for Fluorescent Probes. Chem. Asian J. 2017, 12, 1435-1446. 380. Liu, X. et al. Aziridinyl Fluorophores Demonstrate Bright Fluorescence and Superior Photostability by Effectively Inhibiting Twisted Intramolecular Charge Transfer. J. Am. Chem. Soc. 2016, 138, 6960-6963. 381. Kolmakov, K. et al. A Versatile Route to Red -Emitting Carbopyronine Dyes for Optical Microscopy and Nan oscopy. Eur. J. Org. Chem. 2010, 2010, 3593-3610. 382. Koide, Y., Urano, Y., Hanaoka, K., Terai, T. & Nagano, T. Evolution of group 14 rhodamines as platforms for near -infrared fluorescence probes utilizing photoinduced electron transfer. ACS Chem. Biol. 2011, 6, 600-608. 383. Koide, Y. et al. A reversible near -infrared fluorescence probe for reactive oxygen species based on Te -rhodamine. Chem. Commun. 2012, 48, 3091-3093. 384. Koide, Y. et al. Development of NIR fluorescent dyes based on Si -rhodamine for in vivo imaging. J. Am. Chem. Soc. 2012, 134, 5029-5031. 385. Wang, T. et al. Spirolactonized Si -rhodamine: a novel NIR fluorophore utilized as a platform to construct Si -rhodamine -based probes. Chem. Commun. 2012, 48, 8781-8783. 386. Lukinavicius, G. et al . A near -infrared fluorophore for live -cell super -resolution microscopy of cellular proteins. Nat. Chem. 2013, 5, 132-139. 387. Pastierik, T., Sebej, P., Medalova, J., Stacko, P. & Klan, P. Near -infrared fluorescent 9 -phenylethynylpyronin analogues for bio imaging. J. Org. Chem. 2014, 79, 3374-3382. 93 388. Jia, C. et al. An experimental and computational study on intramolecular charge transfer: a tetrathiafulvalene -fused dipyridophenazine molecule. Chem. Eur. J. 2007, 13, 3804-3812. 389. Velusamy, M., Thomas, K.R.J., Lin, J.T. & Wen, Y.S. Benzo[1,2,5]selenadiazole bridged amines: electro -optical properties. Tetrahedron Lett. 2005, 46, 7647 -7651. 390. Ono, K., Tanaka, S. & Yamashita, Y. Benzobis(thiadiazole)s Containing Hypervalent Sulfur Atoms: Novel Heterocycl es with High Electron Affinity and Short Intermolecular Contacts between Heteroatoms. Angew. Chem. Int. Ed. 1994, 33, 1977-1979. 391. Karikomi, M., Kitamura, C., Tanaka, S. & Yamashita, Y. New Narrow -Bandgap Polymer Composed of Benzobis(1,2,5 -thiadiazole) and Thiophenes. J. Am. Chem. Soc. 1995, 117, 6791-6792. 392. Qian, G. et al. Band Gap Tunable, Donor %Acceptor %Donor Charge -Transfer Heteroquinoid -Based Chromophores: Near Infrared Photoluminescence and Electroluminescence. Chem. Mater. 2008, 20, 6208-6216. 393. Qian, G. et al. Simple and Efficient Near -Infrared Orga nic Chromophores for Light -Emitting Diodes with Single Electroluminescent Emission above 1000 + nm. Adv. Mater. 2009, 21, 111-116. 394. Yao, L. et al. Highly efficient near -infrared organic light -emitting diode based on a butterfly -shaped donor -acceptor chro mophore with strong solid -state fluorescence and a large proportion of radiative excitons. Angew. Chem. Int. Ed. 2014, 53, 2119 -2123. 395. Yang, Y. et al. Near infrared organic light -emitting devices based on donor -acceptor -donor oligomers. Appl. Phys. Let t. 2008, 93, 163305. 396. Qian, G. et al. Synthesis and Application of Thiadiazoloquinoxaline -Containing Chromophores as Dopants for Efficient Near -Infrared Organic Light -Emitting Diodes. J. Phys. Chem. C 2009, 113, 1589-1595. 397. Ishi -i, T., Murakami, K. , Imai, Y. & Mataka, S. Light -harvesting and energy -transfer system based on self -assembling perylene diimide -appended hexaazatriphenylene. Org. Lett. 2005, 7, 3175-3178. 398. Secondo, P. & Fages, F. Design and synthesis of bismacrocyclic hexaazatriphenyle ne derivatives. Org. Lett. 2006, 8, 1311-1314. 399. Luo, M. et al. Rational design, synthesis, and optical properties of film -forming, near-infrared absorbing, and fluorescent chromophores with multidonors and large heterocyclic acceptors. Chem. Eur. J. 2009, 15, 8902-8908. 94 400. Antaris, A.L. et al. A small -molecule dye for NIR -II imaging. Nat. Mater. 2016, 15, 235-242. 401. Thooft, A.M., Cassaidy, K. & VanVeller, B. A Small Push -Pull Fluorophore for Turn -on Fluorescence. J. Org. Chem. 2017, 82, 8842-8847. 402. Hachiya, S., Asai, K. & Konishi, G. -i. Environment -responsive multicolor fluorescent dyes based on nitrophenyl or nitrophenylethynyl oligothiophene derivatives: correlation between fluorescence and *-conjugation length. Tetrahedron Lett. 2013, 54, 3317-3320. 403. Shaya, J., Fontaine -Vive, F., Michel, B.Y. & Burger, A. Rational Design of Push -Pull Fluorene Dyes: Synthesis and Structure -Photophysics Relationship. Chem. Eur. J. 2016, 22, 10627-10637. 404. Hammond, P.R. Laser dye DCM, its spectral properti es, synthesis and comparison with other dyes in the red. Opt. Commun. 1979, 29, 331-333. 405. Tong, H. et al. Color -tunable, aggregation -induced emission of a butterfly -shaped molecule comprising a pyran skeleton and two cholesteryl wings. J. Phys. Chem. B 2007, 111, 2000-2007. 406. Huang, X., Guo, Z., Zhu, W., Xie, Y. & Tian, H. A colorimetric and fluorescent turn -on sensor for pyrophosphate anion based on a dicyanomethylene -4H-chromene framework. Chem. Commun. 2008, 5143-5145. 407. Chen, W., Wright , B.D. & Pang, Y. Rational design of a NIR -emitting Pd(II) sensor via oxidative cyclization to form a benzoxazole ring. Chem. Commun. 2012, 48, 3824-3826. 408. Richard, J.A., Massonneau, M., Renard, P.Y. & Romieu, A. 7 -hydroxycoumarin -hemicyanine hybrids: a new class of far -red emitting fluorogenic dyes. Org. Lett. 2008, 10, 4175-4178. 409. Yuan, L., Lin, W., Yang, Y. & Chen, H. A unique class of near -infrared functional fluorescent dyes with carboxylic -acid -modulated fluorescence ON/OFF switching: rational design, synthesis, optical properties, theoretical calculations, and applications for fluorescence imaging in living animals. J. Am. Chem. Soc. 2012, 134, 1200-1211. 410. Yuan, L. et al. A unique approach to development of near -infrared fluorescent sensor s for in vivo imaging. J. Am. Chem. Soc. 2012, 134, 13510-13523. 411. Chen, H., Lin, W., Cui, H. & Jiang, W. Development of unique xanthene -cyanine fused near -infrared fluorescent fluorophores with superior chemical stability for biological fluorescence im aging. Chem. Eur. J. 2015, 21, 733-745. 95 412. Wei, Y. et al. Design of NIR Chromenylium -Cyanine Fluorophore Library for "Switch -ON" and Ratiometric Detection of Bio -Active Species In Vivo. Anal. Chem. 2016, 88, 1842-1849. 413. Nolde, F. et al. Synthesis and modification of terrylenediimides as high -performance fluorescent dyes. Chem. Eur. J. 2005, 11, 3959-3967. 414. Heek, T., Wurthner, F. & Haag, R. Synthesis and optical properties of water -soluble polyglycerol -dendronized rylene bisimide dyes. Chem. Eur. J . 2013, 19, 10911-10921. 415. Langhals, H. An approach to novel NIR dyes utilising '-effect donor groups. Dyes Pigm. 2003, 59, 109-116. 416. Jiao, C. et al. Bis -N-annulated quaterrylenebis(dicarboximide) as a new soluble and stable near -infrared dye. Org. Lett. 2009, 11, 4508-4511. 417. Wu, D. et al. A core -modified rubyrin with meso -aryl substituents and phenanthrene -fused pyrrole rings: a highly conjugated near -infrared dye and Hg2+ probe. Angew. Chem. Int. Ed. 2008, 47, 193-197. 418. Zhu, X.J., Fu, S.T., Wong, W.K., Guo, J.P. & Wong, W.Y. A near -infrared -fluorescent chemodosimeter for mercuric ion based on an expanded porphyrin. Angew. Chem. Int. Ed. 2006, 45, 3150-3154. 419. Zhu, X., Fu, S., Wong, W. -K. & Wong, W. -Y. A near -infrared fluorescent chemodosimeter for silver(I) ion based on an expanded porphyrin. Tetrahedron Lett. 2008, 49, 1843-1846. 420. Borisov, S.M., Zenkl, G. & Klimant, I. Phosphorescent platinum(II) and palladium(II) complexes with azatetrabenzoporphyrins -new red laser diode -compatible indicators for optical oxygen sensing. ACS Appl. Mater. Interfaces 2010, 2, 366 -374. 421. Barbieri, A ., Bandini, E., Monti, F., Praveen, V.K. & Armaroli, N. The Rise of Near -Infrared Emitters: Organic Dyes, Porphyrinoids, and Transition Metal Complexes. Top. Curr. Chem. 2016, 374, 47. 422. Yan, L., Zhang, Y., Xu, B. & Tian, W. Fluorescent nanoparticles ba sed on AIE fluorogens for bioimaging. Nanoscale 2016, 8, 2471-2487. 423. Yuan, Y., Xu, S., Cheng, X., Cai, X. & Liu, B. Bioorthogonal Turn -On Probe Based on Aggregation -Induced Emission Characteristics for Cancer Cell Imaging and Ablation. Angew. Chem. Int . Ed. 2016, 55, 6457-6461. 96 424. Han, G., Kim, D., Park, Y., Bouffard, J. & Kim, Y. Excimers beyond pyrene: a far -red optical proximity reporter and its application to the label -free detection of DNA. Angew. Chem. Int. Ed. 2015, 54, 3912-3916. 97 DEVELOPING DYE -HCRBPII COMPLEXES INTO NO -WASH BACKGROUND -FREE NEAR -INFRARED TAGS V IA PHOTOSWITCHING. In 0 we describe d the broad scope of a vailable FP variants that cover the entire visible spec trum . These FPs are sufficiently bright and have been applied in a variety of biomedical research fields. TsienÕs review article provides an excellent guide on how to select the optimal FP for a specific experime nt.1 Nonetheless , FPs still have certain pitfalls that may affect the physiology of the POIs or have interference on the experiments due to their suboptimal featur es. 2, 3 For examp le, the Stokes shifts of FPs are usually small. And small Stokes shifts will have significant overlap in the absorption and emission spectra, resulting in self -quenching, or the re -absorption of the emitted photons by adjacent FP molecules. Therefore, the imaging depth and overall emission brightness are severely limited. The emission wavelength of FPs is also a concern. As discussed in Section I.4, near-infrared light is more advantageous for deep tissue and whole body imaging. In the anthozoan family, DsRed derived FPs can emit into far -red but not NIR. Recently biliverdin -binding phytochrome derived NIR FPs have emerged to solve the problem, howeve r, with limited brightness. Because the fluorophore is formed either innately or upon binding with exogenous factors, its maturation time can vary from a couple of minutes to several hours, depending on the FP that is chosen. This mild -to-slow maturation time also limits their application in the real -time imaging of transient biological events. Moreover, the requirement of molecular oxygen in the fluorophore maturation of metazoan FPs (see Section I.1.1) also limits their use in some anaerobic experiment settin gs. 4 Besides, the 98 metazoan FPs share a size of ~ 240 amino acids, and the phytochrome FPs are commonly larger than 55 kDa. Alth ough extensive efforts have been put to develop monomeric FPs, their large sizes and the intrinsic oligomerizing nature lead to the concern of POI perturbation. 5 Recently, it was observed by Landgraf et al. that fused FPs will also cause the mislocalizatio n of POIs. 6 Fluorescent protein tags can be generated by introducing exogenous fluorophores into the protein binding pocket to furnish a complex tag, such as the bilirubin -binding UnaG derived from Japanese eel muscle ,7 and the biliverdin -binding NIR phytochromes (Section I.4.1). Similar strategies have been discussed in Section I.3.2. The combination of genetically addressable protein tag s with photophysically and/or photochemically tunable synth etic ligands should open a wide window for the development of novel imagi ng tags. I.5!Preliminary work towards the hCRBPII -based fluorescent tag In 2012 , our group reported the wavelength regulation of all -trans -retinal when bound to the type II human cellular retinoid binding protein (hCRBP II). 8 hCRBPII complexes with retinal through the imine formation of its active lysine 108 and the aldehyde group of retinal. The complex is protonated under physiological pH to form an iminium, or protonated Schiff base (PSB). The absorption wavelength of this re tinylidene PSB can span from 425 nm to 644 nm by mutating the highlighted key residues in Figure II-1. The ligand -protein interactions that lead to the wide wavelength span are attributed to the collective effect of electrostatic perturbation and steric restraints produced by eith er polar or bulky residues. The tight binding pocket also provides highly reserved water 99 molecules to mediate specific hydrogen -bonding networ ks that further maneuver the ligand-protein interaction. hCRBPII can also tolerate a high level of mutations and maintain its correct fold.9 In this context, by replacing all -trans -retinal with different synthetic dyes, hCRBPII can serve as a robust platform to design novel fluorescent complexes based on various fluorophore -protein interactions. Besides, hCRBPII has certain advantages over FPs and previously discussed protein -based tags. I t is 133 amino acids and about 15 KDa in weight, only half of metazoa n FPs. The ligand conjugation is not oxygen dependent . Both traits can complement the current fluorescent tagging methodologies. As a proof of concept, the merocyanine MCRA was synthesized to conjugate hCRBPI I and type II cellular retinoic acid binding protein (CRABPII), which is also a small Q108K Y19 L117 T53 K40L T51 Q4R58T29 A33 Oall-trans -retinal Lysine NHretinylidene PSB 108Figure 0-1 Crystal structure of hCRBPII KL mutant complexed with all -trans -retinal. Retinal is shown in red. The active lysine residue Q108K is shown in light green. Other key residues engineered to regulate the absorption wavelength of the bound retinylidene are sh own in teal, with oxygen atoms colored red and nitrogen atoms colored blue (PDB ID: 4EXZ). The scheme shows the iminium (PSB) formation between lysine 108 and retinal aldehyde. 100 cytosolic lipid binding protein and is a structural homologue of hCRBPII but with a deeper binding cav ity. 10 Structurally , MCRA has a polyenal tail that is similar to ret inal (highlighted blue in Figure II-2). MCRA binds to both proteins through the similar iminium formation. The formed PSB contains two nitrogens connected through a pentamethine chain that is similar to Cy5 dye (highlighted red in Figure II-2). The optic al features were examined with a variety of mutants. The use of these complexes as potential fluorescent imaging tags was demonstrated by fluorescence imaging with both E. coli and mammalian cells. With extensive protein engineering, the binding time of MCRA to hCRBPII mutants can be shortened to less than 1 minute. The optimized imagin g condition is to use 250 nM MCRA as the staining concentration for HeLa , U2OS, and COS -7 cells and incubate at 37 ¡C for 1 min ,11 then wash with DPBS to remove excess MCRA . Imaging of subcellular compartments such as nuclei and cytosol have been realized by targeting the hCRBPII mutant with signaling peptides like nucleus -localizing sequence (NES) and nucleus -Figure 0-2 Structure of MCRA and its schematic PSB formation with active lysine 132 of CRABPII. The absorption (red) and emission (blue) spectra are of the R132K:R111L (KL) mutant complex with MCRA . 101 excl usion sequence (NES). We were also able to show that the binding rate and pK a can be manipulated by the rational mutagenesis of hCRBP II.11 To this end, the concept of developing fluorescent tags based on the small cytosolic hCRBPII protein proved successf ul. hCRBPII show s potential as a robust platform for this purpose. However, in view of practical application, this is just the beginning. As represented by the CRABPII -KL mutant in Figure II-2, the absorption maxima of the MCRA complexes with either hCRBPII or CRABPII are centered around 600 nm, no matter the nature of mutations in the binding cavit y.12 Another issue is that they often show Stokes shifts smaller than 30 nm. Furthermore, incubating MCRA for a prolonged time leads to non-specific fluorescence background at the same emission wavelength. 11 This background cannot be diminished by extra washing steps and is generated from the non -specific binding of MCRA with other proteins in the cell. In this context, novel fluorescent ligand s are needed to overcome these issues and meet the requirements of real -world live cell imaging experiments. I.6!The requirements of a successful fluorescence im aging tag As an opening remark to the development of novel hCRBPII -based imaging tags, the major requirements from different perspectives can be summarized as followed. Photophysically , a successful system needs to exhibit high extinction coefficient and high fluorescence quantum yield. The product of these two determines the absolute brightness , which is the overarching goal for an imaging tag. A large Stokes shift is also desired. As previously discussed, the large Stokes shift can prevent self -absorptio n of the emitting photons and effectively increase the overall imaging depth into the cell or tissue. 102 In addition, well separated excitation and emission wavelengths can avoid crosstalk between the excitation light source and the emission detector in the m icroscope setup. The third photophysical requirement is the far -red and/or NIR wavelength s. The redder the light, the less scattering is observed in the biological tissue. Moreover, the NIR window provides the longest light penetration depth due to the min imal absorption of haemoglobins and water, as described in Section I.4. And the least energy -intense far -red/NIR excitation light can assure the minimal phototoxicity to the living cells or tissues. Together, these features can secure the highest tissue transpar ency for fluorescence imaging. Chemically, labeling specificity is the major concern. Different from genetically encoded FPs, tags requiring the supplementation of exogenous synthetic ligands always face this challenge. The off -target labeling will cause n on-specific fluorescence signals from proteins other than POIs. The other portion of non -specific fluorescence background comes from the excess dyes or fluorophores that reside in either the hydrophobic compartments or the hydrophilic environments, dependi ng on their chemical natures. To overcome this , extensive washing steps are normally required in a typical dye -staining fluorescence imaging experiment to remove the excess dyes after incubation. These time -consuming washing steps can damage the biological sample, increase the complexity of the experiment, and prevent the signal capture of transient biological events. In this context, fluorogenic ity, or the ability to activate fluorescence after labeling, are much desired to enable no-wash imaging experimen ts. 103 Another important photochemical feature of an advanced imaging tag is the Òon /offÓ switchability of the fluorescence. Extended from the fluorogenicity, if the fluorescence can be turned on and off iteratively by fast photochemical transformations, then the spatiotemporal control of the fluorescence signals can be realized. More importantly, the controllable Òon/offÓ switchability is the key requirement of the tags applied in localization -based super -resolution microscopic technologies like PALM and STOR M.13 Other important chemical features of the tag are fast labeling kinetics and high photostability. In combination with fluorogenicity, the former one enables the fast imaging of biological events. The latter one is essential for allowing the proper imag ing duration with the tag. Mentioned last but not trivial, biologically the dye should elicit least cytotoxicity and high cell permeability. The size of the tag needs to be small so as not to interfere with regular biological functionalities of the tagged POIs. hCRBPII certainly is advantageous in this perspective. I.7!Early attempts in finding fluorescent ligands for hCRBPII What is the optimal fluorophoric ligand of hCRBPII that can meet so many requirements as described in Section I.6? There is no clear answer. It is extremely challenging to predict which molecular scaffold will be fluorescent before its synthesis and examination . Even with the same scaffold, different substituents can lead to drastically different fate of fluorescence. Very recently, the Japan OLED firm Kyulux has secured the license of HarvardÕs machine learning AI Molecular Space Shuttle for the high throughput virtual screening of potential chemical candidates for thermally activated delayed fluorescence (TADF). This AI, developed by a collaborative team between Harvard, 104 Samsung, and MIT, co mbines the state -of-art TD -DFT calculation and machine learning, and is set up on a space of 1.6 million molecules . It can screen for candidates with optimal TADF features such as singlet -triplet energy barrier, r adiative rate, and thus E QE. 14 Unfortunately , it is not feasible to duplicate this approach for the design of suitable sets of fluorescent ligands for hCRBPII with our current computational capacity . Consider ing the shape and size of all -trans -retinal that is embedded inside hCRBPII mutants, the spatial volume of the binding cavity is tight and limit ed (Figure II-3). As described in Section I.4.2, the regular NIR dyes are large scaffolds except for polymethine cyanines. H owever, as exemplified by MCRA , the highly resonating cyanine scaffold distributes the positive charge almost homogenously alon g the chain and has the least bond alternation. The result is minimized dipole moment change when the molecule is excited from S0 to S1, thus featuring the characteristically small Stoke shift. Oall-trans -retinal Figure 0-3 Crystal structure of hCRBPII -KL (Q108K:K40L) mutant, highlighting the cavity for ligand binding (meshed cages). PDB ID: 4EXZ. The bound all -trans -retinal is deleted for clarity. 105 As an important structural feature of retinal , the vinylene group conjugate d with aldehyde can be essential for setting the correct binding trajectory. The (-extension leads to an expected red -shift , for both the absorption and emission wavelengths as long as the main scaffold is fluorescent. Keep these in mind, the first attempt of modifying the MCRA structure for increased Stokes shift was to make isoelectr onic structures of the indolinium head group of MCRA . Shown in Scheme II-1, the first serie s to be tested is the thiacarbocyanine (ThCC) MCRA analogue by substituting the indolium C3 to sulfur. ThCC-1V was synthesized by basic hydrolysis of the Vilsmeier -Haack product of 2,3 -dimethylbenzothiazolium. However, due to very weak electrophilicity of the aldehyde group, none of the regular Grignard, Wittig, and Hornor -Wadsworth -Emmons (HWE) reactions were successful in extend ing the polymethine chain. Alternatively, the condensation between 2,3 -dimethylbenzothiazolium and dianil malonaldehyde followed by basic hydrolysis affor ded ThCC-2V phenylimine with 87% purity mixed with ThCC-2V aldehyde. Complete hydrolysis and further p urification was not fruitful and only yielded degraded side products. NSONSNPhThCC-1V ThCC-2V_imine NNODiAI-1V NNODiAI-2V Scheme 0-1 Structures of thiacarbocyanines ( ThCC s) and diazaindenes ( DiAI s), with different number of vinylene insertions between the head group and the aldehyde. 106 The second series is the diazaindene -derived isoelectronic analogues of ThCC. DiAI-1V and DiAI-2V are synthesized by the HWE reaction of 1 -formyl -3-methyl [2,3a]diazai ndene followed with DIBAL reduction. These reactions yielded different ratio of Z/E isomers. Tremendous effort was invested for the purification of the final products. However, when DiAI-2V was bound to different hCRBPII mutants, the resulting absorption/emission spectra showed no significant difference s from those of MCRA . Table 0-1 Spectral features of selected DiAI -2V/hCRBPII complexes. aRelative quantum yield measurement in percentage. Referenced against Oxazine -170. bLigand -mutant binding half -time, in PBS buffer (pH 7.3) at 23 ¡C. Exponential rising fits were used to derive t 1/2 with protein concentration at 20 !M and ligand concentration of 10 !M. cn.d.: not determined. dWavelengths of absorption shoulders. eEmission maximum, excited at shoulder absorption wavelength. Entry hCRBPII mutant !abs, SB (nm) !em, SB (nm) !abs, PSB (nm) !em, PSB (nm) "PSBat1/2 b(min) 1Q108K:K40L:T51V:T53A:R58Y --5946217.1 n.d. c2Q108K:K40L:T51V:T53S:Y19W:R58W --5976199.2 5773Q108K:K40L:T51V:T53S:Y19W:R58W:T29L --5976198.1 n.d. 4Q108K:K40L:T51V:T53S:Y19W:R58W:L117E --530(500) d5975.3 5.5 5Q108K:K40L:T51V:T53S:Y19W:R58W:L117D --498(530)600n.d. n.d. 6Q108K:K40L:T51V:T53S:Y19W:R58W:L117N --548600n.d. n.d. 7Q108K:K40L:T51V:T53S:Y19W:R58W:L119E --585615n.d. n.d. 8Q108K:K40L:T51V:T53S:Y19W:R58W:V62N 436n.d. 5916187.3 n.d. 9Q108K:K40L:T51V:T53S:Y19W:R58W:V62S --5926188.4 n.d. 10Q108K:K40L:T51V:T53S:Y19W:R58W:V62E --5856175.7 4011Q108K:K40L:T51V:T53S:Y19W:R58W:V62N:L117E 443n.d. 5305983.2 n.d. 12Q108K:K40L:T51V:T53S:Y19W:R58W:V62E:L117E 443n.d. 5305972.5 n.d. 13Q108K:K40L:T51V:T53S:V62E 435578587612n.d. n.d. 14Q108K:K40L:T51V:T53S:L117E 440n.d. 563600n.d. n.d. 15Q108K:K40L:T51A:T53S:V62E 439604583(633) 611(644)en.d. n.d. 16Q108K:K40L:T51A:T53S:V62N --631(590 )64418.1 n.d. 17Q108K:K40L:T53S:V62E 438n.d. 561602n.d. n.d. 18Q108K:K40L:T51C:T53S:V62N --579613n.d. n.d. 107 Summarized in Table II-1, most of the tested mutants exhibited small Stokes shifts with nearly identical absorption and emission maxima of those of MCRA . The absorption maxima are mostly centered around 600 nm, and the Stokes shifts are about 30 ~ 50 nm. Nevertheless, special attention should be paid to some mutant complexes. When residue 117 is mutated to acidic glutamate in hCRBPII, the corresponding complexe shows a blue-shifted absorption maximum at 530 nm (entry 4 in Table II-1). Similarly, the 117D mutant (entry 5) also has the same hypsochromic effect but with the shoulder absorption becoming the maximum. The increased Stokes shift could be due to the electrostatic induction or potential hydro gen-bonding induced by 117E/D. Nonetheless, there is no further structural information for a clear conclusion. Another important observation is highlighted with entry 13 and 15. As shown , when residue 62 is mutated to glutamate, a significant portion of DiAI-2V Schiff base (SB) is formed with absorption maximum around 440 nm. Similar behavior is also observed with some V62N and/or L117E mutants. The SB emission of these two mutants are 578 nm and 604 nm, respectively. Compared with the PSB emission wavelengths centered around 610 nm, the SB emissions are too red -shifted and resemble emission from a PSB with Stokes shifts greater than 140 nm. Although currently there is no regular SB emission to compare with, it is reaso nable to hypothesize that the abnormally large Stokes shifts are rising from excited -state hydrogen bonding process of the imine nitrogen atom. Further observation and evidence will be discussed in 0. The live cell imaging performance of DiAI-2V was tested in HeLa cells. hCRBPII mutant Q108K:K40L:T51V:T53S:Y19W:R58W:L117E (KLVSWW:L117E) was chose n for 108 imaging because of its fast complex formation (t , = 5.5 min). The KLVSWW:L117E gene was cloned into an pECFP -N1 vector and inserted before the ECFP sequence (Figure II-4a). The hybrid construct was then expressed in HeLa cells by transient transfection. a) b) Figure 0-4 Live cell imaging with DiAI -2V and MCRA for performance testing. a) Clone map of hCRBPII mutant KLVSWW:L117E in ECFG -GalT vector. b) Comparison of images of DiAI -2V and MCRA complexed with KLVSWW:L117E in HeLa cells. Cells were stained with 250 nM of each dye and incubated for 10 min at 37 ¡C. Cells were washed with DPBS twice before imaging. ECFP channel: "ex = 458 nm, BP 475 -525. MCRA channel: "ex = 594 nm, LP 615. DiAI -2V channel: "ex = 514 nm, LP 560. BR: bright -field. 109 As shown in Figure II-4b, DiAI-2V was tested against MCRA . The ECFP fluor escence was used as an internal reference for the whole cell staining. However, both MCRA and DiAI-2V showed significant non -specific background after 10 min incubation. As observed in the dye channels, MCRA mostly stains mitochondria and DiAI-2V predominantly stains cell nuclei. To this end, it can be concluded that DiAI-2V is not a suitable dye for the hCRBPII imaging purpose. It is limited by the low quantum yield, low brightness, orange emission color , relative ly small Stokes shift , and most importantly, the lack of specific labeling aptitude. Thus, further structural modification and imaging optimizati on was not pursued. DDDDDADDDA!+!-DAAAM1M2M3cyanine limitmerocyaninecyaninethereforepossibletocrossthisvirtualcyaninelimit 6and,in somerarecases,tostabilizeamerocyaninedyeclosetothis idealcyaninestructure. 7Thiscyaninestructure,extremelydifÞculttoobtainfrom dissymmetricmerocyaninedyes,canbedirectlypreparedby theassociationoftwoidenticaldonoror,morerarely,acceptor end-groupsviaapolymethinechainfeaturinganoddnumber ofsp 2carbonatoms. 8Thesemoleculesspontaneouslyreachthe idealpolymethinestate (IPS)presentinguniquestructuraland spectralcharacteristics(Figure1, M2).Asaresultofthetwo degeneratedresonanceformswiththechargelocalizedatone ortheotherend-group( M1)M3),thepolymethinechainadopts anonalternatingstructurewithanaverageÒoneandahalfÓbond length(BLA )0).Thislackofbondalternationisexperimen- tallyconÞrmedbymostcrystalstructuresreportedinthe literature.8a,9Fromaspectroscopicpointofview,cyaninesin theirIPSarecharacterizedbyasharp,extremelyintense absorptionband(Figure1, M2)resultingfromthereduced vibroniccontributioninthesenonalternatingstructures.This particulartransition,calledsolitonictransitionbyanalogywith dopedpolyacetylene,isshiftedtowardtheNIRspectralrange. Thecompleterationalizationofthistransitionusingtheoretical modelsremainsamatterofdebate;inparticulartheclear assignmentofthehigh-energyshoulderremainsproblematic. 10TheIPSisconservedinthecaseofrelativelyshort !-conjugatedchains(upto9 -13carbonatoms,dependingonthestructure) butisprogressivelylostforlongerchains.Asaconsequence, uponincreasingthechainlength,thesharptransitioncharac- teristicoftheIPSundergoesabathochromicshift( !100nm peradditionalvinylicunit,Figure2,curvea)followedbya deepmodiÞcationofthespectrumwithalargebroadeningof thebandandadecreaseofitsintensity(Figure2,curveb). 11,12Thischain-lengthdependenceofthespectroscopicproperties, generallyreferredasBrookerexperiment, 13wasrationalizedÞrst byTolbertandco-workers 11invokingaÒsymmetrycollapseÓ forlong-chaincyanineelectronicstructuresduetoaPeierls- typedistortion.Insuchacase,thecyanineisnolonger symmetricwhichmeansithaslostitsIPS.Insteaditadoptsa dipolarasymmetricform (Figure2),thatshowsabroadCT transitionandnotthenarrowcyaninetransition.Thissymmetry- loweringdistortionwasrecentlyconÞrmedbytheoretical calculations.10,14Inaddition,Lepkowicz,Hagan,andco-workers reportedthat,foragivencyaninefeaturingan11-carbonatoms (6)(a)Rettig,W.;Dekhtyar,M. Chem.Phys. 2003,293,75Ð90.(b) Ishchenko,A.A.;Kulinich,A.V.;Bondarev,S.L;Knyukshto,V.N. J.Phys.Chem.A 2007,111,13629Ð13627 .(7)(a)Lawrentz,U.;Grahn,W.;Lukaszuk,K.;Klein,C.;Wortmann, R.;Feldner,A.;Scherer,D. ChemsEur.J. 2002,8,1573Ð1587.(b) Wu¬rthner,F.;Archetti,G.;Schmidt,R.;Kuball,H.-G. Angew.Chem., Int.Ed. 2008,47,4529Ð4532.(c)Kulinich,A.V.;Derevyanko,N. A;Ishchenko,A.A.;Bondarev,S.L;Knyukshto,V.N. J.Photochem. Photobiol.,A 2008,200,106Ð113 .(8)Forseveralrecentexamplessee:(a)Bouit,P.-A.;DiPiazza,E.;Rigaut, S.;LeGuennic,B.;Aronica,C.;Toupet,L.;Andraud,C.;Maury,O. Org.Lett. 2008,10,4159Ð4162.(b)Chen,X.;Peng,X.;Cui,A.;Wang, B.;Wang,L.;Zhang,R. J.Photochem.Photobiol.,A 2006,181,79Ð 85.(c)Asato,A.E.;Watanabe,D.T.;Liu,R.S.H. Org.Lett. 2000,2,2559Ð2562.(d)Peng,X.;Song,F.;Lu,E.;Wang,Y.;Zhou,W.; Fan,J;Gao,Y J.Am.Chem.Soc. 2005,127,4170Ð4171 .(9)(a)Yau,C.M.S.;Pascu,S.I.;Odom,S.A.;Warren,J.E.;Koltz, E.J.F.;Frampton,M.J.;Williams,C.C.;Coropceanu,V.;Kuimova, M.K.;Phillips,D.;Barlow,S.;Bre «das,J.-L.;Marder,S.R.;Millar, V.;Anderson,H.L. Chem.Commun. 2008,2897Ð2899.(b)Nagao, Y.;Sakai,T.;Kozawa,K.;Urano,T. DyesPigm. 2007,73,344Ð352. (c)Dai,Z.F.;Peng,B.X.;Chen,X.A. DyesPigm. 1999,40,219Ð 223.(d)Koska,N.A.;Wilson,S.R.;Schuster,G.B. J.Am.Chem. Soc.1993,115,11628Ð11629.(e)Kulpe,S.;Kuban,R.J.;Schulz, B.;Da ¬hne,S. Cryst.Res.Technol. 1987,22,375Ð379 .(10)(a)Egorov,V.V. J.Chem.Phys. 2002,116,3090Ð3102.(b) Kachkovsky,O.D.;Shut,D.M. DyesPigm. 2006,71,19Ð27.(c) Guillaume,M.;Lie «geois,V.;Champagne,B.;Zutterman,F. Chem.Phys.Lett. 2007,446,165Ð169 .(11)Tolbert,L.M.;Zhao,X. J.Am.Chem.Soc. 1997,119,3253Ð3258 .(12)Kachkovski,O.D.;Tolmachov,O.I.;Slominskii,Y.L.;Kudinova, M.O.;Derevyanko,N.O.;Zhukova,O.O. DyesPigm. 2005,64,207Ð216.(13)Brooker,L.G.S.;Sprague,R.H;Smyth,C.P.;Lewis,G.L. J.Am. Chem.Soc. 1940,62,1116Ð1125 .(14)Fabian,J. J.Mol.Struct.(THEOCHEM) 2006,766,49Ð60 .Figure1. Differentresonanceformsproposedforpush -pullchromophoresandrelatedshapeoftheabsorptionspectra(bottom). Figure2. SpectralandelectronicmodiÞcationsbycrossingthecyanine limit.14J.AM.CHEM.SOC. 9VOL.132,NO.12,2010 4329IonPairingEffectinHeptamethineCyanineDyes ARTICLES thereforepossibletocrossthisvirtualcyaninelimit 6and,in somerarecases,tostabilizeamerocyaninedyeclosetothis idealcyaninestructure. 7Thiscyaninestructure,extremelydifÞculttoobtainfrom dissymmetricmerocyaninedyes,canbedirectlypreparedby theassociationoftwoidenticaldonoror,morerarely,acceptor end-groupsviaapolymethinechainfeaturinganoddnumber ofsp 2carbonatoms. 8Thesemoleculesspontaneouslyreachthe idealpolymethinestate (IPS)presentinguniquestructuraland spectralcharacteristics(Figure1, M2).Asaresultofthetwo degeneratedresonanceformswiththechargelocalizedatone ortheotherend-group( M1)M3),thepolymethinechainadopts anonalternatingstructurewithanaverageÒoneandahalfÓbond length(BLA )0).Thislackofbondalternationisexperimen- tallyconÞrmedbymostcrystalstructuresreportedinthe literature.8a,9Fromaspectroscopicpointofview,cyaninesin theirIPSarecharacterizedbyasharp,extremelyintense absorptionband(Figure1, M2)resultingfromthereduced vibroniccontributioninthesenonalternatingstructures.This particulartransition,calledsolitonictransitionbyanalogywith dopedpolyacetylene,isshiftedtowardtheNIRspectralrange. Thecompleterationalizationofthistransitionusingtheoretical modelsremainsamatterofdebate;inparticulartheclear assignmentofthehigh-energyshoulderremainsproblematic. 10TheIPSisconservedinthecaseofrelativelyshort !-conjugatedchains(upto9 -13carbonatoms,dependingonthestructure) butisprogressivelylostforlongerchains.Asaconsequence, uponincreasingthechainlength,thesharptransitioncharac- teristicoftheIPSundergoesabathochromicshift( !100nm peradditionalvinylicunit,Figure2,curvea)followedbya deepmodiÞcationofthespectrumwithalargebroadeningof thebandandadecreaseofitsintensity(Figure2,curveb). 11,12Thischain-lengthdependenceofthespectroscopicproperties, generallyreferredasBrookerexperiment, 13wasrationalizedÞrst byTolbertandco-workers 11invokingaÒsymmetrycollapseÓ forlong-chaincyanineelectronicstructuresduetoaPeierls- typedistortion.Insuchacase,thecyanineisnolonger symmetricwhichmeansithaslostitsIPS.Insteaditadoptsa dipolarasymmetricform (Figure2),thatshowsabroadCT transitionandnotthenarrowcyaninetransition.Thissymmetry- loweringdistortionwasrecentlyconÞrmedbytheoretical calculations.10,14Inaddition,Lepkowicz,Hagan,andco-workers reportedthat,foragivencyaninefeaturingan11-carbonatoms (6)(a)Rettig,W.;Dekhtyar,M. Chem.Phys. 2003,293,75Ð90.(b) Ishchenko,A.A.;Kulinich,A.V.;Bondarev,S.L;Knyukshto,V.N. J.Phys.Chem.A 2007,111,13629Ð13627 .(7)(a)Lawrentz,U.;Grahn,W.;Lukaszuk,K.;Klein,C.;Wortmann, R.;Feldner,A.;Scherer,D. ChemsEur.J. 2002,8,1573Ð1587.(b) Wu¬rthner,F.;Archetti,G.;Schmidt,R.;Kuball,H.-G. Angew.Chem., Int.Ed. 2008,47,4529Ð4532.(c)Kulinich,A.V.;Derevyanko,N. A;Ishchenko,A.A.;Bondarev,S.L;Knyukshto,V.N. J.Photochem. Photobiol.,A 2008,200,106Ð113 .(8)Forseveralrecentexamplessee:(a)Bouit,P.-A.;DiPiazza,E.;Rigaut, S.;LeGuennic,B.;Aronica,C.;Toupet,L.;Andraud,C.;Maury,O. Org.Lett. 2008,10,4159Ð4162.(b)Chen,X.;Peng,X.;Cui,A.;Wang, B.;Wang,L.;Zhang,R. J.Photochem.Photobiol.,A 2006,181,79Ð 85.(c)Asato,A.E.;Watanabe,D.T.;Liu,R.S.H. Org.Lett. 2000,2,2559Ð2562.(d)Peng,X.;Song,F.;Lu,E.;Wang,Y.;Zhou,W.; Fan,J;Gao,Y J.Am.Chem.Soc. 2005,127,4170Ð4171 .(9)(a)Yau,C.M.S.;Pascu,S.I.;Odom,S.A.;Warren,J.E.;Koltz, E.J.F.;Frampton,M.J.;Williams,C.C.;Coropceanu,V.;Kuimova, M.K.;Phillips,D.;Barlow,S.;Bre «das,J.-L.;Marder,S.R.;Millar, V.;Anderson,H.L. Chem.Commun. 2008,2897Ð2899.(b)Nagao, Y.;Sakai,T.;Kozawa,K.;Urano,T. DyesPigm. 2007,73,344Ð352. (c)Dai,Z.F.;Peng,B.X.;Chen,X.A. DyesPigm. 1999,40,219Ð 223.(d)Koska,N.A.;Wilson,S.R.;Schuster,G.B. J.Am.Chem. Soc.1993,115,11628Ð11629.(e)Kulpe,S.;Kuban,R.J.;Schulz, B.;Da ¬hne,S. Cryst.Res.Technol. 1987,22,375Ð379 .(10)(a)Egorov,V.V. J.Chem.Phys. 2002,116,3090Ð3102.(b) Kachkovsky,O.D.;Shut,D.M. DyesPigm. 2006,71,19Ð27.(c) Guillaume,M.;Lie «geois,V.;Champagne,B.;Zutterman,F. Chem.Phys.Lett. 2007,446,165Ð169 .(11)Tolbert,L.M.;Zhao,X. J.Am.Chem.Soc. 1997,119,3253Ð3258 .(12)Kachkovski,O.D.;Tolmachov,O.I.;Slominskii,Y.L.;Kudinova, M.O.;Derevyanko,N.O.;Zhukova,O.O. DyesPigm. 2005,64,207Ð216.(13)Brooker,L.G.S.;Sprague,R.H;Smyth,C.P.;Lewis,G.L. J.Am. Chem.Soc. 1940,62,1116Ð1125 .(14)Fabian,J. J.Mol.Struct.(THEOCHEM) 2006,766,49Ð60 .Figure1. Differentresonanceformsproposedforpush -pullchromophoresandrelatedshapeoftheabsorptionspectra(bottom). Figure2. SpectralandelectronicmodiÞcationsbycrossingthecyanine limit.14J.AM.CHEM.SOC. 9VOL.132,NO.12,2010 4329IonPairingEffectinHeptamethineCyanineDyes ARTICLES thereforepossibletocrossthisvirtualcyaninelimit 6and,in somerarecases,tostabilizeamerocyaninedyeclosetothis idealcyaninestructure. 7Thiscyaninestructure,extremelydifÞculttoobtainfrom dissymmetricmerocyaninedyes,canbedirectlypreparedby theassociationoftwoidenticaldonoror,morerarely,acceptor end-groupsviaapolymethinechainfeaturinganoddnumber ofsp 2carbonatoms. 8Thesemoleculesspontaneouslyreachthe idealpolymethinestate (IPS)presentinguniquestructuraland spectralcharacteristics(Figure1, M2).Asaresultofthetwo degeneratedresonanceformswiththechargelocalizedatone ortheotherend-group( M1)M3),thepolymethinechainadopts anonalternatingstructurewithanaverageÒoneandahalfÓbond length(BLA )0).Thislackofbondalternationisexperimen- tallyconÞrmedbymostcrystalstructuresreportedinthe literature.8a,9Fromaspectroscopicpointofview,cyaninesin theirIPSarecharacterizedbyasharp,extremelyintense absorptionband(Figure1, M2)resultingfromthereduced vibroniccontributioninthesenonalternatingstructures.This particulartransition,calledsolitonictransitionbyanalogywith dopedpolyacetylene,isshiftedtowardtheNIRspectralrange. Thecompleterationalizationofthistransitionusingtheoretical modelsremainsamatterofdebate;inparticulartheclear assignmentofthehigh-energyshoulderremainsproblematic. 10TheIPSisconservedinthecaseofrelativelyshort !-conjugatedchains(upto9 -13carbonatoms,dependingonthestructure) butisprogressivelylostforlongerchains.Asaconsequence, uponincreasingthechainlength,thesharptransitioncharac- teristicoftheIPSundergoesabathochromicshift( !100nm peradditionalvinylicunit,Figure2,curvea)followedbya deepmodiÞcationofthespectrumwithalargebroadeningof thebandandadecreaseofitsintensity(Figure2,curveb). 11,12Thischain-lengthdependenceofthespectroscopicproperties, generallyreferredasBrookerexperiment, 13wasrationalizedÞrst byTolbertandco-workers 11invokingaÒsymmetrycollapseÓ forlong-chaincyanineelectronicstructuresduetoaPeierls- typedistortion.Insuchacase,thecyanineisnolonger symmetricwhichmeansithaslostitsIPS.Insteaditadoptsa dipolarasymmetricform (Figure2),thatshowsabroadCT transitionandnotthenarrowcyaninetransition.Thissymmetry- loweringdistortionwasrecentlyconÞrmedbytheoretical calculations.10,14Inaddition,Lepkowicz,Hagan,andco-workers reportedthat,foragivencyaninefeaturingan11-carbonatoms (6)(a)Rettig,W.;Dekhtyar,M. Chem.Phys. 2003,293,75Ð90.(b) Ishchenko,A.A.;Kulinich,A.V.;Bondarev,S.L;Knyukshto,V.N. J.Phys.Chem.A 2007,111,13629Ð13627 .(7)(a)Lawrentz,U.;Grahn,W.;Lukaszuk,K.;Klein,C.;Wortmann, R.;Feldner,A.;Scherer,D. ChemsEur.J. 2002,8,1573Ð1587.(b) Wu¬rthner,F.;Archetti,G.;Schmidt,R.;Kuball,H.-G. Angew.Chem., Int.Ed. 2008,47,4529Ð4532.(c)Kulinich,A.V.;Derevyanko,N. A;Ishchenko,A.A.;Bondarev,S.L;Knyukshto,V.N. J.Photochem. Photobiol.,A 2008,200,106Ð113 .(8)Forseveralrecentexamplessee:(a)Bouit,P.-A.;DiPiazza,E.;Rigaut, S.;LeGuennic,B.;Aronica,C.;Toupet,L.;Andraud,C.;Maury,O. Org.Lett. 2008,10,4159Ð4162.(b)Chen,X.;Peng,X.;Cui,A.;Wang, B.;Wang,L.;Zhang,R. J.Photochem.Photobiol.,A 2006,181,79Ð 85.(c)Asato,A.E.;Watanabe,D.T.;Liu,R.S.H. Org.Lett. 2000,2,2559Ð2562.(d)Peng,X.;Song,F.;Lu,E.;Wang,Y.;Zhou,W.; Fan,J;Gao,Y J.Am.Chem.Soc. 2005,127,4170Ð4171 .(9)(a)Yau,C.M.S.;Pascu,S.I.;Odom,S.A.;Warren,J.E.;Koltz, E.J.F.;Frampton,M.J.;Williams,C.C.;Coropceanu,V.;Kuimova, M.K.;Phillips,D.;Barlow,S.;Bre «das,J.-L.;Marder,S.R.;Millar, V.;Anderson,H.L. Chem.Commun. 2008,2897Ð2899.(b)Nagao, Y.;Sakai,T.;Kozawa,K.;Urano,T. DyesPigm. 2007,73,344Ð352. (c)Dai,Z.F.;Peng,B.X.;Chen,X.A. DyesPigm. 1999,40,219Ð 223.(d)Koska,N.A.;Wilson,S.R.;Schuster,G.B. J.Am.Chem. Soc.1993,115,11628Ð11629.(e)Kulpe,S.;Kuban,R.J.;Schulz, B.;Da ¬hne,S. Cryst.Res.Technol. 1987,22,375Ð379 .(10)(a)Egorov,V.V. J.Chem.Phys. 2002,116,3090Ð3102.(b) Kachkovsky,O.D.;Shut,D.M. DyesPigm. 2006,71,19Ð27.(c) Guillaume,M.;Lie «geois,V.;Champagne,B.;Zutterman,F. Chem.Phys.Lett. 2007,446,165Ð169 .(11)Tolbert,L.M.;Zhao,X. J.Am.Chem.Soc. 1997,119,3253Ð3258 .(12)Kachkovski,O.D.;Tolmachov,O.I.;Slominskii,Y.L.;Kudinova, M.O.;Derevyanko,N.O.;Zhukova,O.O. DyesPigm. 2005,64,207Ð216.(13)Brooker,L.G.S.;Sprague,R.H;Smyth,C.P.;Lewis,G.L. J.Am. Chem.Soc. 1940,62,1116Ð1125 .(14)Fabian,J. J.Mol.Struct.(THEOCHEM) 2006,766,49Ð60 .Figure1. Differentresonanceformsproposedforpush -pullchromophoresandrelatedshapeoftheabsorptionspectra(bottom). Figure2. SpectralandelectronicmodiÞcationsbycrossingthecyanine limit.14J.AM.CHEM.SOC. 9VOL.132,NO.12,2010 4329IonPairingEffectinHeptamethineCyanineDyes ARTICLES " / nm" / nm" / nmFigure 0-5 Different resonance forms proposed for push -pull chromophores and related shape of the absorption spectra . The charge distribution along the push -pull systems of merocyanines and cya nines are depicted by three states: M 1 (neutral for merocyanine, zwitterionic for cyanine), M 2 (least bond length alteration), and M 3 (bipolar for merocyanine, inverted zwitterionic for cyanine). Merocyanines are featured by one electron rich amine group ( D, donor) and one more electronegative terminal group (A, acceptor) on either ends of the push -pull chain. Typical cyanines bear one neutral amine (D) and one charged iminium (A) on either ends of the push -pull chain. 110 I.8!Approach to NIR emission wit h large Stokes shift Before devising new dye scaffold to reach large Stokes shift, it is important to understand the underlying reason for the small Stokes shift observed with MCRA . As discussed in the beginning of Section I.7, the evenly -dispersed charge along the polymethine chain determines the small Stokes shift of cyanines and merocyanines . Shown in Figure II-5, the electron push -pull merocyanine and cyanine structures can have three resonance limiting forms: the neutral state M1, the zwitterionic state M3, and the M2 state or the Òcyanine limi tÓ. 15 The Òcyanine limitÓ is featured by the least bond length alternation. Its absorption spectrum is characterized by very red -shifted sharp intense band. In contrast, the bipolar M3 state is featured by a very strong electr on donor and acceptor fragments, and a complete intramolecular charge transfer (ICT) occurring in the excited state . This betaine structure has the opposite ma ximized bond length alternation. Because of the continuous symmetry breaking, it is characterized by the broad intense absorption and shows strong solvatochromism. 16, 17 This is the basis of ICT-generated large Stokes shift. Further contemplating the ICT transition (Figure II-6), if strong electron donor and acceptor fragments are installed on the two sides of a (-spacer, the local excitation (LE) of the molecule can elicit a strong dipole moment pointing from the acceptor to the donor moiety. But if the ICT occurs, an opposite dipole moment can be induced by the pair of split charges , minimizing the overall change in the dipole momen t.18 This will commonly result in a smaller energy gap between S 1 and S 0, and hence a bathochromic shift in the emission wavelength. Because the LE happens before the ICT transition, the absorption 111 wavelength is not affected. This explains why the absorption maxima of solvatochromic compounds in various solvents are usually the same. T aking the solvent effect into consideration, the ICT state in S 1 can further be stabilized by the dipole moment reorientation of the surrounding solvent molecules to a different extent. Therefore, a positive solvatochromic (i.e., the emission wavelength shifts bathoch romically wi th increasing solvent polarity) dye can have the most bathochromic shift in emission wavelength when in a polar solvent environ ment. 19 It should be noted that, the above discussion is limited to the planarized ICT (PLICT) transi tion.20 Normally, if the don or and acceptor are not in the same plane in the excited state, a twisted ICT (TICT) can o ccur. 21 This TICT also res ults in large Stokes shift. But the larger the twist, the lower the fluorescence quantum yield , as the radiative decay becomes forbidden .22 Following this idea, a suitable dye scaffold to conjugate with hCRBPII for the far -red/NIR emission wavelength and large Stokes shift should feature a strong ICT character. Like enzymes, the cavity of hCRBPII is filled with highly ordered water molecule, m aking DADAReaction Coordinate S0S1!ELE!EICTEnergy Figure 0-6 Schematic cross section of the ground state and the lowest excited singlet state potential hypersurf aces along the reaction coordi nate corresp onding to the ICT state model. 112 a polar but not an aqueous environment. This environment could be expected to provide the bathochromic shift of the wavelength without a drastic quenching of fluorescence by highly exchangeable protons. Because of the difficulty to predict fluorescence properties, a feasible approach to get a candidate scaffold is to start with a bright ICT dye that has the potential to emit into the red wavelength region. An extensive literature research led to the selection of FR0 scaffold for its extremely high quantum yield and relatively high extinction coeffici ent.23 Shown in Scheme II-2, FR0 has a simple structure. It has a diethylamino group as a strong electron donor, and a formyl group as a strong electron acceptor, spaced by a 9,9 -dimethylfluorenyl fragment. The emission maximum of FR0 spans from 434 nm to 573 nm in a variety of organic solvents ranging from non -polar toluene to very polar methanol. Different spacers were installed between the aldehyde group and the fluorenyl scaffo ld to red -shift the emission wavelength by extending the length of conjugation ( Scheme II-2). Among these, FR-1V exhibits the optimal extinction coefficient, fluorescence quantum yield, and proper emission maxima (Table II-2). Although FR-2V is more red -shifted, its quantum yield is significantly decreased in various NOFR0NOFR-1V NOFR-2V NSOFR-Th Scheme 0-2 FR series of ICT dyes, derivatized from FR0 with different lengths and types of (-spacers. 113 solvents comparing to FR-1V, probably due to more rotation al freedom along the polyene and increased vi brational levels. Besides, the yield of FR-2V is much lower than that of FR-1V due to more Z/E isomers and harder purification. When the diene spacer is locked by sulfur as a thiophene ring to furnish FR-Th, the emission maxima are blue -shifted and close t o those of FR-1V, however with a much smaller extinction coefficient. More interestingly, from non -polar to polar aprotic solvents FR-2V and FR-Th exhibit positive solvatochromism, while a negative solvatochromism is observed when these two are dissolved i n polar protic solvents like ethanol and methanol (i.e., the emission maxima are blue -shifted with increased solvent polarity). It is not clear at this stage the possible Table 0-2 Spectral features of FR dyes in different organic solvents. aAbsolute fluorescence quantum yield measurement using Quantaurus -QY at room temperature. FR0 FR-1V!abs(nm) !em(nm) "a!abs(nm) !em(nm) "Toluene 3964340.703 4164880.759 THF 3924700.802 4145370.796 Ethyl acetate 3894750.662 4105320.820 Dichloromethane 3994970.774 4235750.873 DMF 3985120.775 4206080.691 DMSO 4025260.763 4256140.725 Acetonitrile 3935230.745 4156050.638 Ethanol 3965560.664 420633, 660 0.074 Methanol 3965730.445 4206690.024 FR-2VFR-Th!abs(nm) !em(nm) "!abs(nm) !em(nm) "Toluene 4285590.675 4175030.879 THF 4266030.524 4175630.782 Ethyl acetate 4226000.555 4125610.770 Dichloromethane 435634, 665 0.432 4246040.751 DMF 432633, 669 0.167 4246630.473 DMSO 438633, 672 0.175 4296720.278 Acetonitrile 425630, 670 0.134 4156670.361 Ethanol 433627, 667 0.021 4196040.027 Methanol 4316700.006 4174730.006 114 photochemical transformation that leads to this phenomenon, but very likely the proto nation is involved in the excited hy persurface. In this context, FR-1V stands out as the optimal candidate for further studies. NOFR-1V NH2NNFR-1V -SBFigure 0-7 Normalized spectra of FR-1V in various solvents: (a) absorbance, (b) emission. Normalized spectra of FR-1V-SB ( n-butylamine) in various solvents: (c) absorbance, (d) emission. Tol: toluene; THF: tetrahydrofuran; EA: ethyl acetate; DCM: dichloromethane; DMF: N,N -dimethyl formamide; D MSO: dimethyl sulfone; MeCN: acetonitrile; EtOH: ethanol; MeOH: methanol. Scheme shows the SB formation between FR-1V and n-butylamine. 115 As shown i n Table II-2, the emission wavelength of FR-1V reach es the far -red/NIR region when dissolved in polar protic solvents, although we lack a direct comparison in aqueous solutions due to its very low solubility. Considering the previous discussion of ICT transition, it is expected that the electron withdrawing ability of the acceptor part will decrease when FR-1V forms imine with th e act ive lysine 108 of hCRBPII. This is due to the replacement of oxygen to the less electronegative nitrogen and will lead to the hypsochromic shifts in both absorption and emission maxima to some extent. Following the same argument, the protonation of imine t o iminium can effectively increase the electron withdrawing ability and further lower the band gap to reach more red -shifted wavelengths than those of the original FR-1V aldehyde. To test this idea, N-butylamine is used as a surrogate of the active lysine 108 to bind with FR-1V and form the corresponding SB (Figure II-7). Spectral feature s are summarized in Table II-3. As expected, the n-butylimine exhibits hypsochromic shifts in both absorption and emission. The following up acidification of SB to PSB regains the bathochromic shifts as shown in Figure II-8. The absorption maximum is drama tically Table 0-3 Spectral featur es of FR-1V and its SB in different solvents. FR-1VFR-1V-SBb!abs(nm) !em(nm) "a!abs(nm) !em(nm) "Toluene 4164880.759 3864470.622 THF 4145370.796 3894810.597 Ethyl acetate 4105320.820 3844760.595 Dichloromethane 4235750.873 3914980.685 DMF 4206080.691 3935250.658 DMSO 4256140.725 3965320.749 Acetonitrile 4156050.638 3885240.738 Ethanol 420633, 660 0.074 394510, 675 0.049 Methanol 4206690.024 3935570.065 aAbsolute fluorescence quantum yield measurement. bn-butylimine. 116 red -shifted from 396 nm to 496 nm in DMSO, and from 394 nm to 512 nm in ethanol, respectively. The emission maxima also display over 150 nm bathochromic shift into the NIR region (662 nm for DMSO and 690 nm for ethanol ). The wavelengths are all redder than those of the original FR-1V (Table II-4). We have so far proved the concept of employing ICT to approach NIR emission with large Stokes shift. The next move is to embedded FR-1V into hCRBPII and gain special functionalities while maintaining its optical properties. NNFR-1V -SBHClNNHFR-1V -PSBFigure 0-8 Comparison of (a) absorbance and (b) emission spectra of FR-1V-SB and FR-1V-PSB in DMSO and ethanol. Table 0-4 Wavelength comparison of FR-1V and its SB, PSB. FR-1VFR-1V-SBFR-1V-PSB!abs(nm) !em(nm) !abs(nm) !em(nm) !abs(nm) !em(nm) DMSO 425614396532528662Ethanol 420633,660394510 ( 675)a510690aShoulder emission (see 0 Section I.14 ). 117 I.9!Discovery of photoc hromic dye-hCRBPII complexes As describe d in Section I.7, a large portion of the non -specific fluorescence background is from the non -selective imine formed between the dyeÕs aldehyde group and non -target protein amine groups. It is inevitable to h ave these non -specific labeling due to the high reactivity of the aldehyde group, unless less reactive carbonyl is used instead. Therefore, to develop FR-1V-hCRBPII complexes into useful imaging tags, a different approach is needed to differentiate the emi ssion of target hCRBPII complexes from other non -specific proteins. In ethanol, the FR-1V-PSB ( n-butyliminium) emission is 690 nm, but the SB emission is only at 510 nm. If there is a method to control the complexation of FR-1V exclusively as a PSB with hC RBPII and a SB with other proteins, then the background issue of off -target labeling can be solved by specifically imaging the PSB emission wavelength region . In addition, because of the large emission wavelength difference between FR-1V and its PSB, the f luorescen ce background from the extra dye residing in the hydrophobic cell compartments can be cut off simply by selecting the proper dichroic mirrors in the microscope. This will further enable a no -wash protocol for the live cell imaging. However, the in itial FR-1V and hCRBPII mutants binding assays showed the formation of predominant SB while t he PSB only forms as a limited fraction. Thus, we need an effective method to form PSB in favor of SB, preferably in a controllable manner. The switchable feature of the emitter fluorescence, or the ability to turn on and turn off the fluorescence, is the molecular basis of STORM and PALM. There are different structural approaches to attain this Òon/of fÓ switchability, for example, the spirocyclization, 118 double bond isomerization (either C=C, C=N , or N=N ), and reversible redox add ition.13, 24-26 The most studied small molecular switches are the rapidly photoswitchable azobenzenes and perfluoronated diary ethenes. 25, 27-32 The former has been studied over a few decades, but it is usually limited by its fluorescence efficiency. The latter one is drawing more attention recen tly, 33-35 but is limited by its tunable optic al features. On the contrary, the photoac tivatable and photoswitchable FP rivals provides inspiration to modify our FR-1V/hCRBPII complexes for the desired light -switchable PSB emissi on.36-38 The PA-FPs and PS -FPs have different underlying switching mechanisms . For example, PA-GFP undergoes a Glu222 decaboxylation after photoexcitation and deprotonates its phenol ic oxygen to turn on the green li ght.39 Kaede undergoes a backbone breakage and introduces a double bond to conjugate the His65 (see Section I.1.1). Dreiklang undergoes a reversible hydration/dehydration cycle on the imidazolinone C2 position on 405/365 nm irradiations, respec tively. 40 The most prominent RSFPs are the Dronpa and Padron (revers al of Òdron Ó and ÒpaÓ). 41 Dronpa has an ÒonÓ state that absorbs at 503 nm and emits at 518 nm. It can be converted to a non -fluorescent ÒoffÓ state by intense 488 -nm irradiation. Thus, Dronpa keeps losing its ÒonÓ state emission because the same excitation simultaneously converts it to the ÒoffÓ state. Hence, Dronpa is termed as a negatively photoswit chable RSFP (NS -RSFP). On the contrary, Padron has the opposite behavior and makes it a positively photoswitchable RSFP (PS -RSFP). An intense 503 -nm irradiation can initiate the ÒoffÓ to ÒonÓ transition. T he same light can excite the ÒonÓ state that leads to fluorescence. The exact photoswitching mechanism is 119 still under debate. The exact sequence and time scale of each individual step is not well understood . Nevertheless, both experimental and c omputational studies have demonstrated an ultrafast photoinduc ed cis -trans isomerization of the chromophore that is followed by a rapid ground -state proton transfer (GSPT). This GSPT is accompanied by structural reorganization of both solvent and protein scaff old.42 The two conformations can have ~ 1.5 pK a shift and lead to different protonation states that are responsible for the dark and bright fluorescence (Scheme II-3). The photoswitching mechanism of Dronpa and Padron is intriguing. If this process is recapitulated in FR-1V/hCRBPII complexes, the control of PSB formation will not b e a problem. Besides, the photochemical isomerization will ensure the spatiotemporal control of fluorescence turn ÒonÓ and ÒoffÓ in terms of the PSB NIR emission. In addition, the SB and PSB have a pK a difference that is fitting the photoswitching pattern observed in HONNOSH405 nm488 nmHHNNOSHHONNOSHOHHONNOSHCisTrans Scheme 0-3 Proposed mechanism f or the cis ! trans isomerization of the chromophore in Dronpa and Padron . Bright and dark states correspond to the cis and trans forms of the chromophore, respectively. The two forms are determined by the p Ka difference in the acid -base equili brium . In Dronpa, the cis form is anionic and bright. 120 Dronpa and Padron. Consider the structure of FR-1V and hCRBPII complex, there are two double bonds: the imine C=N and the 3, --conjugate vinylene C=C .,Which double bond is prone to photoisomerization? The photoinduced cis -trans isomerization involves the reorganization of the protein scaffold . Thus, it is reasonable to find possible protein mutations to elicit the desired photoi somerization. The possible gestures of FR-1V inside hCRBPII after covalent imine formation were simulated using AutoDock Vina with a crystal structure of all -trans -retinal (ATR) and hCRBPII KL (Q108K:K40L) mutant complex. Shown in Figure II-9, FR-1V occupies almost the same spatial volume as ATR. The embedd ed depths are also similar. Amino acid residues within 5.5 † distance to FR-1V scaffold are highlighted for mutage nesis screening. Figure 0-9 Flexible d ocking of FR-1V with all -trans -retinal bound hCRBPII KL mutant (PDB ID: 4EXZ). Green stick is t he crystal structure of re tinal . Orange stick is the pose of FR-1V. Docking was performed by AutoDock Vina. Residues !within 5.5 † distance to FR-1V are highlighted in line form. ! 121 By the time of hCR BPII mutagenesis and screening with FR-1V, a light -triggered photochromism was observed in the complexes of ATR with specific CRABPII mutants. In both solution and crystal cases, the ATR/CRABPII complexes can be iteratively converted between a colorless trans-SB form and a blue cis -PSB form ( Figure II-10a).43,44 As shown crystallograhically, the cis -PSB is first formed, followed by thermal isomerization to the trans -SB. It was further demonstrated in single crystals that the same cycle could be driven photochemically, with green light irradiation driving the cis -iminium to the trans -imine, and UV irradiation driving the photocycle in the opposite direction (Figure II-10b). The two forms are stable over a long -time period , enabling the APhotoisomerizingRhodopsinMimicObservedatAtomic ResolutionMeisamNosrati,TetyanaBerbasova,ChrysoulaVasileiou,BabakBorhan, *andJamesH.Geiger *DepartmentofChemistry,MichiganStateUniversity,EastLansing,Michigan48824,UnitedStates *SSupportingInformation ABSTRACT: Themembersoftherhodopsinfamilyof proteinsareinvolvedinmanyessentiallight-dependent processesinbiology.Speci !cphotoisomerizationofthe protein-boundretinylidenePSBataspeci !edwavelength rangeoflightisattheheartofallofthesesystems. Nonetheless,ithasbeendi "culttoreproduceinan engineeredsystem.Wehavedevelopedrhodopsinmimics, usingintracellularlipidbindingproteinfamilymembersas sca#olds,tostudyfundamentalaspectsofprotein/chromo- phoreinteractions.Hereinwedescribeasystemthatspeci !callyisomerizestheretinylideneprotonatedSchi #baseboththermally andphotochemically.Thisisomerizationhasbeencharacterizedatatomicresolutionbyquantitativelyinterconvertingthe isomersinthecrystalboththermallyandphotochemically.Thiseventisaccompaniedbyalargep Kachangeoftheiminesimilar tothep Kachangesobservedinbacteriorhodopsinandvisualopsinsduringisomerization. !INTRODUCTIONIsomerizationoftheretinalchromophoreisthekeyeventina varietyofbiologicalprocesses, 1particularlythosethatinvolve visualperception, 2circadianrhythm, 3channelingionsacrossa membrane,light-inducedphototaxisandsignaling. 4Addition-ally,photoisomerizationofretinal,coupledtotheactionof channelrhodopsinandhalorhodopsintotriggerlight-depend- entiontransportinneurons,hasbeenexploitedinthe burgeoning!eldofoptogenetics. 5Rhodopsinsbindanisomer ofretinalasaprotonatedSchi #base(PSB,iminium)usinga nucleophiliclysineresidueintheirbindingpocket.Though theyhaveavarietyofbiochemicalactivities,mostsharesome commonthemesthatarecriticaltotheirfunction.Firsttheyare characterizedbyaspeci !clightinduced cis!transisomerizationoftheretinylidenemoiety;usuallyeitheran11- cisretinylideneisomerizationineukaryoticopsins(typeIIrhodopsins)ora13- cisretinylideneisomerizationinmicrobialrhodopsins(typeI rhodopsins).1Thisspeci !cisomerizationisdistinctfromthe behaviorofretinylidenePSBsinsolution,whichgivesmixtures ofisomersuponirradiation.Thisisomerizationplacesthe retinylideneSchi #base(SB)intwodistinct,localenviron- ments,directlya #ectingthep Kaoftheiminefunctionality.For example,inmicrobialrhodopsinstheisomerizationoftheC13 doublebondresultsinasmuchasa5 !6unitchangeinthep Kaoftheprotein-boundiminium. 6Intherhodopsinvisual pigment,thisdi #erenceincreasesto8 !11p Kaunits.7DeprotonationofthePSBtoaSchi #base(SB,imine)leads toalargeshiftintheabsorptionspectrumofthepolyenefrom thevisibleregiontotheultraviolet. 1Although,thepasttwo decadeshavewitnessedagreatdealofprogressinunder- standingthemechanismofphotoisomerizationindi #erentrhodopsins,1,8therearesigni !cantchallengesintamingthe isomerizationprocesssuchthatitisobservableinrealtime, withstructurallyinformativetechniques.Tothebestofour knowledge,therearenoreportsthatshowcompletephoto- isomerizationofrhodopsininthecrystallineform, 9andin bacteriorhodopsin,thereiscontroversyregardingthestructure oftheintermediatesidenti !edspectroscopically. 10Amodel systemthatisabletoreproduceanisomerizationeventina controlledmannerwouldallowamuchmoredetailed examinationoftheseprocesses.Herein,wedemonstratethe !rststepstowardthisgoal. Ourworkhasfocusedonthedevelopmentofsoluble,easily manipulatedrhodopsinmimicsderivedfromthereengineering ofcellularretinoidbindingproteinsthatbelongtothe intracelluarlipidbindingfamilyofproteins(iLBP), 11namelyhumancellularretinolbindingproteinII(hCRBPII) 12andhumancellularretinoicacidbindingproteinII(CRABPII). 13Wehaveusedthesesystemstounravelthemechanismof wavelengthtuning,wheretheproteinenvironmentaltersthe absorptionpropertiesofaboundchromophoreoverawide wavelengthrange.Thishasculminatedinthedevelopmentofa setofretinal-boundproteinswithabsorptionmaximathatrange from425to644nm, 12a,13emorethan50nmfurtherthanany ofthenaturalrhodopsinsystems.Thesestudieshaveshedlight onthefundamentalmechanismsofprotein-basedwavelength tuning,withpotentialapplicationsinawiderangeofsystems. ThereengineerediLBPshavealsoshownagreatdealof potentialas $uorescentproteintagsandalsopHresponsive reporters.13e,14Withrespecttothelatter,wehavehadthegood fortuneofidentifyingmutantsofCRABPIIandhCRBPIIthat Received:April9,2016 Published:June16,2016 Articlepubs.acs.org/JACS©2016AmericanChemicalSociety 8802DOI:10.1021/jacs.6b03681J.Am.Chem.Soc. 2016,138,8802 !8808NCRABPII NCRABPII NHCRABPII trans -SB, thermodynamic !abs ~ 360 nm UV light (~360 nm) high pKaheat orgreen light (~ 550 nm) NCRABPII Hlow pKacis -SBtrans -PSBcis -PSB, kinetic !abs > 440 nm a) b) Figure 0-10 Cis-trans imine photoisomerization of all -trans -retinal bound CRABPII complexes. a) The iterative color switching of all -trans retinal/CRABPII complex under UV and green light irradiations. (PDB ID: 4YFP) b) Scheme illustrating the photocycle of retinylidene imine isomerization and the acid -base equilibrium of the thermodynamic SB and kinetic PSB. Figure and scheme are reproduced from ref [43]. 122 determin ation of the pK a difference between both states (~ 5 pK a units) . These observations were similar to those observed in Dronpa and Padron, and further affirmed the feasibility of finding suitable hCRBPII mutants to photoisomerize the FR-1V/hCRBPII complexes . Listed in Table II-5, a no n-exhaustive hCRBPII mutagenesis screening yields a panel of hCRBPII mutants that are capa ble of being photoswitched between two colorful states (see Figure II-11 as represented by FR-1V/ps4). As denoted, the ÒOFFÓ state consists of mainly the SB species, and the ÒONÓ state consists of mainly the PSB species. Although a clear conclusion cannot be drawn as to mutations are critical for the photoswitching phenomenon , however combination of T51V, T53S/C, and Y19W seem important for producing a n efficient photoswitchable mutant. It should also be noted that, Table 0-5 Selected FR-1V/hCRBPII mutants showing fluorescent photochromism. Entry hCRBPII mutant ÒOFFÓ aÒONÓ bÒON/OFFÓ PSB Abs ratio c!abs, SB!em, SB !abs, PSB !abs, PSB !em, PSB ps1 Q108K:K40L:T51V:T53S 3704376376377101.8 ps2 Q108K:K40L:Y19W:R58Y 3954555685726741.2 ps3 Q108K:K40L:T51V:T53S:Y19W:R58L 3804396256297024.4 ps4 Q108K:K40L:T51V:T53S:Y19W:R58Y 3784456006006863.3 ps5 Q108K:K40L:T51V:T53S:Y19W:R58Y :S55Q 3764406166166965.1 ps6 Q108K:K40L:T51V:T53S:Y19W:R58W:L77W 3794546556557303.3 ps7 Q108K:K40L:T51V:T53C:Y19W:R58W:A33W 3774296196006772.0 ps8 Q108K:K40L:T51V:T53S:Y19W:R58W:V62S 3764456066146993.1 ps9 Q108K:K40L:T51V:T53S:Y19W:R58W:V62N 3784516056116962.6 aÒOFFÓ state is designated as the thermodynamic bound state of FR-1V/hCRBPII. It predominantly absorbs in the SB region, with a fractional absorption in the PSB region. bÒONÓ state is designated as the k inetic state of FR-1V/hCRBPII after photo -irradiation, showing a predominant absorption in the PSB region. Wavelengths are recorded in nanometers. cThe ratio is measured by dividing the PSB absorption intensity of the ÒONÓ state over that of the ÒOFFÓ stat e, as an indicator of the photoswitching ability. 123 the mutations do display a pronounced effect in tuning the emission wavelength of FR-1V as a result of its ICT character . Comparing the ÒONÓ state PSB absorption and emission wavelengths betwe en the mutants within Table II-5, one can see the absorption ranges from 572 Ð 655 nm, spanning 83 nanometers. The emission ranges from 674 Ð 730 nm with a span of 56 nanometers. Since the mutations impact the absorption more than the emission, one can conclude that the interactions between the mutants and FR-1V are not solely the result of solvatochromism. These can involve the potential hydrogen -bonding, the ground s tate and excited state deplanarization from molecular twisting, the cis/trans PSB isomerization in the ground state, or even protein reorganization. Shown in Figure II-11 are the absorption and emission spectra of the two photoswitching states of FR-1V/ps4 as a representative of the photoswitching complexes listed in Table II-5. H ere we te rm this complex as CrimFluor -1 ( CF-1). After the thermal binding, CF-1 shows an absorption with a major fraction at 378 nm and a minor fraction at 600 nm. A 5 -second UV irradiation (using a ~ 365 nm handset or a Xenon lamp UV! 5 sec !Yellow light, 10 min, !or RT 30 min !Figure 0-11 Normalized absorption and emission spectra of the two photoswitching states of CF-1 (FR-1V/ps4 complex). ÒOFFÓ state: the colorless thermal equilibrated state after binding. ÒONÓ state: the blue -colored state after UV irrad iation. The equilibrium illustrates the reversible switching under different conditions. 124 equipped with BP 300 -400 band filter) can switch the thermal ÒOFFÓ state to a kinetic ÒONÓ state that has a predominant absorption at 600 nm. Exciting the ÒOFFÓ state at 378 nm results in an intense 445 -nm blue em ission, with a tail at 686 nm (see Section I.16). The 600-nm excitation of the PSB maximum at the ÒONÓ state gives a 686 -nm emission. The ON-to-OFF switching can be driven either by a thermal process (30 min at room temperature in dark) or a photochemical process (10 min irradiation with yellow light). Because of the similarity to the ATR/CRABPII case, the photoswitching cycle is postulated to follow the same steps depicted in Scheme II-4. Initial binding of the fluorophore forms the thermodynamic trans -SB. UV irradiation isomerizes the fluorophore to a cis -SB. This cis -SB is quickly protonated to the NIR -emitting cis -PSB via GSPT due to its high pK a. Y ellow light irradiation or a thermal process can convert the kinetic cis -PSB to a more stable trans -PSB, which is subsequently deprotonated due to its low pK a. We can surmise the change in the pK a of the retinylidene as a functio n of the protein environment from crystal structures . Et2NNLys 108UVvisible lightlighthigh pKalow pKaEt2NNLys 108Et2NNLys 108HEt2NNLys 108HON-state, cis -PSB!abs = 600 nm, !em = 686 nm OFF-state, trans -SB!abs = 378 nm, !em = 445 nm predominant state prior to irradiation predominant state post UV irradiation thermodynamic state kinetic state Scheme 0-4 Postulated photoswitching cycle of CF-1. 125 I.10!Developing CF-1 into a NIR tag for no-wash in vivo imaging CF-1, or the ps4 mutant, is chosen as the optimal candidate for the performance test as a NIR photoswitching tag for live cell imaging. Although ps4 does not have the highest ÒONÓ/ ÒOFFÓ PSB photoswitching ratio, overall it exhibits fast binding rate (~ 20 min for completi on, as compare to ~ 1 h of ps5) and high absorption coef ficient and PSB quantum yield ( comparing to ps3).! First, the bindin g rate of ps4 and FR-1V was measured under pseudo -first order condition s. Shown in Figure II-12a, the initial addition of FR-1V displays an absorption at 420 nm as the free aldehyde residing in the hCRBPII binding cavity. A quick rise at 600 nm is observed within the first 20 min. This 600 -nm PSB absorption then slowly transforms into the 378 -nm SB absorption. Although i t is not possible to rule out the simultaneous formation of the kinetic cis -PSB and thermodynamic trans -SB, however, based on the mechanism of proton -catalyzed imine condensation, it is reasonable to Figure 0-12 Binding of ps4 (20 ! M) and FR-1V in PBS buffer (pH 7.3) over time, incubated at 25 ¡C. (a) Stacked traces of absorbance spectra, FR-1V (7 ! M); (b) Time -course absorbance spectrum monitored at 378 nm, FR-1V (6 ! M). Data points were fitted to an exponen tial rise function. 126 hypothesize a stepwise process composed of the initial cis -PSB formation and subsequent thermal isomerization and deprotonation to trans -SB. Monitoring at the 378 -nm SB absorption along the process, the intensity change perfectly follows an exponential increase as plotted in Figure II-12b, and gives a binding half -time of 8.7 min at 23 ¡C. This further supports the argument of the stepwise SB formation. If otherwise, a bi -exponential rise would be observed. Figure 0-13 Switching fatigue resistance of CF-1. Stacked traces of (a) absorbance spectra, (b) emission spectra, over cycles of alternate UV and yellow light irradiations. For each cycle, excitation at 378 nm lasts 20 s with U -360 UV Bandpass filter (B lue); excitation at 600 nm lasts 15 min with Y -50 (500nm) Longpass filter (Yellow). (c) CF-1 absorbance monitored at 600 nm, and (d) CF-1 emission intensities monitored at the maxima of ÒOFFÓ and ÒONÓ states in non -degassed solutions. 0.6% absorption and 1% emission intensities decreased per cycle. All measurements were in PBS (pH 7.3). 127 The next imp ortant property of a photoswitchable tag is its fatigue resistance, defined as the number of rounds it can be photoswitched before photodegradatio n. Shown in Figure II-13, a solution of CF-1 was subjected to an iterative UV - and yellow -light irradiation. The absorption and emission intensities of the SB and PSB species are monitored at the corresponding wavelength maxima. To mimic the condition of real live cell experiment, the solution of CF-1 was not degassed and supplemented with anti -fading additives. The spectra showed slight decreases in both absorption and emission after multi ple rounds of light irradiation (0.6 % loss in absorption and 1 % loss in emission per cycle ). This is possibly d ue to the oxidation of the conjugate imine moiety as observed in the hepta methine cyanines. 45 It should be noted that for the in vitro fatigue resistance measurement , a 500 W Xenon -Mercury lamp is used with two neutral density filters. The power density of the irradiation is much higher than that in a typical confocal microscopic Figure 0-14 CF-1Õs e mission switching time of the "OFF" and "ON" states. Monitored at the emission maxima under continuous excitation. 128 experiment. Overall , CF-1 has demonstrated excellent photostability and fatigue resistance in a buffered environment. The time duration of the NIR emission in the ÒONÓ state was recorded by monitoring the 686 -nm fluorescence intensity under continuous excitation. The intensity change is fitted to an exponential decrease that follows a first -order rate equation. The half -time is determined as 545 seconds for the PSB emission, and 121 seconds for the SB emission in the ÒOFFÓ state ( Figure II-14). The length of the emission duration should suffice to image a 2048 #2048 frame at the slowest line scann ing rate under standard confocal microscopic setup. A better understand ing of the switching me chanism is necessary before CF-1Õs application in live cell experiment. The OFF -to-ON photoswitching of CF-1 is rapid and takes about 2 -5 seconds. Limit ed by instrumentation, further investigation of the transient Figure 0-15 Temperature -dependent CF-1 ON-to-OFF thermal switching, absorbance monitored at 600 nm. ! 129 events was not pursued. However, the ON -to-OFF switching is much slower. It can be driven both photochemically and thermally. This leads to the question as to what the ratio of the two kinetic r ate constants is . To record the temperature effect on the thermal ON -to-OFF switching, a short UV irradiation was applied to a CF-1 solution sample to secure a complete ÒONÓ state before monitoring the 600 -nm absorption decay in dark. The sample was kept at different temperatures using a coolant -driven thermostat. The temperature was monitored with a thermocouple sensor placed inside the solution. Nitrogen gas was continuously flowed over the cuvette surface to prevent any moisture condensation. As shown in Figure II-15, the thermal switching rate is temperature dependent. The thermal switching half -time were determined by fitting the intensity changes to the previously mentioned first -order exponential deca y. Plotting these time constants over temperature according to the Arrhenius (Equation II-1, 1 ) and Eyring equations (Equation II-1, 2 and 3) , we can derive the corresponding thermochemical coefficients. As calculated, the activation energy derived from the Arrhenius equation is 27.7 kcal "mol -1,!matching well with the enthalpy deri ved from the Eyring plot (27.1 kcal "mol -1, Figure II-16). Following the conventional idea, the thermal switching should lead the kinetic PSB to a thermodynamically more stable state and that usually goes k=Ae!EaRTk=kBThe!"G#RTlnkT=!"H#R1T+lnkBh+"S#Rááááááááááááááááááááááááááááááááááááááá (1) ááááááááááááááááááááááááááááááááá (2) ááááááááááááááááááááá (3) Equation 0-1 Arrhenius and Eyring equations. (1) Arrhenius equation. (2) and (3) Eyring equations. 130 through an entropy decreasing process. However, it is interesting to obtain a positive entropy as 28.2 cal "mol -1"K-1# This m ight reflect the synergistic change in the protein scaffold along the switching p rocess. ! Knowing the thermodynamic constants of the ON -to-OFF thermal switching, we can deduce the photochemical contribution in the yellow light -assisted ON-to-OFF Figure 0-16 Thermal ON -to-OFF switching of CF-1. Rate constants plotted against temperature. (a) Arrheniu s plot. (b) Eyring plot. Figure 0-17 Thermal versus yellow light -assisted (LP 500 filter) ON -to-OFF switching of CF-1, monitored at absorption maximum. Data points collected at 3 -min interval. 131 switching. Shown in Figure II-17, the light -assisted switching seems a lot faster than the dark thermal switching (6.1 min vs. 10.6 min). However, after adjust ing for the small temperature difference, it appears that the photochemical rate constant only contributes to less than 20% of the whole switching process. On the other hand, it suggests that there is a large chemical space for accelerating the photochemical process. In this photoswitchin g system, t he key determinant for mitigating the non -specific fluorescence background is the NIR fluorescence of the kinetic PSB. The SB -PSB Figure 0-18 Working pH range of CF-1. Stack ed traces of emission spectra of ÒOFFÓ (Ex. = 378 nm) and ÒONÓ (Ex. = 600 nm) states: (a) pH 7 to 9; (b) pH 7 to 6. (c) pH -dependent fluorescence intensities of ÒOFFÓ and ÒONÓ states, showing 80% intensities within pH 6.6 ~ 7.6 range. 132 equilibria is affected by the pH environment. Therefore, it is critical to know the proper pH range in which the cis-PSB can stay as a minor fraction before and as a major fraction after the photoswitching , o r simply put, the working pH range .!The SB emission in the ÒOFFÓ state and the PSB emission in the ÒONÓ state were both recorded under various acidities. The fluorescenc e intensities at the corresponding maxima were plotted against the pH value s ( Figure II-18). Up to pH 7.6, CF-1 retains 80% of its ÒONÓ -state NIR emission . At low pH, the system is always on as the trans -SB is protonated to its PSB. Nevertheless, this will not be a problem for imaging under physiological conditions. As an important f act for live cell experiments , the cytotoxicity of FR-1V was evaluated by the MTT assay. The MTT assay relies on the reduction of the yellow tetrazolium dye 3 -(4,5 -dimethylthiazol -2-yl) -2,5-diphenyltetrazolium bromide to a purple formazan by the mitochondr ial reductas e.46 Quantification of the 550 -nm absorbance can reflect the number of viable cells after being exposed to the chemical for a certain period Figure 0-19 MTT assay of FR-1V in (a) HeLa and (b) U2OS cell lines. The absorbance was measured 550 nm, and corrected at 655 nm. Cells were treated with FR-1V for 24 h before MTT staining. Error bars represent the standard deviation of the readouts from ( n = 6) cells. 133 of time . HeLa and U2OS cells were stained with a serial dilution of FR-1V stock solution and incubated for 24 h before the colorimetric measurement. As shown in Figure II-19, FR-1V does not elicit observable cytotoxicity up to 4 !M in U2OS and 8 !M in HeLa cell lines after 24 h. It should be noted that the typical time span of a confocal imaging experiment does not exceed a couple of hours. So, a concentration range of 0.5 ~ 2 !M FR-1V should be practical. As discussed at the beginning of this chapter, one drawback of the FPs is their tendency to oligomerize. It has been shown in a variety of hCRBPII variants , mutations at certain specific positions can lead to a large number of domain -swapped dim ers. 47 In the case of CF-1 for its in vivo application, th e same concern needs to be addressed. The propensity of ps4 dimerization is evaluated by the size -exclusion chromatography of its over -expression in E. coli . As shown in Figure II-20, ps4 remains monomeric in the Figure 0-20 Size exclusion chromatogra phy of hCRBPII mutant ps4 expressed and purified from BL21(DE3)pLys S (buffer: 10 mM Tris "HCl, 2 M NaCl, pH 8.1) using a BioLogic DuoFlow QuadTec 10 system (Bio -Rad) equipped with a Superdex 120 16/600 GL size -exclusion co lumn (GE Healthcare). 134 extract solution up to millimolar concentration (20 mg/mL). O f course, in vivo assessment is necessary to confirm the pure monomeric nature of ps4. To test the performance of CF-1 in vivo , the ps4 fusion constructs with ECFP and EGFP were cloned as illustrated in Figure II-21, respectively. Using the same ECFP -GalT vector of the hCRBPII -KLVSWW:L117E plasmid ( Figure II-4a), t he reporter ECFP was fused on the C -terminus of ps4 for whole cell imaging. To target ps4 at cell nuclei, cytosol, or plasma membranes , the signaling peptides NLS (nuclear localization sequence) , NES (nuclear export sequence) , and CAAX (prenylation tag) were fused to the C-terminus of ps4, respectively , with an N-terminal EGFP as the reporter ge ne.48 The proper FR-1V staining duration was first investigated. As determined in Figure II-12b, the in vitro binding half -time is 8.7 min at 23 ¡C. The binding in the living mammalian cells is accelerated if incubated at 37 ¡C. The nuclei -localized ps4 was then transfected into U2OS cell lines and tested for FR-1V binding under different temperatures as monitored by the NIR PSB fluorescence. As expected, the binding rate increased at 37 ¡C . Shown in Figure II-22b, the CF-1 complexation was completed within 5 min , and the system was switched to the ÒONÓ state to present a bright NIR signal. a) b) Figure 0-21 Schematic maps of the (a) hCRBPII -ECFP and (b) EGFP -hCRBPII -SP fusion constructs. SP: signaling peptides. SP = 3 # NLS (nuclear localization sequence), NES (nuclear export sequence), and CAAX (prenylation tag). 135 Surprisingly, when the cells were stained at 20 ¡C, CF-1 did not form even with prolonged incubation up to 20 min ( Figure II-22a). This gives us an extra handle to control the timing for labeling. Both C - and N -terminal -fused ps4 constructs were transfected and expressed in HeLa, U2OS, and HEK293 cancer cell lines to test the applicability of CF-1. Sati sfactorily, in all cases the photoswitching feature and the ÒONÓ state NIR emission were secured. Figure 0-22 Time serial c onfocal fluorescence images of U2 OS cells expressing construct pFlagCMV2 -EGFP -ps4 -3# NLS, stained with 1 !M FR-1V and incubated at (a) 20 ¡C, and (b) 37 ¡C. Channels: EGFP (Green), "ex = 488 nm, BP 505 -530; CF-1_ON (Red), "switch = 405 n m, "ex = 594 nm, LP 650. Scale bar, 20 !m. 136 Typically, the cells were stained with FR-1V at a final concentration of 1 !M and incubated ECFP/EGFP !CF-1_ON !before switch !CF-1_ON !after switch !CF-1_ON+ BR ![CF-1]-ECFP ![CF-1]-ECFP !EGFP -[CF-1]-NLS !U2OS !HEK293 !HEK293 !Figure 0-23 CF-1 labeling in U2OS and HEK293 cell lines. Left and middle column: cells expressing C -terminal ECFP, showing whole cell fluorescence. Right column: cells expressing N -terminal EGFP, showing nuc lei -localized fluorescence . NLS = nuclear localization sequence. BR: bright -field. Channels: ECFP, "ex = 458 nm, BP 465 -510IR; EGFP, "ex = 488 nm, BP 505 -530; CF-1_ON, "switch = 405 nm, "ex = 594 nm, LP 650; Bright field, "ex = 488 nm. Cells were stai ned with 1 !M FR-1V and incubated at 37 ¡C for 5 min before imaging. No washing steps required. Scale bar, 20 ! m. 137 at 37 ¡C for 5 min. The cells were then directly subjected to the confocal imaging without any washing steps. As can be seen in Figure II-23, the NIR channel (LP650) emission is more detectable in the ECFP construct than the EGFP construct. Presumably, this might EGFP !CF-1_ON !before swi tch !CF-1_ON !after switch !CF-1_ON+ DIC !EGFP -[CF-1]-NLS !EGFP -[CF-1]-NES !EGFP -[CF-1]-CAAX !Figure 0-24 Compartmentalized CF-1_ON labeling in HeLa cells. (a) Labeling of different CF-1 fusion proteins. NLS = nuclear localization sequence. NES = nuclear export sequence. CAAX = prenylation tag. DIC = differential interference contrast. Cells were stained with 2 !M FR-1V and incuba ted at 37 ¡C for 5 min before imaging. No washing steps required. Scale bar, 10 !m. 138 be due to the C -terminal instead of the N -terminal FP fusion that slightly changes th e local pKa of the FR-1V PSB in the hCRBPII cavity. The subcellular labeling of CF-1 is demonstrated in HeLa cells (Figure II-24). The NIR fluorescence intensity has a 10 -fold increase after a 1 -sec 405 -nm laser switching. In the case of plasma membrane labeling, CF-1 is more advantageous than the reporter EGFP. It is observable that EGFP has intense fluorescence in places other than the plasma membrane, possibly due to the unwanted agglomeration. In contrast, CF-1 does not show these off -target signals. Looking at the NIR channel (LP650), it shows residual fluorescence before photoswitching. This signal comes from the small fraction of PSB in the ÒOFFÓ state. Closely examined, the NIR signal in the ÒOFFÓ state matches that in the ÒONÓ state. It should be noted that this residual NIR fluorescence is not the non -specific background Figure 0-25 Labeling specificity of nuclei -localized CF-1 in HeLa cells. Top left: EGFP channel; top right: CF-1_ON channel; bottom left: merged channels + brightfield; bottom right: line profiles of fluorescence intensities in two channels along the blue arrow in the merged image . Scale bars, 20 µm. ! 139 we commonly encounter due to off -target labeling and /or extra dye residing in the hydrophobic compartments. It is the specific signal from the labeled target. A fluorescence line profile is plotted to examine the labeling specificity of CF-1 against the fused EGFP (Figure II-25). As shown, in each pixel along the blue arrow, the NIR fluorescence signal of CF-1 is colocalized with the green fluorescence signal of EGFP, attesting the labeling specificity of CF-1 as high as that of FPs. I.11!Engineering ps4 into a truly fluorogenic tag for spatiotemporal imaging Having demonstrated the imaging capability of CF-1, we next examined the possibility to further engineer CF-1 as a fluorogenic tag w ithout any residual PSB in the ÒOFFÓ state, so that a true ÒdarkÓ NIR channel would be available before photoswitching. This would greatly increase the imaging contrast after photoswitching and would also allow for the spatiotemporal control of the NIR emission . We hypothesized that the installation of more hydrophobic residues in the vicinity of the embedded FR-1V would destabilize the iminium (cationic specie s). This would hav e two potential consequences: 1) lessening or complete suppression of the residual PSB UV! 15 sec !Yellow light, 5 min, !or RT 10 min !Figure 0-26 Normalized absorption and emission spectra of the two photoswitching states of CF-2 (FR-1V/ps5 complex) . 140 absor ption in the ÒOFFÓ state, and 2) an accelerated ON -to-OFF thermal switching rate, by increasing the electrostatic penalty for maintaining a charge d species. After several rounds of altering the ps4 sequence (in collaboration with Dr. Setare Tahmasebi Nick ), ps5 (Q108K:K40L:T51V:T53S:Y19W:R58W:T29L:Q38L:Q128L ) was identified. Illustrated in Figure II-26, t he FR-1V/ps5 complex (designated CF-2) showed negligible PSB absorption in its ÒOFFÓ state, while its absorption and emission maxima, QY and fatigue resistance were similar to those measured for CF-1 (Table II-6 and Figure II-27), but with a 6 -fold increase in the in vitro ON-to-OFF thermal switching rate, and a 4.6-fold increased rate of complex formation ( Table II-6 and Figure II-28). The N-terminal EGFP -fused ps5 constructs with C -terminal signaling peptides were cloned and transfected into HeLa cells. The imaging protocol is similar to that of CF-1. Direct confocal imaging was taken after a quick staining with FR-1V for only 5 min at 37 ¡C, negating any washing st ep. Compare d to the 1 -sec OFF -to-ON switching of CF-1, CF-2 takes 4 seconds to completely switch from ÒOFFÓ to ÒONÓ. Nevertheless, in the Figure 0-27 Switching fatigue resistance of CF-2 monitored at 605 nm absorbance in PBS (pH 7.3). For each cycle, excitation at 378 nm lasts 20 s with U -360 UV Bandp ass filter; excitation at 600 nm lasts 15 min with Y -50 (500 nm) Longpass filter . 141 compartmentalized imaging assays, a truly dark NIR channel is acquired before the photoswitching with a 405 -nm laser (Figure II-29). A similar pixelated line -profile analysis of the colocalized EGFP and CF-2 signals indicates the same level of labeling specificity (Figure II-30). The mean signal -to-background ratio (SBR) of the NIR fluorescence can be gauged in the ÒONÓ state. Taking the nuclei -localized CF-2 images as an example, the mean NIR emission intensity of multiple cell nuclei on different focal planes is divided by the mean background intensity of the dark areas to derive the mean S BR. 49 A r emarkable SBR of 46 is achieved by this analysis ( Figure II-30). Table 0-6 Feature comparison between ps4 and ps5 . Mutant SB!abs / !emPSB!abs / !emQYaSBQYaPSBBinding t1/2 (min) ON !OFFbt1/2 (min) CF-1ps4 378 / 445 600 / 686 0.46 0.16 8.7 7.4 CF-2ps5 382 / 440 604 / 684 0.52 0.13 1.9 1.2 aAbsolute quant um yields measured at pH 7.3. bSwitching half -lives were measured in dark (thermal conversion) at 2 3 ¼C. Figure 0-28 (a) Time -course absorbance spectrum monitored at 382 nm of ps5 and FR-1V binding. Data points were fitted to an exponential rise function. (b) Time -course spectrum of CF-2 ON-to-OFF thermal switching at 23 ¡C. Absorbance monitored at 605 nm. Data points were fitted to an exponential decay function. 142 The ON-to-OFF thermal switching rate is also investigated in vivo and compared between CF-1 and CF-2. Basically, a short 405 -nm laser irradiation was applied to switch the sample cells to the ÒONÓ states. Then a selected region o f interest (ROI) in the field EGFP CF-2_ON before switch CF-2_ON after switch CF-2_ON+ DIC EGFP -[CF-2]-NLS EGFP -[CF-2]-NES EGFP -[CF-2]-CAAX Figure 0-29 Compartmentalized CF-2_ON labeling in HeLa cells. Labeling of different CF-2 fusion proteins. NLS = nuclear localization sequence. NES = nuclear export sequence. CAAX = prenylation tag. DIC = differential interference contrast. Cells were stained with 2 !M FR-1V and incubated at 37 ¡C for 5 min before imaging. No washing steps re quired. Scale bar, 10 !m. 143 of view was monitored for the NIR emission intensity change. Compared to the in vitro rates , the ON -to-OFF switching is accelerated in the live cells even at 20 ¡C (Figure II-31). For CF-1, the h alf-time is shortened from 7.4 min to 2.7 min , while CF-2 goes from 1.2 min to 0.68 min. The faster in vivo switching rate allows for iterative NIR fluorescence turn -on and turn -off in live cells. As demonstrated in Figure II-32, 10 switching cycles were performed with plasma membrane -labeled CF-2 in HeLa cells , showing minimal loss of brightness. As for the ultimate goal of controlling the NIR fluorescence in a spatiotemporal manner, nuclei -localized CF-1 was first tested in vivo . After staining with FR-1V, the cells were kept in the ÒOFFÓ state. A selected cell nucleus was flashed with a 405 -nm switching Figure 0-30 Labeling specificity of nuclei -localized CF-2 in HeLa cells. Top left: EGFP channel; top right: CF-2_ON channel; bottom left: merged channels + DIC; bottom right: line profiles of fluorescence intensities in two channels along the blue arrow in the merged image . Scale bars, 20 µm. Circles and number indicating mean SBR (n = 440). 144 laser in a cross hair mode. Unfortunately, the OFF -to-ON photoswitching of CF-1 is ultrafast and ultra -sensitive. Even a 1 -sec irradiation with th e focused laser beam can Figure 0-31 ON-to-OFF thermal switching kinetics of (a) CF-1 and (b) CF-2 in HeLa cells. Region of interest was excited continuously with 594 nm laser (1.63 !W). Mean ROI intensity was recorded every 6 sec. 1st cycle -ON!1st cycle -OFF !2nd cycle -ON!2nd cycle -OFF !5th cycle -ON!10th cycle -ON!Figure 0-32 In cellulo NIR fluorescence (LP650) ÒON/OFFÓ cycling of HeLa cells expressing EGFP -ps5 -CAAX, showing 10 switching cycles. Arrows: OFF -to-ON, 405 nm laser 4 sec; ON -to-OFF, r.t. 3 min. Scale bar: 10 !m. 145 switch ÒONÓ the whole field of view. This also suggest the photoactivation rate is in the millisecond rang e. Mentioned earlier, CF-2 has a slower OFF -to-ON swi tching rate . Luckily, this inhibits the whole field activation of CF-2 under the same crosshair laser beam, and successfully leads to the spatiotemporal control of NIR fluorescence turn -on as showcased b y the nuclei -localized CF-2 (Figure II-33). The comparison of CF-1 and CF-2 with some well -known far -red/NIR FPs are listed in Table II-7. The NIR brightness is quantified as the product of extinction coefficient at 635 nm and the fractional quant um yield in the range of 700 ~ 900 nm. 50 CF-1 and CF-prior to light activation !fuse EGFP !selective 1 st cell !activation !2nd cell !activation !3rd cell !activation !whole field !activation !Figure 0-33 Spatiotemporal c onfocal imaging of HeLa cells, expressing nuclei -localized EGFP -fused ps5 . Cells were stained with 2 !M FR-1V for 5 min at 37 ¡C and imaged without washing . Red channel: CF-2_ON. Following the arrows: before switch -on; 1 st cell switch -on; 2 nd cell s witch -on; 3 rd cell switch -on; whole field switch -on. Compare with green channel: colocalized reference EGFP. Scale bars, 10 !m. 146 2 have shown superior NIR brightness as 22 -fold and 12 -fold higher than that of Katushka (the dimeric parent of mKate), respectively. I.12!Conclusion and future research directions To date, CF-1 and CF-2 are the first reported photoswitchable NIR tags that employ both a synthetic dye and a fusion protein. The design of CrimFluors accomplishes three complimentary goals towards a superior imaging platform, namely, 1) the use of ICT capable chromophores, 2) generation of a bathochromically -shifted PSB, and 3) the ability to photoactivably turn -on and turn -off the fluorescence signal. The use of ICT capable chromophores leads to Òbackground -freeÓ no -wash NIR f luorescence as a result of the large bathochromic shift of the chromophore upon protonation of the imine, which happens only when bound to the engineered protein, because non -specifically formed imines are not protonated under physiological conditions. The ÒON/OFFÓ fluorescence switch is presumably a result of fluorophore photoisomerization. Compared to RSFPs, CF-1 and CF-2 have larger Stokes shifts and more red -shifted NIR fluorescence. Tested Table 0-7 Comparison of CF-1 and CF-2 with other far -red/NIR FPs. a Katushka E2-Crimson mNeptune eqFP650 eqFP670 CF-1 (ON) CF-2 (ON) Excitation peak (nm) 588605599592605600604Emission peak (nm) 635646649650670686684Fluorescence QY0.34 0.12 0.18 0.24 0.06 0.16 c0.13 c!max (M-1ácm -1) 65,000 58,500 57,500 65,000 70,000 50,500 35,700 Brightness (a.u.) b22,100 7,080 10,350 15,600 4,200 8,080 4,641 !635(M-1ácm -1)1,700 12,640 7,900 4,300 15,700 37,550 27,940 "(700 Ð900 nm) d0.07 0.03 0.05 0.07 0.03 0.07 0.05 Brightness in NIR (a.u.) e119 3793953014712,630 1,400 Relative NIR brightness 13.2 3.3 2.5 4.0 22.1 11.8 aData of Katushka, E2 -Crimson, mNeptune, eqFP650 and eqFP670 are from ref [50]. bCalculated as the product of molar extinction coefficient and quantum yield. cAbsolute quantum yield measured in PBS at pH 7.3. dCalculated from the emission fraction between 700 nm and 900 nm. eCalculated as the product of molar extinction coefficient at 635 nm, and quantum yield in NIR . 147 in different cell lines, CF-1 and CF-2 can be turned to the ÒON Ó state by a 405 -nm laser with a 1 sec - and 3 sec - irradiation time, respectively, and both have good photostability. If optimal NIR brightness is required, then CF-1 is an optimal choice. On the other hand, CF-2 offers a complete dark state with a faster ON -to-OFF thermal switching rate. CrimFluors can c omplement current FPs for live cell imaging and are potential tags for bright deep -tissue imaging . I.12.1!Future structural modification of FR -1V One of the future research directions is further structural optimization of FR-1V for extended lifetimes of the emissive PSB species , such that the NIR brightness can be increased if the radiative dec ay constant does not alter much. Substitution on the 3, --conjugated double bond with electron -donating, electron -withdrawing, and steric functional groups are being attempted . The structur al relationship with the photo-interconversion rates are under investigation. Hopefully, a faster ON -to-OFF photo -conversion rate can be obtained to enable a pure photo -controllable imaging tag for super -resolution microscopy . XNR2R1OR3R4XX = S, Se OMeO OMe (2 equiv.) ZnBr 2/DCM(1 equiv.) 0 ¡C, 15 min rt, 2 h +XR1, R2 = aryl, heteroaryl R3, R4 = EDG, EWG, etc. X = S, Se (or CMe 2 for FR-1V )X = S, (88%) X = Se, (83%) XX = S, Se OMeO OMe (4 equiv.) ZnBr 2/DCM(2 equiv.) 0 ¡C, 15 min rt, 2 h +X = S, (84%) X = Se, (81%) Scheme 0-5 Potential derivatives of FR-1V for improving optical p roperties. 148 Another direction of modificatio n is to replace the C9 of FR-1V fluorenyl to more electron -diffused S, Se, and Si. This type of substitution can increase the conjugation of the scaffold and may further lower the band gap of the derivative. Hopefully, the emissions of these derivatives can be further red -shifted into the long wavelength reg ion of NIR -I, with the minimal trade -off in overall brightness and Stokes shift. One feasible synthetic route is given in Scheme II-5 to construct the hetero -homologue s of FR-1V. The core scaffolds can be furnished by benz-annulation of thiophene or selenophene with 2,5 -dimethoxytetrahydrofur an.51 The installment of the rest side chains should be similar to those of FR-1V. I.12.2!FR-1V/hCRBPII as a dual -emission ratiometric pH probe Now letÕs think in the opposite direction. What can we do with a non -switchable FR-1V/hCRBPII complex? Shown in Table II-6 as an example, the ÒOFFÓ state has a common 450 -nm SB emission with ~ 50% quantum yield. The strong fluorescence can also be utilized for a quantitative imaging measurement in addition to the NIR PSB emission. If the complex is not photoswitchable, then we can simply ascribe the ratio of the SB and PSB presented as per the acid -base equilibrium, whether it involves cis -trans isomerization or not. Therefore, a pH -dependent rat iometric fluorescence readout can be acquired based on the blue and NIR emissions from the equilibrated system. Wavelength ratiometry is always a preferable method in sensing and imaging intermolecular interactions. Especially in UV/Vis and fluorescence sp ectrophotometry, wavelength ratiometry is clearly advantageous in comparison to simple intensity -based absorbance or emission detection by avoiding the misinterpretation from inconsistent 149 concentration of the pigment or emitte r.52 Compare d to normal single -wavelength ratiometry, a dual -emission wavelength ratiometry has even more advantages. For example, when the ratiometry is based on the single -wavelength fluorescence intensity ratio from two different excitation wavelengths, a potential problem is the light source power density fluctuation in the two excitation wave lengths, which will invalidate the ratiometric nature of the readout. On the other hand, a dual -emission wavelength ratiometry is self -calibrated. No matter what fluc tuation may occur with the light source, it will reflect simultaneously in the dual -emission and cause no devi ation in the readout. Other than the compensation of light source fluctuation, a dual -emission sensing system can also simplify the readout if the two emission maxima are spectrally well -resolved. In this case, the deconvolution and integration of the fluorescence spectra is not necessary. Instead, the ratio can directly be derived from the intensities of the two emission maxima. A single wavelength -emission ratiometry is based on the chemical transformation of the initial emitting species to a different emissive species elicited by the intermolecular interaction (e.g., pH change). If the two species have different extinction coefficients and fluores cence quantum yields and significant spectral overlap, it is almost impossible to do the spectral deconvolution. More often than not, the overall ratiometric intensity change is just described as a Òblack boxÓ. The detailed mechanis m is usually ignored on purpose, and a working curve is commonly plotted with experimental results used directly. However, if a dual -emission ratiometric system can be built on the straightforward acid -base 150 equilibrium, then a clear mathematical description can be derived instead of the Òblack boxÓ working curve . In view of this, a mutant -screening was carried out to find proper non -switchable candidates . A high PSB/SB absorbance ratio is desired because the PSB has a lower quantum yield than the SB , and the comparable brightness is essential to give a better dynamic range for the ratio readout. hCRBPII mutant ph1 (Q108K:K40L:T51V:T53S: lgFI455FI708!"#$%&=!pH()'pKa+!R2=0.9938a) b) c) d) Figure 0-34 The prototype of a dual -emission ratiometric pH probe. a) Stacked traces of the absorption spectra of FR-1V/ph1 (Q108K:K40L:T51V:T53S:V6 2E) complex upon iterative light irradiation. UV = U-360 UV Bandpass filter . VIS = Y -50 (500 nm) Longpass filter. Stacked traces of b) SB and c) PSB emission under different pH. d) The correlation of the logarithmic fluorescence ratio and pH. A linear func tion is presented. 151 V62E) was identified as the prototype for the dual -emission ratiometric pH probe. It has a PSB/SB absorbance ratio o f 1.5. It does not show any photoswitching even under extended UV -light irradiation (Figure II-34a). As shown, the logarithmic ratio of the fluorescence intensity of i ts SB and PSB upon pH titration can be perfectly fitted to the linear function ( Figure II-34d) derived from the acid -base equilibrium. Its probing capacity span s a broad pH range over 6.5 ~ 9. There is a 100 -fold increase from the lower limit to the upper limit of the fluorescence intensity ratio . This provid es a wide dynamic range and a high sensitivity. The binding of FR-1V and ph1 has a 3.1 min as the half -time. To apply this prototypic pH probe for live cell application, future work needs to be focused on increasing the quantum yields. Currently, it has a QY of 13% and 8% for SB and PSB, respectively. Another issue to address is the non -specific backg round, because the complex no longer has a mechanism to screen away off-target fluorescence. I.13!Experimental section Materials and g eneral methods All chemicals and solvents were purchased from Sigma -Aldrich and used without further purificatio n. NMR ( 1H, 13C) spectra were recorded with Agilent DirectDrive2 500 MHz spectrometer and referenced with the residual 1H peak from the deuterated solvents. HRMS (ESI) analysis was done by using a Waters Xevo G2 -XS QTOF mass spectrometer (Agilent) and refe renced against Polyethylene Glycol (PEG -400-600). Site -directed mutagenesis and cloning 152 All DNA oligonucleotides used were purchased from Integrated DNA Technologies. Restriction endonucleases were purchased from New England BioLabs. PCR -amplified and restriction enzyme -digested DNA was purified by Wizard ¨ SV Gel and PCR Clean -Up kit (Promega). All plasmids were amplified with QIAprep ¨ Spin Miniprep Kit (QIAGEN). Ligation was performed with T4 DNA Ligase (New England BioLabs). All PCR and ligation products were transformed into XL1 -Blue competent cells (Agilent), which were then plated onto LB -Amp/Tet agar. All site -directed mutagenesis PCR were performed using PfuTurbo DNA Polymerase (Agilent) following a standard TD -PCR protocol. All E. Coli constructs were in the pET -17b vector (Addgene). Mammalian constructs were generated through routine combination of PCR and restriction -enzyme cloning. Cloning -related PCRs were performed with Phusion ¨ High-Fidelity DNA Polymerase (New England BioL abs). ECFP -coded construct was cloned by inserting hCRBPII sequence into the ECFP -GalT vector (Addgene) with XhoI and BamHI sites. EGFP -coded constructs were cloned by inserting hCRBPII into the pFlag -CMV2 vector (Sigma -Aldrich) with NotI and EcoRI sites. EGFP was inserted before hCRBPII with HindIII and NotI sites. Cell nucleus, cytosol, and plasma membrane were targeted by inserting triplicated -NLS (DPKKKRKV), NES (ELAEKLAGLDIN) and CAAX (GKKKKKKSKTKCVIM), respectively, after hCRBPII with EcoRI and BamHI sit es. 48 Protein expression and purification hCRBPII mutants in pET -17b vector were expressed in E. Coli BL21(DE3)pLysS competent cells (Promega). The transformed cells were grown in LB growth medium 153 supplemented with 50 µg/mL ampicillin and 30 µg/mL chloramphenicol at 37 - for 6 h, then induced with 1.0 mM IPTG and further grown at 23 - for 18 h. The cells w ere harvested and lysed with Model 300 V/T Ultrasonic Homogenizer (BioLogics Inc.). hCRBPII protein in the supernatant was collected by ion exchange chromatography (Q Sepharose Fast Flow resin, GE Healthcare). The eluent was desalted by buffer exchange with Tris buffer (10 mM TrisáHCl, pH=8.0) using an ultrafiltration cell under nitrogen flow (~20 psi) through a 10,000 MW cutoff membrane (Ultrafiltration Membrane; material: regenerated cellulose; filter code: YM10; diameter 63.5 mm; NMWL: 10,000; Millipore TM). Further purification was done by Bio -Rad BioLogic DuoFlow QuadTec 10 system or NGC Discover TM system equipped with SOURCE ¨ 15Q column (GE Healthcare) using 10 mM Tris "HCl and 2 M NaCl at pH 8.1, followed by size exclusion chromatography with a Superdex 120 16/600 GL column (GE Healthcare). UV/Vis and fluorescence spectroscopy UV/Vis spectroscopy was performed with Cary 300 Bio UV -Visible Spectrophotometer (Varian) using 1 -cm, 1.0 -mL quartz microcuvettes (Starna Cells). Fluorescence spectroscopy was per formed on a Fluorolog ¨-3 spectrofluorometer (Jobin Yvon, Horiba Scientific) with 1 -cm, 3.5 -mL quartz cuvettes or 1 -cm, 1.0 -mL quartz microcuvettes (Starna Cells). All measurements were taken at ambient temperature (23 ± 1 - ). Extinction coefficients are re ported at "max ()) as averages ( n = 3). Protein samples (20 µM) were measured in PBS solutions. Normalized spectra were shown for clarity unless otherwise noted. 154 Protein extinction coefficient measurement Protein extinction coefficients were measured as previously described by Gill and Vonhippel .53 The theoretical extinction coefficient is calculated based on the following formula: where a, b and c are the numbers of Trp Tyr and Cys residues, respectively, in the sequence of given mutant. The extinction coefficients of the three residues were determined at 280 nm previously ( )Trp = 5690 M -1ácm -1, )Tyr = 1280 M -1ácm -1, )Cys = 120 M -1ácm -1). The concentration of the protein can be measured in 6 M GuanidineáHCl solution (denaturing reagent) as per Lambert -BeerÕs Law: where l is the cuvette path length, c is the concentration of the protein. The absorbance of the same amount of the protein was measured in PBS in the same cuvette. Thus , the experimental extinction coefficient of the protein can be derived by: Absolute fluorescence quantum yield measurement Absolute fluorescence quantum yields ( . ) were recorded at room temperature on a Quantaurus -QY C11347 -11 (Hamamatsu Photonics K.K., Ja pan) equipped with a Xenon lamp and a monochromator as excitation light source, an integration sphere, and a multichannel back -thinned CCD detector. All samples were diluted with PBS solution or corresponding organic solvents ( A < 0.1). Recorded values are average numbers ( n = 5). !Theor=a!!Trp+b!!Tyr+c!!Cys!Exper=Abs280,NatAbs280,Den!!TheorAbs280,Den=!Theor!c!l 155 Relative fluorescence quantum yield measuremen t54 Relative quantum yields were determined with reference to the standard fluorescent dye Oxazine -170 (purchased from Across Organics, Lot# A0098689) . For individual standard or dye -hCRBPII complex, the integrals of corrected emission spectra were plotted against the absorbance at the excitation wavelength at various concentrations. The obtained slopes of protein -fluorophore complexes were compared with tha t of the standard reference. The absorbance of all samples and reference were kept below 0.1 unit to prevent any potential scattering. According to Kasha -VavilovÕs Rule, that generally the quantum yield is independe nt of the excitation wavelen gth,55 all samples were excited at 565 nm to match the profile of Oxazine -170, and the emission data were collected from 575 nm to 800 nm. Temperature was kept constant at 20 ¡C during the course of measurement. The absolute quantum yield of Oxazine -170 was taken as 0. 579 in eth anol.56 The calculation of the sample quantum yield was corrected for the different refractive indices of ethanol ( "=1.362 , 20 ¡C)57 and PBS ( "=1.334 )58: where the subscripts ST and X denote reference and sample, respectively, * is the fluorescence quantum yield, Slope is the gradient of the plot of integrated fluorescence intensity over absorbance, and " the refractive index of the solvent. Dye-hCRBPII binding kinetics !X=!ST"SlopeXSlopeST"!X!ST#$%&'(2 156 Binding kine tics were measured at 23 ¡C. hCRBPII mutant (20 !M in PBS) was mixed with the fluorophore ( 0.8 ~ 1 equivalent) and increase in a bsorbance at the corresponding "max was recorded. Collected data points were fit with a first -order exponential rise function to determine the binding half -time as described below: where A is the absorbance at each time point, Ao is the final absorbance when the complex formation ends, k is the pseudo -first order rate constant, t is the time elapsed, and c is a varying constant which accounts for any time delay from the addition of the fluorophore to the point recording started. Switching fatigue, working pH, and temperature -dependent measurements For in vitro photoswitching fatigue resistance measurement, alternate illumination was performed by using an Hg(Xe) Arc lamp (Oriel TM Instruments) attenuated with two neutral -density filters (Edmund Optics Inc.). Samples in PBS solution were illuminated with U -360 UV Bandpass filter (Edm und Optics) for 20 s ec to switch ÒONÓ , and were illuminated with Y -50 (500nm) Longpass filter (Edmun d Optics) for 15 min to switch ÒOFFÓ . A model MWTC Type -K thermocouple sensor (Omega Engineering Inc.) was used to record the temp erature inside the microcuvette during temperature -dependent measurements. Fluorescence spectral change of samples in PBS solution were measured by titrating with 0.1 M HCl aqueous solution or 0.1 M NaOH aqueous sol ution. pH values were recorded with an ac cumet TM Basic pH meter (Fisher Scientific) equipped with a PerpHect TM ROSS TM Micro Combination pH electrode (Thermo Scientific Orion). A=A0!1"e"kt()+c 157 Mammalian cell culture and transfection All cell lines were cultured at 37 - with 5% CO 2 in DulbeccoÕs Modified EagleÕs Medium (DMEM) supplemented with 1.0 mM sodium pyruvate (Life Technologies), 10% fetal bovine serum (FBS; BioWest), 1% penicillin, and 1% streptomycin (Life Technologies). All cell lines were used without authentication and m ycoplasma detection, since the conclusions of this study made from the images acquired would not be affected. Cells were seeded 1 d before transfection on an ibidi µ -Slide 8 well coverslip (with ibiTreat). Transfection was performed at ~70% confluency with 1.0 µg of plasmid DNA using Lipofectamine TM 3000 reagent (Life Technologies) according to the manufacturerÕs protocol. The medium was replaced after 8 h. Cells were washed with DulbeccoÕs phosphate -buffered saline (DPBS; Life Technologies) twice after ano ther 36 h incubation and incubated in Phenol Red -free RPMI 1640 medium (Life Technologies). Prior to confocal imaging, the fluorophore (in ethanol solution) was added to cells at a final concentration noted in text and incubated at 37 - with 5% CO 2 for 5 m in. All wash ing steps were omitted. Confocal imaging Images were acquired using a Zeiss 510 Meta ConfoCor 3 Laser Scanning Microscope configured on a Zeiss AxioObserver Z1 automated inverted microscope platform, equipped with Argon/2, Diode 405 -30, and He Ne 594 laser sources. DIC images were acquired using a µ -Slide DIC lid (ibidi). S patiotemporal Òturn -onÓ images were taken with a Plan -Apochromat 63x/1.40 Oil DIC M27 objective, all other images were taken with 158 an EC Plan -Neofluar 40x/1.30 Oil DIC M27 obje ctive. ZEN 2009 and LSM Image Browser software (Zeiss) were used to process images. Image acquisition used the following settings. ECFP: excitation 458 nm, main dichroic HFT 458, secondary dichroic NFT 545, and detection channel BP 465 -510IR; EGFP: excitat ion 488 nm, main dichroic HFT 405/488, secondary dichroic NFT 545, and detection channel BP 505 -530; DIC: excitation 488 nm, main dichroic HFT 405/488; Switching: excitation 405 nm; CF: excitation 594 nm, main dichroic HFT 488/594, secondary dichroic NFT 5 45, and detection channel LP 650. All images were scanned in line mode with bit depth of 16 and pixel size of 0.22 µm. Kalman averaging 4 was a pplied. When imaging CrimFluor_ÒOFFÓ and ÒONÓ , the identical microscope settings were used before and after photo switching, and the signal intensities were compared within a normalized dynamic range. Signal -to-background ratio was determined by measuring the mean fluorescence intensity of nuclei relative to adjacent regions outside nuclei in four fields of view using Fiji (n = 440 areas as noted in text ).59 Cell viability assay Cell viability was determined by using the tetrazolium dye 3 -(4,5 -dimethylthiazol -2-yl) -2,5-diphenyltetrazolium bromide (MTT, ThermoFisher Scientifi c).46 HeLa or U2OS cells were seeded at 5,000 cells per well in a 96 -well plate 1 d before the assay. When cells reach ed ~ 80% confluency, FR-1V was applied to cells ( n = 6) at a series of concentrations to span the standard working condition of live cell experiments. Cells were then incubated at 37 - for 24 h before addition of MTT, to examine the toxicity at a prolonged time span, although a typical confocal experiment would not last longer than a 159 few hours. The absorbance of dissolved formazan was recorded at 550 nm and referenced against 655 nm. General synthe tic procedures Commercially available starting materials were obtained from Sigma -Aldrich and were used without further purification unless specified. All moisture sensitive reactions were carried out in flame -dried or oven -dried glassware under an atmosphere of nitrogen or argon. Unless other wise mentioned, solvents were purified as follows: tetrahydrofuran (THF) and diethyl ether (Et 2O) were distilled freshly from the classical sodium/benzophenone ketyl still pot; dichloromethane (DCM), acetonitrile, and toluene were dried over CaH 2 and fresh ly distilled prior to use; dimethylsulfoxide (DMSO), dimethylformamide (DMF), and triethylamine (Et 3N) were distilled from CaH 2 and stored over activated molecular sieves. Chemical shifts were reported relative to the residual solvent peaks. ( 1H-NMR: 5 7.26 ppm for CDCl 3, 5 3.31 ppm for CD 3OD, 5 2.50 ppm for DMSO -d6, respectively. 13C-NMR: 5 77.16 ppm for CDCl 3, 5 49.00 ppm for CD 3OD, 5,39.52 ppm for DMSO -d6, respectively.) Analytical thin layer chromatography (TLC) was performed with pre -coated silic a gel 60 F 254 plates (Analtech, Inc.) Compounds in TLC were visualized upon UV irradiation and various staining techniques, i.e. p-anisaldehyde, potassium permanganate, phosphomolybdic acid in ethanol. Silica gel flash column chromatography was performed w ith Silicycle 40 -60 † (30 ~ 75 µM) silica gel. 160 Synthesis of ThCC 2-Methylbenzothiazole (56 )60 p-Toluenesulfonic acid monohydrate (0.10 g, 0.5 0 mmol, 10 mol%) was added to a mixture of 2 -aminothiophenol ( 55, 0.63 g, 5 .0 mmol, 1 .0 equiv.) and triethyl orthoacetate (0.81 g, 5 .0 mmol, 1 .0 equiv.). The mixture was stirred at room temperature for 1.5 h to reach full conversion. Cold water (10 mL) was added to the mixture under vigorous stirring. The resulting mixture was subjected to acid -base extraction with EtOAc (5 mL # 3). The combined organic layer was dried over Na 2SO4 and concentrated in vacuo to afford a dim yellow oil as the product (0.70 g, 9 4% yield). 1H-NMR (500 MHz, DMSO -d6): 5 (ppm) = 7.97 (d, J = 8.4 Hz, 1 H), 7.83 (d, J = 8.4 Hz, 1 H), 7.48 Ð 7.43 (m, 1 H), 7.38 Ð 7.33 (m, 1H), 2.85 (s, 3 H). ESI -MS calcd [M+H] +: 150.04, found: 150.0. 2,3-Dimethylbenzothiazolium iodide (57) SHNH2CH3C(OEt) 3, 1 equiv. HOTs, 10 mol% neat, r.t., 1.5 h 94 %, WS_I_71 NSNSICH3I, 2 equiv. ACN, 90 ¡C, 4 h sealed tube 89 %, WS_II_46 ClO4OSNNSN1) POCl 3, 1.1 equiv. DMF, 3.3 equiv. 0 ¡C ! r.t., 30 min 2) 70 ¡C, 1.5 h 3) NaClO 4, 1.4 equiv. MeOH, 0 ¡C, overnight 5 M KOH aq. CHCl3, reflux overnight two step 47 % WS_I_88 55565758ThCC-1V Scheme 0-6 Synthesis of ThCC -1V. 161 2-Methylbenzothiazole ( 56, 1.0 g, 6.7 mmol, 1 .0 equiv.) was dissolved in acetonitrile (6 mL) . Then iodomethane (1.9 g, 13.4 mmol, 2 .0 equiv.) was added and the mixture was kept at 90 ¡C for 4 h in a sealed tube. After cooling to room temperature, the preci pitate was collected by filtration, washed with dichloromethane (15 mL # 2), and dried in vacuo to afford a white solid (1.74 g) with 89% yield. 1H-NMR (500 MHz, DMSO -d6): 5 (ppm) = 8.43 (d, J = 8.2 Hz, 1 H), 8.29 (d, J = 8.5 Hz, 1 H), 7.93 Ð 7.87 (m, 1 H), 7.84 Ð 7.78 (m, 1H), 4.20 (s, 3H), 3.17 (s, 3 H). 13C-NMR (125 MHz, DMSO -d6): 5 (ppm) = 177.24, 141.55, 129.22, 128.67, 128.06, 124.44, 116.73, 36.13, 17.03. ESI -MS calcd [M -Cl]+: 164.05, found: 164.0 2-Formylmethylene -3-methylbenzothiazoline (ThCC -1V) Phosphorous oxychloride (58 mg, 0.38 mmol, 1.1 equiv.) was dissolve d in DMF (dry , 83 mg, 1.1 mmol, 3.3 equiv.) at 0 ¡C then warmed to room temperature. After 30 min, the resulting yellow solution was added dropwise to a solution of 57 (100 mg, 0.34 mmol, 1 .0 equiv.) in DMF ( dry , 1 mL) preheated to 70 ¡C over a period of 15 min. The rea ction mixture was stirred for 1.5 h at 70 ¡C. The reaction mixture was cooled to room temperature, poured into methanol (3.5 mL) containing NaClO 4 (59 mg, 0.48 mmol, 1.4 equiv.), and chilled at -20 ¡C overnight. The precipitated deep red crystal s 58 were filtered and stirred in a mixture of chloroform (15 mL) and NaOH aq. (5 M, 7.5 mL) under mild reflux overnight. The chloroform layer was sepa rated and washed with water (2 # 5 mL), dried over Na 2SO4 and concentrated in vacuo . The resulting oil was diluted w ith hexane (10 mL) and crystalized to afford yellow crystal s as the final product (31 mg, 47% yield). 162 1H-NMR (500 MHz, CDCl 3): 5 (ppm) = 9.37 (s, 1 H), 7.58 (d, J = 8.0 Hz, 1 H), 7.37 (t, J = 7.3 Hz, 1 H), 7.22 Ð 7.14 (m, 2H), 5.83 (s, 1H), 3.58 (s, 3 H). 13C-NMR (125 MHz, CDCl3): 5 (ppm) = 1 82.30, 160.46, 139.64, 126.79, 123.30, 122.59, 110.19, 90.72, 32.48. ESI-MS calcd [M+H] +: 192.03, found: 192.0 Attempts of direct o lefination of the ThCC-1V aldehyde group were not successful. Probably due to its electron -rich nature, the aldehyde does not react with Wittig reagent, Grignard reagent, HWE phosphonates, or even the ethoxyvinyl lithium (Scheme II-7). OSNThCC-1V MgBr N. R.Ph3PCH 3BrN. R.N. R.(EtO) 2OPCNN. R.(EtO) 2OPCO2EtN. R.OEt LiScheme 0-7 Unsuccessful attempts for the nucleophilic elongation of ThCC -1V. NSI57NO2NOCeCl3DCM-DMF r.t., 10 d SNNO2NO2NO2NCl1) solvent, reflux 2) 4 M NaOH aq. SNOScheme 0-8 Unsuccessful attempts for the electrophilic elongation of 57. 163 In situ formation of the C2 vinylene o f 57 and subsequent olefination via 3-nitro enamine or N-dinitroaryl pyridinium mediated Zincke rearrangement were not successful either (Scheme II-8). The olefination was finally done via the malondialdehyde dianil condensation (Scheme II-9). However , the subsequent hydrolysis was not successful. It only yielded in a mixture of ThCC-2V and its phenylimine with a 17:83 ratio. Further optimization was not pursued. Malondialdehyde dianil hydrochlorid e61 A solution of 1,1,3,3 -tetramethoxypropane (2.0 g, 12 mmol, 1 .0 equiv.) in hydrochloric acid (1 M, 20 mL) was heated to 50ûC. To it was a dded a solution of aniline (2.3 g, 24 mmol, 2 .0 equiv.) in hydrochloric acid (1 M, 18 mL) in one portion. Upon addition, the mixture gradually turned orange. The reaction reached completion after 4 h. The Scheme 0-9 Attempted synthesis of ThCC -2V and its phenylimine. NSI57HNHNPhPhCl1)1 equiv. Ac2O : AcOH (1:2) 110 ¡C, 3 h 2) Et 2O, reflux, 30 min decant, 3 times 3) NaBF 4, 1.5 equiv. MeOH, reflux, 0 ¡C, overnight, 96 %, WS_II_34 SNNOPhBF4KOH (4 M) reflux, overnight SNO+SNNPh(17 %) (83 %) ThCC-2V ThCC-2V_imine 58HNHNPhPhClMeO OMe OMe OMe PhNH 2, 2 equiv. 1 M HCl aq. 50 ¡C, 4 h59 %, WS_I_141 164 precipitate was collected by filtration, washed with ethyl acetate, and dried in vacuo to afford an orange solid (1.8 g) with 59 % yield. 1H-NMR (500 MHz, CD 3OD): 5 (ppm) = 8.66 (d, J = 11.3 Hz , 2 H), 7.49 Ð 7.43 (m, 4H), 7.38 Ð 7.33 (m , 4 H), 7.31 Ð 7.25 (m, 2 H), 6.23 (t, J = 11.7 Hz, 1 H). 13C-NMR (125 MHz, CD 3OD): 5 (ppm) =159.86, 139.82, 131.17, 127.76, 118.75, 99.31. ESI -MS calcd [M-Cl]+: 223.12, found: 223.1. 3-Methyl -2-((1E,3E)-4-(N-phenylacetamido)buta -1,3-dien-1-yl)benzothiazolium tetrafluoroborate (58) 2,3-Dimethylthiazolium iodide ( 57, 0.70 g, 2.4 mmol, 1 equiv.) was dissolved in a solution of AcOH and Ac 2O (2:1, 30 mL) and heated to 110 ¡C. Malondialdehyde dianil hydrochloride ( 0.70 g, 2.7 mmol, 1.1 equiv.) was then added into the solution in a portion wise manner over 30 min. The mixture was kept at 110 ¡C and vigorously stirred for additional 3 h. The resulting mixture was cooled to 60 ¡C and subjected to vacuum distillation (1.5 mbar) to remove the solvent. The slurry residue was mixed with Et 2O (20 mL) and refluxed for 30 min. The ethereal layer was decanted. Another portion of Et 2O (20 mL) was added, refluxed and decanted. The resulting solid was collected by f iltration, dissolved into a solution of NaBF 4 (0.40 g, 3.6 mmol, 1.5 equiv. , 20 mL of methanol), and crystalized at 0 ¡C overnight to afford red crystals as final product (0.98 g, 97% yield). 1H-NMR (500 MHz, DMSO -d6): 5 (ppm) = 8.56 (d, J = 12.8 Hz, 1 H), 8.32 (d, J = 8.0 Hz, 1 H), 8.24 Ð 8.16 (m, 1 H), 8.09 (d, J = 8.0 Hz, 1 H), 7.81 Ð 7.76 (m, 1 H), 7.70 (t, J = 7.6 Hz, 1 H), 7.72 Ð 7.67 (m, 1 H), 7.67 Ð 7.61 (m, 1H), 7.61 -7.56 (m, 1 H), 7.47 Ð 7.39 (m, 2 H), 7.29 (d, J = 15.0 Hz, 1 H), 5.43 (dd, J1 = 13.3 Hz, J2 = 11.4 Hz, 1H), 4.04 (s, 3H), 165 1.99 (s, 3 H). 13C-NMR (125 MHz, Acetone -d6): 5 (ppm) = 152.27, 145.50, 139.32, 131.42, 130.58, 130.25, 129.50, 129.00, 128.34, 124.61, 116.95, 112.54, 111.88, 36.24, 23.40. ESI-MS calcd [M -Cl]+: 335.12, found: 335.0. Synthesis of DiAI (Diethylphosphono) acetonitril e62 Triethyl phosphite (40 g, 0.24 mol, 1.8 equiv.) was heated to 160 ûC. Chloroacetonitrile (10 g, 0.13 mol, 1 .0 equiv.) was added at this temperature over a period of 1 h. The mixture was held for 3 more hours at 160 ûC until gas evolvement was finished. Sch eme 0-10 Synthesis of DiAI s. NNH2Ac2O (xs) HOTs áH2O1 equiv. 100 ¡C, 5 h91 %, WS_II_11 NNONN1) POCl 3, 1.2 equiv. DMF, 6 equiv. r.t. 45 min 2) r.t. 2 h 3) NaOH aq., reflux, 8 h 73 %, WS_II_20 (EtO) 2OPCN(EtO) 2OP1.2 equiv. NaH, 1.8 equiv. THF, 0 ¡C, then r.t. 4 h 87 %, WS_I_139 NN(E/Z = 5:2)CNNNODIBAL, 2 equiv. THF, -78¡C !r.t., 10 h 47 %, WS_I_126 59606162DiAI-1V 611.2 equiv. NaH, 1.8 equiv. THF, 0 ¡C, then r.t. 4 h 86 %, WS_II_55 NNNNDIBAL, 2 equiv. THF, -78¡C !r.t., dark, 10 h 10 %, WS_II_86 63DiAI-2V CNO(E/Z = 5:2)CN(EtO) 2OPCNClCNP(OEt) 3, 1.8 equiv. 160 ¡C, neat, 4 h89 %, WS_II_15 166 The mixture was distilled over a small Vigreux column (10 cm) under vacuum. T he final fraction with b.p. 115 ~ 120 ûC at 1.5 mbar was collected to afford a colorless oil (20.8 g) as the product (89% Yield ). 1H-NMR (500 MHz, CDCl 3): 5 (ppm) = 4.26 Ð 4.20 ( m, J1 =7.0 Hz, J2 = 1.0 Hz , 4H ), 2.86 ( d, J = 20.6 Hz , 2H ), 1.37 ( t, J = 7.0 Hz , 6H ). 13C-NMR (125 MHz, CDCl 3): 5 (ppm) = 112.76 ( d, J = 11.1 Hz), 64.06 ( d, J = 6.9 Hz), 16.65 ( d, J = 144.1 Hz), 16.42 ( d, J = 5.9 Hz). (2E/Z) -4-Chloro-3-methylbutyl -2-enenitril e63 Chloroacetone (5.0 g, 54 mmol, 1 .0 equiv.) was added to a stirred solution of (diethylphosphono) acetonitrile (11.1 g, 62.7 mmol, 1.16 equiv.) in Et 2O (15 mL in a 100 mL three -necked round bottom flask fitted with a reflux condenser) at room temperature. Solution of NaOH (3.7 M, 17 mL, 1.2 equiv.) was added dropwise over 1 h at room temperature. The reaction mixture came to an ethereal reflux and turned orange. After an additional 4 h of stirring the two -phase mixture was separated. The aqueous phase was extracted with Et 2O (3 # 10 mL) and the combined organic layer was dried over Na 2SO4, filtered and concentrated in vacuo . Vacuum distillation of the residual orange oil afforded a mixture of cis - and trans -4-chloro -3-methylbutyl -2-enenitrile (2.81 g, 32% yield, 50 ~ 60 ¡C at 1.5 mbar, E/Z = 5:2 ) as a colorless oil. ClO(EtO) 2OPCN1.2 equiv. 3.7 M NaOH aq. Et2O, r.t., 5 h 32 %, WS_II_16 ClCN(E/Z = 5:2) 167 E/Z mixture : 1H-NMR (500 MHz, CDCl 3): 5 (ppm) = 5.52 ( s, 1 H), 5.29 ( s, 1 H), 4.27 (s, 2 H), 4.07 ( s, 2 H), 2.15 ( s, 3 H), 2.06 ( s, 3 H). 13C-NMR (125 MHz, CDCl 3): 5 (ppm) =158.30, 158.01, 116.15, 115.36, 99.24 , 99.00, 47.04, 44.41, 21.18, 19.27. (2E/Z )-Diethyl (3 -cyano-2-methylallyl) phosphonate63 A neat mixture of triethyl phosphite (5.23 g, 31.5 mmol, 1.3 0 equiv.) and 4 -chloro -3-methylbutyl -2-enenitrile (2.8 0 g, 24.2 mmol, 1 .00 equiv.) in a 25 mL flask equipped with a stir bar was placed in a preheated bath at 180 ¡C. After stirring overnight, the mixture was allowed to cool to room temperature. Vacuum frac tional distillation was employed and the mixture of geometric isomers of desired phosphonate was collected at 125 ~ 130 ¡C and 1.5 mbar to afforded 3.0 g (yield = 58%, E/Z = 1.1 :1) as a colorless oil. E/Z mixture : 1H-NMR (500 MHz, CDCl 3): 5 (ppm) = 5.25 Ð 5.28 (m, 2 H), 4.09 Ð 4.19 (m, 8 H), 2.96 (d, J = 23.7 Hz, 2 H), 2.71 (d, J = 23.8 Hz, 2 H), 2.19 (dd, J1 = 3.3 Hz, J2 = 0.9 Hz, 3 H), 2.09 (dd, J1 = 3.7 Hz, J2 =1.6 Hz, 3 H), 1.34 (t, J = 7.1 Hz, 6 H), 1.32 (t, J = 7.1 Hz, 6 H). 13C-NMR (125 MHz, CDCl 3): 5 (ppm) = 155.98 (d, J =11.2 Hz), 155.61 (d, J = 11.2 Hz), 116.50 (d, J = 5.4 Hz), 99.49 (d, J = 12.4 Hz), 99.06 (d, J = 12.4 Hz), 62.68 (d, J = 7.1 Hz), 62.62 (d, J = 7.1 Hz), 36.62 (d, J = 135.3 Hz), 34.87 (d, J = 135.3 Hz), 24.57 (d, J = 1.9 Hz), 22.62 ( d, J = 3.1 Hz), 16.53 (d, J = 5.9 Hz), 16.51 (d, J = 5.7 Hz). ESI-MS calcd [M+H] +: 218.09, found: 218.1. ClCN(E/Z = 5:2)P(OEt) 3, 1.3 equiv. 180 ¡C, overnight 58 %, WS_II_19 (EtO) 2OPCN(E/Z = 1.1:1) 168 3-Methyl [2,3a]diazaindene (60) A solution of 2 -picolylamine ( 59, 2.0 g, 18 mmol, 1 .0 equiv.) and p-toluenesulfonic acid monohydrate (3.5 g, 18 mmol, 1 equiv.) in acetic anhydride (12 mL) was heated at 100 ¡C for 5 h. After cooling down to room temperature, the reaction mixture was poured into water (50 mL) and adjusted pH to 10 with 25% NaOH aqueous solution. Three portions of dichloromethane were used to ex tract the aqueous solution (3 # 20 mL). The combined organic layer was washed with brine and dried over Na 2SO4. The removal of solvent afforded a brown oil (2.2 g, 91% yield). 1H-NMR (500 MHz, CDCl 3): 5 (ppm) = 7.66 (d, J1 = 7 Hz, J2 = 1 Hz, 1 H), 7.40 (dt, J1 = 9 Hz, J2 = 1.1 Hz, J3 = 1 Hz, 1H), 7.35 (s, 1 H), 6.68 Ð 6.64 (m, J1 = 9 Hz, J2 = 6.8 Hz, 1H), 6.58 Ð 6.54 (m, 1H), 2.66 (s, 3 H). 13C-NMR (125 MHz, CDCl 3): 5 (ppm) =134.84, 130.27, 120.58, 118.57, 117.81, 112.45, 12.48. ESI -MS calcd [M+H] +: 133.07, found: 133.1. 1-Formyl-3-methyl [2,3a]diazaindene (61) Phosphorous oxychloride (2.78 g, 18.2 mmol, 1.2 0 equiv.) was dissolved in DMF (dry , 6.64 g, 90.8 mmol, 6 .00 equiv.) at 0 ¡C then warmed to room temperature. After 30 min, the formed yellow viscous reagent was ad ded dropwise to a solution of 60 (2.0 g, 15 mmol, 1 .0 equiv.) at 0 ¡C. A mud -like heavy suspension was immediately formed. The mixture was stirred vigor ously at room temperature for 2 h, monitored by TLC (Hexane : EtOAc : Et 3N = 15 : 80 : 5). After complete conversion, NaOH aq. (5 M, 25 mL) and dichloromethane (30 mL) were added to the slurry. The mixture was allowed mild reflux ed overnight. The organic l ayer was separated and combined with additional DCM extractio n 169 (3 # 10 mL), dried over Na 2SO4, and concentrated on rotavap. The crude was purified by flash silica gel chromatography to afford yellow solid (1.3 g, 73% yield). 1H-NMR (500 MHz, CDCl 3): 5 (pp m) = 10.07 (s, 1 H), 8.27 (d, J = 9.2 Hz, 1 H), 7.89 (dt, J1 = 7.0 Hz, J2 =1.0 Hz, 1 H), 7.25 Ð 7.22 (m, 1 H), 6.96 Ð 6.91 (m, 1H), 2.72 (s, 3 H). 13C-NMR (125 MHz, CDCl 3): 5 (ppm) =185.27, 137.22, 129.43, 125.71, 121.78, 119.72, 118.85, 115.20, 12.65. ESI -MS calcd [M+H] +: 161.06, found: 161.0. (2E)-(3-methyl [2,3a]diazaindene -1-yl) acrylonitrile (62) Sodium hydride (60 % in mineral oil, 101 mg, 2.53 mmol, 1.8 0 equiv.) was added to tetrahydrofuran ( dry , 1.5 mL). The mixture was cooled to 0 ¡C. A solution of (diethylphosphono)acetonitrile (299 mg, 1.69 mmol, 1.2 0 equiv.) in tetrahydrofuran ( dry , 1.0 mL) was then added dropwise to the NaH mixture under nitrogen. After hydrogen ceased evolv ing, a solution of 61 (225 mg, 1.41 mmol, 1 .00 equiv.) in tet rahydrofuran ( dry , 8 mL) was added in dropwise over 2 0 min. The reaction mixture was warm ed to room temperature and kept for 4 h. The excess amount of phosphonate was quenched with distilled benzaldehyde (30 mg, 0.2 0 equiv.). The crude was concentrated in vacuo to obtain a mixture ( E/Z = 6.7:1 ). It was further purified by flash chromatography (45% hexane, 50% CH 2Cl2, 5% Et 3N), to afford yellow solid of E-isomer (223 mg, 87% yield). 1H-NMR (500 MHz, CDCl 3): 5 (ppm) = 7.73 (dd, J1 = 7 Hz, J2 = 1 Hz, 1 H), 7.52 (dd, J1 = 9 Hz, J2 = 1 Hz, 1 H), 7.51 (d, J = 15.8 Hz, 1H), 6.99 Ð 6.93 (m, 1 H), 6.76 Ð 6.69 (m, 1H), 6.04 (d, J = 15.8 Hz, 1H), 2.67 (s, 3 H). 13C-NMR (125 MHz, CDCl 3): 5 (ppm) =139.09, 170 137.94, 131.53, 125.72, 122.37, 122.01, 120.34, 117.05, 113.8 1, 89.45, 12.78. ESI -MS calcd [M+H] +: 184.08, found: 184.1. (E)-3-(3-methyl [2,3a]diazaindene -1-yl) acrylaldehyde (DiAI -1V) To a solution of 62 (50 mg, 0.27 mmol, 1 .0 equiv.) in THF ( dry , 0.8 mL) was a dded DIBAL (1 M in hexane, 0.55 mL, 2 .0 equiv.) at -78 ¡C under argon . The reaction mixture was then warm ed to room temperature gradually. The resulting orange solution was stirred at r t for a further 10 h. The reaction was quenched with cold acetone (5 mL) at 0 ¡C. The aluminum emulsion was diluted with water (10 mL ) and treated with saturated solution of Roc helleÕs salt. Extraction with DCM (4 # 10 mL) was employed after 30 min. The combined organic layer was dried and concentrated in v acuo . The crude was purified by flash chromatography (50% hexane, 45% DCM, 5% Et 3N) to afford yellow solid (24 mg, 47% yield). 1H-NMR (500 MHz, CDCl 3): 5 (ppm) = 9.64 (d, J = 8 Hz, 1 H), 7.78 (d, J = 7 Hz, 1H), 7.66 (d, J = 9 Hz, 1 H), 7.65 (d, J = 15.4 Hz, 1H), 7.03 (dd, J1 = 9 Hz, J2 = 9 Hz, 1 H), 6.87 (dd, J = 15.4 Hz, 8 Hz, 1 H), 6.78 (t, J = 7 Hz, 1H), 2.69 (s, 3 H). 13C-NMR (125 MHz, CDCl3): 5 (ppm) =193.42, 142.46, 138.28, 132.83, 126.21, 123.81, 122.89, 122.27, 117.58, 114.07, 12.81. ESI -MS calcd [M+H] +: 187.08, found: 187.1. (2E/Z )-(4E)-3-methyl -5-(3-methyl [2,3a]diazaindene -1-yl) penta -2,4-dienenitrile (63) The procedure was similar to that for the preparation of 62. Instead of (diethylphosphono) acetonitrile, (2 E/Z )-diethyl (3-cyano -2-methylallyl)phosphonate (1.6 g, 171 7.5 mmol, 1.2 equiv.) was used to react with 61 (1.0 g, 6.2 mmol, 1 .0 equiv.). Purification by flash chromatography (36% EtOAc, 61% hexane, 3% Et 3N) and subsequent crystallization in EtOAc afforded yellow crystals (1.2 g, 86% yield) with an isomer ratio of E/Z = 5:2. Further isolation of the two isomers was not fruitful. The whole procedure was carried out in dark under red light. E/Z mixture: 1H-NMR (500 MHz, CDCl 3): / (ppm) = 7.73 Ð 7.66 (m, 1H), 7.66 Ð 7.62 (m, 1HÕ), 7.20 Ð 7.10 (m, 2 H), 7.11 Ð 7.03 (m, 2 HÕ), .6.91 Ð 6.81 (m, 1H, 1 HÕ), 6.70 Ð 6.61 (m, 1H, 1HÕ), 5.27 (s, 1H), 5.07 (s, 1HÕ), 2.67 (s, 3H, 3HÕ), 2.29 (s, 3H), 2.14 (s, 3 HÕ). 13C-NMR (125 MHz, CDCl 3): / (ppm) = 157.08, 156.66, 136.94, 130.27, 130.05, 1 27.65, 127.26, 127.19, 125.15, 124.13, 121.66, 121.59, 120.67, 120.38, 118.77, 118.07, 117.46, 113.33, 95.21, 93.54, 19.46, 16.54, 12.63, 12.60. ESI -MS calcd [M+H] +: 224.11, found: 224.0. (2E/Z )-(4E)-3-methyl -5-(3-methyl [2,3a]diazaindene -1-yl) penta -2,4-dienal (DiAI -2V) The DIBAL reduction procedure was similar to that for the preparation of DiAI-1V. 3 equivalents of DIBAL were used instead of 2 equiv. The whole procedure was performed in the dark room under red light. Th e crude was purified by dry column chromatography and Prep TLC to afford a deep red solid in as an E/Z mixture (12 mg, 10% yield, E/Z = 1:0.4, based on 1D -NOE). Further isolation was not fruitful even with Prep -HPLC. The two isomers quickly isomerize on si lica and alumina. E/Z mixture: 1H-NMR (500 MHz, CDCl 3): 5 (ppm) = 10.40 (d, J = 8.1 Hz, 1 H), 10.11 (d, J = 8.1 Hz, 1 HÕ), 7.69 (d, J = 7.2 Hz, 1H, 1 HÕ), 7.58 (d, J = 9.2 Hz, 1H, 1 HÕ), 7.29 Ð 172 7.24 (m, 1H, 1 HÕ), .7.19 Ð 7.14 (m, 1 H, 1 HÕ), 6.89 Ð 6.83 (m, 1H, 1HÕ), 6.69 Ð 6.64 (m, 1H, 1HÕ), 6.11 (d, J = 8.2 Hz, 1 H), 5.84 (d, J = 8.2 Hz, 1HÕ), 2.68 (s, 3H, 3HÕ), 2.40 (s, 3H), 2.22 (s, 3 HÕ). 13C-NMR (125 MHz, CDCl 3): 5 (ppm) = 191.13, 190.51, 155.32, 137.04, 130.33, 128.26, 127.55, 127.32, 127.10, 125.72, 121.63, 120.65, 117.62, 117.59, 113.52, 29.69, 13.04, 12.58. 1D -NOE (500 MHz, CDCl 3): / (ppm) = 2.40 (10.10). ESI -MS calcd [M+H] +: 227.11, found: 227.0. 173 Synthesis of FR 2-bromo-9H-fluorene (65 )23 Fluorene (4.98 g, 30 .0 mmol) was dissolved in distilled propylene carbonate (60 mL) by heating up to 60¡C shortly. NBS (5.34 g, 30 .0 mmol) was then added to the Scheme 0-11 Synthesis of FRs. OOONBS (1.0 equiv.) (solv.) r.t., 30 min WS_IV_19, 96% BrCH3I (5.0 equiv.) NaOH (10.0 equiv.) DMSO, r.t., 4 h WS_IV_20, 76% Brfuming HNO 3HOAc 0 ¡C!r.t., 2 h WS_IV_37, 73% BrO2NFe (3.0 equiv.) NH4Cl (2.0 equiv.) EtOH, H 2O, N2,85 ¡C, 2 h,WS_IV_33, 99% BrH2NC2H5I (2.5 equiv.) K2CO3 (3.0 equiv.) DMF, 80 ¡C, 5 h,WS_IV_51, 70% BrN1) t-BuLi (2.0 equiv.) THF, -78 ¡C, 1h2) DMF (2.0 equiv.) THF, -78 ¡C, 2 h, then r.t., 1 h, WS_IV_55, 78% NOFR0(EtO) 2OPCN(1.2 equiv.) NaH (1.8 equiv.) THF, 0 ¡C!r.t., 4 h quant., WS_IV_69 NCNDIBAL (3 equiv.) THF, -78 ¡C!r.t., 12 h 26%, WS_IV_76 NOFR-1V (3.0 equiv.) NaH (5.0 equiv.) THF, -30 ¡C!r.t., 6 h 96%, WS_IV_80 NCNFR0DIBAL (3 equiv.) THF, -78 ¡C!r.t., 18 h 19%, WS_IV_85 NFR-2V O(EtO) 2OPCN6465666768697071NBOOPdCl 2 (5 mol%) Dppf (10 mol%) KOAc (5 equiv.) B2Pin 2 (2 equiv.) 1,4-Dioxane, 95 ¡C, 16 h62%, WS_IV_152 NPd(PPh 3)4 (5 mol%) K2CO3 (3 equiv.) DME / H 2O, reflux, 16 h 76%, WS_IV_153 SBrOSOFR-Th (1.5 equiv.) 6972 174 solution in portion over 2 h. The mixture was kept stirred at room temperature. The solution gradually turned yellow and formed precipitates. The reaction was monitored by TLC (hexane). The mixture was then poured into cold water (500 mL) and the precipitate was filtered to afford 7.07 g of an off-white solid (96 % yield ). The crude was used in the next reaction without c rystallization. 1H-NMR (500 MHz, CDCl 3), 5 (ppm) = 7.76 (d, J = 6.6 Hz, 1H), 7.69 (s, 1H), 7.64 (d, J = 8.1 Hz, 1H), 7.54 (dt, J1 = 7.3 Hz, J2 = 1.0 Hz, 1H), 7.51 Ð 7.49 (m, 1H), 7.41 Ð 7.36 (m, 2H), 7.33 (td, J1 = 7.4 Hz, J2 = 1.2 Hz, 2H), 3.89 (s, 2H). 2-bromo-9,9-dimethyl -9H-fluorene (6 6)23 To a mixture of NaOH (4.0 0 g, 100 mmol) and DMSO (50 mL) was added 65 (2.45 g, 100 mmol). The mixture was stirred for 1 h at room temperature. A solution of iodomethane (7.1 0 g, 50.0 mmol) in DMSO (10 mL) was then added to the above mixture. The reaction mixture was stirred for another 4 h at room temperature and monitored by TLC (hexane). The mixture was poured into ice water (100 mL) and extracted with DCM (20 mL # 3). The crude product was purified by fl ash chromatography (hexane) to afford 2.10 g of a white solid ( 76% yield). 1H-NMR (500 MHz, CDCl 3), 5 (ppm) = 7.71 Ð 7.69 (m, 1H), 7.59 (d, J = 8 Hz, 1H) 7.57 Ð 7.54 (m, 1H), 7. 48 Ð 7.46 (m, 1H), 7.44 Ð 7.42 (m, 1H), 7.36 Ð 7.34 (m, 2H), 1.49 (s, 6H). 2-bromo-9,9-dimethyl -7-nitro -9H-fluorene (6 7)23 175 To a solution of 66 (1.90 g, 6.96 mmol) in glacial acetic acid (40 mL) was added fuming nitric acid (6 mL) dropwise at 0 ¡C with vigorous stirring. The r eaction mixture was stir red at room temperature after addition. A y ellowish green precipitate form ed in 30 min. The reaction mixture was poured into 200 mL water at the appearance of dinitro product (5% EtOAc -Hexane, Rf ~ 0.25). The precipitate was filtered, washed thoroughly with water, and recrystallized from acetonitrile (60 mL) to afford light yellow crystals (1.61 g, yield = 73%). 1H-NMR (500 MHz, CDCl 3), 5 (ppm) = 8.29 Ð 8.26 (m, 2H), 7.80 (d, J = 8 Hz, 1H) 7.68 Ð 7.66 (m, 1H), 7.64 (s, 1H), 7.56 Ð 7.54 (m, 1H), 1.55 (s, 6H). 7-bromo-9,9-dimethyl -9H-fluoren -2-amine (68)23 A mixture of 67 (2.5 0 g, 7.86 mmol), iron powder (1.32 g, 23.6 mmol, 3.0 equiv.), and ammonium chloride (840 mg, 15.7 mmol, 2.0 equiv.) in a co -solvent (90 mL of EtOH, 25 mL of H 2O) was heated to reflux (85 ¡C) and kept under mechanical stirring for 2 h. The reaction mixture was then neutralized with saturated NaHCO 3 solution to pH 8 and filtered. The filter cake was was hed with ethanol to dissolve the product. The filtrate was combined and concentrated in vacuo to remove the ethanol. The product was extracted with DCM from the aqueous phase. The organic phase was dried with NaSO 4 and concentrated to affor d a yellow cryst alline solid (2.24 g, yield = 99%). 1H-NMR (500 MHz, CDCl 3), 5 (ppm) = 7.48 Ð 7.47 (m, 2H), 7.44 -7.39 (m, 2H), 6.74 (d, J = 2.1 Hz, 1H), 6.67 (dd, J1 = 8.0 Hz, J2 = 2.2 Hz, 1H), 3.81 (s, 2H), 1.44 (s, 6H). 176 7-bromo-N,N-diethyl -9,9-dimethyl -9H-fluoren -2-amine (69)23 A mixture of 68 (2.0 0 g, 6.94 mmol), ethyl iodide (2.71 g, 17.3 mmol, 2.5 equiv.), potassium carbonate (2.88 g, 20.8 mmol, 3.0 equiv.) in DMF (20 mL) was heated at 80 ¡C for 5 h. The solvent was removed by distillation under reduced pressure. The resulting crude was poured int o water and extract ed with ethyl acetate. The organic phase was combined and dried, and concentrated in vacuo. The crude was purified by flash chromatography with 5% ethyl acetate in hexan e to afford orange crystals (1.67 g, yield = 70%). ESI-MS calcd [M+H ]+: 344.09, found: 343.92. 1H-NMR (500 MHz, CDCl 3), 5 (ppm): 7.51 (d, J = 8 Hz, 1H), 7.44 (d, J = 2 Hz, 1H), 7.41 Ð 7.35 (m, 2H), 6. 86 Ð 6.64 (m, 2H), 3.42 (q, J = 7.0 Hz, 4H), 1.44 (s, 6H), 1.21 (t, J = 7.0 Hz). 7-diethylamino -9,9-dimethyl -9H-fluorene -2-carbaldehyde (FR0)23 To a solution of 69 (500 mg, 1.45 mmol) in dry THF , t-BuLi (1.71 mL of 1.7 M solution in pentane, 2.0 equiv. ) was add ed dropwise at -78 ¡C under argon. The r eaction mixture was stirred for 1 h, and a n orange suspension was f ormed. DMF (212 mg, 2.91 mmol, 2.0 equiv. ) was added dropwise and the formed solution was stirred for additional 2 h at -78 ¡C. The reaction mixture was warmed to room temperature and stirred for 1 h. The reaction was quenched with 2 M HCl aq. and the pH was adjusted to 8. The solution was extracted with EtOAc . The combined o rganic phase was dried with Na 2SO4 and evaporated in vacu o. The crude was purified by flash chromatography ( 5% EtOAc -Hexane ) to yield FR0 (330 mg, 78 %) as a yellow soli d. 177 1H-NMR (500 MHz, CDCl 3), 5 (ppm) = 9.98 (d, J = 0.8 Hz, 1H), 7.90 Ð 7.85 (m, 1H), 7.78 (dt, J1 = 7.8 Hz, J2 = 1.1 Hz, 1H), 7.68 Ð 7.59 (m, 2H), 6.73 Ð 6.66 (m, 2H), 3.46 (q, J = 7.1 Hz, 4H), 1.50 (s, 6 H), 1.30 Ð 1.20 (m, 7H). 13C-NMR (125 MHz, CDCl3): 5 (ppm) = 192.05, 157.33, 153.19, 148.99, 146.90, 133.41, 131.27, 125.20, 122.47, 122.3 5, 118.12, 110.87, 105.06, 46.62, 44.71, 27.24, 12.59. (E)-3-(7-(diethylamino) -9,9-dimethyl -9H-fluoren -2-yl)acrylonitrile (70) To a stirred suspension of NaH ( 13 mg, 0.31 mmol) in tetrahydrofuran (1.0 mL) was added a solution of (diethyl phosphono)acetonitrile (36 mg, 0.21 mmol) in tetrahydrofuran (1 mL). The mixture was kept at 0 ¡C and stirred for 30 min. Then a solution of FR0 (50 mg, 0.17 mmol) in dry tetrahydrofuran (3 mL) was added. The reaction mixture was stirred at ambient temperature for 4 h. The reaction mixture was then poured into cold water and extracted with ethyl acetate. The organic phase was dried and concentrated. Purification by flash c hromatography (6% ethyl acetate in hexane) afforded 56 mg of a yellow solid (yield: quantitative ). 1H-NMR (500 MHz, CDCl 3), 5 (ppm) = 7.55 (dd, J1 = 15.4 Hz, J2 = 8.1 Hz, 2H), 7.48 Ð 7.40 (m, 2H), 7.35 (dd, J1 = 7.9 Hz, J2 = 1.4 Hz, 1H), 6.73 Ð 6.64 (m, 2 H), 5.88 Ð 5.79 (m, 1H), 3.45 (q, J = 7.1 Hz, 4H), 1.47 (s, 6H), 1.23 (t, J = 7.0 Hz, 6H). 13C-NMR (125 MHz, CDCl 3): 5 (ppm) = 156.45, 153.37, 151.25, 148.59, 143.76, 130.07, 127.46, 125.61, 121.81, 120.76, 119.10, 118.53, 110.81, 105.21, 93 .00, 46.64, 44.71, 27.37, 12.61. ESI-HRMS: (calc.) (m/z) calcd for C 22H24N2 [M+H] + 317.2018, found 317.2019. 178 (E)-3-(7-(diethylamino) -9,9-dimethyl -9H-fluoren -2-yl)acrylaldehyde (FR -1V) To a solution of compound 70 (50 mg, 0.16 mmol) in tetrahydrofuran ( dry , 3 m L) was added DIBAL (1 M in THF, 0.5 0 mL) at -78 ¡C under argon. The reaction mixture was then allowed to warm to ambient temperature gradually and was stirred for 12 h. Cold methanol was added drop wise to quench reaction. The mixture was treated with satu rate solution of RochelleÕs salt and extracted with dichloromethane. The combined organic phase was dried and concentrated. Purification by flash chromatography (5% ethyl acetate in hexane) afforded 13 mg of an orange solid (26% y ield). 1H-NMR (CDCl 3, 500 MHz), 5 (ppm) = 9.70 (d, J = 7.8 Hz, 1H), 7.61 Ð 7.47 (m, 5H), 6.78 Ð 6.66 (m, 3H), 3.46 (q, J = 7.1 Hz, 4H), 1.49 (s, 6H), 1.24 (t, J = 7.1 Hz, 6H). 13C NMR (CDCl 3, 125 MHz), 5 (ppm) = 193.83, 156.61, 154.07, 153.38, 148.64, 144.00, 130.54, 128.81 , 126.43, 125.72, 122.01, 121.89, 118.59, 110.82, 105.22, 46.63, 44.72, 27.39, 12.63. ESI -HRMS (m/z) calcd for C 22H25NO [M+H] + 320.2014, found 320.2015. (2E,4E)-5-(7-(diethylamino) -9,9-dimethyl -9H-fluoren -2-yl)-3-methylpenta -2,4-dienenitrile (71) To a sti rred suspension of NaH (102 mg, 2.56 mmol) in dry tetrahydrofuran (2 .0 mL) was added a solution of (2E/Z )-diethyl (3 -cyano -2-methylallyl)phosphonate (333 mg, 1.53 mmol) in dry tetrahydrofuran (1 .0 mL). The mixture was kept at -30 ¡ C and stirred for 1 h . Then a solution of FR0 (0.15 g, 0.51 mmol) in dry tetrahydrofuran (5.0 mL) was added. The reaction mixture was stir red at ambient temperature for 6 h. The reaction mixture was then poured into cold water and extracted with ethyl acetate. The organic 179 phase was dried and concentrated. Purific ation by flash chromatography (5 % ethyl acetate in hexane) afforded 174 mg of an orange solid (96% y ield). 1H-NMR (CDCl 3, 500 MHz), 5 (ppm) = 7.53 (dd, J = 12.9, 8.2 Hz, 2H), 7.46 (d, J = 1.6 Hz, 1H), 7.37 (dd, J = 8.0, 1.6 Hz, 1H), 6.97 (d, J = 15.8 Hz, 1H), 6.88 Ð 6.81 (m, 1H), 6.72 Ð 6.63 (m, 2H), 5.33 Ð 5.29 (m, 1H), 3.44 (q, J = 7.0 Hz, 4 H), 2.29 (d, J = 0.8 Hz, 3H), 1.47 (s, 6 H), 1.22 (t, J = 7.1 Hz, 6H). 13C NMR (CDCl 3, 125 MHz), 5 (ppm) = 157.17, 156.15, 153.33, 148.23, 141.79 , 136.90, 132.14, 127.17, 126.21, 126.13, 121.41, 120.75, 118.56, 118.28, 110.78 , 105.43, 96.55, 46.61, 44.71, 27.47 , 16.72, 12.63. ESI-HRMS (m/z) calcd for C 25H29N2 [M+H] + 357.2331, found 357.2349. (2E,4E)-5-(7-(diethylamino) -9,9-dimethyl -9H-fluoren -2-yl)-3-methylpenta -2,4-dienal (FR -2V) To a solution of compound 71 (0.15 g, 0.42 mmol) in dry THF (8.0 mL) was added DIBAL (1 M in THF, 1.3 mL) at -78 ¡C under argon. The reaction mixture was then a llowed to warm to ambient temperature g radually and was stirred for 18 h. Cold methanol was added drop wise to quench the reaction. The mixture was treated with saturate solution of RochelleÕs salt and extracted with dichloromethane. The combined organic phase was dried and concentrated. Purification by gravity silica chromatography and Prep -TLC (2 % ethyl acetate in hexane) a fforded 28 mg of a red solid (19% y ield). 1H-NMR (CDCl 3, 500 MHz), 5 (ppm) = 10.17 (d, J = 8.2 Hz, 1H), 7.60 Ð 7.50 (m, 3H), 7.42 (dd, J1 = 7.9 Hz, J2 = 1.6 Hz, 1H), 7.16 (d, J = 16.0 Hz, 1H), 6.93 (dd, J1 = 16.0 Hz, J2 = 0.8 Hz, 1H), 6.74 Ð 6.65 (m, 1H) , 6.15 Ð 6.08 (m, 1H) , 3.45 (q, J = 7.1 Hz, 4 H), 180 2.42 (d, J = 1.1 Hz, 3H), 1.50 (s, 6H), 1.23 ( t, J = 7.1 Hz, 6H) . 13C NMR (CDCl 3, 125 MHz), 5 (ppm) = 191.20, 156.22, 155.00, 153.36, 148.24, 141.84, 136.82, 132.51, 129.21, 129.12, 127.36, 126.25, 121.43, 120.82 , 118.58, 110.78, 105.44, 46.62, 44.71, 27.48, 13.13, 12.64. ESI-HRMS (m/z) calcd for C 25H30NO [M+H] + 360.2327, found 360.2336. N,N-diethyl -9,9-dimethyl -7-(4,4,5,5 -tetramethyl -1,3,2-dioxaborolan-2-yl)-9H-fluoren -2-amine (72) To a solution of 69 (0.20 g, 0 .58 mmol) in dry 1,4 -dioxane (30 mL) was added PdCl 2 (6.0 mg, 0.029 mmol, 5 mol%), Dppf (33 mg, 0.058 mmol, 10 mol%), B 2Pin 2 (295 mg, 1.16 mmol, 2.0 equiv.) and potassium acetate (285 mg, 2.90 mmol, 5.0 equiv.). The resulting mixture was flushed with argon, then heated to 95 ¡C and kept stirring for 16 h. The reaction was quenched with water after full conversion of the startin g material monitored with TLC (10% EtOAc in Hexane). The mixture was extracted with EtOAc, and run through a silica pad to remove the palladium and Dppf. The combined organic phase was dried and concentrated in vacuo . The crude was purified by gradient chr omatography with 10 0% Hexane to 5% EtOAc -Hexane. Crystallization in EtOAc afforded 140 mg of a pale yellow crystal ( 62% yield). 1H-NMR (CDCl 3, 500 MHz), 5 (ppm) = 7.80 (s, 1H), 7.76 (d, J = 7.5 Hz, 1H), 7.57 (t, J = 8.0 Hz, 2H), 6.71 (d, J = 2.3 Hz, 1H), 6.67 (dd, J1 = 8.4 Hz, J2 = 2.4 Hz, 1H), 3.44 (q, J = 7.1 Hz, 4H), 1.48 (s, 6H), 1.37 (s, 12H), 1.22 (t, J = 7.1 Hz, 6H). 13C NMR (CDCl 3, 125 MHz), 5 (ppm) = 156.28, 151.82, 148.17, 143.18, 133.96, 128.28, 126.86, 121.45, 181 117.71, 110.61, 105.50, 83.51, 46.65, 44.69, 27.43, 24.91, 23.42, 12.64. ESI-HRMS (m/z) calcd for C 25H35BNO2 [M+H] + 392.2761, found 392.2776. 5-(7-(diethylamino) -9,9-dimethyl -9H-fluoren -2-yl)t hiophene-2-carbaldehyde (FR -Th) To a solution of 72 (50 mg, 0.13 mmol) and 5-bromo -2-thiophene-carb oxaldehyde (37 mg, 0.19 mmol, 1.5 equiv.) in DME (3.0 mL) was added Pd(PPh 3)4 (8.0 mg, 6.4 !mol, 5 mol% ) and potassium carbonate (53 mg, 0.38 mmol, 3.0 equiv.). Water (0.4 mL) was added as a co -solvent. The mixture was heated to reflux for 15 h. The resulting crude was run through a silica pad to remove palladium and extracted with EtOAc. The organic phase was dried and concentrated. The crude was purified by flash chromatography with 5% EtOAc in Hexane and crystalized in EtOAc to afford 29 mg of a bright -red crystal ( 76% yield). 1H-NMR (CDCl 3, 500 MHz), 5 (ppm) = 9.88 (s, 1H), 7.75 (d, J = 4.0 Hz, 1H), 7.6 6 Ð 7.55 (m, 4H), 7.42 (d, J = 4.0 Hz, 1H), 6.74 Ð 6.66 (m, 2H), 3.45 (q, J = 7.1 Hz, 4H), 1.51 (s, 6H), 1.24 (q, J = 7.1, 6.1 Hz, 6H). 13C NMR (CDCl 3, 125 MHz), 5 (ppm) = 182.65, 156.10, 155.91, 153.58, 148.32, 141.88, 141.28, 137.68, 129.44, 125.95, 125.62, 123.09, 121.49, 120.2 7, 118.74, 110.79, 105.39 , 46.79, 44.71, 27.46, 12.63. ESI-HRMS (m/z) calcd for C 24H26NOS [M+H] + 376.1735, found 376.1729. 182 REFERENCE S 183 REFERENCES 1. Shaner, N.C., Steinbach, P.A. & Tsien, R.Y. A guide to choosing fluorescent proteins. Nat. Methods 2005, 2, 905-909. 2. McKinney, S.A., Murphy, C.S., Hazelwood, K.L., Davidson, M.W. & Looger, L.L. A bright and photostable photoconvertible fluorescent protein. Nat. Methods 2009, 6, 131-133. 3. Durisic, N., Cuervo, L.L. & Lakadamyali, M. Quantitative super -resolution microsco py: pitfalls and strategies for image analysis. Curr. Opin. Chem. Biol. 2014, 20, 22-28. 4. Costantini, L.M. & Snapp, E.L. Fluorescent Proteins in Cellular Organelles: Serious Pitfalls and Some Solutions. DNA Cell Biol . 2013, 32, 622-627. 5. Zhang, M. et a l. Rational design of true monomeric and bright photoactivatable fluorescent proteins. Nat. Methods 2012, 9, 727-729. 6. Landgraf, D., Okumus, B., Chien, P., Baker, T.A. & Paulsson, J. Segregation of molecules at cell division reveals native protein locali zation. Nat. Methods 2012, 9, 480-482. 7. Kumagai, A. et al. A Bilirubin -Inducible Fluorescent Protein from Eel Muscle. Cell 2013, 153, 1602-1611. 8. Wang, W. et al. Tuning the Electronic Absorption of Protein -Embedded All -trans -Retinal. Science 2012, 338, 1340-1343. 9. Gunasekaran, K., Hagler, A.T. & Gierasch, L.M. Sequence and structural analysis of cellular retinoic acid -binding proteins reveals a network of conserved hydrophobic interactions. Proteins Struct . Funct . Bioinf . 2004, 54, 179-194. 10. Vasile iou, C. et al. Elucidating the exact role of engineered CRABPII residues for the formation of a retinal protonated Schiff base. Proteins Struct . Funct . Bioinf . 2009, 77, 812-822. 11. Berbasova, T. in PhD Thesis. Michigan State University 2014. 12. Yapici, I. et al. ÒTurn -OnÓ Protein Fluorescence: In Situ Formation of Cyanine Dyes. J. Am. Chem. Soc. 2015, 137, 1073-1080. 184 13. Dempsey, G.T., Vaughan, J.C., Chen, K.H., Bates, M. & Zhuang, X. Evaluation of fluorophores for optimal performance in localiza tion-based super -resolution imaging. Nat. Methods 2011, 8, 1027-1036. 14. GŠmez -Bombarelli, R. et al. Design of efficient molecular organic light -emitting diodes by a high -throughput virtual screening and experimental approach. Nat. Mater. 2016, 15, 1120-1127. 15. D−hne, S. & Radeglia, R. Revision der Lewis -Calvin -Regel zur charakterisierung vinyloger. Polyen - und polymethin−hnlicher verbindungen. Tetrahedron 1971, 27, 3673-3693. 16. Bouit, P. -A. et al. Continuous Symmetry Breaking Induced by Ion Pairing Ef fect in Heptamethine Cyanine Dyes: Beyond the Cyanine Limit. J. Am. Chem. Soc. 2010, 132, 4328-4335. 17. Tolbert, L.M. & Zhao, X. Beyond the Cyanine Limit: + Peierls Distortion and Symmetry Collapse in a Polymethine Dye. J. Am. Chem. Soc. 1997, 119, 3253 -3258. 18. Szydlowska, I., Kyrychenko, A., Nowacki, J. & Herbich, J. Photoinduced intramolecular electron tr ansfer in 4 -dimethylaminopyridines. Phys. Chem. Chem. Phys. 2003, 5, 1032-1038. 19. Maksim, V.S., Vladimir, N.B. & Stefan, W.H. Fluorescent dyes with large Stokes shifts for super -resolution optical microscopy of biological objects: a review. Methods Appl . Fluoresc . 2015, 3, 042004. 20. Haberhauer, G., Gleiter, R. & Burkhart, C. Planarized Intramolecular Charge Transfer: A Concept for Fluorophores with both Large Stokes Shifts and High Fluorescence Quantum Yields. Chem. Eur. J. 2016, 22, 971-978. 21. Sasaki, S., Drummen, G.P.C. & Konishi, G. -i. Recent advances in twisted intramolecular charge transfer (TICT) fluorescence and related phenomena in materials chemistry. J. Mater . Chem. C 2016, 4, 2731-2743. 22. Liu, X. et al. Aziridinyl Fluorophores Demons trate Bright Fluorescence and Superior Photostability by Effectively Inhibiting Twisted Intramolecular Charge Transfer. J. Am. Chem. Soc. 2016, 138, 6960-6963. 23. Kucherak, O.A., Didier, P., M”ly, Y. & Klymchenko, A.S. Fluorene Analogues of Prodan with Su perior Fluorescence Brightness and Solvatochromism. J. Phys . Chem. Lett. 2010, 1, 616-620. 24. Fern⁄ndez -Su⁄rez, M. & Ting, A.Y. Fluorescent probes for super -resolution imaging in living cells. Nat. Rev. Mol. Cell Biol. 2008, 9, 929-943. 185 25. Dugave, C. & D emange, L. Cis %Trans Isomerization of Organic Molecules and Biomolecules: + Implications and Applications. Chem. Rev. 2003, 103, 2475-2532. 26. Kwon, J. et al. RESOLFT nanoscopy with photoswitchable organic fluorophores. Sci. Rep. 2015, 5, 17804. 27. Bossi, M., Belov, V., Polyakova, S. & Hell, S.W. Reversible Red Fluorescent Molecular Switches. Angew. Chem. Int. Ed. 2006, 45, 7462-7465. 28. Cusido, J., Deniz, E. & Raymo, F.M. Fluorescent Switches Based on Photochromic Compounds. Eur. J. Org. Chem. 2009, 2009, 2031-2045. 29. Herder, M. et al. Improving the Fatigue Resistance of Diarylethene Switches. J. Am. Chem. Soc. 2015, 137, 2738-2747. 30. Jeong, Y. -C., Park, D.G., Lee, I.S., Yang, S.I. & Ahn, K. -H. Highly fluorescent photochromic diarylethene with an exce llent fatigue property. J. Mater. Chem. 2009, 19, 97-103. 31. Jeong, Y. -C., Yang, S.I., Ahn, K. -H. & Kim, E. Highly fluorescent photochromic diarylethene in the closed -ring form. Chem. Commun. 2005, 2503-2505. 32. Uno, K. et al. In Situ Preparation of Highly Fluorescent Dyes upon Photoirradiation. J. Am. Chem. Soc. 2011, 133, 13558-13564. 33. Fılling, J. et al. Synthesis and Characterization of Photoswitchable Fluorescent Silica Nanoparticles. Small 2008, 4, 134-142. 34. Nevskyi, O., Sysoiev, D., Oppermann, A., Huhn, T. & Wıll, D. Nanoscopic Visualization of Soft Matt er Using Fluorescent Diarylethene Photoswitches. Angew. Chem. Int. Ed. 2016, 55, 12698-12702. 35. Roubinet, B. et al. Carboxylated Photoswitchable Diarylethenes for Biolabeling and Super -Resolution RESOLFT Microscopy. Angew. Chem. Int. Ed. 2016, 55, 15429-15433. 36. Subach, F.V. et al. Red Fluorescent Protein with Reversibly Photoswitchable Absorbance for Photochromic FRET. Chem. Biol. 2010, 17, 745-755. 37. Tiwari, D.K. et al. A fast - and positively photoswitchable fluorescent protein for ultralow -laser -power RESOLFT nanoscopy. Nat. Methods 2015, 12, 515-518. 38. Zhang, X. et al. Highly photostable, reversibly photoswitchable fluorescent protein with high contrast ratio for live -cell superresolution microscopy. Proc. Natl. Acad. Sci. U. S. A. 2016, 113, 10364-10369. 186 39. Lukyanov, K.A., Chudakov, D.M., Lukyanov, S. & Verkhusha, V.V. Photoactivatable fluorescent proteins. Nat. Rev. Mol. Cell Biol. 2005, 6, 885-891. 40. Zhou, X.X. & Lin, M.Z. Photoswitchable fluorescent proteins: ten years of colorful chemistry and exciting applications. Curr. Opin. Chem. Biol. 2013, 17, 682-690. 41. Acharya, A. et al. Photoinduced Chemistry in Fluorescent Proteins: Curse or Blessing? Chem. Rev. 2017, 117, 758-795. 42. Warren, M.M. et al. Ground -state proton transfer in the phot oswitching reactions of the fluorescent protein Dronpa. Nat. Commun. 2013, 4, 1461. 43. Nosrati, M., Berbasova, T., Vasileiou, C., Borhan, B. & Geiger, J.H. A Photoisomerizing Rhodopsin Mimic Observed at Atomic Resolution. J. Am. Chem. Soc. 2016, 138, 8802-8808. 44. Berbasova, T. et al. Light -Activated Reversible Imine Isomerization: Towards a Photochromic Protein Switch. ChemBioChem 2016, 17, 407-414. 45. Nani, R.R., Kelley, J.A., Ivanic, J. & Schnermann, M.J. Reactive species involved in the regioselective photooxidation of heptamethine cyanines. Chem. Sci. 2015, 6, 6556-6563. 46. Mosmann, T. Rapid colorimetric assay for cellular growth and survival: Appli cation to proliferation and cytotoxicity assays. J. Immunol . Methods 1983, 65, 55-63. 47. Assar, Z. et al. Domain -Swapped Dimers of Intracellular Lipid -Binding Proteins: Evidence for Ordered Folding Intermediates. Structure 2016, 24, 1590-1598. 48. Uttamap inant, C. et al. A fluorophore ligase for site -specific protein labeling inside living cells. Proc. Natl. Acad. Sci. U. S. A. 2010, 107, 10914-10919. 49. Grimm, J.B. et al. A general method to fine -tune fluorophores for live -cell and in vivo imaging. Nat. Methods 2017, 14, 987-994. 50. Shcherbo, D. et al. Near -infrared fluorescent proteins. Nat. Methods 2010, 7, 827 -829. 51. Rafiq, S.M., Sivasakthikumaran, R. & Mohanakrishnan, A.K. Lewis Acid/Brınsted Acid Mediated Benz -Annulation of Thiophenes and Electron -Rich Arenes. Org. Lett. 2014, 16, 2720-2723. 52. Demchenko, A.P. Practical aspects of wavelength ratiometry in the studies of intermolecular interactions. J. Mol . Struct . 2014, 1077, 51-67. 53. Gill, S.C. & von Hippel, P.H. Calculation of protein extincti on coefficients from amino acid sequence data. Anal . Biochem . 1989, 182, 319-326. 187 54. Wrth, C., Grabolle, M., Pauli, J., Spieles, M. & Resch -Genger, U. Relative and absolute determination of fluorescence quantum yields of transparent samples. Nat. Protoc. 2013, 8, 1535-1550. 55. Turro, N.J., Ramamurthy, V., Cherry, W. & Farneth, W. The effect of wavelength on organic photoreactions in solution. Reactions from upper excited states. Chem. Rev. 1978, 78, 125-145. 56. Rurack, K. & Spieles, M. Fluorescence Quantum Yields of a Series of Red and Near-Infrared Dyes Emitting at 600 %1000 nm. Anal. Chem. 2011, 83, 1232-1242. 57. Wohlfarth, C. in Ch48: Refractive index of ethanol. Refractive Indices of Pure Liquids and Binary Liquid Mixtures . Landolt-Bırnstein Ð Group III Condensed Matter. Vol . 47 , (ed. Lechner , M. D. ), 108-115 (Springer -Verlag Berlin Heidelberg ; Springer Materials; 2008). 58. Schoch, R.L., Kapinos, L.E. & Lim, R.Y.H. Nuclear transport receptor binding avidity triggers a self -healing collapse transition in FG -nucleoporin molecu lar brushes. Proc. Natl. Acad. Sci. U. S. A. 2012, 109, 16911-16916. 59. Schindelin, J. et al. Fiji: an open -source platform for biological -image analysis. Nat. Methods 2012, 9, 676-682. 60. Barbero, M., Cadamuro, S. & Dughera, S. The efficient o -benzenedi sulfonimide catalysed synthesis of benzothiazoles, benzoxazoles and benzimidazoles. Arkivoc 2012, 262-279. 61. Gerowska, M., Hall, L., Richardson, J., Shelbourne, M. & Brown, T. Efficient reverse click labeling of azide oligonucleotides with multiple alkyn yl Cy -Dyes applied to the synthesis of HyBeacon probes for genetic analysis. Tetrahedron 2012, 68, 857-864. 62. Joachim, G., Gıtz, N., Jaedicke, H., Mayer, G. & Rack, M., Process for the Preparation of Pyridine Derivatives. WO/2005/063780, 2005, DOW Agrosciences LLC. 63. Sovdat, T. et al. Backbone Modification of Retinal Induces Protein -like Excited State Dynamics in Solution. J. Am. Chem. Soc. 2012, 134, 8318-8320. 188 EXCITED STATE PROTON TRANSFER OF FR PHOT OBASES AND DEVELOPING DYE -HCRBPII COMPLEXES INTO B RIGHT, LONG STOKES S HIFT, FAST -FORMING TAGS VIA EXC ITED -STATE HYDROGEN BONDING. The NIR photoswitch described in 0 solves the fluorescence background issue. Although the molecular brightness of its NIR emission is comparatively high, the absolute fluorescence quantum yield (0.16 of CF-1) is still lower th an those of the synthetic dyes listed in Section I.4.2. The partial reason is that CrimFluors have significant ly larger Stokes shifts than those polymethine or polyene dyes with higher quantum yields. Along the charge transfer coordinate, the larger momentum change is usually correlated to a longer timescale that may allow more non -radiative decay pathways to happen and lead to lower QYs. In particular, the imine -isomerization of CF-1/2 is highly probable to happen in the excited state after local excitation. This excited -state photochemical transformation should result in more molecular motion prior to the radiative decay and appears to be a trade -off for the NIR fluorescence photoactivation. Even the non -switchable mutant complex ( Section I.12.2) may undergo transient excited -state isomerization but not end in a stabilized ground -state isomeric form. Thus, a fundamental question arises : can we minimize molecular motions in the excited state to achieve superior quantum yield, not by structural rigidification bu t preferably via a photochemical event, mean while retaining the large Stokes shift? 189 I.14!Discovery of intermolecular ES PT in the solution study of FR0 While investigating the spectroscopic behavior of FR-1V in various organic solvents, its precursor FR0 was also studied to ma ke a fair comparison and to validate Figure 0-1 Normalized spectra of FR0 in various solvents: (a) absorbance, (b) emission. Normalized spectra of FR0-SB (n-butylamine) in various solvents: (c) absorbance, (d) emission. Tol: toluene; THF: tetrahydrofuran; EA: ethyl acetate; DCM: dichloromethane; DMF: N,N -dimethyl forma mide; DMS O: dimethyl sulfone; A CN: acetonitrile; EtOH: ethanol; MeOH: methanol. Scheme shows the SB formation between FR0 and n-butylamine. NOFR0NH2NNFR0-SB 190 the strong solvatochromism is present along with ICT character. As shown in Figure III-1, FR0 does show the reported strong solvat ochromism in its fluorescence when exposed in various solvents. The absorption maxim um of FR0 is center ed around 400 nm, and the emission maxim um span s a range of 139 nm (from 434 nm in toluene to 573 nm in methanol). FR-1V absorption is center ed at 420 nm, while its emission spans a wider range from 488 nm (in toluene ) to 669 nm (in methanol ). This is expected as the bathochromic shifts in FR-1V is a re sult of its conjugation, which also makes the excited state more prone to solvent relaxation , leading to a stronger emission solvatochromism. Notably, the absorption maxima of both FR0 and FR-1V are blue-shifted accordingly w hen the n-butylimine Schiff bases are formed. However, the emission of FR0-SB is remarkably different from that of FR-1V-SB ( Figure II-7). In non -protic solvents, FR-1V-SB ( n-butylimine) display s a normal positive solvatochromism of emission which spans a range of 85 nm from 447 nm in toluene to 532 nm in DMSO, while the emission of FR0-SB ( n-butylimine) spans a much narrower range of 36 nm from 430 nm (in toluene ) an-Butylimine Schiff base. bAbsolute fluorescence quantum yield measurement using Quantaurus -QY at room tempera ture. Table 0-1 Spectral comparison of FR0 -SB and FR-1V-SB in organic solvents. ET(30) FR0 -SBaFR-1V-SBa!abs(nm) !em(nm) "b!abs(nm) !em(nm) "Toluene 33.9 3734290.457 3864470.662 THF 37.4 3684300.699 3894810.597 Ethyl acetate 38.1 3684280.623 3844760.595 Dichloromethane 40.7 3734420.758 3914980.685 DMF 43.2 3734600.748 3935250.658 DMSO 45.1 3754660.784 3965320.749 Acetonitrile 45.6 3684600.751 3885240.738 Ethanol 51.9 372463, 630 0.402 3945100.049 Methanol 55.4 374463, 633 0.169 3935570.065 191 to 466 nm (in DMSO ). In protic solvents (e.g., ethanol and methanol), FR0-SB not only shows the normal fluorescence centered around 460 nm, it also displays a strong emission band at 630 nm with much greater bathochromic shift ( Figure III-1d, Table III-1). A closer look at the correlation between the emission maxima and the solvent polarity is presented in Figure III-2. As shown, the emission wavelengths are converted to wavenumbers and correlated with ReichardtÕs E T(30) solvent polarity sca le,1 which is a measurement of the linear free -energy between the ground - and excited -state of ReichardtÕs betaine. FR0 displays a nice linear correlation in the energy terms, proving its characterized wavelength solvatochromism. Similarly, FR-1V also shows a nice linear correlation. In the case of their corresponding SBs ( n-butylimine), FR-1V-SB demonstrates a decent correlation. Interestingly, by selecting and plotting the short -wavelength (SW) emission maxima in ethanol and methanol, FR0-SB also exhibits the same level of the linear free -energy relationship. This indicates that the origin of the Figure 0-2 Solvatochromism of FR0 , FR-1V, and their SBs as evident by their emission wavelengths in different solvents. Plots illustrate the correlation between wavenumbers and solvent polarities. The short wavelength emission maxima of FR0 -SB in ethanol and methano l were selected for the plot. ReichardtÕs E T(30) scale is applied. 192 abnormal long -wavelength (LW) emission bands in alcohols is different from the solvent polarity effect. There is no clear correlation between the fluorescence quantum yields and solvent polarity as plotted in Figure III-3. However, general trend line should be presented in the aprotic solvents if wavelength solvatochromism is observed in this sub -group of solvents against E T(30) values. Because the ReichardtÕs sol vent scale takes into account H-bond interactions between the betaine and protic solvents, a clearer evidence of potential H -bond in the ground -state of FR0-SB would be the two distinct trend lines with aprotic and protic solvent groups, respectively. The protic solvents examined are limited to ethanol and methanol in Figure III-2 and Figure III-3, and thus more protic solvents need to be evaluated to provide a more concrete conclusion. Nonetheless, in Figure III-3 we can observe a rough trend among aprotic solvents that shows the increase in QY with increasing solvent polarity (from toluene to acetonitrile), although not quite linear. It will be interesting to know if a linear trend line is apparent with protic solvents. Due to the structural similarity of FR0-SB and FR-1V-SB, one might expect the same propensity of QY decreasing from methanol to ethanol, however, a large increase in QY of FR0-SB is Figure 0-3 QY as a function of solvent polarity of FR0 -SB and FR-1V-SB. 193 observed. It is not clear if the QY increase is related to the LW emission or simply due to the non-linear effect s. As shown in Figure III-4b, the 630 -nm LW emission band originat es from the 373 -nm excitation as does the 463 -nm SW emission band. Acidification of FR0-SB in ethanol with either hydrochloric acid (aq.) or perchloric acid (EtOH) solutions gives the protonated iminium ( FR0-PSB) with the absorption maximum at 488 nm ( Figure III-5a). Further acidification can protonate FR0-SB at the diethylamino nitrogen atom instead of the imine and yiel ds an ammonium that shows blue -shifted absorption and emission maxima at 345 nm and 410 nm, respectively, and is quickly hydrolyzed or ethanolyzed. The emission spectrum of the 488 -nm absorbing FR0-PSB perfectly overlaps with the LW emission band of FR0-SB in ethanol solution ( Figure III-5b), proving that the LW emission band is radiated from the protonated FR0-SB in the excited state. Considering the solution environm ent of FR0-SB and the direct excitation of 373 nm that pumps the excited state a) b) Figure 0-4 (a) Absorption and emission of FR0 -SB in ethanol. (b) Excitation spectra of FR0 -SB in ethanol at 460 nm and 630 nm, corrected against Rhodamine -B. 194 of FR0-SB, the abnormal LW emission phenomenon can clearly be described as an excited -state intermolecular proton transfer (ESiPT) process. This ESiPT process is not only observed in ethanol and methanol. In fact, solvents that provide exchangeable protons can all induce this LW emission. Shown in Figure III-6, in all 6 mono -/di-alcohols tested, FR0-SB exhibits the LW emission , with a varying peak maximum. It is tempting to linearly correlate the observed LW maxima with the E T(30) values of the solvents, which actually gives an R 2 coefficient of 0.84. Howev er, a closer look at both the SW and LW emission bands reveals more complex structures as featured by shoulders and plateaus. A peak fitting of the experimental spectra performed with a second -derivative analysis gives four Voigtian -type traces in Figure III-6. In all cases, the SW and LW emission bands can be deconvoluted into two traces. The envelope spectra of the four traces in each solvent present a perfect fit for the experimental results , a) b) NNFR0-SBHClEtOH NNHFR0-PSBFigure 0-5 (a) Stacked absorption traces of FR0 -SB acidification with hydrochloric acid. (b) Absorption and emissi on spectra of acidified FR0 -SB. 195 satisfying shoulders and plateau features . Although there is no clear physical insight to Figure 0-6 Absorption and emission spectra of FR0 -SB in different alcoholic solvents. (a) methanol, (b) ethanol, (c) 1 -propanol, (d) 2 -propanol, (e) n-amyl alcohol, (f) ethylene glycol. Traces labeled with Peak -i (1 ~ 4) are fluorescence spectra simulated by second derivative fitting with a Voigtian ampl itude distribution using PeakFit TM. The Envelope traces are the convolution of multiple Peak -i, showing a perfect overlap with the experimental spectra. a) c) e) f) d) b) 196 support this type of peak fitting, it is reasonable to conclude that the real transitions and/or emissive species should be no less than the 4 traces deciphered here. Tentatively, it might be plausible to ac cept the two traces for each emission band, given the possibility of imine isomerization in the excited states. More interestingly, dissolving FR0-SB in glacial acetic acid gives the absorption spectrum in Figure III-7a. The 342 -nm absorption peak corresponds to the protonation of the diethylamino group, with t he coexistence of the PSB iminium absorbing at 472 nm. Excitation at 342 nm leads to the ammonium emission at 405 nm and the LW emission centered at 622 nm, which corresponds to the iminium emission as shown in Figure III-7b with the excitation at 472 nm. This suggests the ESiPT not only happens on the neutral imine nitrogen atom, but also can have a double proton transfer at the excited state, namely, the deprotonation of the ammonium and the protonation of the imine. a) b) c) Figure 0-7 Absorption and emission spectra of FR0 -SB dissolved in acetic ac id. (a) Absorption spectrum, showing ammonium peaks at 342 nm and iminium peaks at 472 nm. (b) Emission spectra excited at 472 nm. (c) Emission spectra excited at 342 nm. In (b) and (c), emission traces labeled with Peak -i are simulated by second derivativ e fitting with a Voigtian amplitude distribution using PeakFit TM. The Envelope traces are the convolution of multiple Peak -i. 197 Thus far, the exceptional LW emission bands of FR0-SB observed in protic solvents can be ascribed to two types: 1) a single ESiPT in alcoholic solvents; and 2) a double ESiPT in organic acidic solvents ( Scheme III-1). In both scenarios, the excited state proton transfer on the imine nitrogen atom is responsible for the LW emission. The emission intensity of the LW band is variable in different alcohols, indicating a possible correlation with the H -bond donor strength instead of the commonly studied solvent polarity. In the case of 2 -propanol and n-amyl alcohol, the SW and LW emission bands are not well resol ved. This may suggest the impact of the steric hindrance projected by the aliphatic side chains. The two traits here further suggest that a preset H -bonding may be already present between the solute and solvent in the ground state. I.15!Time-resolved study of FR0-SB ESiPT Time -resolved spectroscopic studies were carried out to provide in -depth understanding of the excited state photodynamics of FR0-SB ESiPT. This part of the work results from a strong collaboration with Dr. Muath Nairat from Prof . Marcus Dantus Õ research gr oup (Michigan State University ). a) b) NNFR0-SBORH!NHN!em ~ 630 nm !!em ~ 450 nm FR0-PSBNNFR0-SBESiPT h"NNFR0-SBORH!NHN!em ~ 630 nm !!em ~ 405 nm FR0-PSBNNHFR0-ammonium ESiPT h"ROHHdoublesingle !abs ~ 370 nm !abs ~ 340 nm Scheme 0-1 Schematic transformation of (a) single and (b) double ESiPT of FR0 -SB in protic solvents. 198 As described in Section I.14, the ESiPT behavior of FR0-SB denotes the increased basicity in the excited state such that the proton abstraction from the solvent molecule can happen. This type of molecules that show weak basicity in the ground state but elicit an increased basicity upon light excitation is defi ned as a ÒphotobaseÓ (pK a* > pKa). The underlying dynamics and principles of the key excite d state proton transfer process have been the long -standing interest ever since Fırster and WellerÕ s early work. 2, 3 In this field, p hotoacids (pK a* < pK a) are more commonly reported. Both intra - and inter -molecular ESPT of photoacids have been observed and survey ed.4-7 The most and earliest studied photoacids are substituted 3- and --naphthols. 8, 9 More recently, it was found th at substitution with strong electro n withdrawing groups on hydroxyarenes can generate ÒsuperÓ photoacids, which are denoted by drastic pK a change in the excited state (greater than 6 units ) and have a negative pK a*. The most investigated is 8 -hydroxypyren -1,3,6-trisulfonic acid (HPTS), or pyr anine.10-13 Conversely , studies on photobases are scarce and limited to heterocyclic nitrogen -containing compounds such as acridi ne,14 3-styrylpyridi ne,15 aminoanthraquino ne,16 quinolines, 17 dibenzophenazi ne,18 7-azaind oles, 19, 20 as well as carbonyl compounds such as xanthon e,21 curcum in,22 or bifunctional naphthol photoacids in which a basic moiety becomes moderately more basic upon photo excitat ion.23, 24 An interesting theoretical investigation is recently reported by LischkaÕs g roup. They applied different high -level computation al methods to study the excited -state properties and environmental effects of retinal PSBs. It was found that the feasibility of the proton transfer process from the PSB to its surrounding solvent partner is significantly 199 enhanced in the excited state, meaning that an ESPT step could be initiated already in the early stage of photodynamics of retinal PSB which functions as a photo acid.25 This, however, is in contrast to our observation for the ESPT in FR0-SB. Acetonitrile and ethanol were selected to represent non -proton transfer and proton -transfer solvents for ultrafast spe ctroscopic studies, respectively. The steady -state absorption and emission spectra of FR0-SB/PSB in these two solvents are aligned in Figure III-8 for a clearer view. The time -correlated single -photon counting (TCSPC) measurements of FR0-SB/PSB in acetonitrile and ethanol are summarized in Figure III-9. The emission in ACN is featu red by a single exponential decay with a lifetime with 2.11 ns. This SW emission lifetime is sharply decreased to ca. 89 ps in ethanol. The decrease Figure 0-8 Steady -state absorption (solid) and emission (shaded) spectra of FR0 -SB in (a) acetonitrile (ACN) and (b) ethanol (EtOH). The spectra of FR0 -PSB (acidified by perchloric acid in ethanol) are also overlaid. ! 200 in SW emission lifetime can be explained by the competing intermolecular proton transfer of FR0-SB in the e xcited state. On the other hand, the LW emission of FR0-SB in ethanol has a lifetime of 1.31 ns, which is in the same order of magnitude as the SW emission lifetime in ACN. Comparing the FR0-PSB TCSPC trace with that of FR0-SB LW emission trace, the two have an identical single -exponential decay as plotted in Figure III-9 inset, further validating the origin of the FR0-SB LW emission fr om the excited state proton transfer. An isotope study was carried out with deuterated ethanol (ethanol -d6, EtOD). The absorption spectra in EtOH and EtOD are identical. The emission spectrum in EtOD has a less prominent LW emission. With a heavier deuterium, more vibrational crossover is expected and correspondingly the SW and LW emission band are less resolved (Figure Figure 0-9 TCSPC traces with single exponential fits for FR0 -SB near the emission maxima when dissolved in ACN and EtOH. Inset shows that the red emission trace at 650 nm of FR0 -SB is identical to the FR0 -PSB emission. 201 III-10a). Because the fluorescence lifetime is measuring the population decay of the excited state species, the ratio of decay time should directly reflect the ratio of the kinetic rates of the proton transfer reaction. In the deuterated ethanol, the LW emission lifetime increases from 1.24 ns to 2.00 ns, showing a noticeable isotope effect of 1.6 (Figure III-10b). Comparing methanol and methanol -d4, a more prominent primary isotope effect Figure 0-10 Isotopic effect of FR0 -SB in EtOH and ethanol -d6 (EtOD). (a) Steady -state absorption and emission spectra of FR0 -SB in EtOH and EtOD. (b) TCSPC decay curves of FR0 -SB LW emission along with single expo nential decay fits in EtOH and EtOD . An isotope effect of 1.6 is observed in EtOD . a) b) 202 of 1.9 is observed ( Figure III-11). Together, these results validate the determining role of the proton transfer in the origin of this LW emission. It is now clear that FR0-SB is acting as a photobase in the excited state in protic solvents. The change in pK a upon excitation is determined using the Fırster equatio n:2, 12 where pK aand pK a are the excited state and gr ound state logarithmic acidity constants, respectively. R is the ideal gas constant, T is the temperature in Kelvin, and h6,is the energy of 0 -0 electronic transition of FR0-SB and FR0-PSB. The 0 -0 transition energies are estimated as 24,450 cm -1 and 17,890 cm -1 from the corresponding crossover points in the absorption and emission spectra of FR0-SB and FR0-PSB, respectively ( Figure III-8b). The calculated pK a increase of FR0-SB is 13.8 units . While pK a changes up to 14 units have been reported for photoaci ds, 26 to t he best of our knowledge the largest pKa change reported for a photobase is 10.8 for 5 -aminoquinoli ne.17 With the 0pKa in hand, pKa!"pKa=h!1"h!2()/2.3RTEquation 0-1 Fırster e quation for calculating the excited -state pK a difference. Figure 0-11 TCSPC decay curves of FR0 -SB LW emission with single expone ntial decay fits in Me OH and methanol -d4 (MeOD). An isotope effect of 2.0 is observed in Me OD. 203 calculation of the excited state pK a* requires an estimate for the ground state pK a of FR0-PSB. Although a good estimate for the pK a in EtOH can be derived from the pK a of the imine in water, this was hampered by the insoluble nature of FR0 -SB in water. No te that the pK a of ammonium salt is generally elevated by ~3 units when dissolved in ethanol instead of wat er. 27 Therefore, we resorted to an indirect measurement of the pK a via quantifying th e mole fraction of each species by multi -variant linear regression with known a) b) Figure 0-12 (a) Transient absorption spectra of FR0 -SB at various time intervals after excitation in EtOH . Labeled arrows show the steps during the ESPT process. (b) Energy progression during the proton transfer process . 204 extinction coefficients as a function of wavelengths (UV/Vis analysis), generated between the acid -base reaction of FR0-SB with 3-naphthylammonium in EtOH as described in a previously reported procedu re. 28 FR0-PSB shows a 0.23 unit lower pK a than that of 3-naphthylammonium in EtOH. The pK a of 3-naphthylammonium in water is 3.9, thus, its pKa in EtOH is estimated at 6.9. Considering the difference in pK a values for a -naphthylammonium and FR0-PSB, the pKa of the latter in EtOH is estimated to be 6.7. This results in an estimated excited state pK a* of 21 for FR0-SB. A better understanding of the mechanism and associate dynamics of the FR0-SB excited state proton transfer is pr ovided by the transient absorption spectra (TAS). As shown in Figure III-12, only a decrease in the absorption (positive signal) around 500 nm (20,000 cm -1) was observ ed upon excitation. This is ascribed to the excited state absorption (ESA) of FR0-SB. The ESA signal decays quickly and is accompanied by a transient negative signal centered at 588 nm (17,000 cm -1). This transient negative signal gives rise to another negative signal that initially centers at 630 nm (15,870 cm -1) and shifts to lower energy over time. The negative signal at 630 nm persists longer than 500 ps and is assigned as the stimulated em ission (SE) from FR0-PSB. Accordingly, the transient signal at 588 nm is ascribed to the SE from an intermediate species that is formed during the intermolecular proton transfer process . The intermediate can be viewed as a caged complex between FR0-SB and the solvent molecule (ethanol) with a partial transfer of the proton ( vide infra ) prior to the full proton transfer process . This feature is commonly observed in the excited state intermolecular proton transfer of photoaci ds. 26, 29 205 In the absence of proton transfer (e.g. in ACN), the transient absorption of FR0-SB shows rapid ESA and SE emergence soon after excitation , since both correspond to the non -protonated form (Figure III-13a). Moreover, both signals are long lived and can be fit to a biexponential decay . A long component of ca. 1.89 ns corresponding to the lifetime of the excited state is observed, together with a short 24 ~ 30 ps component that can be ascribed to an intramolecular response from the molecule upon excitation (Figure III-13, b and c). The 24 ps decay component is also observed in the SE of FR0-PSB a) b) c) Figure 0-13 (a) Transient absorption of FR0 -SB in ACN , showing the long -lived excited state absorption and stimulated emission signals from the non -protonated form with biexponential fits at the frequencies of (b) excited state absorpt ion and (c) stimulated emission signal s. 206 (acidified with HClO4/EtOH ), confirming its nature as an intramolecular mode from FR0-SB regardless of its protonation st atus ( Figure III-14). A four -level sequential global analysis model can be used to provide more complete description of the experimental transient absorption da ta of FR0-SB in EtO H.30 The resulting evolution associated spectra (EAS) are provided in Figure III-15. The model shows that the first component (black trace) , which is the ESA of FR0-SB centered around 500 nm (20,000 cm -1), decays with a 15.8 ps time constant. This timescale is in agreement with the average dielectric relaxation of EtO H.31 This decay is accompanied with a rise in the second component (red trace) that features a broad SE signal centered at 588 nm (17,000 cm -1) and decays with a 73.9 ps time constant. Based on these time constants, we infer that the dielectric relaxation of the solvent is coupled with the formation of an intermediate species prior to the full proton transfer step. This intermediate can be viewed as a complex with a partially transferred proton or a highly polarized H -bonding state Figure 0-14 Stimulated emission signal from the transient absorption of FR0 -PSB in acidic E tOH. Two decay components are observed, with a fast one corresponding to an intramolecular response from FR0 -SB, and a slow one corresponding to the excited state lifetime. ! 207 between the base and the solvent. Characterized by the FR0-PSB SE feature with maxima around 630 -660 nm (15,870 ~ 15,150 cm -1), the final step is best described with two EAS components (blue and purple traces) . Considering the similarities between the last two EAS components 3 and 4, we contemplate them as a single step for the formation of the fully protonated FR0-PSB. The two exponential components are better fit for the non-exponential decay of the protonated form that can undergo geminate recombination after the proton tran sfer. 32, 33 The last EAS component is long lived and decays with a a) b) Figure 0-15 (a) Evolution associated spectra (EAS) of the four levels used in the sequential global analysis model to describe the proton transfer dynamics of FR0 -SB in EtOH. (b) Population kinetics of the four levels used in the global analysis model. 208 1.15 ns time constant . This value agrees with the previously determined lifetime of FR0-PSB using TCSPC ( Figure III-9). Similar photophysical behavior is observed when FR0-SB is dissolved in MeOH (Figure III-16). The global analysis model shows that the ESA signal from the non -protonated form decays to form the partially proton transferred complex on a 2.9 ps timescale, which also agrees with the av erage dielectric solvation time of MeO H.31 The protonated FR0-PSB is formed on a 27.1 ps timescale according to the global analysis model. ! A further validation study of the global analysis model was done by viewing the transient absorption traces at selec ted wavelengths of FR0-SB in EtOH and ethanol -d6 (EtOD). Observed through t he SE signal at 570 nm , the dielectric solvent relaxation that couples with the intermediate formation is formed with a time constant of 15.2 p s that close ly agrees with the interme diate formation time determined by the global analysis Figure 0-16 Transient absorpt ion trace showing the energy pro gression during the excited state FR0 -SB proton transfer process in MeOH. ! 209 model (15.8 ps). Nonetheless in EtOD, we notice an isotope effect of 1.5 ( vide supra ), further confirming that the dielectric solvent relaxation is coupled with partial hydrogen bonding formation and dissociation ( Figure III-17a). The SE from the fully protonated FR0-PSB (i.e., t he second trace at 650 nm) shows its formation is associated with a time constant of 57.3 ps. This value is ascribed to the dissociation of the intermediate that leads to FR0-PSB formation and is more accurately validated by the global analysis as a time constant of 73.9 ps. A primary isotope effect of 2.0 is observed in the EtOD trace at this wave length. The two strong isotope effects observed both before and after the proton transfer step clearly elucidate the ÒcontactÓ or ÒcovalencyÓ feature of the geminate ion complex intermedia te,34 which can be more accurately described as excited state hydrog en bonding (ESHB) complex. The isotope effect in methanol -d4 (MeOD) is more pronounced. In both of the intermediate and the a) b) Figure 0-17 Transient absorption traces at (a) 570 nm where stimulated emission from the intermediate formation is observed, a nd (b) 650 nm where stimulated emission from FR0-PSB can be seen while dissolved in EtOH (black) and EtOD (red). Biexponential decay constants are given in the inset along with !the pre -exponential factors in parentheses. An isotope effect of 1.5 is observed during the formation of the partially proton transferred intermediate , while an isotope effect of 2 .0 is observed during the formation of final protonated FR0-PSB. 210 protonated FR0 -SB formations, a primary isotope effect (greater than 2 ) is observed (Figure III-18). Overall, the scheme of excited state intermolecular proton transfer (ESiPT) from EtOH to FR0-SB is depicted in Figure III-19 and can be summarized as following: (i) The lifetime of the non -protonated FR0-SB in aprotic solvent, where proton transfer is not feasible, can be deduced from the emission lifetime at 480 nm in ACN solution. (ii) In protic solvents, a highly polarized H -bonding complex is formed on a timescale that competes with the solvent dielec tric relaxation, with the observation of an isotope effect of 1.5 ~ 2 during the formation of this intermediate. (iii) The final proton transfer step occurs on a 73.9 ps timescale and involves the dissociation of the partially -transferred proton to produce the protonated FR0-PSB. An isotope effect of 2 is observed for this step. (iv) The protonated FR0-PSB is long l ived and emit with a 1.31 ns lifetime. a) b) Figure 0-18 Transient absorption traces at (a) 580 nm where stimulated emission from the intermediate formation is observed, a nd (b) 660 nm where stimulated emission from FR0 -PSB can be seen while dissolved in MeOH (black) and MeOD (red). Biexponential decay constants are given in the in set along with the pre -exponential factors in parentheses. An isotope effect of 2 is observed during the formation of the intermediate as well as in the final protonated form formation . 211 Having established the super photobasicity of FR0-SB in protic solvents , we explore the ability of this compound to abs tract protons from stronger C -H bonds. We notice that there is no clear sign of proton transfer from C -H bonds, even with low pK a solvents such as chloroform . On the other hand, N -H compounds such as imidazole were deprotonated in the presence of FR0-SB up on excitation (Figure III-20). This is not surprising since the bond dissociation energy of chloroform C -H bond is much higher than that of the solvated O -H or N -H bonds. T hese results may infer the importance of H -bonding between the base and the solvent in the ground state for ES iPT to be feasible. 400nm,adecreaseinabsorption(positivesignal)around500nm(20000cm!1)wasobservedinitially,whichisattrib-utedtoexcitedstateabsorption(ESA)ofFR0-SB.TheESAsignaldecaysquickly(arrowAinFigure3a)andisaccom-paniedbyatransientnegativesignalcenteredat588nm(17000cm!1,arrowB)thatgivesrisetoanothernegativesignalinitiallycenteredat630nm(15870cm!1,arrowC)andshiftstolowerenergyovertime.Thenegativesignalat630nmpersistslongerthan500psandisassignedasthestimulatedemission(SE)fromFR0-PSB.Thetransientsignalat588nmisascribedtotheSEfromanintermediatespeciesthatformsduringthecourseoftheintermolecularprotontransfer.TheintermediateisthoughttobeacagedcomplexbetweenFR0-SBandthesolventmoleculewithapartialtransferoftheproton(videinfra)priortothefullprotontransferprocessthatisobservedintheintermolecularprotontransferofphotoacids.[3a,20]Transientabsorptionintheabsenceofaprotontransferevent,suchasinACN,showsthatESAandSEappearsoonafterexcitation(FigureS6)sincebothcorrespondtothenon-protonatedform.BothESAandSEsignalsarelong-livedandcanbefittoabiexponentialdecaywithalongcomponentofabout2.1nscorrespondingtothelifetimeoftheexcitedstateandashort26to30pscomponentthatisascribedtoconformationalchangesofthemoleculeuponexcitation.The26psdecaycomponentisalsoobservedwhenexcitingFR0-PSB(inEtOH/HClO4),confirmingitsnatureasanintramolecularmodefromFR0-SBregardlessofitsproto-nationstatus(FigureS7).Amorecompletedescriptionoftheexperimentaltran-sientabsorptiondatainEtOHisbestobtainedusingafour-levelsequentialglobalanalysismodel.[30]TheevolutionassociatedspectraobtainedareprovidedinFigureS8.Asstatedpreviously,excitationinACN(I!III,Figure4)leadstoalong-livedexcitedstate(2.11ns),whileprotontransferinEtOHdrasticallyreducestheexcitedstatelifetimeofIIIintheproticsolvent(89psinEtOH).Themodelshowsthatthefirstcomponent,whichistheESAofFR0-SBcenteredaround500nm(20000cm!1),decayswitha15.8pstimeconstant(seeIII!IV,Figure4),whichagreeswiththepreviouslydeduced"18psfromsteadystatespectraandtheTCSPCmeasuredlifetime.ThistimescaleisinagreementwiththeaveragedielectricrelaxationofEtOH(16ps).[31]ThisdecayisaccompaniedwithariseinthesecondcomponentfeaturingabroadSEsignalcenteredat588nm(17000cm!1),presumablythepartiallyprotonatedstatedepictedasIV,decayingwitha73.9pstimeconstanttoyieldtheexcitedPSBformofFR0(V).ItisworthnotingthatthepartiallyprotonatedstateIVisnotemissiveandisonlyobservedbystimulatedemission.Basedonthesetimeconstants,weinferthatthedielectricrelaxationofthesolventiscoupledwiththeformationofanintermediatespeciespriortothefullprotontransferstep.CharacterizedbytheFR0-PSBSEfeaturewithmaximaaround630Ð660nm(15870Ð15150cm!1),thefinalstepisbestdescribedwithtwocomponents.Consideringthesimilaritiesbetweenthelasttwocomponents(seeFigureS7,forspectraofcomponentsthreeandfour)westipulatethemassolvationoftheprotonatedformV.Thisstepoccursona211pstimescale,whichagreeswiththeobservedrisetimeintheTCSPCdataat650nm(197ps,seeFigureS2b).Thelastcomponentislonglivedanddecayswitha1.15nstimeconstant,avalueinagreementwiththeprotonatedFR0-SBlifetime(1.31ns)thatwaspreviouslydeterminedusingTCSPC(Figure2c).SimilarphotophysicalbehaviorisobservedwhenFR0-SBisdissolvedinMeOH(FigureS9).TheglobalanalysismodelshowsthattheESAsignalfromthenon-protonatedformdecaystoformthepartially-transferredprotoncomplexona2.9pstimescale,alsoinagreementwiththeaveragedielectricsolvationtimeofMeOH(5ps).[31]TheprotonatedFR0-SBisformedona27.1pstimescaleaccordingtotheglobalanalysismodel.AsdepictedinFigure5,theglobalanalysisdeducedpathwaywasfurtherconfirmedusingthetransientabsorptiontracesatselectedwavelengthsofFR0-SBinEtOHandEtODFigure3.a)TransientabsorptionspectraofFR0-SBatvarioustimeintervalsafterexcitationinEtOH.LabeledarrowsshowthestepsduringtheESPTprocess.b)Energyprogressionduringtheprotontransferprocess.Figure4.TheobservedintermolecularESPTdynamicsinEtOHalongwiththeassociatedtimeconstantsforthestepsasobtainedfromglobalanalysis(black)andtheTCSPCdata(colored).AngewandteChemieCommunications14744www.angewandte.org!2018Wiley-VCHVerlagGmbH&Co.KGaA,WeinheimAngew.Chem.Int.Ed.2018,57,14742Ð14746Figure 0-19 The observed intermolecular ESPT dynamic s in EtOH along with the associated time constants for the steps as obtained from global analysis (black) and the TCSPC data (colored). ! 212 The ground state H-bond requirement may also be related to the lifetime of the excited state super photobase and diffusion of the proton donor . I.16!Recapitulate ESHB in hCRBPII complexes to develop bright, long Stokes Shift, and fast -forming imaging tag I.16.1!ESHB of photoacids and the applications Both intermolecular and intramolecular excited state proton tran sfer, or more generally excited state hydrogen bonding, are hot topics in fundamental research as well as application design. 35, 36 The representative molecules showing excited state intramolecular proton transfer (ESIPT) are hydroxyflavones and hydroxychromon es. 37-39 These scaffolds commonly have a photoacidic hydroxyl group that can transfer the proton to an adjacent carbonyl group, rea lizing a light -triggered keto -enol tautomerization in the excited state. Actually, compounds bearing an H -bonding donor (HBD) that can form a five - or six -membered ring transition state with an adjacent H -bonding acceptor (HBA) C(imidazole) Figure 0-20 Normalized fluorescence spectra of FR0 -SB in the acetonitrile solution of imidazole. The arrow shows the LW emission band (630 -660 nm) emerging with the addition of imidazole (from blue to red), indicating the proton transfer from imidazole to FR0 -SB. 213 can demonstrate this type of ESiPT pheno menon. 40-43 In line with this mechanism, ES iPT essentially features a dual fluorescence of the normal and the tautomer form, respectively. This feature has been applied to probe the environment basicity since the dual emission ratio reflects th e strength of the intramolecular proton transfer which is affected by the hydrogen bond accepting ability of the environm ent.44 This concept has been further developed to sense anions such as acetate and fluoride that can establish strong ESHB with the hyd roxyl group and therefore inhibit the ES iPT proce ss.45-49 ESiPT has also been used to design materials that can entirely frustrate the energy transfer between a large band-gap donor and a small band -gap acceptor such that a concentration -independent white-light-emitting single molecule is feasibl e.50 On the other hand, ESiPT is the critical step for wildtype GFP to fluoresce. Mentioned in Section I.1, the initially matured GFP chromophore p-HBI is in a neutral phenol form. Upon a 390 -nm photon excitation, the phenol is deprotonated via a water mediated proton wire composing Ser205 and Glu222 and results in a highly fluorescent phenolate emitting green light.51 This intermolecular ESPT pathway was also studied with conformationally locked model compounds in solution and revealed the important role of ESPT in excited state photoisomerizati on,52 which is the key transformation observed for RSFP Dron pa.53 In another model study with synthetic imidazole fluorophores, it was shown that ESPT -coupled excited state charge transfer (ESCT) may be responsible for the observed emission pH sensitivity in many FPs due to the charge swapping between the two imidazole nitrogen atoms on the excited sta te.54 214 Intermolecular ESPT has been widely used to increase the Stokes shift of FPs. ESPT is suggested as the mechanism responsible for the large Stokes shifts observed in LSSmKates RFPs based on crystallograp hic and mutagenesis analys es. 55 In viewing the hydrogen bonding between the DsRed -chromophore hydroxyl group and Glu160 in LSSmKate1, the proton relay involving the hydroxyl, Ser158 and the Asp160 carboxylate in LSSmKate2, and also the proton relay compose d of the hydroxyl, Ser143 and Asp158 in mKeima, similar ESPT pathways were engineered for other DsRed RFPs based on these templates. It was successfully shown by VerkhushaÕs group that the Stokes shifts were all greatly enlarged by over 100 nm in mNeptune, mCherry, mStrawberry, mOrange, and mK O.56 Similar strategies have been adopted in intramolecular cases with synthetic analogues to develop DNA label s.57 Recently, a phenylbenzothiazole derivative HABT was designed to bind to a tyrosine kinase to establish a hydrogen bonding network with neighboring T766 and active -site water to facilitate ESPT for probing the protein binding site by the change in Stokes shif t.58 I.16.2!ESHB in dye/hCRBPII complexes Described thus far in this section, all of the large Stokes shift strategies are based on photoacidic hydroxyls or N -Hs. In Section I.14 and I.15 we described the super photobasic behavior of FR0-SB and its apparent larger Stokes shift. Is it possible to recapitulate this photobase initiated ESPT or ESHB in the hCRBPII scaffold? Actually, the phenomenon was already observed in complexes of DiAI-2V/Q108K:K40L:T51V: T53S:V62E and DiAI-2V/Q108K:K40L:T51A:T53S: V62E ( Table II-1). However, the key mutations of hCRBPII that facilitate ESHB were not clear at that time. 215 During the course of engineering hCRBPII mutants for regulating the e mission wavelengths of ThioFluor-1, Dr. Elizabeth M. Santos found a mutant Q108K:K40E:T53A: R58L:Q38F:Q4F (KEALFF) that emits at around 600 nm with a 390 -nm excitation after binding with ThioFluor-1. The apparent 200 nm Stokes shift cannot be explained by the direct emission of ThioFluor-1/KEALFF SB. The emission wavelength is also blue -shifted about 100 nm from the regular ThioFluor-1/hCRBPII PSB emission maxima .59 Contemplating the super photobasicity of FR0-SB and its corresponding LW emission, the emiss ion character of ThioFluor-1/KEALFF presents a similar feature and is ascribed to ESPT, or more accurately ESHB due to the blue -shift in the emission wavelength that is not the signature of protonated complexes. Collaboratively, this was investigated to scope the generality of the ESHB phenomenon in engineered hCRBPII with different fluorophores that could form a neutral imine in the ground state. A more detailed description of the related protein engineering investigation can be found in Dr. SantosÕ PhD th esis. The important mutations for ESHB of FR dyes can be visualized in the docking simulation of FR-1V in a model hCRBPII scaffold ( Figure III-21). In principle, an H -bonding acceptor located within a proper distance to the FR-1V imine nitrogen atom should be able to stabilize the proton on the nitrogen atom via a partial H -bond and he nce facilitate the ESHB to proceed. With this consideration and accounting for the average H -bonding length, a radius of 5 ~ 6 † is scanned from the FR-1V imine nitrogen in the docking model. As shown in Figure III-21, residues within 5 † radius are highlighted in green. These include W106 and E117. The former is conservative towards mutagenesis. 216 The latter shows enhanced ESHB with ThioFluor-1 but impedes the binding with FR-1V. Residues at 6 † radius are highlighted in red, including L93, Q4, L40, I42, T51, and F64. Consider ing the elongation of short aliphatic residues to glutamate, four other mutations that can potentially fit in 6 † radius are also simulated and highlighted in magenta, including V62E, T53E, L115E, and L119E. Mutagenesis at these positions with HBA residues (i.e., glutamate, aspartate, and histidine) along with other mutations leads to a panel of mutants that are subsequently screened over all of the FR dyes. Not surprisingly, the expressible mutants, predominantly as glutamate s, show various extent of ESHB emissions with different compounds ( Figure III-22). It is notable that the LW emission band of ESHB by exciting the SB of the Figure 0-21 Potential residues facilitating excited state hydrogen bonding on FR-1V imine nitrogen atom . Residues within 5 † to the nitrogen atom are shown in green. Residues within 6 † distance are shown in red and magenta. The pose of FR-1V is simulated with a flexible docking method by using AutoDock Vina. The protein matrics is extracted from hCRBPII mutant KL (PDB ID: 4EXZ). FR-1V:carbon (cyan), nitrogen (blue), hydrogen (grey). Magenta residues are simulated with PyMOL. 217 fluorophores are all blue -shifted as compared to the corresponding PSB emissions. In the case of FR0/Q108K:K40E:T53A:R58L:Q38F:Q4F (psES235 ) complex, the SB shows two resolved absorption maxima . The excitation of these two maxima (383 and 413 nm) give slightly different ESHB emission spectra with similar quantum yield s (0.75 and 0.80, respectively) . Int erestingly, the long -wavelength excitation actually leads to the short -wavelength emission (Figure III-22). The complex of FR0/Q108K:K40D:T53A:R58L: Q38F:Q4F:V62E (dimer) has an ESHB emission with signifi cant bathochromic shift of ~ Figure 0-22 Representative FR/hCRBPII complexes showing excited state hydrogen bonding long wavelength emissions. Mutant Q108K:K40D:T53A:R58L:Q38F:Q4F:V62E is in the dimeric form. 218 80 nm. This dimer effect has also been observed elsewhere. The ESHB energy transfer efficiency is determined by both the ligand and the protein host. In some cases, the complexes even show the distinguishable shoulder band in t he long -wavelength region. These observations together suggest the existence of a complex proton network in these dye/hCRBPII complexes that may have varying solvent reorganization time constants in the ESHB processes to cause the different ratio of the LW emission components, as discussed in Section I.15. I.16.3!A bright ESHB tag with large Stokes shift and fast -forming kinetics Considering application for live cell imaging, FR-1V/psES235 complex was selected for its optimal emission wavelength ("7%,8,595 nm) and high fluorescence quantum yield (*,8,0.72). FR0/psES235 exhibits a 92% quantum yield with an emission maximum around 510 nm . The excitation spectrum of FR-1V/psES235 is shown in Figure Figure 0-23 Excitation spectrum of FR-1V/psES235 600 nm emission. The red trace is the excitation spectrum corrected against Rhodamine -B. The green trace is the absorption spectrum. The blue trace is the normalized brightness cu rve, plotting the product of # and $ with each square as a data point. 219 III-23. Interestingly, the ESHB emission is not only observed by 392 -nm excitation but also by irradiation at 280 nm, although the energy transfer efficiency from the 280 -nm absorbing tryptophans is much lower than the FR0 SB as demonstrated by the comparison between the absorption spectrum and the normalized brightness curve. This might suggest the near -field energy transfer between W106 and FR-1V imine in the excited state. The binding of FR-1V and psES235 is instantaneous. Under the condition of pseudo -first order measurement ( Figure III-24a), it only takes 0.18 s ec as the binding half-time. In the more accurate second -order rate constant measurement, it is showing the ra te at the level of 106 M-1s-1 (Figure III-24b). The ultrafast binding kinetics of FR-1V/psES235 is critical for live cell imaging. Because of the sub -minute binding, the a) b) Figure 0-24 Binding kinetics of FR-1V with psES235 . a) Pseudo -first order binding of FR-1V (10 !M) with psES235 (20 !M). The binding was monitored at the 595 nm ESHB emission maximum. An exponential fit gives the bi nding half -time as 0.18 s. b) Second -order rate constant measurement. Variable concentrations of FR-1V were used to bind 100 nM of psES235 . Observable rate constants were derived from single exponential fits. The second -order rate constant was then derived from the linear regression of observable pseudo first order rate constants as a function of concentration. All measurements were taken in PBS buffer (pH = 7.3). 220 required cell staining time is only limited by FR-1VÕs permeation and diffusion into the cells. This fast rate can differentiate the binding of FR-1V with psES235 from other nonspecific cellular proteins, thus providing a feasible mechanism to reduce n onspecific fluorescence background by removing the excess amount of FR-1V before its reaction with nonspecific targets. To test the imaging performance of FR-1V/psES235 in cellulo , the gene of psES235 is cloned into pFlag -CMV2 vectors with a N -terminal EGFP sequence and C -terminal localization peptides (i.e., NLS, NES, and CAAX) as described in 0. The high brightness of FR-1V/psES235 allows a final staining concentration of 500 nM of FR-1V, EGFP -psES235 -NLS !EGFP -psES235 -NES !EGFP -psES235 -CAAX !EGFP !FR-1V/psES235 !Figure 0-25 Compartmentalized FR-1V/psES235 imaging in live HeLa cells. NLS = nuclear localization sequence. NES = nuclear export sequence. CAAX = prenylation tag. Cells were stained wi th 500 nM FR-1V and incubated at 37 ¡C for 1 min. Cells were washed 3 times with DPBS before imaging. Scale bar, 10 !m. 221 which is much lower than the amount required for the photoswitching mutants. The fast binding enables a quick labeling protocol with 1 min staining at 37 ¡C followed by 3 wash steps (Figure III-25). In all sub -cellular imag es, the confocal red channel (LP615) shows bright emission from the long -wavelength ESHB fluore scence of the complex. The line profile analysis demonstrates high signal specificity of the red fluorescence that is colocalizing with the reference EGFP signal ( Figure III-26). To this end, an imaging tag featuring high brightness, large Stokes shift, and fast labeling kinetics has been established. EGFP -psES235 -NLS !EGFP -psES235 -NES !EGFP -psES235 -CAAX !Figure 0-26 ESHB fluorescence specificity of FR-1V/psES235 . Top row shows the merged EGFP and LP615 channels in cell nuclei, cytosol, and plasma membrane -localized HeLa cells. Bottom row shows the line profile of colocalized EGFP and E SHB signal along the white arrows in the top images. Scale bar, 10 !m. 222 I.17!Dual-color live cell imaging enabled by a single dye Current multicolor live cell imaging is enabled by combining multiple FPs or synthetic dyes that have separat e excitation or emission channels in the confocal microscope. A good example of spontaneously labeling six FPs is given in the beginning of Section I.2. However, multicolor labeling with synthetic dyes require the choice of different dye scaffolds and necessitates multiple rounds of staining and washing. On the other hand, the 200 nm difference in the excitation wavelengths (405 -nm Diode vs 594 -nm HeNe) and 100 -nm difference in emission maxima (595 nm vs 704 nm) of the ESHB FR-1V/psES235 and the photoswitching FR-1V/ps4 tags provide the opportunity to achieve dual -color imaging by u sing just a single dye (FR-1V) with conventional dichroic settings. Or more generally, the dual -color imaging is achievable if two hCRBPII mutants can be properly expressed in the target cell and give the desired ESHB and PSB excitation/emission, respectiv ely, without any interfering nonspecific background. The initial attempts to co -express ps4 and psES235 in HeLa cells for dual -color imaging was not successful. For unknown reasons, the expression of the two mutants was problematic. Besides, the binding ra tes of FR-1V with the two mutants are different (1.9 min vs 0.18 s), which results either in insufficient labeling time for ps4 or in increasing the nonspecific background for psES235 . Moreover, the absolute fluorescence quantum yields also have 4 -fold dif ference, resulting in incomparable sensitivity of the two imaging channels. In this context, it is necessary to find a mutant with comparable binding kinetics and quantum yield to psES235 that can form a stable FR-1V PSB after thermal binding. 223 Mutant psST6 1 (Q108K:K40H:T53A:R58L:Q38F:Q4F) was identified after mutagenesis and screening. As shown in Figure III-27, the complex of FR-1V/psST61 has a predominant PSB absorpti on at 516 nm, with a small portion of SB absorption at 398 nm. Excitation at 516 -nm leads to the signature PSB emission centered at 630 nm with a shoulder at 650 nm. Interestingly, this spectrum is almost identical to the emission spectrum of FR0-PSB in HC lO4/EtOH ( Figure III-8b). The excitation at 398 -nm emits a typical SB fluorescence at 450 nm, however, with a small fraction at 630 nm as the ESHB emission. Nonetheless, the quantum yield of the PSB emission is 39% . Together with a 2-fold stronger absorption, the brightness of the PSB emission is estimated as 14 -fold higher than that of the unwanted ESHB emission. The binding of FR-1V with psST61 is also fast ( Figure III-28). The binding assay displays a half -time of 6.8 s ec, not as fast as the 0.18 s of psES235 , but closer in the magnitude than the 1.9 min observed with ps4. The second -order rate constant is determined as 2.5 # 10 4 M-1"s-1, only about 50 -fold smaller than that of psES235 . Figure 0-27 Absorption and emission spectra of FR-1V/psST61 . 224 To test dual -color imaging, HeLa cells are co -transfected with pFlag -CMV2 vectors encoding EGFP -psES235 -NES and EGFP -psST61 -NLS, respectively. Ideally, t he ESHB emission of FR-1V/psES235 should be localized in the cytosol and the PSB emission of FR-1V/psST61 should be observed only inside cell nuclei. Shown in Figure III-29b, the PSB emission is colored red and delineated inside the cell nuclei as expected. However, in the cyan -colored ESHB channel, the fluorescence is not only observed in cell cytosol but also in the nuclei ( Figure III-29a). In fact, the latter signal is colocalized with signals from the PSB emission channel, indicating that it originates from the ESHB emis sion of FR-1V/psST61 . Although this ÒleakyÓ fluorescence of FR-1V/psST61 is observed in the ESHB channel, it should be noted that the ESHB emission from the two constructs are clearly delineated rather than mingled, as well as in the reference EGFP channel . As a proof of concept, this imaging assay successfully demonstrate d the feasibility to label two a) b) Figure 0-28 Binding kinetics of FR-1V/psST61 . a) Binding of FR-1V (10 !M) and psST61 (10 !M) monitored at the 630 nm fluorescence maximum. The intensity curve is fit with a single exponential function. b) Second -order rate constant m easurement. Variable concentrations of FR-1V were used to bind 100 nM of psST61 . Observable rate constants were derived from single exponential fits. The second -order rate constant was then derived from the linear regression of observable pseudo first orde r rate constants as a function of concentration. All measurements were taken in PBS buffer (pH = 7.3). 225 sub -cellular compartments with FR-1V and different hC RBPII tags . Clearly resolved fluorescence signals have been acquired (Figure III-29d). The high intensity of the ÒleakyÓ ESHB emission of FR-1V/psST61 is somewhat unexpected since it does not show this level of ESHB in vitro . This might be caused by the pH difference or expr ession variability between E. coli and mammalian cancer cells. Nonetheless, the crystal structure of FR-1V/psST61 provides some understanding of the unwanted spectral behavior. It shows two orientations of the key K40H mutation with a 50:50 ratio ( Figure III-30). The population with a short distance (2.9 †) between the a!b!c!d!Figure 0-29 Dual -color live cell imaging with FR-1V in HeLa cells co-transfected with pFlag -CMV2 vectors encoding EGFP -psES235 -NES and EGFP -psST61 -3xNLS. a) ESHB channel: "ex = 405 nm, LP615. b) PSB channel: "ex = 514 nm, LP650. c) Reference EGFP channel: "ex = 488 nm, BP505 -530. d) Merged image of a) and b), showing well -defined edges of cell nuclei and surrounding cytosol. Cells were stained with 500 nM of FR-1V and incubated at 37 ¡C for 1 min, then washed with DPBS for 3 times before imaging. Scale bars, 20 !%&, 226 imidazole and the imine nitrogen atoms is susceptible to ESHB. Therefore, a logical approach to suppress this unwanted ESHB is to force His40 to adopt the flipped conformation . As shown in Figure III-31, Phe 130 and Gln128 are in close distance to the cyan conformer of His40. These two positions can be mutated to polar or anionic residues to facilitate H -bonding with His40 and lock its geometry. On the other side, I le42 is close Figure 0-31 Proposed key re sidues for mutagenesis to suppress K40H ESHB. Distances are measured from the FR-1V/psST61 crystal structure. Figure 0-30 Crystal structure of FR-1V/psST61 complex, showing two trajectories of the 40H residue in a 50:50 population ratio. The closer conformer has a 2.9 † distance between the histidine and the imine nitrogen atom. Protein crystallized by Dr. Alireza Ghanbarpou r. 227 to the green conformer of His40 . It can be mutated to bulky aromatic residues so that the green His40 can be pushed away to the cyan pose. Mutagenesis work is currently in progress and followed up by Rahele Esmatpoursalmani . I.18!Conclusion and significance A donor -acceptor ICT -type fluorenyl imine is identified as a super photobase in protic solvents, featuring a rare and strong dual -emission phenomenon origina ting from the intermolecular excited state proton transfer process between the solute and surrounding solvent molecules. Steady -state and ultraf ast spectroscop ic techniques show a partially transferred H -bonding complex formation on a picosecond timescale as an intermediate step during the ESPT process. Investigations with other carbon and nitrogen based solvents suggest a ground -state H -bonding complex as the prerequisite. A more general excited state hydrogen bonding process is realized in dye/hCRBPII complexes, featuring large Stokes shifts and high fluorescence quantum yields. The optimal complex is engineered into a fast -forming tag for live c ell imaging. The emitting process can be contemplated as the counterpart of the ESPT pathway in GFP -like FPs such that the ESHB is initiated by a photobase instead of a photoacid. A variant of the ESHB mutants is identified to feature a stable PSB absorption and emission. Combining the ESHB and PSB mutants, a proof -of-concept dual -color imaging system is established with the single dye FR-1V and demonstrated feasible. Further engineering effort is in progress to optimize the dual -color labeling system. 228 I.19!Experimental section General procedures for protein mutagenesis, cell -based assays, and steady -state spectroscopi c measurement are as described in Section I.13. The transient absorption experiments were performed by Dr. Muath Nairat and are described briefly as followed. General procedure of transient absorption experiment The t ransient absorption setup consists of a regeneratively amplified Ti:Sapphire laser (Legend, Coherent, Santa Clara, CA) producing femtosecond pulses at a repetition rate of 1 kHz. The pulses were compressed to their transform -limit duration with a pulse sha per utilizing the Multiphoton Intrapulse Interference Phase Scan (MIIPS) approach. The output was centered at 800 nm with a duration of 40 fs. The pulses were split using a beam splitter into an arm that was used to generate the second harmonic signal cent ered at 400 nm using a &-BBO crystal which served as the pump. The probe arm was sent into an optical delay line and was focused on a 2 mm YAG crystal to generate a white light continuum that extends from 450 to 900 nm. A 680 nm short pass filter was used to filter the continuum probe pulses. The two arms were non -collinearly combined and focused using a 25.0 cm lens into a 1 mm quartz cuvette. The probe signal was collected using a compact CCD (QE65000, Ocean Optics). Temporal dispersion of the probe pulse was measured by cross correlation and the observed chirp was used to correct the transient absorption data during global analysis . Global analysis of the results was carried out using Glotaran software. The synthetic procedure for FR0-SB is detailed in the supporting information along with NMR data. 229 General procedure of pK a measurement by UV/Vis spectrophotometry It is feasible to use an acid with known pK a value to titrate the pK a of an unknown acid. This can be generalized to the base case as well. The general acid case is used to derive the method as followed. Considering two partially ionized acids in the same solution with a given volume of V, the equilibria of the two acids a re in the forms: where the footnotes of 1 and 2 denote the known acid and the unknown sample, and HA is the acid and A - is the conjugate base. K a is the ionization constant. Therefore, at any pH in the equilibria, it has the overall equilibrium: With the given V, it can be rewritten as: and for a given solution, it has: If define: HA1H++A1-Ka1HA2H++A2-Ka2Ka1Ka2=[A1!][A2!]"[HA2][HA1]!"pKa=pKa2#pKa1=log[A1#][HA1]$[HA2][A2#]%&'()*!pKa=pKa2"pKa1=lognA1"nHA1#nHA2nA2"$%&'()nAi=nAi!+nHAi(i=1,2)!i=nHAinHAi+nAi! 230 there is: For a binary acidic solution with a certain pH, there is always: that is, and can be rewritten as: where # is the extinction coefficient as a function of wavelength and A is the absorption as a function of wavelength that can be determined by spectroscopic measurements. Specifically, and can be determined with the pure solutions of the acids and their conjugate bases. And with known values, can be determined by multi -variant linear regression analysis. Similarly, if the binary solution is basified to completely form conjugate bases, there is: and the two-variant linear regression with the following model can simply give !pKa=log1"!1()#!2!1#1"!2()A!l=[A1!]""A1!!+[HA1]""HA1!+[A2!]""A2!!+[HA2]""HA2!Vl!"#$%&A!=nA1'("A1'!+nHA1("HA1!+nA2'("A2'!+nHA2("HA2!Vl!"#$%&A!=nA1'"A1(!+#1'nA1"HA1!("A1(!()+nA2'!A2("+#2'nA2!HA2"(!A2("()!Ai!"!HAi"nAi!iVl!"#$%&A!=nA1'"A1(!+nA2'"A2(!S=[i!!A!!f!Ai!!()]2nAi 231 Synthesis and characterization of FR0 -SB To a solution of FR0 (10.5 mg, 35.8 !mol) in ethanol (3.0 mL) was added n-butylamine (100 !L, 28.2 equiv.). The resulting solution w as stirred at room temperature for 4 h to complete the reaction. The solvent and the excess amount of n-butylamine was removed in vacuo to afford 12.5 mg light yellow solid (yield = quant.). 1H NMR (CDCl 3, 500 MHz), / (ppm): 8.30 (s, 1H), 7.83 (d, J = 1.3 Hz, 1H), 7.61 Ð 7.51 (m, 3H), 6.74 Ð 6.65 (m, 2H), 3.67 Ð 3.60 (m, 2H), 3.45 (q, J = 7.1 Hz, 4H), 1.72 (p, J = 7.2 Hz, 2H), 1.50 (s, 6H), 1.47 Ð 1.39 (m, 2H), 1.23 (t, J = 7.0 Hz, 6H), 0.98 (t, J = 7.4 Hz, 3H). 13C NMR (CDCl 3, 125 MHz), / (ppm): 161.52, 156.42, 153.14, 148.22, 142.73, 133.14, 128.59, 126.33, 121.49, 120.85, 118.09, 110.74, 105.49, 61.64, 46.70, 44.70, 33.22, 27.43, 20.52, 13.97, 12.64. ESI -MS: (calc.) (m/z) calcd for C 24H33N2 [M+H] + 349.2644, found 349.2643. 232 REFERENCE S 233 REFERENCES 1.!Reichardt, C. Solvatochromic Dyes as Solvent Polarity Indicators. Chem. Rev. 1994, 94, 2319-2358. 2. Weller, A. Quantitative Untersuchungen der Fluoreszenzumwandlung bei Naphtholen. Zeitschrift fr Elektrochemie, Berichte der Bunsengesellschaft fr physikalische Chemie 1952, 56, 662-668. 3. Fırster, T. Die pH oÀAbh−ngigkeit der Fluoreszenz von Naphthalinder ivaten. Zeitschrift f p›r Elektrochemie und angewandte physikalische Chemie 1950, 54, 531-535. 4. Pines, E. & Pines, D. in Ch7: Proton Dissociation and Solute -Solvent Interactions Following Electronic Excitation of Photoacids . Ultrafast Hydrogen Bonding Dyna mics and Proton Transfer Prosesses in the Condensed Phase . (eds. Elsaesser , T. & Bakker , H. J. ), 155-184 (Springer Netherlands ; Dordrecht; 2002). 5. Douhal, A., Lahmani, F. & Zewail, A.H. Proton -transfer reaction dynamics. Chem. Phys. 1996, 207, 477-498. 6. Arnaut, L.G. & Formosinho, S.J. Excited -state proton transfer reactions I. Fundamentals and intermolecular reactions. J. Photochem . Photobiol . A 1993, 75, 1-20. 7. Ireland, J.F. & Wyatt, P.A.H. in Acid -Base Properties of Electronically Excited States of Organic Molecules. Advances in Physical Organic Chemistry . Vol. 12 , (ed. Gold , V. ), 131-221 (Academic Press ; 1976). 8. Agmon, N. Elementary Steps in Excited -State Proton Transfer. J. Phys. Chem. A 2005, 109, 13-35. 9. Jankowski, A., Dobryszycki, P., Lipi $ ski, J. & Stefanowicz, P. Excited -State Proton Transfer in Hydroxynaphthaldehydes Covalently Bound to Proteins. J. Fluoresc. 1998, 8, 103-113. 10. Heo, W. et al. Coherent intermolecular proton transfer in the acid -base reaction of excited state pyranine. Phys. Chem. Chem. Phys. 2017, 19, 18243-18251. 11. Finkler, B. et al. Highly photostable "super" -photoacids for ultrasensitive fluorescence spectroscopy. Photochem. Photobiol. Sci. 2014, 13, 548-562. 234 12. Tolbert, L.M. & Solntsev, K.M. Excited -State Proton Transfer: + From Constrained Systems to ÒSuperÓ Photoacids to Superfast Proton Transfer. Acc. Chem. Res. 2002, 35, 19-27. 13. Clower, C., Solntsev, K.M., Kowalik, J., Tolbert, L.M. & Huppert, D. Photochemistry of ÒSuperÓ Photoacids. 3. Excited -State Proton Transfer from Perfluoroalkylsulfonyl -Substituted 2 -Naphthols. J. Phys. Chem. A 2002, 106, 3114-3122. 14. Ryan, E.T., Xiang, T., Johnston, K.P. & Fox, M.A. Absorption and Fluorescence Studies of Acridine in Subcritical an d Supercritical Water. J. Phys. Chem. A 1997, 101, 1827-1835. 15. Favaro, G., Mazzucato, U. & Masetti, F. Excited state reactivity of aza aromatics. I. Basicity of 3 -styrylpyridines in the first excited singlet state. J. Phys . Chem. 1973, 77, 601-604. 16. Yatsuhashi, T. & Inoue, H. Molecular Mechanism of Radiationless Deactivation of Aminoanthraquinones through Intermolecular Hydrogen -Bonding Interaction with Alcohols and Hydroperoxides. J. Phys. Chem. A 1997, 101, 8166-8173. 17. Driscoll, E.W. , Hunt, J.R. & Dawlaty, J.M. Photobasicity in Quinolines: Origin and Tunability via the SubstituentsÕ Hammett Parameters. J. Phys . Chem. Lett. 2016, 7, 2093-2099. 18. Dey, D., Sarangi, M.K., Ray, A., Bhattacharyya, D. & Maity, D.K. Excited state hydrogen b onding fluorescent probe: Role of structure and environment. J. Lumin . 2016, 173, 105-112. 19. Wu, Y. -S. et al. Water -Catalyzed Excited -State Proton -Transfer Reactions in 7 -Azaindole and Its Analogues. J. Phys. Chem. B 2015, 119, 2302-2309. 20. Shen, J. Y. et al. Probing water micro -solvation in proteins by water catalysed proton -transfer tautomerism. Nat. Commun . 2013, 4, 2611. 21. Vogt, B.S. & Schulman, S.G. Reversible proton transfer in photoexcited xanthone. Chem. Phys . Lett. 1983, 97, 450-453. 22. Akul ov, K. et al. Acid Effect on Photobase Properties of Curcumin. J. Phys. Chem. A 2014, 118, 2470-2479. 23. Ditkovich, J., Pines, D. & Pines, E. Controlling reactivity by remote protonation of a basic side group in a bifunctional photoacid. Phys. Chem. Chem. Phys. 2016, 18, 16106-16115. 235 24. Ditkovich, J., Mukra, T., Pines, D., Huppert, D. & Pines, E. Bifunctional Photoacids: Remote Protonation Affecting Chemical Reactivity. J. Phys. Chem. B 2015, 119, 2690-2701. 25. Aquino, A.J.A., Barbatti, M. & Lischka, H. Excited -State Properties and Envir onmental Effects for Protonated Schiff Bases: A Theoretical Study. ChemPhysChem 2006, 7, 2089-2096. 26. tt, A. et al. A Comprehensive Self -Consistent Spectrophotometric Acidity Scale of Neutral Br¿nsted Acids in Acetonitrile. J. Org. Chem. 2006, 71, 2829-2838. 27. Goodhue, L.D. & Hixon, R.M. Electron -Sharing Ability of Organic Radicals. VII. Dissociation Constants of a Series of Organic Acids and Amines in Water, Methanol and Ethanol. J. Am. Chem. Soc. 1934, 56, 1329-1333. 28. Kumpulainen, T., Lang, B., R osspeintner, A. & Vauthey, E. Ultrafast Elementary Photochemical Processes of Organic Molecules in Liquid Solution. Chem. Rev. 2017, 117, 10826-10939. 29. Simkovitch, R. et al. Comprehensive Study of Ultrafast Excited -State Proton Transfer in Water and D 2O Providing the Missing RO ÐáááH+ Ion-Pair Fingerprint. J. Phys. Chem. A 2014, 118, 4425-4443. 30. Snellenburg, J.J., Laptenok, S.P., Seger, R., Mullen, K.M. & van Stokkum, I.H.M. Glotaran: A Java -Based Graphical User Interface for the R Package TIMP. J. Sta t. Softw. 2012, 49, 1-22. 31. Horng, M.L., Gardecki, J.A., Papazyan, A. & Maroncelli, M. Subpicosecond Measurements of Polar Solvation Dynamics: Coumarin 153 Revisited. J. Phys. Chem. 1995, 99, 17311-17337. 32. Agmon, N., Pines, E. & Huppert, D. Geminate re combination in proton oÀtransfer reactions. II. Comparison of diffusional and kinetic schemes. J. Chem. Phys . 1988, 88, 5631-5638. 33. Pines, E. & Huppert, D. Observation of geminate recombination in excited state proton transfer. J. Chem. Phys . 1986, 84, 3576-3577. 34. Paul, B.K., Ganguly, A. & Guchhait, N. Quantum chemical exploration of the intramolecular hydrogen bond interaction in 2 -thiazol -2-yl-phenol and 2 -benzothiazol -2-yl-phenol in the context of excited -state intramolecular proton transfer: A focus on the covalency in hydrogen bond. Spectrochim . Acta , Part A 2014, 131, 72-81. 35. Zhao, G. J. & Han, K. L. Hydrogen Bonding in the Electronic Excited State. Acc. Chem. Res. 2012, 45, 404-413. 236 36. Han, K. L. & Zhao, G. J. (eds.) Hydrogen Bonding and Transfer in the Excited State . Vol. I & II, (John Wiley & Sons, Ltd. ; 2010). 37. Dziuba, D. et al. Rational Design of a Solvatochromic Fluorescent Uracil Analogue with a Dual -Band Ratiometric Response Based on 3 -Hydroxychromone. Chem. Eur. J. 2014, 20, 1998-2009. 38. Klymchenko, A.S. & Demchenko, A.P. Multiparametric probing of intermolecular interactions with fluorescent dye exhibiting excited state intramolecular proton transfer. Phys. Chem. Chem. Phys. 2003, 5, 461-468. 39. Sengupta, P.K. & Kasha, M. Excited state p roton -transfer spectroscopy of 3 -hydroxyflavone and quercetin. Chem. Phys . Lett. 1979, 68, 382-385. 40. Hubin, P.O., Laurent, A.D., Vercauteren, D.P. & Jacquemin, D. Investigation of ESIPT in a panel of chromophores presenting N -H[three dots, centered]N intramolecular hydrogen bonds. Phys. Chem. Chem. Phys. 2014, 16, 25288 -25295. 41. Fang, X. et al. Synthesized Blue Fluorescent Protein Analogue with Tunable Colors from Excited -State Intramolecular Proton Transfer through an N ÐHáááN Hydrogen Bond. J. Phys. C hem. Lett. 2014, 5, 92-98. 42. Zhao, J., Ji, S., Chen, Y., Guo, H. & Yang, P. Excited state intramolecular proton transfer (ESIPT): from principal photophysics to the development of new chromophores and applications in fluorescent molecular probes and lumi nescent materials. Phys. Chem. Chem. Phys. 2012, 14, 8803-8817. 43. Formosinho, S.J. & Arnaut, L.G. Excited -state proton transfer reactions II. Intramolecular reactions. J. Photochem . Photobiol ., A 1993, 75, 21-48. 44. BilokinÕ, M.D. et al. Dual -Fluorescen ce Probe of Environment Basicity (Hydrogen Bond Accepting Ability) Displaying no Sensitivity to Polarity. J. Fluoresc. 2008, 19, 545-553. 45. Wu, J., Liu, W., Ge, J., Zhang, H. & Wang, P. New sensing mechanisms for design of fluorescent chemosensors emergi ng in recent years. Chem. Soc. Rev. 2011, 40, 3483-3495. 46. Wu, Y. et al. Fluorescence Sensing of Anions Based on Inhibition of Excited -State Intramolecular Proton Transfer. J. Org. Chem. 2007, 72, 62-70. 47. Peng, X., Wu, Y., Fan, J., Tian, M. & Han, K. Colorimetric and Ratiometric Fluorescence Sensing of Fluoride: + Tuning Selectivity in Proton Transfer. J. Org. Chem. 2005, 70, 10524-10531. 237 48. Zhang, X., Guo, L., Wu, F. -Y. & Jiang, Y. -B. Development of Fluorescent Sensing of Anions under Excited -State Intermolecular Proton Transfer Signaling Mechanism. Org. Lett. 2003, 5, 2667-2670. 49. Choi, K. & Hamilton, A.D. A Dual Channel Fluorescence Chemo sensor for Anions Involving Intermolecular Excited State Proton Transfer. Angew. Chem. Int. Ed. 2001, 40, 3912-3915. 50. Park, S. et al. A White -Light-Emitting Molecule: Frustrated Energy Transfer between Constituent Emitting Centers. J. Am. Chem. Soc. 2009, 131, 14043 -14049. 51. Acharya, A. et al. Photoinduced Chemistry in Fluorescent Proteins: Curse or Blessing? Chem. Rev. 2017, 117, 758-795. 52. Baranov, M.S. et al. Conformationally Locked Chromophores as Models of Excited -State Proton Transfer in Fluore scent Proteins. J. Am. Chem. Soc. 2012, 134, 6025-6032. 53. Fron, E. et al. Ultrafast Excited -State Dynamics of the Photoswitchable Protein Dronpa. J. Am. Chem. Soc. 2007, 129, 4870-4871. 54. Berezin, M.Y., Kao, J. & Achilefu, S. pH -Dependent Optical Prope rties of Synthetic Fluorescent Imidazoles. Chem. Eur. J. 2009, 15, 3560-3566. 55. Piatkevich, K.D. et al. Monomeric red fluorescent proteins with a large Stokes shift. Proc. Natl. Acad. Sci. U. S. A. 2010, 107, 5369-5374. 56. Piatkevich, K.D., Malashkevich , V.N., Almo, S.C. & Verkhusha, V.V. Engineering ESPT Pathways Based on Structural Analysis of LSSmKate Red Fluorescent Proteins with Large Stokes Shift. J. Am. Chem. Soc. 2010, 132, 10762-10770. 57. Wenge, U. & Wagenknecht, H. A. Synthetic GFP Chromophore and Control of Excited -State Proton Transfer in DNA: An Alternative Concept for Fluorescent DNA Labels with Large Apparent Stokes' Shifts. Synthesis 2011, 502-508. 58. Mancini, D.T., Sen, K., Barbatti, M., Thiel, W. & Ramalho, T.C. Excited -State Proton Tr ansfer Can Tune the Color of Protein Fluorescent Markers. ChemPhysChem 2015, 16, 3444-3449. 59. Santos, Elizabeth M. in PhD thesis. Michigan State University 2017. 238 MINIMIZING IMAGING B ACKGROUND OF DYE -HCRBPII TAGS VIA PHOTOINDUCED ELECTRO N TRANSFER QUENCHING . In the previous two chapters , we have shown two different strategies to diminish the nonspecific fluorescence background, namely, ICT -promoted NIR emission with photoisomerization and ultrafast binding featured in ESHB mutants . The former requires specific mutants to keep the FR-1V/hCRBPII complexes in the imine state to prevent NIR emission from nonspecific ally formed iminium s. The latter takes advantage of the unique sub -second binding kinetic s facilitated by spec ial ESHB mutants. Both systems rely on specific photochemical transformations of their excited state s to mitigate nonspecific background . On the other hand, a broad range of dye/hCRBPII complexes will find use in: 1) utilizing the ICT character of FR-1V to make triple - or multi -color imaging tag (see Section I.17); 2) develop ing FR-1V/hCRBPII into a dual -channel ratiome tric pH probe (see Section I.12.2). However, the two applications cannot be realized with merely the two aforementioned photochemical systems, sinc e n one of the two strategies will be suitable for the removal of unwanted fluorescence in the two scenarios that require the incorporation of other dye/hCRBPII complexes . In this context, a more general approach is needed to control the nonspecific background while allowing a no-wash protocol . I.20!Current strategies to enable no -wash live cell imaging Fluorogenicity, i.e., the ability to keep the fluorophore in dark (or a spectrally resolved emission state other than that for imaging) before a specifically designed event to turn on the fluorescence, can be achieved by either physical or chemical mechanisms. 239 I.20.1!Fluorogenicity based on physical interaction with environments The photophysical approach usually relies on the enviro nmental change of the fluorophores to restore the fluorescence that is initially quenched by various non-chemical mechanis ms.1 As a mimicry of GFP fluorophore , which shows strong emission inside the restricted protein scaffold but completely non -emis sive in free environmen ts, 2 molecular rotors that become fluorescent as a result of the restriction of internal conversion (RIR) have been used in the design of fluorogens . The classical examples are DCVJ and its derivatives. These compounds show strong vi scosity -dependent fluorescence and have been used to visualize the fluid shear stress on the membrane of endothelial cell layer s.3 The most representative RIR dye s for no -wash imaging are the ligands of RNA aptamer , PYP- and FAP -tags (Section I.2.2.2).4-6 Example s of coupling cyanines as fluorescent antennae to malachite green of which the fluorescence can be activated by FAP has also been repo rted. 7, 8 Very recently, a new family of synthetic 1,3,4 -oxadiazole aryl fluorosulfate s has been designed to specifically bind to intraorganellar transthyretins and realize no-wash imaging in living cells and Caenorhabditis elegan s.9 Besides t he fluorescence restoration by RIR , dyes featuring fluorescence quenching by polarity -dependent TICT and PLICT , which is generally termed as ÒsolvatochromismÓ, have also been used to develop no -wash imaging tags. 10, 11 One example is the push -pull type fluorene derivatives with excitation/emission wavelengths more red -shifted than Laurdan. These dyes show largely increased fluorescence quantum yields after moving from hydrophilic to lipophilic environments, thus being used as imaging tags for lipid domains and plasma membrane s.12 The red -region 240 solvatochromic dye Nile Red has also been developed into fluorogenic tag s for plasma membrane imaging. It has negligible fluorescence in polar aqueous environments but is highly fluorescent after being located to apolar lipid membranes by either SNAP tagging or ligand -receptor bi nding.13, 14 Applic ation of other know n solvatochromic dyes with the same strategy are also widely repor ted.1, 15 Strictly s peaking, this approach is not truly fluorogenic because the se fluorophores are not completely dark in polar environments . Another non -chemical strategy for no -wash labeling is to make use of aggregation -caused quenching (ACQ). Basically, the fluo rophores can stack into H -aggregates or self -assemble to nanoparticl es at high concentration and quench fluor escence . Yoshii et al . reported that by tethering TMR and an eDHFR inhibitor with a 15 -carbon alkyl phenylalanine sulfonamide linker , the rhodamine -based tag can self -template into nanoparticles and have a dynamic asse mbly -disassembly equilibrium inside living cells. When disassembled, this tag can bind to the dihyrofolate reductase and emit green fluorescence originating from TMR.16 Sim ilar strategy has been applied to a Cy5 derivativ e.17 Ajay aghoshÕs group reported an asymmetric NIR squarain e dye assembly that does not need a binding ligand . In in vitro studies, the NIR fluorescence is quenched due to the self -assembly and consequent ACQ. After disassembly, t his squaraine dye can undergo a Michael addition with cysteine -rich proteins to trigger a green fluorescen ce. 18 Further progress was reported by Klymchenko and co -workers. A 3,3 -dimethyl indoline -based symmetric squaraine was synthesized and made into a series of dimers wit h PEG linkers of different lengths that form H -aggregates in aqueous surroundings and have their fluorescence quenched . The PEGylated squaraine dimer s 241 were tethered to carbetocin and able to bind to the oxytocin receptor overexpressed on the su rface of HEK cells. After binding, the dimers were inserted into lipid layers and subsequently de -aggregate to turn on squaraine fluorescen ce. 19 Alt ernatively, aggregation -induced emission (AIE) has been used for no -wash imaging. The non -fluorescent AIEgen TPE has bee n tethered to target -recognition ligands or bioorthogonal click moieties and delivered to mammalian cell membrane or inside bacteria. The AIEgen is brought to close vicinity by elevated local concentration and hence emits l ight.20, 21 An interesting type of AIEgen s is reported by LiuÕs group . In their design, ESIPT has been incorporated with A IE. 22 An acetyl group can be hydrolyzed by esterase to reveal a phenolic hydroxyl group that facilitates ESIPT to the imine of an AIE moiety at the ortho positio n, turning on bright and red -shifted emission. Interestingly, only the diethyl amino variant demonstrates this AIE effect because steric hindrance of the ethyl groups prevents the close stacking of the molecules that can cause (-( stacking. Accordingly, dimethyl amino variant lost this AI E.22 I.20.2!Fluorogenicity based on chemical modification of fluorophores The fluorescence restoration mechanisms described in Section I.20.1 do not involve any chemical transformation of the fluorophores . Chemical modifications of these fluorogens are solely for targeting purpose or improvement of physical properties. However, as discussed, these approaches usually cannot afford complete fluorescence quenching of the fluorogens because of the inability of a clearly differentiated local environment in living organisms. In this context, various chemical modifications are 242 introduced onto the fluorophore scaffolds so that the fluorescence can be quenched and reactivated in a controllable manner. Often t hese mod ifications incorporate different photophysical mechanisms to suppress the fluorophoreÕs emission, such as TICT for biarsenical probes like FlAsH and ReAsH. 23 A re cent report by Wang et al. also incorporated RIR with this trivalent arsenical EDT moiety, sho wing that the emission of the D -A-D type benzopyrylium dye could be restored due to the high viscosity elicited in the local environment after binding of the arsenic with a vicinal dithiol peptide (VD P).24 Spirolactonization is another commonly adopted che mical transformation on xanthene -type dyes t o afford switchable fluorophore. An ortho -carboxyl substituted on the phenyl ring at the meso position cyclizes with the meso carbon under apolar conditions and forms a non -emissive spirolactone. In polar or low pH environment, the spirolactonization is reversed and a fully conjugated xanthene core is restored , bringing back the emission. This strategy has been applied to fluorescein, rhodamine, and silicon -rhodamine derivativ es. 25, 26 No matter how intricate the fluorogens are designed, targeting specificity is always desired but difficult . Aforementioned mechanisms incorporated for fluorogenicity are not for this purpose. I n this context, fluorophores are now commonly couple d to target -recognition ligands that can quench the fluorescence and undergo designed reactions to restore the ir emission properties. These fluorogens are termed quenched activity -based probes (qA BPs).27, 28 The strategies being used to stop fluorescence quenching are normally enzyme -based reaction s. For example, acetyl ester groups have been installed on the position 2 and 7 of fluorescein or the meso position of BODIPYs to invert the 243 electron -donating ability and stop emission. An esterase is able to hydrolyze the es ters and correct the electron flow to restore emissio n.29 A similar approach is reported to amidate rhodamine and rhodol amine groups with amino acids and later on to activate the fluorescence by carboxypeptidas es. 30 Alt hough utilizing an enzymatic reactio n to deprotect the enzyme substrate from the fluorophore scaffold is a feasible way, this approach limits the choice of dye scaffolds since a chemical modifiable functional group is necessary with the requirement that its presence can quench fluorescence. Therefore , a more general strategy to develop fluorogens is to append a quencher group on to the fluorophore. Consequently, a variety of quenching mechanisms have been applied to design the quencher part. FRET has been applied in the early designs as quencher groups. In many qABPs, a second dye or a black -hole quencher that has spectral overlap with the observing dye is tethered with the reporter by a lengthy linker , which also contains an enzyme reactive motif. Numerous FRET -based fluorogens have been published to probe the activities of different enzym es. 31 For a broad application scope, SNAP -tag has been developed into FRET -based qAB Ps.32 Recently, AIEgen has also been successfully incorporated to FRET -qABPs to image c aspase activit y.33 FRET is a long -range energy transfer process . From a fundamental point of view, Fırster resonance energy transfer is facilitated by dipole -dipole interaction, or a Coulombic interaction. Basically, the excited donor *D with an excited el ectron can undergo periodic harmonic oscillations along the molecular framework, creating an oscillating electric dipole. In turn it produces an oscillating electric field in the space 244 around *D, which in classical theory can drive a nearby acceptor A into resonance with matched oscillating frequency in its natural excited state. This frequency matching prerequisite is described by spectral overlap and sufficient oscillating strength. Because the energy transfer efficiency decrease is inversely proportional to the sixth power of the distance between donor (D) and acceptor (A), it usually requires th is distance to be within 1-10 nm, which makes the linker lengthy. Besides, the requirement for spectral -overlapped donor -acceptor pair also limits the choice of r eporters. On the other hand, in contrast to the long -range Fırster -type energy transfer, Dexter energy transfer (DET) features a short -range energy transfer process between donor and acceptor. Different from the transition allowed Fırster process, for forbidden transition the Co ulombic interaction is negligible and energy transfer is due to the electron Thisjournalis ©TheRoyalSocietyofChemistry2016 J.Mater.Chem.C, 2016,4,2731--2743| 2733The1ETstateisassociatedwithapairoffrontierorbitals, i.e.theHOMOandLUMOwhereasingleelectronistransferred fromtheHOMOtotheLUMO.AsforDÐAsystems,anelectron istransferredfromthedonororbital(HOMO)totheacceptor orbital(LUMO)uponphotoexcitationresultinginabiradicaloid pair(Fig.2a).Frontierorbitalinteractionincreasestheexcita- tionenergyrequiredtoreachthe 1ETstateandconsequentlya perpendicularconformationminimisestheexcitationenergy. Ataperpendicularconformation,therelativeenergylevelofthe 1ETstatecanbeapproximatedbysubtractingtheelectron Fig.1 Jab!on«skidiagramsofvariousenergy/electrondonorÐacceptor(DÐA)systems.(a)Fo ¬rsterResonanceEnergyTransfer(FRET). y:anglebetween vectorsofthedonoremissionandtheacceptorabsorption;(b)DexterEnergyTransfer(DET);(c)TwistedIntramolecularChargeTransfer(TICT) dynamics.11UponexcitationfromtheGS,theLEstateequilibratesrapidlywiththeTICTstateafterfastchargetransfer.GS=groundstate;GS D=ground statedonor;GS A=groundstateacceptor;ES D=excitedsingletstatedonor;ES A=excitedsingletstateacceptor;LE=locallyexcitedstate;R=effective DÐAdistance. Fig.2 (a)Preferredgeometriesof 1LEand 1ETstates.Schematicenergydiagramsofthe 1LE(blue), 1ET(green)and 1CT(red)stateswhen:(b) 1LEand 1ETcharacterarecomparabletoeachother,(c)whenthestericrestriction(black)isintroducedtotwisttheDÐAjunction( e.g.analkylgroupat ortho-position)and(d)whentheDÐAjunctionismadecoplanar( e.g.acarbonbridge). JournalofMaterialsChemistryC ReviewOpen Access Article. Published on 11 February 2016. Downloaded on 09/02/2018 21:37:46. This article is licensed under a Creative Commons Attribution-NonCommercial 3.0 Unported Licence.View Article OnlineFigure 0-1 Jablo $ ski diagrams of various energy/electron donor -acceptor (D -A) systems. a) Fırster Resonance Energy Transfer (FRET). ': angle between donor emission and acceptor absorption. b) Dexter Energy Transfer (DET). Artwork copied from ref [10]. 245 energy transfer through orbital wavefunction overlap, of which the rate decreases exponentially to the distance between donor and acceptor. Thus, t he effective energy transfer dis tance is smaller than 1 nm, making the design of the fluorogen more compact (Figure IV-1). Weissleder and co -workers have attached a clickable tetrazine to coumarins for standard bioorthogonal trans -cyclooctene labeling. The emission of coumarins is quenched by this tetrazine moiety and can be restored after click reaction. Mechanistically, t he fluorescence intensity of these native fluorogens is independent of solvent polarity, indicating the pivotal role of Dexter -type through -bond energy transfer (TBE T). 34 It should be noted that, both Fırster and Dexter theories predict a direct dependence on spectral overlap integral and require orbital orientation. However, in th e recent decade, modifications of Dexter theory for short -range intramolecular electron energy transfer have been derived and manifested experimentally. These include orientational -dependent exchange interaction, super exchange, and more generally through -bond interaction (a.k.a. TBET), where energy transfer happens with D and A separation much larger than the sum of their van der Waals radii while selection rule exclude the possibility of dipole -dipole interaction. This type of intramolecular electron energy transfer is especially efficient in rigidly linked Donor -bridge -Acceptor molecules whereas the D -A distance is greater than 1 nm for dipole -dipole forbidden processes. Detailed discussion and description of these physical processes is beyon d the scope of this research. Interested readers are referred to SpeiserÕs review article and TurroÕs book of photochemistry. 35, 36 Pioneered by Kevin BurgessÕ research group, this type of through 246 bond interaction has been utilized to develop various throu gh bond energy transfer cassette for fluorescent molecules with large pseudo -Stokes shift, as well as different analytes sensing system s.37 This TBET process has also been applied as quenching mechanism and expanded to tetrazoles and different dye scaffold s, including BODIPY, Acedan and xanthe nes. 38, 39 Different from excited state energy transfer, static quenching involves ground -state complexation of the emitter and an absorber, which is usually substituted aryl s or amines. This particular quenching mecha nism has been applied to the development of second and third generation no -wash PYP -tags. The PYP ligand -tethered emitter can form a ground -state complex with the cinnamic acid ligand or a nitrobenzene moiety to have its emission diminished. After binding with PYP -tag, the complex is dissociated, and the emission is restor ed.40, 41 A similar design was reported for a Golgi -targeting Òon -offÓ COX -2 probe based on folding and unfolding between ANQ and indomethacin .42 Without either the need for spectral or wave function overlap, or ground state complexation, photoinduced electron transfer (PeT) is the most widely adopted quenching mechanism for the design of fluoroge ns. 43-45 Bas ically, the emitter is tethered to a quencher moiety via a covalent and non -conjugated spa cer. An enzymatic ligand can be incorporated to the spacer so that the quencher is removed after enzymatic digestion, hence the emission is restored. The quencher can also be designed as part of the recognition unit of the peptidic or protei c labeling tag. Upon binding, a specifically placed nucleophile inside the peptidic/proteic tag can react with the quencher and stop the quenching process (for example see Figure I-19). 247 There are two related mechanisms of PeT . The first one involves the participation of an empty low -lying orbital of the quencher (Figure IV-2a). Briefly, after the HOMO electron of the fluorophore is excited into the corresponding LUMO, instead of the radiative decay back to the HOMO that gives fluorescence, the excited electron is transferred to the empty orbital of the quencher moiety, thereby leading to the fluorescence quenching. The electron ca n later on be transferred back to the HOMO of fluorophore following a nonradiative path to restore the stable ground state. Alternatively, a high -lying filled quencher orbital can also participate in a PeT mechanism. As shown in Figure IV-2b, the filled high -lying quencher orbital can transfer one electron to the excited HOMO of fluorophore, therefore stopping the radiative decay pathway. The excited electron in the LUMO of the fluorophore can transfer back to the emptied quencher orbital nonradiatively to form the stable ground state. HOMO LUMO HOMO LUMO HOMO LUMO HOMO LUMO HOMO LUMO HOMO LUMO h!h!flurophore quencherflurophore quencherflurophore quencherflurophore quenchera) b) Figure 0-2 PeT mechanisms. a) Participation of fluorophore HOMO -LUMO and an empty quencher orbital. b) Participation of fluorophore HOMO -LUMO and a filled quencher orbital. 248 Because the fluorophore and the quencher are electronically coupled to each other weakly, PeT provides freedom to select the proper fluorophore and quencher pairs. During recent decades, numerous reports of utilizing PeT have emerged for the development of fluorescent Òon -offÓ tags a nd pro bes. 46 PeT has also been incorporated with other photophysical/photochemical mechanisms to develop probes, such as ESIPT, spirocyclization, and ICT, et c.47-49 There is a wide scope of PeT quencher part s, depending on the match of redox potential with the fluorophores. The common ones include nitrophenyl, trifluoroacyl, anisol e,50 aryl aldehyde s,51 aryl amine s,52 quinone,53 indanedione,54 and maleimi de,55 piperidi ne,56-58 to name a few. Even --Gal has been used as a PeT quencher for fluorescei n.59 From a publication view, t he most important and prevailing quencher is the maleimide. Besides, t he maleimide group has been widely used in bioconjugation reactions of various functional molecules with targeting peptides or proteins via a thi ol-initiated Michael reaction. An impressive design is reported by KeillorÕ s group, where a coumarin is attached with two substituted maleimide quenchers to yield PeT fluorescence quenching. Taking a strategy similar to biarsenical FlAsH , the two maleimide s bind with the adjacent cysteines of the VDP and have their LUMO elevated, capable of stopping PeT quenching and regenerating coumarin fluorescence (see Section I.3.4 and Scheme I-8c).60 MaleimideÕs quenching mechanism has been substantially investigat ed.61, 62 Two maleimide moieties were attached on a dansyl fluorophore. A laser flash photolysis study showed that there was no triplet -state deactivation of the excited dansyl, excluding the possibility of intersystem crossing (ISC). A low -temperature fluorescence intensity study 249 at 77K further excluded the possible quenching by internal conversion (IC). Both steady -state a nd time -resolved Stern -Volmer analysis proved the dynamic quenching nature of this ma leimide quenching and validated its electron -transfer nature. A further validation was performed by the quenching efficiency change of meso -aryl substituted tetramethyl BODIPYs with maleimide installed on the ortho -, meta -, and para -positions of the aryl group. A s expected, a distance -dependent fluorescence quenching of the BODIPY was observed, with the ortho -substitution as the most quenched variant, further manifesting the electron -transfer nature of the maleimide quenchi ng.63 I.21!Designing a PeT -quencher tethered FR0 as a dark stain Since FR-1V demonstrates outstanding properties as a fluorescent lead in many aspects, a maleimide PeT quencher tethered with FR-1V is directly proposed following the discussion in Section 0 as a starting point (Figure IV-3). As shown, a maleimide is covalently tethered to the donor amine group of FR-1V through a methylene unit. Hence, the PeT quencher part is kept close to the fluorophore and has the flexibilit y to facilitate the desired energy transfer process. A DFT calculation was performed with Q1FR-1V in vacuum with the basis set 6 -31+G to roughly estimate the feasibility of PeT quenching with this scaffold . A further TD -DFT calculation will be necessary to more accurately gauge the orbital distribution in different solvents. The n-butylimine and n-butyliminium derivatives of Q1FR-1V were also NNOOOQ1FR-1V Figure 0-3 Structure of model PeT -quenched FR fluorophore. 250 calculated for their energy diagrams to check the feasibility of PeT quenching. The results are plotted in Figure IV-4. For the free aldehyde Q1FR-1V, the HOMO and LUMO+1 orbitals reside on the fluorophore framework while the LUMO localizes on the maleimide moiety. This indicates the potential donor -type PeT mechanism ( Figure IV-2a) as the HOMO electron will populate the LUMO+1 orbital after local excitation and subsequently relaxes into the LUMO on the maleimide. DFT calculation of the n-butylimine analog (denoted as Q1FR-1V_SB ) showed that the relative positions of the HOMO, LUMO, and LUMO+1 remain the same. This is exciting because it indicates the feasibility of PeT !"#$%&' !(#(%&' !)#)%&' !)#(%&' !"#(%&' !*#+%&' !,#+%&' !)#,%&' !)#-%&' ./0/ 120/ 120/3* !"#$ %"&!"#$ %"&'() !"#$ %"&'*() Figure 0-4 Orbital diagram of Q1FR -1V and the corresponding n-butylimine ( Q1FR -1V_SB ), n-butyliminium ( Q1FR -1V_PSB ), calculated with DFT/B3LYP/6 -31+G in vacuum. The orbital views on the left side are of Q1FR -1V. The color code of the right -side energy diagram is: blue (HOMO of FR-1V/SB/PSB ), green (LUMO of maleimide), and red (LUMO of FR-1V/SB/PSB ). 251 quenching of any protein -bound fluorophore in the SB form, regardless of specific or non-specific binding, if the environment al effect s are ignored. Therefore, this simple maleimi de quencher can potentially eliminate the non -specific fluorescence background from both the residing free aldehyde after staining and the non -specific imines formed with other proteins in vivo . The energy diagram of the n-butyliminium analog (denoted as Q1FR-1V_PSB ) was further calculated to examine the PeT quenching ability of the NIR emission of FR-1V if any non -specific PSB were to be formed. The DFT results show that under protonation the LUMO and LUMO+1 change order , therefore none of th e donor -PeT or the acceptor -PeT would be operational . !"#" $%#" $%#"&' ()*+,-. ('*/,-. (0*+,-. (0*),-. ()*1,-. ('*/,-. (0*',-. (+*/,-. ('*2,-. NOONOOClNOOMeO !"#$%& '"($)* !"!"+, !"-./ Figure 0-5 Orbital diagram of the n-butylimine of FR-1V tethered with maleimides bearing different substituents. Left -side orbital views are shown with non -substituted maleimide. The color code on the right -side diagram is: blue (HOMO of FR-1V_SB ), green (LUMO of maleimides), and red (LUMO of FR-1V_SB ). 252 The quenched fluorophore Q1FR-1V is supposed to react with a specific ally engineered cysteine residue inside the binding cavity of hCRBPII to achieve fluorescence recovery , and hence labeling specificity. Non -substitu ted maleimide is quite reactive towards different nucleophiles in biological settings and lacks chemoselectivity. In view of this, substitution on the maleimide will be necessary to temper its reactivity. As such, chlorine - and methoxy -substituted maleimid es are calculated to roughly represent the EWG - and EDG -modified maleimides. Shown in Figure IV-5, there is no change in the relative order of LUMO and LUMO+1 of quenc her-tethered FR-1V n-butylimine. Further calculation was performed to screen a broader scope of substituents for one that !"#" $%#" $%#"&' NOONOOClNOOMeO !"!"#$ !"%&' NOOO2NNOONCNOOF3C!"()*+" ,-./ ()*+,-. (/*),-. (/*0,-. !"#0 !"#*1 !"0%2 ()*),-. (1*1,-. (/*2,-. ()*+,-. (/*),-. (/*',-. ()*3,-. (/*+,-. (/*),-. (/*+,-. (/*1,-. ()*3,-. (/*+,-. (/*3,-. ()*3,-. Figure 0-6 Orb ital diagram of the n-butyliminium of FR-1V tethered with maleimides bearing different substituents. Left -side orbital views are shown with non -substituted maleimide. The color code on the right -side diagram is: blue (HOMO of FR-1V_ PSB), green (LUMO of mal eimides), and red (LUMO of FR-1V_PSB ). 253 potentially quenches the PSB emission. As shown in Figure IV-6, when protonated as iminium, the order of LUMO orbitals of FR-1V_PSB and maleimides are inverted except for the maleimide substituted with strongly electro n withdrawing nitro group. In anothe r word, except for the nitro -substituted maleimide, all the other maleimides cannot quench the fluorescence of FR-1V_PSB through the PeT process. However, the nitro substituent on the maleimide will greatly increase its electrophilicity and thus decrease chemoselectivity. To test the feasibility of PeT quenching, Q1FR0 was synthesized to tether a non -substituted maleimide to the amine group of FR0 via a methylene unit (Scheme IV-1). DFT calculation showed that even with a higher HOMO of FR0 and its corresponding n-butylimine SB, the relative order of the maleimide HOMO did not change (data not shown). Scheme 0-1 Synthesis of Q1FR0 . NOOHONHOOCH2O, 37%NaOH aq.pH 5.1r.t., 30 min WS_VI_23, 79% PhMe, 3† MS Dean-Stark, reflux, 16 h WS_VI_8, 53% BrH2NBrHNNOOBrNNOORRIor R 2SO4R = Me, Et BrHNOBrHN68Ac2O, AcOH 60¡C, 2 hWS_VI_22, 83% 1) BH 3¥Me 2S, THF, reflux, 6 h 2) NaHCO 3 aq. 1 hWS_VI_28, quant. BrNNOODean-Stark, O/N WS_VI_30, 79% HNODean-Stark, O/N WS_VI_35, 81% 2) DMF, -78 ¡C, 1 h then r.t. 2 h WS_VI_34, 64% NNOOO1) t-BuLi, -78 ¡C, THF, 1 h Q1FR073747576777873,PhMe, reflux 73,PhMe, reflux 1) t-BuLi, -78 ¡C, THF, 1 h (68) 2) DMF, -78 ¡C, 1 h then r.t. 2 h 254 To synthesize Q1FR0, initially N-hydroxymethyl maleimide 73 was coupled with 68 through an SN2 reaction using a Dean-Stark apparatus. Nonetheless, the subsequent alkylation of t he secondary amine 74 was not successful with either methyl iodide, ethyl iodide, or dimethyl sulfate, only yielding over alkylated byproducts . Arylamine 68 was subjected to acylation and borane reduction to afford 76. The monoethyl amine was then coupled with 73 to install the N-methylmaleimide. After obtaining 77, a metal -halogen exchange and subsequent formylation were attempted to furnish Q1FR0, but met with no succes s. The maleimide group could not tolerate the strong alkali condition. Instead, 76 was first formylated , and then coupled with 73 to afford Q1FR0. Compare d to the original FR0, the absorption and emission spectra of Q1FR0 are slightly blue shifted as recorded in ethanol (Figure IV-7). The absorption maximum is at 375 nm, 20 nm blue-shifted as compared to FR0. The emission maximum is at 535 nm when excited at 375 nm, also 20 nm more blue . The extinction coefficient of Q1FR0 is 31,600 M-1"cm-1, which is smaller than the 43,000 M-1"cm-1 calculated for FR0.!Figure 0-7 Extinction coefficient, absorption and emission spectra of Q1FR0 in ethanol. 255 Shown in Figure IV-8, n-propane thiol wa s used as the surrogate of cysteine to react with Q1FR0 and check the propensity of fluorescence recovery after stopping the PeT quenching. Excess equivalent of n-propane t hiol wa s added to a solution of Q1FR0 in ethanol at ambient temperature. As expected, a more than 4 -fold increase in the fluorescence intensity wa s observed. The reaction with n-propane thiol wa s rapid and the aldehyde group remain ed intact . Next, the fluorescence recovery was recorded with two sequential addition s of n-butylamine and n-propane thiol. Shown in Figure IV-9a, Q1FR0 was first subjected to the Michael addition with n-propane thiol. After the maximal fluorescence recovery of the aldehyde was obtained, excess amount of n-butylamine was added to probe the fluorescing ability of Q1FR0 upon imine formation. A gradual formation of the n-butylimine was observed over 1 h and was similar to the binding mode of Q1FR0 with just n-butyl amine , indicating the least interference of the maleimide moiety on the imine condensation ( Figure IV-9b, c ). Not surprisingly, excitation at 357 nm resulted in the signatur e dual emission featuring the regular short -wavelength SB emission at 443 nm NNOOOSHEtOH, 25¡C NNOOOS(xs)Q1FR0Q1FR0_PSH Figure 0-8 Fluorescence recovery of Q1FR0 after addition of n-propane thiol (PSH). 256 and the long -wavelength ESPT emission at 597 nm (Figure IV-9d). However, when the Figure 0-9 Fluorescence recovery of Q1FR0 tested by sequential addition of n-butylamine (nBA) and n-propane thiol (PSH) in diff erent order. a) Reaction scheme of two sequential addition orders. b) Stacked traces of time -dependent absorption of Q1FR0 after addition of excess nBA. c) Stacked traces of time -dependent absorption of Q1FR0_PSH adduct after addition of excess nBA. b) and c) showed the same timescale of imine condensation of nBA with Q1FR0 and Q1FR0_PSH . d) Emission spectrum of Q1FR0 after the sequential addition of PSH then nBA. e) Emission spectrum of Q1FR0 after the sequential addition of nBA then PSH. d) and e) were re corded with the same concentrations of Q1FR0 , PSH, and nBA. a) b) c) d) e) NNOOOSHEtOH, 25¡C NNOOOS(xs)Q1FR0Q1FR0_PSH NH2EtOH, 25¡C NNOONSBu(xs)Q1FR0_nBA_PSH NNOONBuQ1FR0_nBANH2EtOH, 25¡C (xs)SHEtOH, 25¡C (xs) 257 addition order of n-propane thiol and n-butylamine was reversed (Figure IV-9a), the addition of n-propane thiol no longer elicited any fluorescence enhancement even after 40 min ( Figure IV-9e). The results suggested that when excess amount of n-butylamine was initially added, both the maleimide and the aldehyde were reacted. Because the two experiments were carried out with the same concentrations of Q1FR0, n-propane thiol, and n-butylamine, the ESPT emission intensity at 597 nm lead us to speculate that the maleimide reacted with n-butylamine . In Figure IV-9d, the fluorescence intensity was ca. 107 CPS and should be considered as the maximal intensity of the non-quenched imine fluorescence . While in Figure IV-9e, the corresponding fluorescence intensity was 3 times less . It is clear that n-propane thiol could not react with the maleimide, since n-butylamine had already reacted, and the Michael produ ct with n-butylamine could also partially stop the PeT quenching but not to the full extent. The reason for the poor resolution of the 443-nm and 597 -nm peaks in Figure IV-9d was not clear. Q1FR0 has low stability in protic solvents . Shown in Figure IV-10a, equal amount of Q1FR0 was dissolved in ethanol and DMSO and samples were kept EtOH O/N EtOH fresh DMSO fresh NNOOOQ1FR0HNNOOOHHNNOOO+78Figure 0-10 Low stability of Q1FR0 . a) Picture shows the fluorescence of Q1FR0 samples stored in ethanol at 4 ¡C overnight (O/N), freshly prepared in ethanol and DMSO. Samples are excited with UV handset. b) Possible degradation pathway. a) b) 258 under different conditions. When excited with a 365 -nm UV -light handset, the freshly -made samples showed dim green fluorescence (peak at 535 nm). However, an overnight sample stored in ethanol , even at 4 ¡C , led to significant increase in fluorescence inte nsity. A degradation pathway similar to Hofmann elimination is plausible ( Figure IV-10b). I.22!Modification of the maleimide quencher and in vitro , in vivo studies A secon d version of donor -PeT quenched fluorophore Q2mFR0 was synthesized. The maleimide moiety was substituted with a methoxy group to decrease the reactivity towards amines , and the methylene linker connecting the maleimide and FR0 was extended to an ethylene unit to prevent Hofmann -type elimination. The synthesis started with methoxybromination of maleimide, followed by alkaline elimination in methanol. Concomitantly, methoxide addition to the target molecule 80 happened in the strongly basic conditio n. Separate acidic elimination was carried out to re -introduce the double bond (Scheme IV-2). 3-Methoxy maleimide 80 was then chloroethylated on the nitrogen with 1 -bromo -2-chloroethane to give 81, which was then subjected to an SN2 reaction with 78 for obtaining Q2mFR0. However, use of various bases and solvents under different temperatures failed to deliver the product, returning the starting material 78 (Scheme IV-3). The m ore reactive N-bromoethy l maleimide 82 was prepared in a similar manner NHOOMeO NHOONHOOMeO MeO 1) Br 2, MeOH RT, 16 h 2) Na, MeOH RT, 20 h WS_VI_65, 95% TsOH, tolueneDean-Stark, 48 h WS_VI_67, 88% 7980Scheme 0-2 Synthesis of 3 -methoxy maleimide. 259 and was used to couple with 78. The N-bromoethyl maleimide 82 was also subjected to the Finkelstein halogen exchange reaction to yield N-iodomethyl maleimide 83. N-hydroxyethyl maleimide was prepared via the reaction of maleimide with ethylene oxide . Nonetheless, none of these functionalized N-ethyl maleimides found success in coupl ing with 78, results in degraded maleimides and recovered 78 (Scheme IV-4). To trouble -Scheme 0-3 First attempt to synthesize Q2mFR0 . 78NHOOMeO NOOMeO ClBrCl+K2CO3TBAB neat, 40 !WS_VI_81, 85% 8081NOQ2mFR0NOOMeO HNO+NOOMeO Cltemp. base, solvents 81Scheme 0-4 Attempts to synthesize Q2mFR0 with different functionalized N-ethyl maleimides. NHOOMeO NOOMeO BrBrBr+K2CO3TBAB neat, 40 !WS_VI_80, 75% NOOMeO BrNaI (5 equiv.) Acetone reflux, O/N WS_VI_86, 94% NOOMeO I78NOQ2mFR0NOOMeO HNO+NOOMeO X82, X = Br, 83, X = I, 84, X = OH 82828380+NOOMeO OHNHOOMeO OTHF40 !, O/NWS_VI_97, 30% 8084conditions 260 shoot the reactivity problem, precursors of 78, 76 and 68, were reacted with 83, respectively (Scheme IV-5). Surprisingly, even with the least sterically hindered 68 did not succumb to the reaction . Possibly, these functionalized N-ethyl maleimides are not stable enough to allow the nucleophilic a ttack of fluorenyl amines. Knowing this, the synthesis of Q2mFR0 was redesigned via the stepwise modifications of the fluorene part. Bromoethylation of 78 proceeded with low conversion rate and yielded with the alkyl bromide 85. The S N2 reaction of 85 with maleimide 80 furnished Q2mFR0 with a moderate yield ( Scheme IV-6). Compared to Q1FR0, Q2mFR0 has a slightly higher extinction coefficient (32,000 M-1"cm-1). The abso rption centers at 388 nm and is 13 nm red -shifted as compared to HNBr+NOOMeO I83degraded 83 and recovered 7676H2NBr+NOOMeO I83degraded 83 and recovered 6868Scheme 0-5 Testing S N2 reactions of maleimide 83. Scheme 0-6 Final synthesis of Q2mFR0 . K2CO3, DMF 80 !, O/NWS_VI_103, 58% b.r.s.m. NNOOMeO HNOBrBrNOO80 (1.1 equiv.) K2CO3 (1.5 equiv.) DMF, 80 !, O/N WS_VI_104, 58% Q2mFR07885Br 261 Q1FR0. Correspondingly, the emission maximum is at 543 nm and is 8 nm red -shifted (Figure IV-11a). When different amount s of n-propane thiol were added , no change was observed over time in the absorption spectra ( Figure IV-11b). Contrarily, the emission intensity d isplayed a significant increase after the addition of thiol, validating the feasibility of the same donor -PeT quenching ( Figure IV-11c). The fluorescence recovery rati o was quantified with two metrics. Shown in Figure IV-11d, the recovery ratio was measured in the absolute height of the emission maximum at 543 nm, giving a 9.3 -fold increase after thiol addition. Alternatively, a m ore accurate measurement was done with the integration a) b) c) d) e) Figure 0-11 a) Absorption and emission spectra of Q2mFR0 in ethanol. b) Stacked absorption traces of Q2mFR0 added with n-propane thiol (PSH) over time. c) Emission spectra of Q2mFR0 before and after addition of n-propane thiol. d) Fluorescence recovery measured peak intensity. e) Fluorescence recovery measured by whole spectra integration. 262 of overall fluorescence spectra, resulting in a 9.9 -fold increase after thiol addition ( Figure IV-11e). With eit her measurement, greater than 90% PeT quenching efficiency was validated. The chemoselectivity was again tested via different addition sequences of n-butylamine and n-propane thiol ( Scheme IV-7). The n-butylimine formation s of Q2mFR0 before and after reaction with n-propane thiol effectively have the same binding kinetics . The thiol adduct of Q2mFR0 did not change the absorption spectrum of the n-butylimine as evidenced by the same abso rption maximum ( Figure IV-12a, b). Two samples of Q2mFR0 in ethanol were used to react with n-butylamin e, or first with n-propane thiol then n-butylamine , respectively. Both exhibited the signature ESPT emission of FR0 at 620 nm along with the regular imine emission at 450 nm ( Figure IV-12c, d). It should be noted that equal amount and concentration of Q2mFR0 were used in the two sam ples, therefore a direct comparison of fluorescence intensity at the peak maxima should reveal Scheme 0-7 Chemoselectivity and fluorescence recovery of Q2mFR0 with sequential addition of n-butylamine (nBA) and n-propane thiol (PSH). a) PSH then nBA. b) nBA then PSH. SHEtOH, 25 ¡C (xs)NH2EtOH, 25 ¡C (xs)Q2mFR0_nBA_PSH NH2EtOH, 25 ¡C (xs)SHEtOH, 25 ¡C (xs)NNOOMeO OQ2mFR0NNOOSOQ2mFR0_PSH MeO NNOOMeO NQ2mFR0_nBABuNNOOSNMeO Bu 263 the PeT quenching efficiency for both imine and iminium fluorescence . With the addition of only n-butylamine, the 620-nm fluorescence intensit y was 1.9 million C PS. However, if the thiol adduct was first formed and then followed by addition of n-butylamine, the fluorescence intensity at 620 nm was over 7.6 million CPS , giving a 4 -fold recovery ratio. Hence, the PeT quenching efficiency of the imine and iminium is abou t 75%. The a) b) c) d) e) Figure 0-12 Chemoselectivity and fluorescence recovery of Q2mFR0 with sequential addition of n-butylamine (nBA) and n-propane thiol (PSH). a) Stacked absorption traces of Q2mFR0 with excess nBA over time. b) Stacke d absorption traces of Q2mFR0_PSH adduct with excess nBA over time. a) and b) were recorded in ethanol with same amount of Q2mFR0 , nBA, and PSH. c) Absorption and emission spectra of Q2mFR0_nBA . d) Absorption and emission spectra of the final solution of Q2mFR0_PSH added with nBA. c) and d) were recorded in ethanol with same starting concentration of Q2mFR0 . e) Emission spectra of Q2mFR0 in ethanol with sequential addition of nBA then PSH. 264 chemoselectivity of the aldehyde towards thiol over amine was tested by the sequential addition of n-butylamine then followed by addition of n-propane thiol. Evidenced in Figure IV-12e, a greater than 5 -fold increase in the fluorescence intensit y was observed after thiol addition. This also confirmed that the PeT quenching efficiency was over 80% since all the measurements were carried out in the same cuvette. Next, flexible docking of Q2mFR0 was used to propose positions inside hCRBPII for introduction of cysteines that can react with the 3 -methoxymaleimi de after imine formation. The top 9 poses are displayed in Figure IV-13. As shown, all residues within 5 † distance to the maleimide double bond are highlighted. Among these, L36, R58, S55, A33, and T53 were selected for mutation to cysteine. To demonstrate PeT quenching efficiency in vitro , the ESPT mutant psES235 was selected as the template for Figure 0-13 Flexible docking results of Q2mFR0 in hCRBPII KL mutant (PDB ID: 4EXZ). Residues within 5 † distance to the maleimide double bond carbon atoms are highlighted in cyan and stick form. Q2mFR0 is covalently attached to the Q108K residue. The top 9 poses are stacked in and displayed line form. 265 mutagenesis as it featured both long -wavelength emission and ultrafast imine formation kinetics. Before cysteine mutations on the listed positions, residues C95, C121, and C126 of psES235 were mutated to alanines to avoid any possible reactions with Q2mFR0 outside the binding cavity. Nonetheless , the resulting tripl e mutant and its derivatives (i.e., A33C, S55C, L36C, R58C, and Q108L) were not express ed solubl y. Hence, cysteine mutations were directly carried out on psES235 . Summarized in Table IV-1, A33C, S55C, and R58C mutations showed protein expressions with different ratio of hCRBPII monomer over domain -swapped dimer . T29C mutant of psES235 completely altered the expression pattern with only dimer harvested. For a consistent comparison of the spectral features with psES235 , only monomers were used in the following experiments. Table 0-1 Oligomeric nature of psES235 cysteine mutations. MUTANT Soluble expression monomer dimer psES235 Q108K:K40E:T53A:R58L:Q38F:Q4F 100%-psWS44 Q108K:K40E:T53A:R58L:Q38F:Q4F:C95A:C121A:C126A no expression psWS45 Q108K:K40E:T53A:R58L:Q38F:Q4F:C95A:C121A:C126A:A33C no expression psWS47 Q108K:K40E:T53A:R58L:Q38F:Q4F:C95A:C121A:C126A:S55C no expression psWS48 Q108K:K40E:T53A:R58L:Q38F:Q4F:C95A:C121A:C126A:L36C no expression psWS50 Q108K:K40E:T53A:R58C:Q38F:Q4F:C95A:C121A:C126A no expression psWS49 Q108L:K40E:T53A:R58L:Q38F:Q4F:C95A:C121A:C126A no expression psWS51 Q108K:K40E:T53A:R58L:Q38F:Q4F:A33C 68%32%psWS52 Q108K:K40E:T53A:R58L:Q38F:Q4F:S55C 54%46%psWS53 Q108K:K40E:T53A:R58C:Q38F:Q4F 91%9%psWS54 Q108K:K40E:T53A:R58L:Q38F:Q4F:L36C > 99% -psWS55 Q108K:K40E:T53A:R58L:Q38F:Q4F :T29C -> 95% Monomers and dimers were separated by size exclusion chromatography. The isolated yields of oligomers were determined by NanoDrop microvolume spectrophotometer with default protocol. The template mutant psES235 is shaded blue. Mutants that cannot express a re shaded grey. Soluble proteins are shaded green. 266 The imine formation of Q2mFR0 with psES235 and its expressible mutants (i.e., psWS54 and monomers of psWS51 , psWS52 , and psWS53 ) share the same observable pattern. The absorption and emission spectra of psWS54 are shown in Figure IV-14 as an example . It should be noted that the imine formation with psES235 , psWS51 , psWS52, and psWS53 is fast and takes less than 2 min. With psWS54 , the imine also forms quickly with ~ 70% conversion upon addition of Q2mFR0, although the maximal intensity at 374 nm is observed over 1.5 h. The excitation at 374 nm g ives an exclusive long-wavelength a) b) c) d) e) Figure 0-14 Absorption and emission spectra of Q2mFR0 and psES235 cysteine mutants. a) Stacked absorption traces of Q2mFR0 and psWS54 over time. b) Emission spectrum of Q2mFR0 /psWS54 with 374 nm excitation. Emission spectra with excitation at corresponding absorption maxima at different time points after Q2mFR0 addition to c) psWS51 monomer, d) psWS52 monomer, e) psWS53 monomer. 267 ESPT emission centered at 538 nm. In contrast, the complex of Q2mFR0/psES235 is essentially non -fluorescent. Shown in Figure IV-14c-d, different ratio of regular imine emission and ESPT emission is observed in different mutant complexes, with a more obvious imine emission presented in psWS53 . The fluorescence recovery ch ecked at different time points post Q2mFR0 addition shows a gradual increase , validating the effective inhibition of the maleimide PeT quenching by cysteine addition. A closer look at the fluorescence recovery kinetics i s provided in Figure IV-15. Q2mFR0/psES235 was used as the control of baseline fluorescence before recovery. As seen, its emission intensity was consi stently small over 4 h. Among the four cysteine mutants, psWS54 showed the fastest recovery kinetics and the highest recovery ratio. The bar graph p lotted in Figure IV-15b shows that even after 30 min, psWS54 can recover over 80% of the ESPT fluorescence. The pseudo -first order fluorescence recovery rates can be derived from Figure IV-15a by fitting the curves to an exponential function. As shown in Table IV-2, psWS54 also possesses the highest quantum yield. a) b) Figure 0-15 Fluorescence re covery of Q2mFR0 /psWS51 -54. a) Fluorescence recovery kinetics monitored at 540 nm. b) Fluorescence recovery ratio of different cysteine mutants over psES235 monitored at 540 nm at different time points. 268 High-resolution mass spectrometry (HRMS) was used to investigate the stoichiometry of the reaction s between Q2mFR0 and hCRBPII mutants. There are several possibilities after the addition of Q2mFR0 to a solution of hCRBPII mutants: i) non-covalent residence of Q2mFR0 in the binding cavity; ii) Michael addition inside the cavity without imine formation; iii) both the aldehyde and maleimide moieties react as desired; iv) multiple Michael additions with cysteines on the exterior of hCRBPII (i.e., C95, Table 0-2 QY and fluorescence recovery kinetics of psES325 and psWS51 -54. Fluorescence quantum yields are determined by absolute measurement. The emission recovery half -time is determined by fitting the fluorescence recovery kinetic curves to an exponential rise function. Mutant QYaFluorescence recovery half -time t 1/2 (min) bpsES235 0.05 -psWS51 0.40 88.7 psWS52 0.31 45.2 psWS53 0.49 154.5 psWS54 0.67 15.0 NNOOMeO OQ2mFR0HSCys -hCRBPII- Lys NH2NNOOMeO OShCRBPII- Lys NH2HSCys -hCRBPII- Lys NH2NNOOMeO NShCRBPII NaBH 3CN(500 nM) PBS, r.t., 3 h NNOOMeO NHShCRBPII Michael adduct (M.A.) Imine formation (I.F.) Reductive amination (R.A.) Scheme 0-8 Detectable adducts of Q2mFR0 /hCRBPII in HRMS. A sequential step of reductive amination after both imine formation and Michael addition was used to derivatize the complex into a more stable form for mass detection. 269 C121, and C126). There is no difference in HRMS for i) and ii), but one can tell from the aldehyde fluorescence recovery if the Michael addition happens or not. Scenario iii) and iv) can be differentiate by HRMS data and the spectral change in the ESPT emission as well. A brief description of the possibilities is presented in Scheme IV-8. The imine or iminium formed from Q2mFR0 and hCRBPII complexation was easily hydrolyzed under ESI analysis condition. Hence, reductive amination was carried out with NaCNBH 3 in aqueous solution to reduce the imine double bond and furnish more stable derivatives for mass detection ( Scheme IV-8). As shown in Table IV-3, no adduct m/z peak was observed with samples under native condition (indicated by I.F. , imine formation ). Instead, after reductive amination, correct m/z peaks were found with the correspo nding mutants. Importantly, there was no detection of any mutant adducted with Table 0-3 HRMS of Q2mFR0 /hCRBPII adducts. Apo protein [M]++ Q2mFR0 (free/M.A.) [M+1] ++ Q2mFR0 (I.F.) [M]++ Q2mFR0 (R.A.) [M+1] +psES235 calc. 15671.7 16089.9 16071.9 16073.9 exp. 15671.5 N.O. N.O. N.D. psWS51 calc. 15703.8 16123.0 16104.0 16107.0 exp. 15704.0 16123.0 N.O. 16107.0 psWS52 calc. 15687.8 16107.0 16088.0 16091.0 exp. N.D. N.D. N.D. N.D. psWS53 calc. 15661.7 16079.9 16061.9 16064.9 exp. 15662.0 16081.0 N.O. 16065.0 psWS54 calc. 15661.7 16079.9 16061.9 16064.9 exp. 15662.0 16081.0 N.O. N.D. Data collected with ESI + mode on a Waters Xevo G2 -XS QTOF mass spectrometer (Agilent). All samples were run through a BetaBasic column (CN, 10 # 1 mm, 5 !M) with acetonitrile and 0.1% formic acid in H 2O as the mobile phase. N.O.: not observed. N.D.: not determined. 270 multiple Q2mFR0, ensuring the 1:1 stoichiometry of the maleimide adduction. Further evidence of the regio -selectivity with the interior cysteine was provided with Q2mFR0/psES235 as no Michael adduct was observed under native condition. Based on these HRMS observations, we can conclude that as a proof of principle, Q2mFR0 successfully bound cys teine variants of psES235 (psWS51 to psWS54 ) via the Michael addition of the thiol group on to the maleimide ring , subsequently with the imine formation between Q108K and the aldehyde moiety. As validated with spectroscopic measurement, the unbound Q2mFR0 is in the ÒdarkÓ state. Its complex with the above EGFP BP505 -530psWS54 BP505 -575IR DIC+psWS54 15 min 30 min Figure 0-16 Confocal imaging of HeLa cells expressing pFlag -CMV2 construct: EGFP -psWS54 -3xNLS. Kalman averaging 4 was applied. The refere nce EGFP channel was excited with a 488 nm laser and collected with a BP505 -530 bandpass filter. The psWS54 channel was excited with a 405 nm laser and collected with a BP505 -575IR bandpass filter. Same excitation power, gain and offset levels were applied . Cells were stained with 2 !M Q2mFR0 and incubated at 37 ¡C for indicated time. No washing steps were applied before imaging. Scale bar: 20 !m. 271 psES235 variants successfully restore s ESPT emission of the resultant imines after quencher deactivation by the cysteines. These in vitro analyses have paved the way for testing PeT -based no -wash in cellulo imaging. psWS54 was cloned into the pFlag2 -CMV vector s encoding N -terminal EGFP as a reference reporter and C -terminal signaling peptides as described in 0. The nuclei -localizing construct was expressed in the living HeLa cells and a no -wash i maging protocol was tested. As shown in Figure IV-16, cells were stained with Q2mFR0 at a final concentration of 2 !M. Samples were directly subjected to imaging at different time duration. After 15 min of Q2mFR0 incubation, the fluorescence of the dye/hCRBPII complex was observed . A dim but obvious fluorescence background was presented in the non -transfected cells, as more clearly observed in the overlay of DIC and h CRBPII channels. This non -specific background became more profound with extended incubation time, and was directly observable in the hCRBPII channel. Consider ing the in vitro EGFP BP505 -530psWS54 BP530-600DIC+psWS54 Figure 0-17 Confocal imaging of HeLa cells expressing pFlag -CMV2 construct: EGFP -psWS54 -3xNLS. The reference EGFP channel was excited with a 488 nm laser and collected with a BP505 -530 bandpass filter. The psWS54 channel was excited with a 405 nm laser and collected with a BP530 -600 instead of a BP505 -575IR bandpass filter. Cells were stained with 2 !M Q2mFR0 at room temperature for 35 min. No washing steps were applied before imaging. Scale bar: 20 !m. 272 emission spectrum of Q2mFR0/psWS54 (Figure IV-14b), the emission dichroic was set to BP530 -600 to avoid the crosstalk of EGFP channel. The temperature for dye -staining was lowered to 23 ¡C to decrease non -specific binding of the maleimide quencher. Nevertheless, severe non -specific background was observed without washing ( Figure IV-17). The non -specicific fluorescence background caused by unwanted reactions of the maleimide quencher was further estimated with non -transfected HeLa cells . Cells were stained with Q2mFR0 and kept at room temperature. It was observed that with extended time of incubation, the non -specific background gradually increas ed (Figure IV-18). This non-specific background is probably caused by the restoration of quenched fluorescence due to side reactions of the maleimide with biological thiols, such as GSH . Different reductive sulfur species that are common in biological settings need to be examined along with Q2mFR0 in vitro to identify the origin of the background fluorescence. I.23!Conclusion and future directions A fluorogenic dye Q2mFR0 has been design ed based on the PeT quenching by a tethered substituted maleimide moiety for no -wash live cell imaging. The fluorescence 15 min 25 min 35 min DIC + BP505 -575IR Figure 0-18 Non -specific fluorescence background of Q2mFR0 at different time points. Q2mFR0 (2 !M) was used to stain HeLa cells at room temperature. The field of view was excited with a 405 nm laser and images were acquired with a BP 505-575IR bandpass filter. No washing steps were applied. Scale bar: 20 !m. 273 recovery is facilitated by the nucleophilic addition of the maleimide with specifically positioned cysteine resides inside the hCRBPII binding cavity. More importantly , the recovered fluorescence signal for microscopic imaging is the emission from the ESPT process. Hence, three triggering mechanisms have been incorporated to obtain fluorogenicity for this system, namely, i) donor -type PeT quenching, ii) imine formation of the active lysine and the fluorophoreÕs aldehyde, and subsequent iii) long -wavelength emission from the ESPT process. Modification of Q2mFR0 is needed to increase its fluorogenicity, i.e., the reaction specificity towards engineered cysteine of hCRBPII . It is hypothesized that non -specific reaction of the maleimide is caused by GSH. Therefore, two strategies are currently adapted for the optimization. The first involves the installment of two substituted maleimide moieties on the scaffold of FR0. Presuma bly, the first non -specific GSH adduct will sterically hinder further addition of GSH to the second maleimide, thus keeping the fluorophore quenched before its specific reaction inside hCRBPII cavity. In turn, two cysteines are needed to be engineered insi de the cavity to accommodate the two maleimides. The second stra tegy is to further increase the PeT quenching efficiency of the substituted maleimide while introducing more steric bulk to reduce its reactivity towards non-specific thiols. It is evident tha t the introduction of methoxy effectively decreases the electrophilicity of the maleimide and keeps the quenching at the same time. It can be reasoned that with different alkoxy substitutions, the reactivity and quenching efficiency of the maleimide will b e affected both electronically and sterically. A quick method for 274 screening the proper alkoxy substituent can be proposed ba sed on Stern -Volmer dynamic quenching analysis (Equation IV-1). Briefly, different alkoxy -substituted maleimides can be mixed with FR0 at various concentrations. For a given maleimide, the PeT quenching rate kq can simply be derived by plottin g the ratio of QY mixture over QY FR0l against the variable concentrations of the maleimide. The lead maleimide can then be tethered to 85 for in vivo testing. Apparently, the two st rategies can be combined. I.24!Experimental section Materials and general methods All chemicals and solvents were purchased from Sigma -Aldrich and used without further purification. NMR ( 1H, 13C) spectra were recorded with Agilent DirectDrive2 500 MHz spectrometer and referenced with the residual 1H peak from the deuterated solvents. HRMS (ESI) analysis was done by using a Waters Xevo G2 -XS QTOF mass spectrometer (Agilent) and referenced against Polyethylene Glycol (PEG -400-600). General protocols of site -directed mutagenesis, cloning, protein expression and purification, mammalian cell culture and transfection, UV/Vis and fluorescence spectroscopies, dye -hCRBPII binding kinetics, absolute quantu m yield measurement, 1!x=1!0+kq[Q]!e0!ex=1+kq!0[Q]ááááááááááááááááááááááááááááááááááá (1) ááááááááááááááááááááááááááááááááá (2) Equation 0-1 Stern -Volmer equations. .,is the fluorescence life -time. kq is dynamic quenching rate. [Q] is the concentration of maleimides. * is the fluorescence quantum yield. Footnote Ò0Ó denotes the solution with only FR0 . Footnote ÒxÓ denotes the mixed solutions. 275 and live cell confocal imaging can be referred to the description of general methods in Section I.13. N-methyloyl maleimide (7 3)64 To a suspension of maleimide (1.0 g, 0.010 mmol) in an aqueous solution of NaOH (pH 5.1, 1.0 mL ) was added 37% formalin (0.33 g, 0.011 mmol). The mixture was stirred at room temperature for 30 min. A white crystalline solid precipitated from the mixture. The solid was filtered and recrystallized from ethyl acetate to afford white crystal (1.03 g, 79% yield ). 1H NMR ( 500 MHz , DMSO -d6), / (ppm): 7.06 (s, 2H), 6.2 7 (t, J = 7.0 Hz, 1H), 4.76 (d, J = 7.0 Hz, 2H). 13C NMR (126 MHz, DMSO -d6) / (ppm): 171.05, 135.37, 60.06. ESI-MS: (calc.) (m/z) calcd for C5H5NO3 [M+ H]+ 128.03, found 128.10. 7-Bromo-N-maleimidomethyl -9,9-dimethyl -9H-fluoren -2-amine (74) To a solution of 68 (1.00 g, 3.46 mmol) in anhydrous toluene (30 mL) was added 73 (530 mg, 4.16 mmol) and 3 † MS. The reaction was performed in a Dean -Stark apparatus and was reflux ed overnight. After full conversion, the reaction mixture was concentrated in vacuo and the raw product was purified by gradient chromatography with 15 ~ 20% ethyl acetate in hexane to afford a light yellow solid (733 mg, 53% yield). 1H NMR (500 MHz, DMSO -d6) / (ppm): 7.64 (d, J = 1.9 Hz, 1H), 7.52 (dd, J1 = 8.2 Hz, J2 = 2.5 Hz, 2H), 7.39 (dd, J1 = 8.1 Hz, J2 = 1.9 Hz, 1H), 7.03 (s, 2H), 6.94 (d, J = 2.1 Hz, 1H), 6.79 (t, J = 7.1 Hz, 1H), 6.73 (dd, J1 = 8.3 Hz, J2 = 2.2 Hz, 1H), 4.86 (d, J = 7.0 276 Hz, 2H), 1.34 (s, 6H). 13C NMR (126 MHz, DMSO -d6) / (ppm): 171.73, 155.20, 155 .14, 146.98, 139.16, 135.71, 135.18, 130.09, 127.40, 126.11, 121.61, 120.72, 118.42, 112.41, 106.63, 46.98, 46.92, 27.32, 14.55. ESI-MS: (calc.) (m/z) calcd for C20H17BrN2O2 [M+H] + 397.05, found 397.13. N-(7-bromo-9,9-dimethyl -9H-fluoren -2-yl)acetamide (75) To a solution of 68 (300 mg, 1.04 mmol) in glacial acetic acid (3 mL) was added acetic anhydride (531 mg, 5.20 mmol). The solution was stirred at 60 ¡C for 2 h. The resulting mixture was then neutralized with aqueous sodium car bonate solution and extracted with ethyl acetate. The combined organic phase was concentrated in vacuo . The crude was purified by flash column with 30% ethyl acetate in hexane to afford a white solid (285 mg, 83% yield). 1H NMR (500 MHz, Chloroform -d) / (ppm): 7.73 (d, J = 2.0 Hz, 1H), 7.62 (d, J = 8.2 Hz, 1H), 7.56 Ð 7.50 (m, 2H), 7.44 (dd, J1 = 8.1 Hz, J2 = 1.9 Hz, 1H), 7.36 (dd, J1 = 8.1 Hz, J2 = 2.0 Hz, 1H), 2.22 (s, 3H), 1.47 (s, 6H). 13C NMR (126 MHz, Chloroform -d) / (ppm): 193.48, 155.62, 154.39, 137.77, 137.55, 134.35 , 130.08, 126.08, 120.98 , 120.46, 118.78, 114.40, 109.99, 47.25, 26.99, 24.82. ESI-MS: (calc.) (m/z) calcd for C17H16BrNO [M+H ]+ 330.05, found 330.09. 7-Bromo-N-ethyl -9,9-dimethyl -9H-fluoren -2-amine (76) To a solution of 75 (440 mg, 1.33 mmol) in anhydrous THF (25 mL) was added borane -dimethyl sulfide (152 mg, 190 !L, 2.00 mmol). The reaction mixture was refluxed 277 for 6 h. Saturated sodium bicarbonate (10 mL) was then added and the mixture was refluxed for another 1 h. The final mixture was diluted with water and extracted with ethyl acetate. The organic phase was combined and dried with anhydrous sodium sulfate. The solvent was removed under reduced pressure to afford a white solid (420 mg, yield = quant.) 1H NMR (500 MHz, Chloroform -d) / (ppm): 7.54 (d, J = 8.1 Hz, 1H), 7.49 (d, J = 1.7 Hz, 1H), 7.45 (d, J = 8.0 Hz, 1H), 7.40 (dd, J = 8.0, 1.8 Hz, 1H), 6.89 (s, 1H), 6.83 (s, 1H), 3.29 (q, J = 7.1 Hz, 2H), 1.44 (s, 6H), 1.34 (t, J = 7.2 Hz, 3H). 13C NMR (126 MHz, Chloroform -d) / (ppm): 155.18, 154.79, 148.42, 138.89, 129.81, 128.00, 125.74, 121.02, 119.82, 118.68, 111.94, 106.82, 46.85, 38.85, 27.26, 14.81. ESI-MS: (calc.) (m/z) calcd for C17H18BrN [M+H ]+ 316.07, found 316.12. 7-Bromo-N-ethyl -N-maleimidomethyl -9,9-dimethyl -9H-fluoren -2-amine (77 ) To a solution of 76 (160 mg, 0.506 mmol) in anhydrous toluene (20 mL) was added 73 (130 mg, 1.01 mmol). The reaction was refluxed at 130 ¡C overnight with a Dean -Stark apparatus . The reaction mixture was then purified by flash chromatography with 10% ethyl acetate in hexane to afford a bright red solid (170 mg, 79% yield). 1H NMR (500 MHz, DMSO -d6) / (ppm): 7.65 (d, J = 2.0 Hz, 1H), 7.58 (d, J = 8.4 Hz, 1H), 7.54 (d, J = 8.0 Hz, 1H), 7.39 (dd, J1 = 8.1 Hz, J2 = 1.9 Hz, 1H), 7.18 (d, J = 2.4 Hz, 1H), 7.05 (s, 2H), 6.90 Ð 6.83 (m, 1H), 5.03 (s, 2H), 3.59 (q, J = 7.0 Hz, 2H), 1.38 (s, 6H), 1.11 (t, J = 7.0 Hz, 3H). ESI-MS: (calc.) (m/z) calcd for C 22H21BrN2O2 [M+H] + 425.08, found 425.17. 278 7-(Ethylamino) -9,9-dimethyl -9H-fluorene -2-carbaldehyde (78) To a solution of 76 (30 mg, 0.095 mmol) in anhydrous THF (3 mL) was added tert -butyllithium (0.14 mL, 1.7 M in pentane, 0.24 mmol) at -78 ¡C under argon. The reaction mixture was stirred at -78 ¡C for 1 h. A solution of DMF (17 mg, 0.24 mmol) in anhydrous THF was added and the mixture was stirred for another 1 h. The mixture was then gradually warm ed to room temperature and kept for another 2 h. The reaction was quenche d with chilled methanol and extracted with ethyl acetate. The organic phase was combined, dried over anhydrous sodium sulfate , and concentrated in vacuo . The crude was purified by flash chromatography with 15% ethyl acetate in hexane to afford a yellow solid (16 mg, 64% yield). 1H NMR (500 MHz, Chloroform -d) / (ppm): 9.99 (s, 1H), 7.88 (d, J = 1.4 Hz, 1H), 7.78 (dd, J1 = 7.8 Hz, J2 = 1.4 Hz, 1H), 7.66 (d, J = 7.8 Hz, 1H), 7.60 (d, J = 8.2 Hz, 1H), 6.68 (d, J = 2.1 Hz, 1H), 6.63 (dd, J1 = 8.3 Hz, J2 = 2.2 Hz, 1H), 3.26 (q, J = 7.1 Hz, 2H), 1.48 (s, 6 H), 1.32 (t, J = 7.2 Hz, 3 H). 13C NMR (126 MHz, Chloroform -d) / (ppm): 192.09, 157.32, 153.26, 149.46, 146.70, 133.73, 131.16, 127.31, 122.47, 122.44, 118.41, 112.32, 106.52, 46.61, 38.70, 27.13, 14.70. ESI-MS: (calc.) (m/z) calcd for C18H19NO [M+H] + 266.15, found 266.24. 7-((Maleimidomethyl) (ethyl )amino)-9,9-dimethyl -9H-fluorene -2-carbaldehyde (Q1FR0 ) The procedure is similar to the preparation of 77. To a solution of 78 (15 mg, 0.057 mmol) in dry toluene (5 mL) was added 73 (14 mg, 0.11 mmol). The reaction was carried 279 out with a Dean -Stark apparatus overnight. The crude product was purified with 20% ethyl acetate in hexane to afford an orange solid (17 mg, 81% yield). 1H NMR (500 MHz, Chloroform -d) / (ppm): 10.00 (s, 1H), 7.93 Ð 7.88 (m, 1H), 7.80 (dd, J1 = 7.8 Hz, J2 = 1.5 Hz, 1H), 7.68 (dd, J1 = 12.7 Hz, J2 = 8.1 Hz, 2H), 7.23 (d, J = 2.4 Hz, 1H), 6.99 (dd, J1 = 8.5 Hz, J2 = 2.5 Hz, 1H), 6.74 (s, 2H), 5.18 (s, 2H), 3.75 (q, J = 7.1 Hz, 2H), 1.52 (s, 6H), 1.27 (t, J = 7.0 Hz, 3 H). 13C NMR (126 MHz, Chloroform -d) / (ppm): 192.10, 171.13, 157.14, 153.62, 147.19, 146.33, 134.45, 134.01, 131.01, 127.75, 122.61, 122.29, 118.77, 112. 45, 107.59, 54.61, 46.87, 45.38, 27.10, 12.63. ESI-MS: (calc.) (m/z) calcd for C23H22N2O3 [M+H ]+ 375.17, found 375.25. 3,3-Dimethoxypyrrolidine -2,5-dione (79 )65 Bromine (12.4 g, 4.0 0 ml, 77.3 mmol) was added dropwise to a solution of maleimide (5.0 0 g, 51.5 mmol) in methanol (200 m L) at 0 p=C. The reaction mixture was stirred for 16 h at RT, then concentrated in vacuo . The crude was dissolved in methanol (75 m L) and added dropwise to a solution of metal sodium (4.74 g, 206 mmol) in methanol (200 m L). After 20 h the reaction mixture was concentrated in vacuo and diluted with EtOAc (100 m L). The mixture was n eutralized by slow addition of HCl aq. (6 M) , then diluted into water and extracted with EtOAc . The combined organic phase was washed with brine and dried over anhydrous NaSO4. Concentration in vacuo afforded a white solid (7.8 g, 95% yield). 1H NMR (500 MHz, Chloroform -d) / (ppm): 8.18 (s, 1H), 3.44 (s, 6H), 2.86 (s, 2H). ESI-MS: (calc.) (m/z) calcd for C6H9NO4 [M+H ]+ 160.06, found 160.15. 280 3-Methoxy maleimide (80 )65 To a solution of 79 (7.8 g, 44 mmol) in toluene (200 mL) was added TsOH "H2O (0.84 g, 4.4 mmol). The reaction was refluxed with a Dean -Stark apparatus for 24 h. Toluene (80 mL) was removed and the flask was replenished with fresh toluene. The reaction mixture was refluxed for another 24 h before being concentrated in vacuo . The crude was purified by flash chromatography with 40 ~ 60% EtOAc i n Hexane to afford a white flak7 solid (5.5 g, 88% yield). 1H NMR (500 MHz, Chloroform -d) / (ppm): 7.28 (s, 1H), 5.44 (d, J = 1.5 Hz, 1H), 3.94 (s, 3H). 13C NMR (126 MHz, Chloroform -d) / (ppm): 169.46, 165.27, 161.08, 97.45, 59.07. ESI-MS: (calc.) (m/z) calcd for C5H5NO3 [M+H ]+ 128.03, found 128.10. N-chloroethyl -3-methoxy maleimide (81) Powdered solid of 80 (50 mg, 0.39 mmol) was mixed with anhydrous potassium carbonate (163 mg, 1.2 mmol) and tetrabutylammonium bromide (TBAB, 13 mg, 0.039 mmol). T o the mixture was added 1 -bromo -2-chloro ethane ( 680 mg, 4.7 mmol) and the suspension was stirred at 40 p=C for 3 h. The crude was purified by flash chromatography with 30% EtOAc in hexane to afford a yellow solid (64 mg, 85% yield). 1H NMR (500 MHz, Chloroform -d) / (ppm): 5.44 (s, 1H), 3.94 (s, 3H), 3.85 (t, J = 6.3 Hz, 2H), 3.66 (t, J = 6.3 Hz, 2H). 13C NMR (126 MHz, Chloroform -d) / (ppm): 169.56, 165.18, 160.86, 96.39, 59.03, 40.85, 39.04. ESI-MS: (calc.) (m/z) calcd for C7H8ClNO3 [M+H ]+ 190.03, found 190.13. 281 N-bromoethyl -3-methoxy maleimide (82) The procedure is similar to the preparation of 81. Powdered solid 80 (50 mg, 0.39 mmol) was mixed with anhydrous K 2CO3 (163 mg, 1.2 mmol), TBAB (13 mg, 0.039 mmol), and 1,2 -dibromoethane ( 890 mg, 4.7 mmol). Flash chromatography afforded an off-white solid (69 mg, 75% yield). 1H NMR (500 MHz, Chloroform -d) / (ppm): 5.44 (s, 1H), 3.95 (s, 3H), 3.92 (t, J = 6.6 Hz, 2H), 3.51 (t, J = 6.7 Hz, 2H). 13C NMR (126 MHz, Chloroform -d) / (ppm): 169.46, 165.06, 160.87, 96.39, 59.04, 38.93, 28.28. ESI-MS: (calc.) (m/z) calcd for C7H8BrNO3 [M+H ]+ 233.98, found 234.10. N-iodoethyl -3-methoxy maleimide (83) To a solution of 82 (100 mg, 0.427 mmol) in anhydrous acetone (5 mL) was added sodium iodide (320 mg, 2.14 mmol). The reaction mixture was refluxed overnight and then concentrated in vacuo . The crude was purified by flash chromatography with 30% EtOAc in hexane to afford an off-white solid (113 mg, 94% yield). 1H NMR (500 MHz, Chloroform -d) / (ppm): 5.44 (s, 1H), 3.95 (s, 3H), 3.89 (t, J = 7.2 Hz, 2H), 3.30 (t, J = 7.2 Hz, 2H). 13C NMR (126 MHz, Chloroform -d) / (ppm): 169.32, 164.89, 160.89, 96.38, 59.04, 39.67 , 0.18 . ESI-MS: (calc.) (m/z) calcd for C7H8INO3 [M+H] + 281.96, found 282.05. N-hydroxyethyl -3-methoxy maleimide (84) 282 A solid of 80 (50 mg, 0.39 mmol) and a solution of ethylene oxide (0.20 mL, 2.5 M, 0.41 mmol) in THF (2.5 mL) was mixed and heated to 40 p=C in a sealed tube overnight. A light orange solid was filtered and purified by flash chromatography with 50% EtOAc in hexane, affording 20 mg (30% yield). 1H NMR (500 MHz, Chloroform -d) / (ppm): 5.44 (s, 1H), 3.94 (s, 3H), 3.77 ( m, 2H), 3.73 Ð 3.66 (m, 2H), 2.16 (s, 1H). 13C NMR (126 MHz, Chloroform -d) / (ppm): 170.63, 165.93, 160.98, 96.31, 60.96, 59.02, 40.53. ESI-MS: (calc.) (m/z) calcd for C7H9NO4 [M+H ]+ 172.06, found 172.15. 7-((2-Bromoethyl)(ethyl)amino) -9,9-dimethyl -9H-fluorene -2-carbaldehyde (85) To a solution of 78 (130 mg, 0.490 mmol) in dry DMF (5 mL) was added 1,2 -dibromoethane (460 mg, 2.45 mmol ). The mixture was stirred and heated at 80 p=C overnight. The crude was washed with water and extracted with EtOAc. The combined organic phase was dried over anhydrous sodium sulfate and concentrated in vacuo . The title product was purified by flash chroma tography with 1 0% EtOAc in hexane to afford an orange solid (68 mg, 61 % yield b.r.s.m.). 1H NMR (500 MHz, Chloroform -d) / (ppm): 10.00 (s, 1H), 7.90 (d, J = 1.4 Hz, 1H), 7.81 ( m, 1H), 7.67 ( m, 2H), 6.73 (s, 2H), 3.80 (t, J = 7.9 Hz, 2H), 3.59 Ð 3.45 (m, 4H), 1.51 (s, 6H), 1.27 (t, J = 7.1 Hz, 4H). 13C NMR (126 MHz, Chloroform -d) / (ppm): 192.06, 157.42, 153.34, 146.36 , 133.86, 131.17, 130.90 , 122.79, 122.65, 122.53, 118.55, 111.06, 105.46, 52.69, 46.80, 45.63, 28.24, 27.20, 12.56. ESI-MS: (calc.) (m/z) calcd for C20H22BrNO [M+H] + 372.09, found 372.17. 283 7-((3-Methoxy -maleimidoethyl)(e thyl )amino)-9,9-dimethyl -9H-fluorene -2-carbaldehyde (Q2mFR0 ) To a solution of 85 (4.0 mg, 0.011 mmol) in dry DMF (1 mL) was added 80 (2.7 mg, 0.022 mmol) and potassium carbonate (7.4 mg, 0.54 mmol). The mixture was stirred and heated at 80 p=C overnight. The resulting mixture was washed with water and extracted with ethyl acetate. The c ombined organic phase was dried and concentrated in vacuo . The title compound was purified by prep -TLC with 40% EtOAc in hexane to afford a light yellow solid (2.6 mg, 58% yield). 1H NMR (500 MHz, Chloroform -d) / (ppm): 9.98 (s, 1H), 7.89 (d, J = 1.3 Hz, 1 H), 7.78 (dd, J1 = 7.8 Hz, J2 = 1.5 Hz, 1H), 7.64 ( m, 2H), 6.94 Ð 6.88 (m, 1H), 6.77 (d, J = 8.3 Hz, 1H), 5.36 (s, 1H), 3.87 (s, 3H), 3.74 (dd, J1 = 8.6 Hz, J2 = 6.1 Hz, 2H), 3.57 (dd, J1 = 8.6 Hz, J2 = 6.1 Hz, 2H), 3.48 (q, J = 7.1 Hz , 2H), 1.51 (s, 7H), 1.23 (t, J = 7.1 Hz , 3 H). 13C NMR (126 MHz, Chloroform -d) / (ppm): 192.08, 169.97, 165.53, 160.94, 157.47, 153.46, 146.57, 133.69, 131.11, 122.52, 122.49, 118.40, 111.40, 106.02, 97.49, 96.28, 59.06, 58.95, 46.79, 34.54, 29.70, 27.14, 12.40. ESI-MS: (calc.) (m/z) calcd for C25H26N2O4 [M+H] + 419.1971, found 419.1973. 284 REFERENCE S 285 REFERENC ES 1.!Klymchenko, A.S. Solvatochromic and Fluorogenic Dyes as Environment -Sensitive Probes: Design and Biological Applications. Acc. Chem. Res. 2017, 50, 366-375. 2. Walker, C.L. et al. Fluorescence imaging using synthetic GFP chromophores. Curr. Opin. Chem. Bio l. 2015, 27, 64-74. 3. Haidekker, M.A. & Theodorakis, E.A. Molecular rotors -fluorescent biosensors for viscosity and flow. Org. Biomol. Chem. 2007, 5, 1669-1678. 4. Ludovic, J. & Arnaud, G. Fluorogen -based reporters for fluorescence imaging: a review. Methods Appl . Fluoresc . 2015, 3, 042007. 5. Hıvelmann, F., Gaspar, I., Ephrussi, A. & Seitz, O. Brightness Enhanced DNA FIT -Probes for Wash -Free RNA Imaging in Tissue. J. Am. Chem. Soc. 2013, 135, 19025-19032. 6. Hori, Y. & Kikuchi, K. Protein labeling with fluorogenic probes for no -wash live -cell imaging of proteins. Curr. Opin. Chem. Biol. 2013, 17, 644-650. 7. Bruchez, M.P. Dark dyes Ðbright complexes: fluorogenic prote in labeling. Curr. Opin. Chem. Biol. 2015, 27, 18-23. 8. Szent -Gyorgyi, C., Schmidt, B.F., Fitzpatrick, J.A.J. & Bruchez, M.P. Fluorogenic Dendrons with Multiple Donor Chromophores as Bright Genetically Targeted and Activated Probes. J. Am. Chem. Soc. 2010, 132, 11103-11109. 9. Baranczak, A. et al. A Fluorogenic Aryl Fluorosulfate for Intraorganellar Transthyretin Imaging in Living Cells and in Caenorhabditis elegans. J. Am. Chem. Soc. 2015, 137, 7404-7414. 10. Sasaki, S., Drummen, G.P.C. & Konishi, G. -i. R ecent advances in twisted intramolecular charge transfer (TICT) fluorescence and related phenomena in materials chemistry. J. Mate r. Chem. C 2016, 4, 2731-2743. 11. Haberhauer, G., Gleiter, R. & Burkhart, C. Planarized Intramolecular Charge Transfer: A Con cept for Fluorophores with both Large Stokes Shifts and High Fluorescence Quantum Yields. Chem. Eur. J. 2016, 22, 971-978. 12. Shaya, J. et al. Turn -on Fluorene Push ÐPull Probes with High Brightness and Photostability for Visualizing Lipid Order in Biomembranes. ACS Chem. Biol. 2017, 12, 3022-3030. 286 13. Prifti, E. et al. A Fluorogenic Probe for SNAP -Tagged Plasma Membrane Proteins Based on the Solvatochromic Molecule Nile Red. ACS Chem. Biol. 2014, 9, 606 -612. 14. Karpenko, I.A. et al. Red Fluorescent Turn -On Ligands for Imaging and Quantifying G Protein -Coupled Receptors in Living Cells. ChemBioChem 2014, 15, 359-363. 15. Liu, T. -K. et al. A Rapid SNAP -Tag Fluorogenic Probe Based on an Environment -Sensitive Fluorophore for No -Wash Live Cell Imaging. ACS Chem. Biol. 2014, 9, 2359-2365. 16. Yoshii, T., Mizusawa, K., Takaoka, Y. & Hamachi, I. Intracellular Protein -Responsive Supramolec ules: Protein Sensing and In -Cell Construction of Inhibitor Assay System. J. Am. Chem. Soc. 2014, 136, 16635-16642. 17. Hou, T. C., Wu, Y. -Y., Chiang, P. Y. & Tan, K. T. Near -infrared fluorescence activation probes based on disassembly -induced emission cya nine dye. Chem. Sci. 2015, 6, 4643-4649. 18. Anees, P., Sreejith, S. & Ajayaghosh, A. Self -Assembled Near -Infrared Dye Nanoparticles as a Selective Protein Sensor by Activation of a Dormant Fluorophore. J. Am. Chem. Soc. 2014, 136, 13233-13239. 19. Karpenk o, I.A. et al. Fluorogenic Squaraine Dimers with Polarity -Sensitive Folding As Bright Far -Red Probes for Background -Free Bioimaging. J. Am. Chem. Soc. 2015, 137, 405-412. 20. Yuan, Y., Xu, S., Cheng, X., Cai, X. & Liu, B. Bioorthogonal Turn -On Probe Based on Aggregation -Induced Emission Characteristics for Cancer Cell Imaging and Ablation. Angew. Chem. Int. Ed. 2016, 55, 6457-6461. 21. Zhao, E. et al. Light -Enhanced Bacterial Killing and Wash -Free Imaging Based on AIE Fluorogen. ACS Appl. Mater. Interfaces 2015, 7, 7180-7188. 22. Peng, L. et al. Rational Design of a Red -Emissive Fluorophore with AIE and ESIPT Characteristics and Its Application in Light -Up Sensing of Esterase. Anal. Chem. 2017, 89, 3162-3168. 23. Schneider, A.F.L. & Hackenberger, C.P.R. Fluo rescent labelling in living cells. Curr. Opin. Biotechnol. 2017, 48, 61-68. 24. Wang, Y. et al. Development of a red fluorescent light -up probe for highly selective and sensitive detection of vicinal dithiol -containing proteins in living cells. Chem. Sci. 2016, 7, 518-524. 287 25. Lukinavi 1ius, G. et al. Fluorogenic Probes for Multicolor Imaging in Living Cells. J. Am. Chem. Soc. 2016, 138, 9365-9368. 26. Lukinavi 1ius, G. et al. A near -infrared fluorophore for live -cell super -resolution microscopy of cellular proteins. Nat. Chem. 2013, 5, 132-139. 27. Zhang, Q., Liu, H. & Pan, Z. A general approach for the development of fluorogenic probes suitable for no -wash imaging of kinases in live cells. Chem. Commun. 2014, 50, 15319-15322. 28. Blum, G. et al. Dynamic imaging of protease activity with fluorescently quenched activity -based probes. Nat. Chem. Biol . 2005, 1, 203-209. 29. Kim, S., Kim, H., Choi, Y. & Kim, Y. A New Strategy for Fluorogenic Esterase Probes Displaying Low Levels of Non -specific Hydrolysis. Chem. Eur. J. 2015, 21, 9645-9649. 30. Kuriki, Y. et al. Establishment of Molecular Design Strategy To Obtain Activatable Fluorescent Probes for Carboxypeptidases. J. Am. Chem. Soc. 2018, 140, 1767 -1773. 31. Mizukami, S., Watanabe, S., Akimoto, Y. & Kikuchi, K. No -Wash Protein Labeling with Des igned Fluorogenic Probes and Application to Real -Time Pulse -Chase Analysis. J. Am. Chem. Soc. 2012, 134, 1623-1629. 32. Sun, X. et al. Development of SNAP -Tag Fluorogenic Probes for Wash -Free Fluorescence Imaging. ChemBioChem 2011, 12, 2217-2226. 33. Yuan, Y., Zhang, R., Cheng, X., Xu, S. & Liu, B. A FRET probe with AIEgen as the energy quencher: dual signal turn -on for self -validated caspase detection. Chem. Sci. 2016, 7, 4245-4250. 34. Meimetis, L.G., Carlson, J.C.T., Giedt, R.J., Kohler, R.H. & Weisslede r, R. Ultrafluorogenic Coumarin -Tetrazine Probes for Real -Time Biological Imaging. Angew. Chem. Int. Ed. 2014, 53, 7531-7534. 35. Speiser, S. Photophysics and Mechanisms of Intramolecular Electronic Energy Transfer in Bichromophoric Molecular Systems: + Solution and Supersonic Jet Studies. Chem. Rev. 1996, 96, 1953-1976. 36. Turro, N. J., Ramamurthy, V. & Scaiano, J. C. Modern Molecular Photochem istry of Organic Molecules. (University Science Books; Sausalito, California; 2010). 37. Fan, J., Hu, M., Zhan, P. & Peng, X. Energy transfer cassettes based on organic fluorophores: construction and applications in ratiometric sensing. Chem. Soc. Rev. 2013, 42, 29-43. 288 38. Li, Z. et al. Tetrazole Photoclick Chemistry: Reinvestigating Its Suitability as a Bioorthogonal Reaction and Potential Applications. Angew. Chem. Int. Ed. 2016, 55, 2002-2006. 39. Knorr, G., Kozma, E., Herner, A., Lemke, E.A. & Kele, P. New Red -Emitting Tetrazine -Phenoxazine Fluorogenic Labels for Live -Cell Intracellular Bioorthogonal Labeling Schemes. Chem. Eur. J. 2016, 22, 8972-8979. 40. Kamikawa, Y. et al. Design of a protein tag and fluorogenic probe with modular structure for live -cell imaging of intracellular proteins. Chem. Sci. 2016, 7, 308 -314. 41. Hori, Y., Nakaki, K., Sato, M., Mizukami, S. & Kikuchi, K. Development of Protein -Labeling Probes with a Redesigned Fluorogenic Switch Based on Intramolecular Association for No -Wash Live -Cell Imaging. Angew. Chem. Int. Ed. 2012, 51, 5611-5614. 42. Zhang, H. et al. An Off ÐOn COX -2-Specific Fluorescent Probe: Targeting the Golgi Apparatus of Cancer Cells. J. Am. Chem. Soc. 2013, 135, 11663-11669. 43. de Silva, A.P., Moody, T.S. & Wright, G.D. Fluorescent PET (Photoinduced Electron Transfer) sensors as potent analytical tools. Analyst 2009, 134, 2385 -2393. 44. Callan, J.F., de Silva, A.P. & Magri, D.C. Luminescent sensors and switches in the early 21st century. Tetrahedron 2005, 61, 8551-8588. 45. Mart™nez -M⁄Œez, R. & SancenŠn, F. Fluorogenic and Chromogenic Chemosensors and Reagents for Anions. Chem. Rev. 2003, 103, 4419-4476. 46. Daly, B., Ling, J. & de Silva, A.P. Current developments in fluorescent PET (photoinduced electron transfer) sensors and switches. Chem. Soc. Rev. 2015, 44, 4203-4211. 47. Men, Y. et al. Rational design and synthesis of fast -response NBD -based fluorescent probes for biothiols. Tetrahedron Lett. 2015, 56, 5781-5786. 48. Georgiev, N.I., Sakr, A.R. & Bojinov, V.B. Design and synthesis of a novel PET and ICT based 1,8 -naphthalimide FRET b ichromophore as a four -input Disabled ÐEnabled -OR logic gate. Sens . Actuators , B 2015, 221, 625-634. 49. Yang, X. -F. et al. A dual emission fluorescent probe enables simultaneous detection of glutathione and cysteine/homocysteine. Chem. Sci. 2014, 5, 2177 -2183. 289 50. Zhang, W., Ma, Z., Du, L. & Li, M. Design strategy for photoinduced electron transfer -based small -molecule fluorescent probes of biomacromolecules. Analyst 2014, 139, 2641-2649. 51. Yang, C. et al. An aldehyde group -based P -acid probe for selective fluorescence turn -on sensing of cysteine and homocysteine. Biosens . Bioelectron . 2016, 80, 17 -23. 52. Dubey, R.K., Knorr, G., Westerveld, N. & Jager, W.F. Fluorescent PET probes based on pe rylene -3,4,9,10-tetracarboxylic tetraesters. Org. Biomol. Chem. 2016, 14, 1564-1568. 53. Nawimanage, R.R., Prasai, B., Hettiarachchi, S.U. & McCarley, R.L. Rapid, Photoinduced Electron Transfer -Modulated, Turn -on Fluorescent Probe for Detection and Cellula r Imaging of Biologically Significant Thiols. Anal. Chem. 2014, 86, 12266-12271. 54. Shiraishi, Y., Nakamura, M., Kogure, T. & Hirai, T. Off -on fluorometric detection of cyanide anions in an aqueous mixture by an indane -based receptor. New J. Chem. 2016, 40, 1237-1243. 55. Okitsu, K., Misawa, T., Shoda, T., Kurihara, M. & Demizu, Y. Development of an ON/OFF switchable fluorescent probe targeting His tag fused proteins in living cells. Bioorg. Med. Chem. Lett. 2017, 27, 3417-3422. 56. Zhao, Y. et al. Benzimi dazo[2,1 -a]benz[de]isoquinoline -7-one-12-carboxylic acid based fluorescent sensors for pH and Fe3 +. J. Photochem . Photobiol ., A 2016, 314, 52-59. 57. Lee, J.J. et al. A PET -based fluorometric chemosensor for the determination of mercury( II) and pH, and hydrolysis reaction -based colorimetric detection of hydrogen sulfide. Dalton Trans . 2016, 45, 5700-5712. 58. Liu, Z., Zhou, Y., Du, L. & Li, M. Novel intramolecular photoinduced electron transfer -based probe for the Human Ether -a-go-go-Related Gene (hERG) potassium channel. Analyst 2015, 140, 8101-8108. 59. Kamiya, M. et al. An Enzymatically Activated Fluorescence Probe for Targeted Tumor Imaging. J. Am. Chem. Soc. 2007, 129, 3918-3929. 60. Chen, Y. et al. Coumarin -Based Fluorogenic Probes for No -Wash Protein Labeling. Angew. Chem. Int. Ed. 2014, 53, 13785-13788. 61. Guy, J. et al. Convergent Preparation and Photophysical Characterization of Dimaleimide Dansyl Fluorogens: + Elucidation of the Maleimide Fluorescence Quenching Mechanism. J. Am. Chem. Soc. 2007, 129, 11969-11977. 290 62. Girouard, S., Houle, M. -H., Grandbois, A., Keillor, J.W. & Michnick, S.W. Synthesis and Characterization of Dimaleimide Fluorogens Designed for Specific Labeling of Proteins. J. Am. Chem. Soc. 2005, 127, 559-566. 63. Matsumoto, T., Urano, Y., Shoda, T., Kojima, H. & Nagano, T. A Thiol -Reactive Fluorescence Probe Based on Donor -Excited Photoinduced Electron Transfer: + Key Role of Ortho Substitution. Org. Lett. 2007, 9, 3375-3377. 64. Vail, S.L. & Pierce, A.G. Limitations for the addition of amides to formaldehyde and glyoxal. J. Org. Chem. 1972, 37, 391-393. 65. Booker -Milburn, K.I., Baker, J.R. & Bruce, I. Rapid Access to Azepine -Fused Oxetanols from Alkoxy -Substituted Maleimides. Org. Lett. 2004, 6, 1481-1484.