1) PROBING THE WAVELENGTH REGULATION OF RHODOPSIN PIGMENTS VIA
DE NOVO PROTEIN ENGINEERING OF A RHODOPSIN MIMIC AND;
2)
ENGINEERING CRABPII AS A RETINAL ISOMERASE AND A PROTEIN FUSION
TAG
By
Kin Sing Stephen Lee

A DISSERTATION
Submitted to
Michigan State University
in partial fulfillment of the requirements
for the degree of
DOCTOR OF PHILOSOPHY
Chemistry
2011

i

ABSTRACT
1) PROBING THE WAVELENGTH REGULATION OF RHODOPSIN PIGMENTS VIA
DE NOVO PROTEIN ENGINEERING OF A RHODOPSIN MIMIC AND;
2) ENGINEERING CRABPII AS A RETINAL ISOMERASE AND A PROTEIN FUSION
TAG
By
Kin Sing Stephen Lee
After several decades of research, the mechanism of wavelength regulation in human
rhodopsins has not been fully elucidated at a molecular level. Although the gene sequences of
color rhodopsin (Blue, Green and Red) and Rod rhodopsin have been released and the crystal
structure of bovine rhodopsin has been resolved, the important protein-ligand interactions that
lead to wavelength modulation have not been dissected in details.

In order to tackle the

problems arisen using rhodopsin to study the theories behind wavelength regulation, a de novo
approach has been taken by engineering a protein surrogate as a rhodopsin mimic to bind alltrans-retinal as a protonated Schiff base. Throughout the study with cellular retinoic acid
binding protein II (CRABPII) as a rhodopsin mimic, we discovered that a fully embedded
chromophore is needed for the system to respond to amino acid changes. Therefore, a shorter
chromophore, C15 aldehyde, has been used instead of a full-length all-trans-retinal. Based on the
electrostatic potential calculations on color opsins models and the rhodopsin crystal structure, we
hypothesize that the electrostatic potential projected on the bound chromophore from the protein
plays an important role on wavelength regulation. To investigate the hypothesis, the crystal
structure of C15 aldehyde with CRABPII mutant R132K:R111L:L121E:R59W was used as a
template and a series of mutants has been generated and tested. We are able to modulate the

ii

absorption maxima of the bound C15 aldehyde PSB from 380 nm to 424 nm based on the change
of the overall electrostatic potential projected on the bound chromophore by the protein.
Protein engineering has been developed for decades to solve the problem in different
aspect of science. In human eyes, upon absorption of a photon, isomerization of 11-cis-retinal
specifically to all-trans-retinal leads to rhodopsin conformational changes which induce signal
transduction and allow us to see. Rhodopsin has a high affinity to 11-cis-retinal and poor
binding to all-trans-retinal which allows rhodopsin to reload with 11-cis-retinal and the vision
cycle continues. We would like to apply the same principle of design to re-engineer CRABPII
into retinal isomerase upon absorption of specific wavelength.

During the process of re-

engineering CRABPII into a retinal isomerase, we found that CRABPII has an ability to
isomerize 11-cis-retinal to all-trans-retinal without irridation with light. We, therefore, conduct
a series of experiments to study the mechanism of isomerization and propose that the
isomerization is the result of conjugated addition of a nucleophile, rotation and elimination of the
nucleophile.
Recently, protein fusion tags play an important role in biological research for their ability
to probe the biological processes in Nature. Green fluorescent protein (GFP) is one of the
commonly used tags. However, it is not without limitations. In order to provide a protein fusion
tag that is orthogonal to GFP, we would like to re-engineer CRABPII as a fluorescent protein
through incorporation of suitable chromophores. We have successfully re-engineered CRABPII
to bind to a merocyanine as a protonated Schiff base.

The resulting protein-merocyanine

complex is red shifted to ~590 nm from 465 nm and fluoresces at ~615 nm with the quantum
yield ~10%.

iii

ACKNOWLEDGMENTS
Throughout my six years of graduate study at Michigan State University, I have had a lot
of great and happy memories. Overall, these six years not only enriched my knowledge in
bioorganic chemistry but also helped me to develop as a scientist. However, nothing would had
happened without the following people. I would like to take this chance to say thank you to all
of them.
My supervisor, Babak Borhan, is a great scientist and teacher. I was fortunate to have him
as my supervisor throughout my graduate study at Michigan State University. He was always
helpful and very patient to show me everything that I needed to be a true scientist. When I first
joined the laboratory, I have so many weaknesses that prevented me from being a scientist.
Babak tried many ways, both criticism and encouragement, to fix my problems. On the other
hand, he is a kidult. He always brings a lot of fun to the laboratory and helps us to really enjoy
exploring science in the laboratory. I did enjoy my four years in this laboratory and I would like
to thank him for everything he did. I will definitely miss these days.
Chryssoula is my best friend and she played a huge role in my graduate study. She was
always willing to help and is a very good listener and problem solver. She was like a bridge
between my mind and Babak’s mind for my first two years in Babak’s laboratory and translated
my “singlish” to a more understandable language. Besides, she was a very good company to
enjoy great food. We always dined out together with Kit and Justas and it was always fun. I
cannot use any words to describe what she has done to help me and I will treasure this friendship
forever.
I would like to thank to our collaborator Dr. James Geiger and his group. They are great
collaborators and provided us with a lot of beautiful crystal structures to accelerate our research.

iv

Jim is a very intelligent and interesting scientist. He always provides many new ideas and the
bio-meeting that we had every week was one of the meetings that I did not want to miss because
it was always fun and I learnt a lot during these meetings.
My labmates: Aman, Arvind, Calvin, Camille, Carmin, Kumar, Mercy, Wenjing, Roozbeh,
Sarah, Tanya, Toyin, Dr. Li Xiaoyong and former labmates: Dr. Daniel Whitehead, Dr. Stewart
Hart, Dr. Montserrat Rabago-Smith, Dr. Jennifer Schoemaker, Dr. Jun Yan, Dr. Tao Zhang, Dr.
Xiaofei Jia are all fun people and we always have fun in and outside the laboratory. I do
appreciate their friendship and the fun working environment they created. Also, I would like to
thank the undergraduates: Yoomi, Farid, Kaveri and Ruba. They helped me to conduct some
experiments and were always helpful and efficient.
I also would like to thank Professor Chang. He was my undergraduate supervisor and my
second reader at Michigan State University. He gave me many advices especially for my first
two years and helped me to get use to living in US. In addition, the dinners in his home with his
wife, Mrs. Chang, are some of the best memories I have in US.
In addition, I would like to acknowledge Dr. John Frost as my supervisor for the first two
years of my graduate study. Although the time was short, I did learn a lot, especially in the field
of biosynthesis and microbial engineering. I also appreciate the friendship with Frost’s group
members: Justas, Dongming, Lao Li, Jingtao, Wei, Heather, Mapitso, Mo, Ningqing, Jihane,
Brad and Jingsong.
My Hong Kong friends: Kit, Fei, Harrison, Sewa and SinC that I met in East Lansing
helped me to enjoy my stay in East Lansing. We always had fun and enjoyed great chinese food
every week. There are so many great and fun memories between us and I will treasure them
forever.

v

Last but not least, I would like to say thank you to my parents for their support,
understanding and encouragement throughout my six years of graduate study. Lastly, I am so
grateful to meet my wife, Suet Yi Lo, at Michigan State University. She is always understanding
and supportive. She took care of me very well and helped me to get through some tough and sad
periods of life during my graduate study. I would like to express my appreciation from the
bottom of my heart for her love.

vi

TABLE OF CONTENTS
List of Tables ............................................................................................................................... xii

List of Figures............................................................................................................................. xvi

List of Schemes ......................................................................................................................... xxix

KEY TO SYMBOLS OR ABBREVIATIONS ....................................................................... xxx

Probing the Wavelength Regulation of Rhodopsin Pigments via de Novo Engineering of a
Rhodopsin mimic .......................................................................................................................... 1

1.1. Introduction ....................................................................................................................... 1

1.1.1. The Importance of G-Protein Coupled Receptors ............................................................... 1

1.1.2. History of Vision Science .................................................................................................... 3

1.1.3.

The Process of Vision .......................................................................................................... 4

1.1.4. The Science of Color Vision in Humans ........................................................................... 11

1.1.5. Wavelength Regulations in Other Microbial Rhodopsins ................................................. 23

1.1.6. Re-engineering a Protein Mimic to Probe the Wavelength Regulation in Rhodopsin ..... 29

1.2. Probing the Wavelength Regulation using Re-engineered CRABPII as a Rhodopsin
Mimic ....................................................................................................................................... 35

1.2.1. Re-engineering CRABPII into Rhodopsin Protein Surrogate ........................................... 35

1.2.2.Effect of Electrostatic Potential on Wavelength Regulation in Rhodopsin ......................... 39

1.2.3.Probing the Wavelength Regulation in Rhodopsin using CRABPII–Triple Mutant
R132K:R111L:L121E ................................................................................................................... 43

1.2.3.1. Further Investigations on Wavelength Modulation and Retinal Binding in CRABPII at
Position 76 and 56......................................................................................................................... 47

1.2.3.2. Investigating the Effect of Mutation at position 59 on Wavelength Regulation of
CRABPII ....................................................................................................................................... 52

1.2.4. Probing the Wavelength Regulation in Rhodopsin using CRABPII and C15 aldehyde ... 54

1.2.4.1. Synthesis of C17 aldehyde and C15 aldehyde.................................................................. 57

1.2.4.2. Binding of C17 aldehyde with R132K:R111L:L121E mutants ...................................... 57

1.2.4.3. Binding of C15 aldehyde with CRABPII mutant R132K:R111L:L121E ....................... 60

1.2.4.4.
...........
 Exploring the Effect of Mutations at Position 59 on the Wavelength Modulation in
CRABPII using C15 aldehyde ....................................................................................................... 62

1.2.4.4.1. Investigating the Element for Bound C15 aldehyde PSB Stabilization ....................... 65

1.2.4.4.2. Studying the Effect of Electrostatic Potential on Wavelength Regulation based on
CRABPII R59 Mutant. ................................................................................................................. 67

1.2.4.5. Probing the Effect of Negative Charged Residues on Wavelength Regulation in
CRABPII ....................................................................................................................................... 69

1.2.4.6. Exploring the Effect of Mutation at Position 76 on Wavelength Regulation in CRABPII
using C15 aldehyde........................................................................................................................ 71

1.2.4.7. Study the Effect of Residues Around the Bound C15 aldehyde PSB.............................. 72

1.2.4.8. Probing the Wavelength Regulation in CRABPII Mutant R132K:R111L ..................... 77

vii

1.2.4.8.1. Discovery the Alternate Binding Site for C15 aldehyde in CRABPII ......................... 78

1.2.4.9. Fluorescence as a Tool to Identify the Relative Position of the Bound C15 aldehyde ... 85

1.2.4.10. Study the effect of BTP on C15 aldehyde binding ........................................................ 88

1.2.4.11. Study the effect of Tyr134 on bound C15 aldehyde absorption and binding ................ 93

1.3. Conclusion ............................................................................................................................ 95

1.4. Materials and Methods .................................................................................................... 96

i.Protein Mutagenesis, Bacterial Expression and Purification ...................................................... 96

i.Protein Mutation ......................................................................................................................... 96

iv. Dpn I Digestion of the Amplification Products .................................................................... 102

v.DNA Transformation of PCR product ..................................................................................... 102

vi.DNA Purification .................................................................................................................... 103

vii. Sample preparation for DNA sequencing ............................................................................ 104

viii. Protein Expression of CRABPII Mutants........................................................................... 105

ix.Protein Purification of CRABPII Mutants .............................................................................. 105

x. FPLC Protocol: ...................................................................................................................... 106

B.Reductive Amination Protocol ................................................................................................ 107

C. Extinction Coefficient Determination ................................................................................... 109

D. General Procedure for UV Binding Study ............................................................................ 111

E.Fluorescence Titration ............................................................................................................. 112

F.Molecular Modeling ................................................................................................................ 116

G.Chemistry ................................................................................................................................ 116

Synthesis of (2E,4E)-3-methyl-5-(2,6,6-trimethylcyclohex-1-enyl)penta-2,4-dienenitrile (1-25)...
......................................................................................................................................... 117

Synthesis of (2E,4E)-3-methyl-5-(2,6,6-trimethylcyclohex-1-enyl)penta-2,4-dienal (1-23) ..... 119

Synthesis of (2E,4E,6E)-5-methyl-7-(2,6,6-trimethylcyclohex-1-enyl)hepta-2,4,6-trienenitrile (126)
......................................................................................................................................... 120

Synthesis of (2E,4E,6E)-5-methyl-7-(2,6,6-trimethylcyclohex-1-enyl)hepta-2,4,6-trienal (1-22) ..
......................................................................................................................................... 121

Bibliography .......................................................................................................................... 124

Development of in vivo Color Screening System for Probing the Long Range Interactions
on Wavelength Regulation of Rhodopsin in CRBPII ............................................................ 136

2.1. Introduction ................................................................................................................... 136

2.2. Study the Second Shell Interactions on Wavelength Regulation using Engineered
Rhodopsin Mimics ................................................................................................................ 143

2.2.1. Design of Color Screening Method for CRBPII .............................................................. 150

2.2.2. Testing the in vivo Activity of BCDOX and Biosynthesis of β-carotene ........................ 155

2.2.3. Cloning of CRBPII into pCWori (+) Vector ................................................................... 160

2.2.4. Cloning of Gene Cassette (LacIq-Ptac-CRBP / LacIq-Ptac-CRBPII (KS) (Q108K:K40S) /
LacIq-Ptac-CRABPII / LacIq-Ptac-CRABPII (KLE) (R132K:R111L:L121E)) into pBADBCDOX Plasmid and Evaluation of the Color Screening Methods .......................................... 161

2.4. Conclusion ..................................................................................................................... 166

2.5. Materials and Methods ................................................................................................. 168

A. Plasmid Manipulation ........................................................................................................... 168


viii

i. Plasmid Purification................................................................................................................ 168

ii. Sample preparation for DNA sequencing:............................................................................. 168

iii. Polymerase Chain Reaction (PCR) Conditions .................................................................... 170

iv. Primer Design for PCR and Mutagenesis: ............................................................................ 170

v. DNA Purification using Agarose Gel: ................................................................................... 172

B. Cloning .................................................................................................................................. 173

i) Cloning (CRABPII, CRABPII mutants (R132K:R111L:L121E), CRBPII and CRBPII mutant
(Q108K:K40S)) into pCWori(+) ................................................................................................ 173

®
ii. Cloning of β-carotene monooxygenases (BCDOX) into pBAD-TOPO ............................. 176

iii. Cloning of LacIq-Ptac-CRBPII, LacIq-Ptac-CRBPII (Q108K:K40S), LacIq-Ptac-CRABPII
and LacIq-Ptac-CRABPII (R132K:R111L:L121E) into pBAD-BCDOX................................... 179

C. Protein Expression Analysis.................................................................................................. 182

D. HPLC analysis of in vivo Biosynthesis of β-carotene and all-trans-retinal......................... 182

E. Color Screening of CRBPII, CRABPII and both mutants .................................................... 183

Bibliography .......................................................................................................................... 185

Engineering CRABPII and CRBPII into a Retinal isomerase ............................................. 190

3.1. Introduction ................................................................................................................... 190

3.1.1. Isomerization Triggers Biological Processes ................................................................... 190

3.1.2. Biological Processes Triggered by Light Induced Isomerization .................................... 193

3.1.2.1. Phytochromes................................................................................................................ 193

3.1.2.2. Xanthopsins (Photoactive Yellow Protein) ................................................................... 194

3.1.2.3. Rhodopsin ..................................................................................................................... 196

3.1.2.4. Isomerization for Recycling 11-cis-Retinal in Vertebrates and Invertebrates .............. 198

3.1.2.4.1. Vertebrates ................................................................................................................. 198

3.1.2.4.2. Invertebrates ............................................................................................................... 205

3.2. Engineering CRABPII into Retinal Isomerases ......................................................... 206

3.2.1. Design Principles ............................................................................................................. 206

3.2.2. Synthesis of 11-cis-retinal ............................................................................................... 210

3.2.3. Identification of CRABPII Candidate for 11-cis-Retinal Isomerase ............................... 210

3.2.3.1. UV-vis study ................................................................................................................. 210

3.2.3.2. Prediction of 11-cis-retinal Binding Position in CRABPII .......................................... 214

3.2.3.3. Comparison of the Binding Constant with Both 11-cis-retinal and all-trans-Retinal
across Different CRABPII Mutants ............................................................................................ 215

3.2.3. Discovery of Isomerization Activity in CRABPII ............................................................ 218

3.2.4. Attempt to Dissect the Mechanism of 11-cis-retinal Isomerization ............................... 220

3.2.4.1. Is a Small Fraction of the Protein Responsible for the Isomerization?......................... 220

3.2.4.2. Investigating the Relationship between Isomerization and PSB or SB Formation ...... 221

3.2.4.3. Probing the Possible Michael Addition-Isomerization Mechanism ............................. 223

3.2.5. Catalytic Isomerization in CRABPII ............................................................................... 230

3.2.6. Attempted to Decrease the Isomerization Activity in CRABPII ........................... 232

3.3. Searching for Another Counterion for Bound 11-cis-retinal PSB in CRABPII ..... 234

3.4. human Cellular Retinol Binding Protein II as a Retinal Isomerase......................... 236

3.5. Synthesis of Ring Locked 11-cis-retinal Analog for Crystallography...................... 240

3.6.1. Synthesis of Ring-locked 11-cis-retinal Analog .............................................................. 241


ix

3.5.2. Comparison of UV-vis Spectroscopic and Fluorescent Spectroscopic Study between 11cis-retinal and its Ring-locked Analog 3-18 with CRABPII Mutants ........................................ 242

3.6. Conclusion ..................................................................................................................... 245

3.7. Material and Methods .................................................................................................. 247

A) Biology ................................................................................................................................. 247

v.Protein Mutagenesis, Bacterial Expression and Purification ................................................... 247

i. Protein Mutation on pET-17b vector ...................................................................................... 247

ii. DNA Transformation and Purification .................................................................................. 248

iii. Protein Expression and Purifications.................................................................................... 248

B. Protein Characterization and Binding Assays ....................................................................... 249

i. Calculation Extinction Coefficients of CRABPII mutants ..................................................... 249

ii. Reductive Amination ............................................................................................................. 249

iii. UV-vis and Fluorescent Spectroscopic Study ...................................................................... 249

iv. HPLC study .......................................................................................................................... 249

B. Chemistry: ............................................................................................................................. 251

Synthesis of diethyl 3-(trimethylsilyl)prop-2-ynylphosphonate ................................................. 252

Synthesis of trimethyl((5E)-4-methyl-6-(2,6,6-trimethylcyclohex-1-enyl) hexa-3,5-dien-1ynyl)silane (3-11) ........................................................................................................................ 253

Synthesis of 1,3,3-trimethyl-2-((1E)-3-methylhexa-1,3-dien-5-ynyl)cyclohex- 1-ene .............. 254

Synthesis of (E)-3-iodobut-2-en-1-ol.......................................................................................... 255

Synthesis of (E)-tert-butyl(3-iodobut-2-enyloxy)dimethylsilane ............................................... 255

Synthesis of tert-butyl(((2E,8E)-3,7-dimethyl-9-(2,6,6-trimethylcyclohex-1-en-1-yl)nona-2,6,8trien-4-yn-1-yl)oxy)dimethylsilane (3-13) ................................................................................. 256

Synthesis of trimethyl((5E)-4-methyl-6-(2,6,6-trimethylcyclohex-1-enyl)hexa-3,5-dien-1ynyl)silane (3-14) ........................................................................................................................ 257

Synthesis of 11-cis-retinol .......................................................................................................... 259

Synthesis of 11-cis-retinal .......................................................................................................... 260

The synthesis of 3-methoxy-2-cyclohexen-1-one (3-20)............................................................ 261

Synthesis of (E)-6-(2-hydroxy-4-(2,6,6-trimethylcyclohex-1-enyl)but-3-en-2-yl)-3methoxycyclohex-2-enone (3-21) ............................................................................................... 262

Synthesis of (E)-6-(2-hydroxy-4-(2,6,6-trimethylcyclohex-1-enyl)but-3-en-2-yl)-3methoxycyclohex-2-enol (3-26) ................................................................................................. 263

Synthesis of (E)-4-(2-hydroxy-4-(2,6,6-trimethylcyclohex-1-enyl)but-3-en-2-yl)cyclohex-2enone (3-22) ................................................................................................................................ 264

Synthesis of (E)-4-((E)-4-(2,6,6-trimethylcyclohex-1-enyl)but-3-en-2-ylidene)cyclohex-2-enone
(3-23) ......................................................................................................................................... 264

Synthesis of ethyl 2-((E)-4-((E)-4-(2,6,6-trimethylcyclohex-1-enyl)but-3-en-2-ylidene)
cyclohex-2-enylidene)acetate (3-24) .......................................................................................... 265

Synthesis of 2-((E)-4-((E)-4-(2,6,6-trimethylcyclohex-1-enyl)but-3-en-2-ylidene)cyclohex-2enylidene)ethanol (3-27) ............................................................................................................. 266

Synthesis of 2-((E)-4-((E)-4-(2,6,6-trimethylcyclohex-1-en-1-yl)but-3-en-2-ylidene)cyclohex-2en-1-ylidene)acetaldehyde .......................................................................................................... 267

Bibliography .......................................................................................................................... 270

Engineering CRABPII as a Protein Fusion Tag .................................................................... 278

4.1. Introduction ................................................................................................................... 278


x

4.2. Reengineering CRABPII into a Chromophoric and Fluorescent Protein ............... 289

4.2.1. Binding of Merocyanine dyes to CRABPII Mutants ....................................................... 290

4.2.1.1. Synthesis of Merocyanine 4-10, 4-11 and 4-12 ............................................................ 293

4.2.1.2. UV-vis Spectroscopic Study with Merocyanines .......................................................... 297

4.2.1.2. Modulating the Absorption of the Chromophoric Protein-CRABPII ........................... 302

4.2.1.3. Fluorescent Properties of Merocyanine with CRABPII mutants.................................. 307

4.2.2. Incorporation of Azulenic Aldehyde Chromophore with CRABPII ............................... 311

4.2.2.1. Synthesis of Azulenic Aldehyde Chromophore 4-20 and 4-21 .................................... 313

4.2.2.2. UV-vis Spectroscopic Study with Chromophore 4-20 and 4-21 .................................. 315

4.3. Conclusion ..................................................................................................................... 318

4.3. Materials and Methods ................................................................................................. 320

A. Biology .................................................................................................................................. 320

i. Protein Mutagenesis, Bacterial Expression and Purification .................................................. 320

ii. Protein Characterization and Binding Assays ....................................................................... 321

iii. Emission Spectrum Measurement ........................................................................................ 321

iv. Fluorescence Quantum Yield Measurement and Calculation............................................... 323

B. Chemistry .............................................................................................................................. 325

Synthesis of (2E,4E)-4-(1,3,3-trimethylindolin-2-ylidene)but-2-enenitrile (4-15) .................... 326

Synthesis of (2E,4E)-4-(1,3,3-trimethylindolin-2-ylidene)but-2-enal (4-16) ............................. 327

Synthesis of (2E,4E,6E)-6-(1,3,3-trimethylindolin-2-ylidene)hexa-2,4-dienenitrile (4-............ 328

Synthesis of (2E,4E,6E)-6-(1,3,3-trimethylindolin-2-ylidene)hexa-2,4-dienal (4-12) ............... 329

Synthesis of (2E,4E,6E)-3-methyl-6-(1,3,3-trimethylindolin-2-ylidene)hexa-2,4-dienenitrile (414)
......................................................................................................................................... 330

Synthesis of (2E,4E,6E)-3-methyl-6-(1,3,3-trimethylindolin-2-ylidene)hexa-2,4-dienal (4-10) 332

Synthesis of (2E,4E,6E,8E)-3-methyl-8-(1,3,3-trimethylindolin-2-ylidene)octa-2,4,6-trienenitrile
(4-19) ......................................................................................................................................... 333

Synthesis of (2E,4E,6E,8E)-3-methyl-8-(1,3,3-trimethylindolin-2-ylidene)octa-2,4,6-trienal (412)
......................................................................................................................................... 334

Synthesis of Azulene-1-carbaldehyde (4-22).............................................................................. 334

Synthesis of (2E,4E)-5-(azulen-1-yl)-3-methylpenta-2,4-dienenitrile (4-23) ............................ 335

Synthesis of (2E,4E)-5-(azulen-1-yl)-3-methylpenta-2,4-dienal (4-20) ..................................... 336

Synthesis of (2E,4E,6E,8E)-9-(azulen-1-yl)-3,7-dimethylnona-2,4,6,8-tetraenenitrile (4-24) .. 337

Synthesis of (4E,6E,8E)-9-(azulen-1-yl)-3,7-dimethylnona-2,4,6,8-tetraenal (4-21) ................ 338

Bibliography .......................................................................................................................... 341


xi

LIST OF TABLES
Table 1-1. Kd and λmax for Retinal Bound to CRABPII Mutants. ............................................... 70

Table 1-2. λmax and Kd of all-trans-retinal with R132K:R111L:L121E mutants having a
negatively charged residue installed at different positions. ......................................................... 78

Table 1-3. λmax and Kd of all-trans-retinal with R132K: R111L:L121E mutants having V76
mutation. ....................................................................................................................................... 82

Table 1-4. λmax and Kd of all-trans-retinal with R132K:R111L:L121E mutants having T56
mutation. ....................................................................................................................................... 83

Table 1-5. λmax and Kd of all trans-retinal with R132K:R111L:L121E mutants having R59
mutation. ....................................................................................................................................... 88

Table 1-6. λmax of all-trans-C17 aldehyde with R132K:R111L:L121E mutants having R59
mutations. ...................................................................................................................................... 92

Table 1-7. λmax and Kd of all-trans-C15 with R132K:R111L:L121E mutants having R59
mutation. ....................................................................................................................................... 96

Table 1-8. λmax and Kd of all-trans-C15 with R132K:R111L:L121E mutants having negative
residues installed at different positions. ..................................................................................... 105

Table 1-9. λmax and Kd of all-trans-C15 with R132K:R111L:L121E mutants having negative
residues installed at different positions. ..................................................................................... 106

Table 1-10. λmax and Kd of all-trans-C15 with R132K:R111L:L121E mutants having mutation at
179

Arg111.

................................................................................................................................... 108


Table 1-11. λmax and Kd of all-trans-C15 with R132K:R111L:L121E mutants having negatively
charged residues installed at different positions. ....................................................................... 111

Table 1-12. λmax and Kd of all-trans-C15 with R132K:R111L mutants..................................... 113

Table 1-13. Relative fluorescence at saturation for different CRABPII mutants. ..................... 123

Table 1-14. Study on the Effect of BTP buffer. .......................................................................... 124

Table 1-15. λmax and Kd of all-trans-C15 with Tyr134 mutants. ............................................... 129

Table 1-16. PCR recipe for CRABPII........................................................................................ 131


xii

Table 1-17. PCR cycle for CRABPII ......................................................................................... 131

Table 1-18. PCR Primers ............................................................................................................ 97

Table 1-19. Recipe for sequencing CRABPII: ........................................................................... 105

Table 1-20. FPLC Method for CRABPII Purification: ............................................................. 107

Table 1-21. Online Desalting Protocol using column (Beta-basic CN, 10 x 1 mm, 5 µm) ....... 109

Table 1-22. Extinction Coefficient of CRABPII mutants ........................................................... 110

Table 2-1. Absorption of CRABPII R59 mutants with C15 aldehyde and all-trans-retinal. ...... 153

Table 2-2. Absorption of CRBPII Mutants bound to all-trans-retinal. ..................................... 155

Table 2-3. Recipe for Sequencing for Gene in Chapter 2.......................................................... 170

TM

Table 2-4. PCR Recipe for Phusion

DNA Polymerase ......................................................... 170


Table 2-5. Primers for gene amplification with restriction site (underline) installed: ............. 171

Table 2-6. Primers for site-directed mutagenesis with mutated codon underline: ................... 171

Table 2-7. PCR Conditions for Site-directed Mutagenesis ....................................................... 172

Table 2-8. PCR Recipe for Cloning into pCWori(+)................................................................. 173

Table 2-9. Gene Digestion for CRABPII/CRBPII and pCWori ................................................. 174

Table 2-10. CIP Recipe for pCWori(+) ..................................................................................... 174

Table 2-11. Ligation Recipe for Cloning into pCWori(+)......................................................... 175

®

Table 2-12. Gene Digestion Recipe for Cloning into pBAD-TOPO 1 .................................... 176

®

Table 2-13. Gene Digestion Recipe for Cloning into pBAD-TOPO 2 .................................... 177

®

Table 2-14. CIP Recipe for Cloning into pBAD-TOPO .......................................................... 177

®

Table 2-15. Gene Ligation Recipe for Cloning into pBAD-TOPO ......................................... 178

Table 2-16. Digestion Recipe for Cloning into pBAD-BCDOX ................................................ 179

Table 2-17. CIP Recipe for Cloning into pBAD-BCDOX ......................................................... 180

Table 2-18. Gene Ligation Recipe for Cloning into pBAD-BCDOX ......................................... 181


xiii

Table 3-1. 11-cis-Retinal and all-trans-retinal tested with different CRABPII mutants with
engineered R132K....................................................................................................................... 210

Table 3-2. The binding constant of different CRABPII mutants with 11-cis-retinal and all-transretinal. ......................................................................................................................................... 215

Table 3-3. Isomerization activity of different CRABPII mutant. ............................................... 217

Table 3-4. Isomerization study of R132K:R111L:L121E with different equivalent of 11-cisretinal. ......................................................................................................................................... 219

Table 3-5. Study the Effect of NaCl on Isomerization in R132K:R111L:L121E with 11-cisretinal. ......................................................................................................................................... 220

Table 3-6. Effect of deuterated solvent on isomerization in R132K:R111L:L121E. ................. 223

Table 3-7. Study of the effect of glycerol on retinal isomerization in CRABPII. ...................... 226

Table 3-8. Study the pH effect on isomerization in CRABPII.................................................... 227

.......

Table 3-9. Study the turnover ability for 11-cis-retinal isomerization in R132K:R111L:F15Y.
..................................................................................................................................................... 229

Table 3-10. Study the turnover ability for 11-cis-retinal isomerization in R132K:R111L:L121E
and WT-CRABPII........................................................................................................................ 230

Table 3-11. Study the effect of T56V and T54V mutation on isomerization activity. ................ 232

Table 3-12. Study of 11-cis-retinal isomerization activity using CRBPII. ................................ 235

Table 3-13. UV-vis study of 11-cis-retinal with CRBPII and its mutants. ................................ 236

Table 3-14. UV-vis study of 11-cis-retinal ring-locked analog with CRABPII mutants. .......... 242

Table 3-15. Binding Study of 11-cis-retinal ring-locked analog 3-18 with CRABPII mutants. 243

Table 3-16. Primers Sequence for CRABPII Mutation in Chapter 3 ........................................ 248

Table 4-1. UV-vis absorption of mero-cyanines with CRABPII mutant (R132K: R111L: L121E).
..................................................................................................................................................... 298

Table 4-2. UV-vis absorption of merocyanine 4-10 with CRABPII mutant. ............................. 301

Table 4-3. UV-vis absorption of merocyanine 4-10 with CRABPII mutant. ............................. 303

Table 4-4. λmax of merocyanine 4-10 PSB in different solvent and corresponding polarity of the
solvent. ........................................................................................................................................ 306


xiv

Table 4-5. UV-vis absorption of merocyanine 4-11 with CRABPII mutant
(R132K:R111L:L121E). .............................................................................................................. 308

Table 4-6. Quantum yield of merocyanine 4-10 PSB and merocyanine 4-10 with CRABPII
mutant (R132K:R111L:L121E)................................................................................................... 312

Table 4-7. UV-vis study of azulenic aldehyde 4-20 with CRABPII mutant
(R132K:R111L:L121E). .............................................................................................................. 318
Table 4-8. UV-vis study of azulenic aldehyde 4-20 with CRABPII mutant (R132K:R111L). ... 325

Table 4-9. Primers for Mutagenesis in Chapter 4 ..................................................................... 320


xv

LIST OF FIGURES
Figure 1-1. Crystal Structures of Human A2A Adenosine Receptor (Green) and Bovine
Rhodopsin (Cyan) showing the 7- helical domains. (For interpretation of the references to
color in this and all other figures, the reader is referred to the electronic version of this thesis
(or dissertation).) ............................................................................................................................ 1

Figure 1-2. Structure of 11-cis-retinal. ......................................................................................... 2

Figure: 1-3. Anatomy of eye. ......................................................................................................... 4

Figure 1-4. Structure of rod cell and cone cell. ............................................................................ 5

Figure 1-6. Binding site of rhodopsin with Lys296 forming PSB with retinal and counterion
Glu113............................................................................................................................................. 6

Figure 1-5. Stereoview of rhodopsin with retinal. ........................................................................ 6

Figure 1-7. Retinals found in different organisms. ....................................................................... 7

Figure 1-8. Photoisomerization Cycle of Rhodopsin. ................................................................... 8

Figure 1-9. Signal Transduction Cascade Upon Photo Absorption by Rhodopsin (Rh:
Rhodopsin; PDE: Phosphodiesterase; Gt- : Transducin- subunit; Gt: Transducinsubunit; GDP: Guanosine Diphosphate; GTP: Guanosine Triphosphate; cGMP: cyclic
Guanine monophosphate; GMP: Guanine monophosphate. ........................................................ 10

Figure 1-10. UV-vis spectrum of rod and pigmented rhodopsins. .............................................. 11

Figure 1-11. Absorption maxima of 11-cis-retinal and its Schiff base and protonated Schiff
base. .............................................................................................................................................. 12

Figure 1-12. Proposed mechanism for wavelength regulation in rhodopsin.............................. 13

Figure 1-13. 11-cis-Retinal, dihydroretinal and tetrahydroretinal analogs with absorption
before and after incubation with rhodopsin and the opsin shift. .................................................. 15

Figure 1-14. Retinal analogs with different charged group along the polyene and the
a
b
corresponding UV-vis absorption ( in ethanol; treated with Et3N in ethanol). ........................ 16

Figure 1-15. Crystal structure of rhodopsin (11-cis-retinal in dark blue) and models of color
rhodopsins (11-cis-retinal in cyan for blue rhodopsin, 11-cis-retinal in green for green
rhodopsin and 11-cis-retinal in red for red rhodopsin) with closed residues in different colors. 18

Figure 1-16. Scheme shows the relative contributions from the counterion E113, rhodopsin (Rh)
and its residues (at certain distance around bound 11-cis-retinal PSB) on modulating the energy

xvi

of bound chromophore based on Complete Active Space Second-Order Perturbation (CASPT2).
The blue arrow indicates the blue shift induced by the counterion. The red arrow shows the red
shift that is caused by the protein around 3.0 to 3.5 Å from the bound retinal PSB. ................... 20

Figure 1-17. C6, C7-Ring-lock analog of all-trans-retinal with corresponding max of their
PSB and bound to bacteriorhodopsin. .......................................................................................... 22

Figure 1-18. Stereoview of different microbial rhodopsins with the corresponding absorptions.
....................................................................................................................................................... 24

Figure 1-19. Signal transduction of sensory rhodopsin. (sRhII: Sensory Rhodopsin II; HtrII:
Transducer; CheA: Histidine Kinase). ......................................................................................... 25

Figure 1-20. Important residues around sensory rhodopsin II (Blue) and bacteriorhodopsin
(Green) with residues different between bacteriorhodopsin and sensory rhodopsin II labeled
(residues in parenthesis correspond to bacteriorhodopsin). ....................................................... 27

Figure 1-21. Crystal structure of bacteriorhodopsin (Green) and sensory rhodopsin II (Cyan)
with the water-mediated interaction between Arg82 (Bacteriorhodopsin) or Arg72 (Sensory
Rhodopsin II) and PSB of bound all-trans-retinal........................................................................ 28

Figure 1-22. Stereoview of CRABPII (Green) with retinoic acid (Cyan) ................................... 31

Figure 1-23. Crystal structure of fatty acid binding protein (Green) with 1-anilinonaphthalene
8-sulfonic acid (Blue).................................................................................................................... 32

Figure 1-24. Structure of all-trans-retinal (1-17), all-trans-retinoic acid (1-20) and all-transretinol (1-21) ................................................................................................................................. 32

Figure 1-26. Retinoic acid binding site in CRABPII with interaction of Glu73 with Trp109 and
Arg111 shown. .............................................................................................................................. 33

Figure 1-25. Crystal structure of CRABPII with retinoic acid (Blue) with important interactions
shown in dash lines. ...................................................................................................................... 33

Figure 1-30. Binding site of mutant R132K:R111L:L121E with bound all-trans-retinal PSB
stabilized by counterion Glu121. .................................................................................................. 37

Figure 1-29. Acid-base titration of mutant R132K:R111L:L121E with all-trans-retinal
monitored by UV-vis spectroscopy and the calculated pKa of the bound retinal PSB in mutant
R132K:R111L:L121E. .................................................................................................................. 37

Figure 1-31. Scheme for reductive amination and the mass spectrum of mutant
R132K:R111L:L121E and mutant R132K:R111L:L121E with all-trans-retinal after reductive
amination with an adduct peak shown (+ 268 m/z). ..................................................................... 38

Figure 1-32. Binding site of rhodopsin with 11-cis-retinal showing a trans-imine. ................... 39


xvii

Figure 1-33. Electrostatic potential calculation (APBS suite) of blue, rod, green, and red opsin
(calculations of blue, green, and red are based on available homology models) projected on the
Van der Waals surface of retinal (the electrostatic potential of the retinylidene chromophore was
set to zero as to only illustrate the electrostatic contribution of the protein). The chromophore is
divided into three segments; the qualitative average score for each segment represents the
overall electrostatic potentials that lead to the wavelength regulation of each opsin. ................ 41

Figure 1-34. Stereoview of R132K:R111L:L121E with residues that are 4 Å away from the
bound all-trans-retinal PSB in cyan color .................................................................................... 43

Figure 1-35. Position of Arg59, Ala32, Thr56 and Val76 relative to the bound all-trans-retinal
in mutant R132K:R111L:L121E. .................................................................................................. 44

Figure 1-36. Left: UV-vis spectrum of all-trans-retinal titrated with CRABPII mutant
R132K:R111L:L121E:T56R; Right: Deconvolution of the UV-vis spectrum of all-trans-retinal
with CRABPII mutant R132K:R111L:L121E:T56R. .................................................................... 45

Figure 1-37. Stereoview of CRABPII mutant R132K:R111L:L121E crystal structure with Ala36,
Ala32 and Phe15 shown (Cyan). .................................................................................................. 46

Figure 1-38. Crystal structure of CRABPII mutants F15W (Green), R132K:R111L:A32E
(Purple) and WT-CRAPBII. .......................................................................................................... 47

Figure 1-39. Stereoview of model mutant R132K:R111L:L121E with T56E and V76E mutations
shown. ........................................................................................................................................... 48

Figure 1-40. Model of CRABPII mutants R132K:R111L:L121E: T56D (left) and
R132K:R111L:L121E:T56E (right). ............................................................................................. 50

Figure 1-41. Sequence alignment of CRABPII with proteins belong to Fatty Acid Binding
Protein (FABP); hCRABPI: human CRABPI; hCRABPII: human CRABPII; rCRBPI: rat
Cellular Retinol Binding Protein (CRBP) I; rCRBPII: rat CRBPII; hCRBPIII: human CRBPIII;
hCRBPII: human CRBPII; zCRBP: zebra CRBP; hCRBPIV: human CRBP IV; hAFABP: human
adipocyte FABP; mFABPIV: mouse FABP IV; hBFABP: human Brain-type FABP; hMFABP:
human muscle FABP; rIFABP: rat Intestinal FABP; hLFABP: human Liver FABP; zLFABP:
zebra Liver FABP; hFABPI: human FABP I. ............................................................................... 51

Figure 1-42. Crystal structure of R132K:R111L:L121E with all-trans-retinal PSB (Purple) and
Arg59 shown. ................................................................................................................................ 52

Figure 1-43. UV-vis spectra of R132K:R111L: L121E:R59E titrated with all-trans-retinal. .... 52

Figure 1-44. Stereoview of overlay between apo-structure of CRABPII mutant
R132K:R111L:L121E and model of CRABPII mutant R132K:R111L:L121E:V76E. .................. 54

Figure 1-45. Left: Overlay of holo-structure between CRABPII (Green) and chicken LFABP
(Cyan); Right: Overlay of apo-structure between CRABPII (Green) and chicken LFABP (Cyan).
....................................................................................................................................................... 54

xviii

Figure 1-46. Crystal structure of all-trans-retinal with CRABPII mutant R132K:R111L:L121E
with protein surface (Black) and solvent exposed -ionone ring (Cyan) shown. ....................... 55

Figure 1-47. Electrostatic potential calculation using APBS on CRABPII mutant
R132K:R111L:L121E. Left: Calculation based on vacuum environment; Right: Calculation
based on water environment (dielectric constant 78 was used). .................................................. 55

Figure 1-48. All-trans-retinal 1-17 and its short analogs (C17 aldehyde 1-22 and C15 aldehyde
1-23). ............................................................................................................................................. 55

Figure 1-49. Overlay between holo-structure of R132K:R111L:L121E with all-trans-retinal
(Purple) and energy minimized models of R132K:R111L:L121E with C17 aldehyde (Green) and
C15 aldehyde (Blue) (protein surface in black shadow). .............................................................. 56

.

Figure 1-51. Absorption spectrum of R132K:R111L:L121E with C15 aldehyde 1-23 (Right) and

C17 aldehyde 1-22 (Left) showing the PSB formation. ................................................................. 59

Figure 1-52. Overlay between crystal structure of R132K:R111L:L121E with all-trans-retinal
(Purple) and model of R132K:R111L:L121E with C17 aldehyde (Green). .................................. 60

Figure 1-53. Overlay between crystal structure of R132K:R111L:L121E with all-trans-retinal
(Purple) and model of R132K:R111L:L121E with C15 aldehyde (Blue). ..................................... 60

Figure 1-54. UV-vis spectra of mutant R132K:R111L:L121E:R59E titrated with C15 aldehyde.
....................................................................................................................................................... 62

Figure 1-55. Deconvoluted spectrum of R132K:R111L:L121E with R59 mutations.................. 63

Figure 1-56. Stereoview of overlay between crystal structure of R132K:R111L:L121E with alltrans-retinal (Cyan) and R132K:R111L:L121E:R59W with C15 aldehyde (C15) (Green) ........... 64

Figure 1-57. Overlay of crystal structures between all-trans-retinal in R132K:R111L:L121E
and C15 aldehyde in R132K:R111L:L121E:R59W. ...................................................................... 65

Figure 1-58. Crystal structure of R132K:R111L:L121E:R59W with C15 aldehyde with Glu121
and BTP shown. ............................................................................................................................ 65

Figure 1-59. Overlay of R132K:R111L:L121E (Blue) and R132K:R111L:L121E:R59E (Cyan)
with holo-structure of R132K:R111L:L121E:R59W (Green). All three structures contain BTP at
the same position........................................................................................................................... 66

Figure 1-60. Right: Overlay of R132K:Y134F:R111L:L121E apo-structure (Purple) with
R132K:R111L:L121E:R59W holo-structure (Green) with ordered water shown in purple sphere;
Left: Overlay of R132K:Y134F:R111L:L121E:T54V apo-structure (Cyan) with

xix

R132K:R111L:L121E:R59W holo-structure (Green) with ordered waters shown in Cyan spheres.
....................................................................................................................................................... 67

Figure 1-61. Electrostatic potential around bound C15 aldehyde in different CRABPII mutants
(Left: R132K:R111L:L121E:R59; Middle: R132K:R111L:L121E:R59W; Right:
R132K:R111L:L121E:R59E) calculated by APBS package with the corresponding absorption
shown. ........................................................................................................................................... 68

Figure 1-62. Crystal structure of R132K:R111L:L121E:R59W with C15 aldehyde. Negative
charges are installed at the position highlighted in blue.............................................................. 69

Figure 1-63. Crystal structure of C15 aldehyde bound to mutant R132K:R111L:L121E:R59W. ...
....................................................................................................................................................... 72

Figure 1-64. Left: Overlay between holo-structure of R132K:R111L:L121E:R59W with C15
aldehyde (Cyan) and apo-structure of R132K:R111:L121E (Yellow) with Glu121 showed in two
conformations (Conf.): Glu Conf. A overlapped with Glu121 in R132K:R111L:L121E:R59W;
Glu Conf. B hydrogen bonding with Arg 111 in R132K:R111:L121E; Right: The crystal structure
of R132K:R111LE:R59W with BTP showed. The ordered water is not in the plane of PSB. ...... 74

Figure 1-65. Electrostatic potential around the bound C15 aldehyde in different CRABPII
mutants (Left: R132K:R111L:L121E; Middle: R132K:R111L:L121D; Right:
R132K:R111L:L121Q) calculated by the APBS package. ............................................................ 76

Figure 1-66. Stereoview of crystal structure of R132K:R111L:A32E with C15 aldehyde (Green).
....................................................................................................................................................... 79

Figure 1-67. Stereoview of overlay holo-structure of R132K:R111L:L121E:R59W (Green) and
R132K:R111L:A32E (Cyan) with bound C15 aldehyde. ............................................................... 80

Figure 1-68. Overlay of holo-strcuture of R132K:R111L:A32E (Cyan) and R132K:R111L:
L121E:R59W (Green) with bound C15 aldehyde (C15) (in R132K:R111L:A32E: Orange; in
R132K:R111L:L121E:R59W: purple) and BTP (Purple)............................................................. 80

Figure 1-69. Bound C15 aldehyde PSB (Orange) stabilized by the carbonyl group of Ser37. ... 81

Figure 1-70. The optimum CNO angle for PSB stabilization by counterion is 108˚. ................. 81

Figure 1-71. Relative position of Arg59 and Phe15 with C15 aldehyde in
R132K:R111L:L121E:A32E. ........................................................................................................ 82

Figure 1-72. PSB stabilization in R132K:R111L:T54E by water-mediated interactions with
Ala36 and Ser12............................................................................................................................ 82


xx

Figure 1-73. Comparison of PSB environment in R132K:R111L:L121E:R59W and
R132K:R111L:A32E. .................................................................................................................... 83

Figure 1-74. Overlay of the bound C15 aldehyde in R132K:R111L:T54E (Green) with bound
C15 aldehyde in R132K:R111L:A32E (Orange). .......................................................................... 84

Figure 1-75. Overlay of all-trans-retinal (Rt) in R132K:R111L:L121E (Blue), C15 aldehyde
(C15) in R132K:R111L:L121E:R59W (Purple) and C15 in R132K:R111L:T54E (Green) with Glu
121 (Purple) in R132K:R111L:L121E and R132K:R111L:L121E:R59W and Leu121 (Green) in
R132K:R111L:T54E shown. ......................................................................................................... 85

Figure 1-76. Fluorescence quenching of C15 aldehyde with R132K:R111L:L121E:R59W (top
curve) and R132K:R111L:T54E (bottom curve)........................................................................... 86

Figure 1-77. Position of Trp (Cyan) in CRABPII relative to bound C15 aldehyde in
R132K:R111L:A32E (Green) and in R132K:R111L:L121E:R59W (Purple). .............................. 87

Figure 1-78. a) C15 aldehyde with R132K:R111L:L121E:R59E in BTP buffer at pH 7; b) C15
aldehyde with R132K:R111L:L121E:R59E in PBS buffer at pH 7; c) C15 aldehyde with
R132K:R111L:L121E:R59W in BTP buffer at pH 7; d) C15 aldehyde with
R132K:R111L:L121E:R59W in PBS buffer at pH 7. .................................................................... 90

Figure 1-79. Dissociation constant of BTP with R132K:R111L:L121E:R59W (Right) and
R132K:R111L:L121E:R59E (Left) calculated based on the absorbance change of UV-vis
spectrum of C15 aldehyde with CRABPII mutant titrated by BTP. ............................................... 92

Scheme 1-2. Proposed mechanism of Schiff base formation in R132K:R111L:L121E with alltrans-retinal. ................................................................................................................................. 93

Figure 1-80. Deconvoluted spectrum of C15 aldehyde with R132K:R111L:L121E:R59D. ...... 111

Figure 2-1. Crystal structure (Right) and anaglyph stereo picture (Right) of Nacetylneuraminate lyase (Left) showing the active site lysine (Purple) and mutated residues
(Blue)........................................................................................................................................... 139

Figure 2-2. Crystal structure of bacteriorhodopsin with hightlighted residues around the bound
retinal PSB (Residues <5 Å away are shown in purple while residues >5 Å away are shown in
cyan)............................................................................................................................................ 140

Figure 2-3. Crystal structure of bacteriorhodopsin (Green) and sensory rhodopsin II (Cyan)
with the water-mediated interaction between Arg82 (Bacteriorhodopsin) or Arg72 (Sensory
Rhodopsin II) and PSB of bound all-trans-retinal...................................................................... 141

Figure 2-4. Scheme for directed evolution. ............................................................................... 142


xxi

Figure 2-5. Human pancreatic lipase hydrolyzed triacylglycerol to diacylglycerol and fatty acid
regiospecifically. ......................................................................................................................... 143

Figure 2-6. Scheme for DNA shuffling. ..................................................................................... 145

Figure 2-8. Overlay of C15 aldehyde with CRABPII mutants (R132K:R111L: L121E:R59W)
(Blue) and all-trans-retinal with CRABPII mutants (R132K:R111L: L121E) (Green). ............. 146

Figure 2-7. Scheme describing a screening system for subtilisin BPN’. .................................. 146

Figure 2-9. Overlay crystal structure of CRBPII (Green) with all-trans-retinol and CRABPII
(Blue) with all-trans-retinal. ....................................................................................................... 147

Figure 2-10. Crystal structure of CRBPII with all-trans-retinol (Blue) and the anaglyph picture
on the right. ................................................................................................................................. 148

Figure 2-11. Structure of all-trans-retinol and all-trans-retinal. ............................................. 148

Figure 2-12. Crystal structures of all-trans-retinal in CRBPII (Green, left) and all-trans-retinal
PSB in CRABPII mutant (R132K:R111L:L121E, cyan, right). .................................................. 149

Figure 2-13. Titration of CRBPII mutants Q108K:K40L with all-trans-retinal. ...................... 150

Figure 2-14. Electrostatic potential projected on 11-cis-retinal in rod rhodopsin and color
rhodopsins using APBS Package. Top: Electrostatic potential calculation from the residues >5
Å away from the retinal binding site; Middle: residues <5 Å away from the retinal binding site;
Bottom: Electrostatic potential calculation of whole protein. .................................................... 151

Figure 2-15. Projection of electrostatic potential by CRBPII mutants on all-trans-retinal (Left:
Q108K:K40L; Right: Q108K:K40L:D61R:D63R:T120R:E118R). ............................................ 152

Figure 2-16. Proposed in vivo color screening method for CRBPII mutants. .......................... 154

Figure 2-17. Biosynthesis of

-carotene in vivo. ..................................................................... 155


Figure 2-18. MEP Pathway in E. coli. ...................................................................................... 156

Figure 2-19. Proposed mechanism of

-carotene cleavage into all-trans-retinal. ................. 157


Figure 2-20. Elucidation of the mechanism of

-carotene cleavage using labeled oxygen. ... 158


Figure 2-21. in vivo test of pOrange in different bacteria strain. ............................................. 159

Figure 2-22. HPLC trace for chromophore a) Extracted from XL-1 Blue transformed with
pOrange; b) Extracted from XL-1 Blue transformed with pOrange and pBAD-BCDOX-CRBPII
(KS). ............................................................................................................................................ 160

Figure 2-23. Transcription mechanism of PBAD promoter. ...................................................... 161

xxii

Figure 2-24. Cell pellet of XL-Blue transformed with pOrange + pBAD-BCDOX (Left: with
addition of L-arabinose; Right: without addition of L-arabinose). ............................................ 162

Figure 2-25. Cloning strategy for CRBPII. ............................................................................... 163

Figure 2-26. Cloning strategy for plasmid containing BCDOX and CRBPII (pBAD-BCDOXCRBP). ........................................................................................................................................ 165

Figure 2-27. a) SDS-PAGE analysis of protein expression of BCDOX and CRBPII in XL-1 Blue
(Lane 1: XL-1 Blue; Lane 2: XL-1 Blue expressed BCDOX (with addition of L-arabinose); Lane
3: XL-1 Blue expressed CRBPII (with addition of IPTG); Lane 4: bio-rad SDS-PAGE size
marker (Broad-range)(From Bottom: 6.5 KDa, 14.4 KDa, 21.5 KDa, 31 KDa, 45 KDa, 66.2
KDa, 97.4 KDa, 116.3 KDa, 200 KDa); Lane 5: Pure CRABPII; b) expand image of SDS-PAGE
around 15 KDa; pointer indicated CRBPII protein. .................................................................. 166

Figure 2-28. Absorption of CRBPII (Q108K:K40L) at pH 7.15 and pH 8.23. ......................... 167

Figure 2-29. Cell pellets of XL-1 Blue transfected with pOrange and pBAD-BCDOX-CRBPII
(Left: without inducer; Middle: Addition of IPTG; Right: Addition of L-arabinose)................. 167

Figure 2-30. Cell pellet of XL-1 Blue/pOrange with CRPBII (KS), WT-CRBPII, WT-CRABPII or
CRABPII (KLE) expressed together with BCDOX. .................................................................... 168

Figure 2-31. Absorption of all-trans-retinal with CRBPII mutant (Q108K:K40S) at pH 7.73. 169

Figure 3-1. Isomerization changes the geometrical relationship with surrounding area. ....... 196

Figure 3-2. Schematic diagram of cell membrance. ................................................................. 197

Figure 3-3. Isomerization leads to a better pack cell membrane. ............................................. 197

Figure 3-4. Chromophores in different light sensing proteins. 3-1. phytochromobilin in
phytochromes where R1 is the repeating unit of the shown dipyrrole; 3-2. Retinal in rhodopsins
(isomerization occurs at C11-C12 in vertebrates and invertebrates rhodopsin and at C13-C14 in
microbial rhodopsin; 3-3. coumaric acid in xanthopsins. ......................................................... 198

Figure 3-5. Schematic diagram of the domain included in bacterial phytochrome Cph1 and its
response regulator Rcp1. GAF: Chromophore binding domain; HKD: histidine kinase domain.
..................................................................................................................................................... 199

Figure 3-6. Crystal structure of bacteriophytochrome bound to the chromophore (Cyan)
through thio-linkage.................................................................................................................... 200

Figure 3-7. Stereoview of photoactive yellow protein with trans-p-coumaric acid shown....... 201

Figure 3-8. Schematic diagram for photoactive yellow protein light cycle. ............................. 202

Figure 3-9. Structure of rod cell................................................................................................ 203


xxiii

Figure 3-10. 11-cis-Retinal cycle in retina. (scALDH stands for short-chain alcohol
dehydrogenase/reductase; LRAT stands for lecithin retinol acyl transferase.) ......................... 204

Figure 3-11. Proposed addition-elimination mechanism for isomerization in RPE65. ............ 206

Figure 3-12. Second proposed mechanism for isomerization in RPE65................................... 207

Figure 3-13. Amine analogs of retinal as inhibitors for RPE65. .............................................. 207

Figure 3-14. Stereoview of RPE65 colored according to secondary structure (Helix: Cyan; sheet: Purple; Loop: Pink). The hydrophobic tunnel is located in the middle of the structure. 208

Figure 3-15. Iron binding site in RPE65 with histidine coordinated to iron shown (Red) ....... 208

Figure 3-16. Proposed mechanism involves iron activated deacylation followed by
isomerization and hydration of the cationic intermediate. ......................................................... 209

Figure 3-17. Schematic diagram of squid photoreceptor. ........................................................ 211

Figure 3-18. Schematic diagram of squid vision system. .......................................................... 212

Figure 3-19. Design principle for reengineering CRABPII to all-trans-retinal isomerase. ..... 214

Figure 3-20. HPLC trace of purified 11-cis-retinal. ................................................................. 215

Figure 3-21. Acid base titration of 11-cis-retinal (11zRt) with CRABPII a) mutant
R132K:R111L:L121E (KLE) and b) R132K:R121E (KE). ......................................................... 218

Figure 3-22. Stereoview of overlay of crystal structure of R132K:R111L:T54E with C15
aldehyde (C15) and 11-cis-retinal (11zRt) docked model. .......................................................... 219

Figure 3-23. Structure and corresponding length of all-trans-retinal (3-2), 11-cis-retinal (3-16)
and C15 aldehyde (3-17). ............................................................................................................ 220

Figure 3-24. The distance between Lys132 and the Phe65 and Ile63 (Purple) in crystal structure
R132K:R111L:T54E with C15 aldehyde. .................................................................................... 221

Figure 3-25. HPLC trace of a) 11-cis-retinal; b) 11-cis-retinal with R132K: R111L:L121E after
30 mins of incubation in dark at rt; c) 11-cis-retinal with R132K: R111L:F15Y after 30 mins of
incubation in dark at rt. .............................................................................................................. 223

Figure 3-26. Proposed mechanism of isomerization in CRABPII. ........................................... 228

Figure 3-28. Ordered water molecules within the retinal binding pocket in
R132K:R111L:L121E. ................................................................................................................ 229

Figure 3-27. Hydrogen bonding network around the bound 11-cis-retinal in rhodopsin. ....... 229


xxiv

Figure 3-29. Rate of isomerization of R132K:R111L:F15Y in PBS buffer and deuteriated PBS
buffer. .......................................................................................................................................... 231

Figure 3-30. Rate of SB formation in deuterated PBS buffer and PBS buffer. ......................... 232

Figure 3-31. Ordered water network in CRABPII mutant R132K:R111L:T54E. ..................... 233

Figure 3-32. Stereoview of 11-cis-retinal (Green) model showing Thr54 and Thr56 (Purple)
closed to C11-C12 of the bound 11-cis-retinal. .......................................................................... 238

Figure 3-33. Model of 11-cis-retinal with R132K: R111L:L121E with Ser12 and Ser37 showed
(Purple). ...................................................................................................................................... 240

Figure 3-34. UV-vis spectra of R132K:R111L: F15Y:S12D and R132K:R111L:F15Y:S37D with
11-cis-retinal. .............................................................................................................................. 240

Figure 3-35. Crystal structure of retinal with CRBPII with the ordered water inside the binding
site shown. ................................................................................................................................... 241

Figure 3-36. Crystal structure of CRBPII with bound all-trans-retinal (tRt) (Blue) and docked
11-cis-retinal (11zRt) (Green). ................................................................................................... 241

Figure 3-37. Electrostatic potential calculation on CRBPII mutants Q108K:K40L (Left) and
Q108K:K40L:T51V:T53C (Right) using APBS package. ........................................................... 243

Figure 3-38. Overlay model of docked 11-cis-retinal in CRBPII with holo-structure of all-transretinal with CRBPII. ................................................................................................................... 244

Figure 3-39. UV-vis spectra of 11-cis-retinal with R132K:R111L:T51V:T53C:R58W. ........... 245

Figure 3-40. UV-vis spectra of ring-lock analog 3-18 with mutant R132K:R111L:F15Y and
mutant R132K:R111L:L121E. .................................................................................................... 248

Figure 3-41. Acid-base titration of 11-cis-retinal with R132K:L121E and R132K:R111L:L121E.
..................................................................................................................................................... 249

Figure 3-42. Crystal structure of cellular retinal binding protein (left) and the 11-cis-retinal
binding site (right). ..................................................................................................................... 252

Figure 4-1. Schematic diagram of His-tag bound with Nickel (II) nitrilotriacetic acid (Ni-NTA).
..................................................................................................................................................... 286

Figure 4-2. Crystal structure of chimeric protein 2-Adrenergic G protein-coupled receptor (
2AR) (Green) and lysosome (T4L) (Cyan). ............................................................................. 287

Figure 4-3. Schematic spectra of green fluorescence protein. (Blue: absorption spectrum;
Green: fluorescent spectrum) ..................................................................................................... 288


xxv

Figure 4-5. Mechanism of fluorophore 4-5 formation in green fluorescent protein................. 290

Figure 4-6. Two forms of fluorophore in green fluorescent protein. ........................................ 291

Figure 4-7. Fluorophore in green fluorescent protein showing Glu222 closed to Ser65 and
Tyr66. .......................................................................................................................................... 291

Figure 4-8. a) Schematic diagram for FRET. Left: Light irradiation of fluorescent donor;
Right) Fluorescent donor is excited and the energy is transferred to fluorescent acceptor nonradiantly through excited dipole-dipole interaction between fluorescent donor and acceptor.
Fluorescence of fluorescent donor disappears and fluorescent acceptor is excited and emits red
fluorescence. b) Sketch spectra shows the fluorescent spectrum from fluorescent donor
overlapping with the absorption spectrum of fluorescent acceptor. c) Sketch graph shows the
fluorescent intensity of fluorescent acceptor is indirectly proportional to the distance between
fluorescent donor and acceptor (r). ............................................................................................ 293

2+

Figure 4-9. Scheme for Ca

detection using GFP. ................................................................. 294


Figure 4-10. Schematic diagram of fluorescent activated cell sorting system.......................... 295

Figure 4-11. Strategy for engineered CRABPII as chromophoric or fluorescent protein. R
stands for any heterocycle. ......................................................................................................... 297

Figure 4-12. Structure of cyanine dye (4-8) and merocyanine (4-9). ....................................... 299

Figure 4-13. Joblonski diagram. ............................................................................................... 299

Figure 4-14. Structure of merocyanine 4-10 and 4-11 with the corresponding max and max
after bound with bacterio-rhodopsin. ......................................................................................... 300

Figure 4-15. Schematic diagram of formation cyanine dye upon binding of merocyanine with
bacteriorhodopsin as a PSB........................................................................................................ 301

Figure 4-16. UV-vis spectra of merocyanine 4-10 with CRABPII mutant R132K:R111L:L121E.
..................................................................................................................................................... 305

Figure 4-17. Solvent exposed lysines in R132K: R111L:L121E. .............................................. 306

Figure 4-18. UV-vis spectrum of mutant R132L:R111L:L121E incubated with a) merocyanine
4-10 and b) merocyanine 4-11 in PBS buffer at room temperature. .......................................... 306

Figure 4-19. UV-vis spectrum of R132K:R111L:L121E:R59E with a) merocyanine 4-10 and b)
merocyaine 4-11.......................................................................................................................... 307

Incubation at room temperature in PBS buffer (pH = 7) ........................................................... 308

Figure 4-20. Merocyanine 4-10 binding site in mutant R132K:R111L:L121E:R59W. ........... 308


xxvi

Figure 4-21. a) Overlay crystal structure of R132K:R111L:L121E:R59W with C15 aldehyde
(blue) and R132K:R111L:L121E:R59W with merocyanine 49 (purple); b) Overlay crystal
structure of R132K:R111L:L121E with all-trans-retinal (cyan) and R132K:R111L:L121E:R59W
with merocyanine 4-10 (purple).................................................................................................. 309

Figure 4-22. a) Crystal structure of C15 aldehyde with R132K:R111L:L121E:R59W; b) Crystal
structure of merocyanine 4-10 with R132K:R111L:L121E:R59W. ............................................ 309

Figure 4-24. Merocyanine binding site in R132K:R111L:L121E:R59W. The positions with
tryptophan were engineered are shown (Cyan) with the corresponding max. ........................ 311

Figure 4-23. UV-vis spectra of merocyanine 4-10 with R132K:R111L. ................................... 311

Figure 4-26. Emission spectra of different mutants excited at 283 nm. (KLE stands for
R132K:R111L:L121E). ............................................................................................................... 312

Figure 4-28. UV-vis spectra of merocyanine 4-10 PSB with different solvents ........................ 314

Figure 4-27. UV-vis spectra of merocyanine 4-10 and merocyanine 4-10 with
R132K:R111L:L121E (KLE) mutants having tryptophan installed at different position (Figure 424). .............................................................................................................................................. 314

Figure 4-29. UV-vis spectra of merocyanine 4-11 with mutants R132K:R111L:L121E:V76W
(KLE:V76W), R132K:R111L:L121E:T56W (KLE:T56W) and R132K:R111L:L121E:R59W
(KLE:R59W). .............................................................................................................................. 316

Figure 4-30. Fluorescent spectra of merocyanine 4-10 and its PSB excited at different
wavelength and different concentration...................................................................................... 316

Figure 4-31. Emission Spectra of merocyanine 4-10, merocyanine 4-10 PSB and merocyanine
4-10 with mutant R132K:R111L:L121E:R59W (R132K:R111L: L121E:R59W) excited at 570 nm.
..................................................................................................................................................... 317

Figure 4-32. Fluorescence of mutant R132K:R111L: L121E:R59W (KLE:R59W),
R132K:R111L:L121E: T56W (KLE:T56W) and R132K:R111L:L121E:V76W (KLE:V76W) with
merocyanine 4-10 irradiated at 571 nm. .................................................................................... 318

Figure 4-33. Fluorescence of merocyanine 4-10 PSB in buffer with different amount of glycerol.
..................................................................................................................................................... 319

Figure 4-34. Azulene and its resonance form. .......................................................................... 320

Figure 4-35. Structure of azulenic aldehyde chromophore and its resonance structure after
bound with bacteriorhodopsin. ................................................................................................... 320

Figure 4-36. UV-vis spectra of azulenic aldehyde 4-20 with R132K:R111L:L121E. ............... 324


xxvii

Figure 4-37. UV-vis spectrum of R132K:R111L:L121E with azulenic aldehyde 4-20 at pH 6.8.
..................................................................................................................................................... 325

Figure 4-38. Model of azulenic aldehyde 4-20 with R132K:R111L:L121E:R59W max of the
glutamate at 36 and 59 mutants is showed in parenthesis.......................................................... 326

Figure 4-39. Crystal structure of C15 aldehyde in alternate binding pocket with KL:T54E
showing Arg59 position. ............................................................................................................. 326

Figure 4-40. Crystal structure of C15 aldehyde in R132K:R111L:T54E with Thr54 and Phe15
shown. ......................................................................................................................................... 328


xxviii

LIST OF SCHEMES
Scheme 1-1. Synthesis of C15 aldehyde 1-23 and C17 aldehyde 1-22. ........................................ 58

Scheme 3-1. Synthesis of 11-cis-retinal..................................................................................... 216
Scheme 3-2. Synthesis of ring-locked 11-cis-retinal analog 3-18. ............................................ 247

Scheme 4-1A. Synthesis of merocyanine 4-10, 4-11 and 4-12. ................................................. 303

Scheme 4-1B. Synthesis of merocyanine 4-10, 4-11 and 4-12. ................................................. 304

Scheme 4-2. Synthesis of azulenic aldehyde chromophore 4-20. .............................................. 322

Scheme 4-3. Synthesis of azulenic aldehyde chromophore 4-21. .............................................. 323


xxix

KEY TO SYMBOLS OR ABBREVIATIONS
11zRt

11-cis-retinal

Å

angstrom

ºC

degree Celsius

εr

dielectric constant

λ

lamba

%

percentage

∠

angle

Amino Acids
Ala, A

Alanine

Arg, R

Arginine

Asn, N

Asparagine

Asp, D

Aspartic acid

Cys, C

Cysteine

Gln, Q

Glutamine

Glu, E

Glutamic acid

His, H

Histidine

Leu, L

Leucine

Lys, K

Lysine

Phe, F

Phenylalanine

Pro, P

Proline

Thr, T

Threonine

Trp, W

Tryptophan

xxx

Tyr, Y

Tyrosine

Val, V

Valine

Abs

absorbance

Amp

Ampicillin

APBS

Adaptive Poisson-Boltzmann Solver

BC

Before Christ

BCDOX

β-carotene deoxygenases

bRh

bacteriorhodopsin

BTP

1,3-bistris(hydroxy-methyl)methyl-amino)propane

ca.

approximately

cGMP

cyclic guanine monophosphate

Clm

Chloramphenicol

cm

centimeter

CNO

carbon-nitrogen-oxygen

CRABPI

cellular retinoic acid binding protein I

CRABPII

cellular retinoic acid binding protein II

CRALBP

cellular retinal binding protein

CRBPII

cellular retinol binding protein II

Cti

cis-trans isomerase

d.d.

distillation deionization

Desol.

desolvation

DIBAL-H

diisobutylaluminium hydride

DNA

deoxyribonucleic acid

xxxi

Eq.

equivalent

EtOAc

ethyl acetate

EtOH

Ethanol

FABP

Fatty Acid Binding Protein

FPLC

fast protein liquid chromatography

GFP

Green fluorescent protein

GMP

guanine monophosphate

GPCRs

G-protein coupled receptors

G-protein

guanine nucleotide-binding protein

GTP

guanine triphosphate

h

hour(s)

H2 O

Water

hAFABP

human adipocyte fatty acid binding protein

hBFABP

human Brain-type fatty acid binding protein

hCRABPI

human cellular retinoic acid binding protein I

hCRABPII

human cellular retinoic acid binding protein II

HOMO

highest occupied molecular orbital

hCRBPII

human cellular retinol binding protein II

hCRBPIII

human cellular retinol binding protein III

hCRBPIV

human cellular retinol binding protein IV

hFABPI

human fatty acid binding protein I

hMFABP

human muscle fatty acid binding protein

HKD

histidine kinase domain

xxxii

HPLC

High-performance liquid chromatography

HWE

Horner-Wadsworth-Emmons

iLBP

intracellular lipid bindning protein

IPTG

isopropyl-β-D-1-thiogalactopyranoside

kcal

kilocalories

Kd

dissociation constant

KDa

kilodalton

LB

Lysogeny broth

LDA

lithium diisopropylamide

LFABP

liver fatty acid binding protein

LRAT

lecithin retinol acyl transferase

LUMO

lowest unoccupied molecular orbital

MALDI

matrix-assisted laser desorption/ionization

MCP

microchannel plate detector

min

minute(s)

mM

milimolar

mol

mole

mFABPIV

mouse fatty acid binding protein IV

mRFP

monomeric red fluorescent protein

MW

molecular weight

m/z

mass to charge

N.D.

not determined

nm

nanometer

xxxiii

nM

nanomolar

OD

optical density

PBS

phosphate saline buffer

PCR

polymerases chain reaction

PDE

phosphodiesterases

PSB

protonated Schiff base

RAR

retinoic acid receptor

rCRBPI

rat cellular retinol binding protein I

rCRBPII

rat cellular retinol binding protein II

rLFABP

rat liver fatty acid binding protein

Rh

Rhodopsin

rIFABP

rat intestinal fatty acid binding protein

rpm

round per minute

rt

room temperature

RXR

retinoid ‘X’ receptor

SB

Schiff base

scALDH

short-chain alcohol dehydrogenase/reductase

SDS-PAGE

sodium dodecyl sulfate polyacrylamide gel electrophoresis

sRh

sensory rhodopsin

TBAF

tetrabutylammonium floride

TM

trademark

TMS

tetramethylsilyl-

TOF

time of flight

xxxiv

tRt

all-trans-retinal

µM

micromolar

µm

micrometer

µg

microgram

UV-vis

Ultraviolet-visible

zCRBP

zebra cellular retinol binding protein

zLFABP

zebra liver fatty acid binding protein

xxxv

Chapter One
Probing the Wavelength Regulation of Rhodopsin Pigments via de Novo Engineering of a
Rhodopsin mimic
1.1. Introduction
1.1.1. The Importance of G-Protein Coupled Receptors

15

Figure 1-1. Crystal Structures of Human A2A Adenosine Receptor
49

(Green) and

Bovine Rhodopsin (Cyan) showing the 7-α helical domains. (For interpretation of
the references to color in this and all other figures, the reader is referred to the
electronic version of this dissertation.)

1

1-5

G-protein coupled receptors (GPCRs),

comprise a large 7-α helical transmembrane

protein family in eukaryotes and they are responsible for a variety of different biological
6-8

functions in mammalians (vision, odor, taste, memory, mood, etc.).

GPCRs generally detect

molecules outside the cells and transmit the signals across the membrane. This process initiates
signal transduction pathways inside the cell and leads to cellular response. Activation of GPCRs
modulates intracellular physiology and signal transduction across the cells; as such, GPCRs have
become an important druggable target. As an example, GlaxoSmithKline’s Zantac (ranitidine)
9

targets stomach acid production. It is a histamine H2-receptor antagonist. When histamine H2receptor on parietal cells (stomach epithelium cells) in the stomach is activated by histamine, the
proton pump is initiated and produces acid inside the stomach. Thus, inhibition of histamine H2receptor can stop or alleviate acid production. About 350 of 800 GPCRs found in humans are
10,11

potential pharmaceutical targets

and ca. 50% of drugs on the market target GPRCs. Despite

the significant interest from the pharmaceutical industry, the structural studies of GPCRs are far
12,13

lagged behind.

The crystallization of transmembrane GPCRs is more difficult than

crystallizing soluble proteins. There are only two GPCR proteins that have been crystallized upto-date, human adenosine A2A receptor and rhodopsin
(Figure 1-1).

14,15

As a result, most studies on GPCRs
O

have focused on theoretical analyses and mutagenesis

1-1

studies.
Figure 1-2. Structure of 11-cisretinal.

2

1.1.2. History of Vision Science
GPCRs are responsible for detection of external stimuli, particularly small molecules or
hormones. However, the members of the GPRC subfamily called rhodopsins are able to detect
light in the form of photons rather than a compound. Rhodopsins are responsible for initiating
vision in both vertebrates and invertebrates. Vision is one of the most complex and fascinating
processes in humans. It allows us to assimilate information from the environment. The science
of vision dates back to ca. 400 B.C. when Democritus proposed that vision is based on a solid
16

imprint on air.

After that, there were two main theories proposed by Plato and Euclid, and

Aristotle and Galan. Plato (ca. 350 B.C.) proposed the intromission theory: Vision was thought
17

to be based on rays irridiated from eyes that interact with objects.

This theory was followed by

the extramission theory proposed by Aristotle: Vision is the process during which eyes receive
18

reflected light from an object, and sun is the source of light.

The debate between the above

two hypotheses continued until Kepler wrote a detailed theoretical explanation of the optics of
16,19

the eye in 1604.

16,20

In 1706, Newton wrote a book called Optiks.

Newton first proposed

that light is composed of particles, that also have wave properties, which later were known as
photons. This was a revolutionary idea about light and vision during that time. About a century
later, the photoreceptor molecule called rhodopsin was identified by Franz Boll and Willy
16,21-25

Kuehne in the 1870s.

Eight decades later, George Wald from Harvard University

identified a single chromophore, retinal (Figure 1-2), as the central piece of vision.
30-32

discovery brought him a Nobel prize in medicine and physiology in 1967.

3

26-29

This

Since then, a

large amount of effort has been put to study the visual signal transduction pathway in a
molecular level, with great accomplishments and discoveries during the last 30 years.
1.1.3

The Process of Vision
Throughout 30 years of research, it is

Vitreous
Chamber
Iris

believed that vision starts when the eyes
receive light reflected from the object or Optical
Nerve
directly from the object. The received light

Cornea
Lens

passes through the cornea and is focused by
the lens. The rays then pass through the
vitreous chamber, a jelly-liked medium, and

Retina
Figure: 1-3. Anatomy of eye.

are focused on the retina (Figure 1-3).
Retina, where signal transduction from light into neural (electrical) signal occurs (Figure 133-37

3) has numerous rod (about 90 millions) and cone (about 7 millions) cells.

Rod cells, as the

name suggests have a cylindrical, rod shape (Figure 1-4). They are responsible for night vision
38-40

and equally distributed along the retina.

Cone cells have a similar structure and function

like rod cells except that their outer segment is cone shaped (Figure 1-4).

38-41

Cone cells are

localized at the fovea and are responsible for color vision. There are three kinds of cone cells
that are mainly differentiated by the pigment (red, green and blue) they contain in the outer
42-44

segment.

The cones and rod are different in their neuron circuitry. Multiple rod cells

converge on a single interneuron that helps to amplify the signal with the cost of losing visual
acuity while one cone cell connects to a single bipolar cell. The convergency of the rod cells

4

allows them to produce reliable neuron signals in dim light environment while cones require
much more intense light to generate signal from bipolar cell. As a result, rod cells are about 100
45

times more sensitive than cone cells.

The outer segment of both rod and cone cells consists of

about 1500 thin phospholipid bilayer membranes called disks (Figure 1-4).

The visual

Cone Cell
Outer Segment

Epithelium
Rod Cell
Outer Segment

Cytoplasm

Disk
Rhodopsin

Cytoplasm
Rod Cell
Inner Segment

46

Cone Cell
Inner Segment

Mitochondria

Extracellular

Nucleus
Synapse
Light

Light
Rod Cell

Cone Cell

Figure 1-4. Structure of rod cell and cone cell.
pigments are located within these disks.

47

These visual pigments are on the outer side, lying on

the pigment epithelium. The inner segment of the cell consists of different organelles and the
cell nucleus while the cell terminus forms a synapse. During cell activation by the absorption of
light, neurotransmitters (glutamates), are released from the synapses. The signals are transmitted
48

to other nerve cells by using neurotransmitters and finally to the brain for signal interpretation.

5

Figure 1-5. Stereoview of rhodopsin with retinal.
The visual pigments within the disks are called opsins (Figure 1-4). Opsins (39 KDa) are
49

7-α helical trans-membrane proteins and belong to the family of GPCRs
conserved disulfide bond (Figure 1-5).

1,50-52

with a highly

Additionally, a tripeptide sequence

(Glu(Asp)/Arg/Tyr(Trp), which is involved in G protein activation, is found at the cytoplasmic
border of helix 3.

53

Also, there are light-

dependent phosphorylation sites (Ser and
Thr) at the C-terminal tail for GPCRs
54

deactivation.

Lys296
Lys296

Opsin is bound to the

3.4Å
3.4Å

chromophore, retinal or its derivatives,
Glu113
Glu113

through the formation of a protonated Schiff
49,55-57

base (PSB) with a Lys296

at helix 7

and is stabilized by counterion Glu113 Figure 1-6. Binding site of rhodopsin with
Lys296 forming PSB with retinal and
counterion Glu113.

6

(Figure 1-6).

49,58-60

The bound chromophore is fully

covered by the protein within the lipid bilayers.
1-1

The

O

holo-form is referred as Rhodopsin. 11-cis-retinal is the
chromophore found in human while 3-dehydro-11-cisO

1-2

retinal is commonly found in fish, amphibia and reptiles
and some of the insects use 3-hydroxy-11-cis-retinal
(Figure 1-7).

O

61,62

The visual transduction in humans occurs when a

1-3

photon is absorbed by the 11-cis-retinal PSB in
HO

Rhodopsin. The absorption of a photon results in the

Figure 1-7. Retinals found in
different organisms.

isomerization of 11-cis-retinal to all-trans-retinal in
63,64

femto-seconds time scale.

During the isomerization,

subsequent intermediates were observed and characterized through UV-vis and other
spectroscopic techniques (Figure 1-8).

65-68

At the metarhodopsin II state, 11-cis-retinal PSB
68,69

from rhodopsin is fully isomerized to all-trans-retinal Schiff base (SB).

This isomerization

results in a conformational change in rhodopsin leading to the activation of G-protein
(transducin) that initiates a cascade of enzymatic reactions resulting in signal transduction
(Figure 1-8).

35,68

7

11

12
11
NH+
Lys296

11-cis-, 498 nm

12
Photo, 570 nm
"highly distorted all trans"

Lys296

11

11
O

N
H+

12

12

Batho-, -140 ˚C, 543 nm
"distorted trans"

all trans-retinal
+

ns

Opsin

Lumi, -40 ˚C
497 nm
ms
meta-I, -15 ˚C
480 nm

11
N
12
meta-II, 0 ˚C, 380 nm
all trans SB
Signal Transduction
Cascade

Figure 1-8. Photoisomerization Cycle of Rhodopsin.

8

Lys296

Upon absorption of a photon, 11-cis-retinal isomerizes to a highly distorted all-trans67,69

retinal called photorhodopsin that absorbs at 555 nm.

The highly distorted photorhodopsin

h!

Rh

Rh*

Rh*
Gt-!

Gt-!
Gt-#

Gt-#

GDP

PDE PDE
"
"

PDE PDE
"
"

PDE PDE
#
!

Gt-!

GDP

GTP

PDE PDE
"
"
PDE PDE
#
!

PDE PDE
#
!

x2

GTP

Gt-#

Rh*

Figure 1-9.

Signal Transduction Cascade Upon Photo
cGMP

Absorption

by

Rhodopsin

(Rh:

Rhodopsin;

GMP

PDE:

Phosphodiesterase; Gt-α : Transducin-α subunit; Gt-βγ : PDE
!

Ca2+ influx
STOP

PDE
#

Transducin-βγ subunit; GDP: Guanosine Diphosphate;
relaxes to bathorhodopsin. Further relaxation results in lumirhodopsin and metarhodopsin-I. A
proton transfer from the protonated Schiff base to the counterion Glu113 leads to the

9

intermediate called metarhodopsin II (meta-II) where the signal transduction cascade starts and
all-trans-retinal is released from rhodopsin (Figure 1-8).

65,68,69

At the metarhodopsin-II state, the extramembrane loop on the cytoplasmic side is exposed,
which leads to its interaction with transducin (Figure 1-9).

70

Transducin, a heterotrimeric

guanine nucleotide-binding protein (G-protein), consists of three subunits (α, β and γ) and is
50,71-74

responsible for mediating second messenger production inside the cell.

The activation of

transducin is through random collision via two-dimensional diffusion of rhodopsin and
75

transducin on the disk membrane.

The cytoplasmic loop of rhodopsin interacts with the α-

subunit of transducin that leads to the displacement of guanine diphosphate by guanine
triphosphate (GTP). This causes the destabilization of the heterotrimer (Figure 1-9). The
destabilized transducin dissociates and active GTP-α subunit (GTP:Gt-α) binds the γ-subunit of
the phosphodiesterases (PDE).

The binding between GTP:Gt-α and γ-subunit of the

phosphodiesterases removes the inhibitory effect of γ-subunit on α and β unit of PDE. The PDE
hydrolyzes cyclic guanine monophosphate (cGMP) to guanine monophosphate (GMP) lowering
the cellular concentration of cGMP (Figure 1-10). cGMP is a gated-ligand for opening Ca

2+

ion

channels across the membrane. A subtle decrease in cGMP concentration leads to the closure of
2+

Ca

ion channel that results in a depolarization of the membrane; thus, an electrochemical
48,50,75,76

signal is generated.

The signal is transmitted through the synaptic end of the cell to the

brain through nerves. The process is ceased by the phosphorylation of metarhodopsin-II by
rhodopsin kinase on C-terminus and the phosphorylated rhodopsin is completely inactivated by

10

77,78

binding with arrestin.

The overall process transforms light energy to electrochemical
79

signals and amplifies the signals. One metarhodopsin-II is able to activate 300 transducins
80

while each activated PDE can hydrolyze about 1000 cGMPs before deactivation.

The signal

amplification and high quantum yield (0.67) allows us to see continuously even in dark
81

Normalised Absorption

environment.

1

Blue opsin
(420 nm)

Rhodopsin
Red opsin
(498 nm)
(560 nm)
Green opsin
(530 nm)

0.5

400

500
Wavelength (nm)

600

700

Figure 1-10. UV-vis spectrum of rod and pigmented rhodopsins.
1.1.4

The Science of Color Vision in Humans
In addition to the numerous research programs focused on understanding the visual

transduction process on a molecular level, there have been a great amount of work done to
elucidate the mechanism of color vision. Color vision is an amazing process that allows us to see
the colorful world; differentiation of colors in eyes is based on the energy (frequency) of photons
(Figure 1-10). In the fifth century BC, black and white were believed to be the most important
11

colors and all colors were
O

!max = 380 nm

1-1

thought to be derived from
16

black and white.

It was

because no color is as light
N

!max = 365 nm

1-4

as white and no color is as
dark as black. No one had
argued

NH+

!max = 440 nm

rationale
Newton

1-5

with

the

until

above
1704.

conducted

a

breakthrough experiment at
Figure 1-11. Absorption maxima of 11-cis-retinal and its
that time when he was able
Schiff base and protonated Schiff base.
to

generate

prismatic

spectrum from white light. This proved that white light is not unified but compounded.
th

20

th

During 18 and 19 centuries, color vision was explained by the specific interaction between
different colors with different “particles” on retina.
sensitivities towards different colors.

The theory was based on different

Instead of infinite number of retina “particles” with

different color, Young proposed that there were only three such “particles” for color vision, later
82

identified as three color rhodopsins in the 1800s.

This was further elaborated as the

trichromatic theory by Ewald Hering. The mechanism of color vision became more interesting
when Wald and co-workers identified 11-cis-retinal as the only chromophore in the visual
26-28

pigments.

11-cis-Retinal and its SB with n-butylamine absorb at 380 nm and 365 nm,

12

H

H
H
NH+

H
NH+

Lys

Lys296

NH+

H
H
!max = 440 nm

H

O

H
!max = 380 nm
Figure 1-12. Proposed mechanism for wavelength regulation in rhodopsin.
respectively, in ethanol (Figure 1-11). While the 11-cis-retinal PSB absorbs at 440 nm in
ethanol, 11-cis-retinal bound with different color pigments and rhodopsin shows an absorption
spectrum ranging from 400 nm to 600 nm, allowing us to see color throughout the whole visible
region (Figure 1-11).

50,68,83

The spectrum shift induced by the visual pigments from 11-cis-

retinal PSB in ethanol is called the “opsin shift”. The 200 nm absorption shift for one single
chromophore is due to specific interactions of 11-cis-retinal with the different rhodopsins.

13

Although several hypotheses have been suggested regarding the wavelength regulation in
rhodopsin, even after five decades of studies, the mechanism is still not completely clear.
However, it is believed that the wavelength regulation is based on various degrees of charge
delocalization and conjugation allowed along the PSB-bound polyene inside the protein (Figure
1-12):
1) The interactions between the retinal PSB and the counterion: Strong interaction between the
retinal PSB and the counterion causes localization of the positive charge on the PSB
84-88

nitrogen, thus, minimizing the charge delocalization.

2) The planarity of the polyene of retinal: Changing the planarity of the polyene system affects
89-91

the degree of conjugation of retinal PSB leading to blue shift or red shift.

3) The presence of another charged residue along the polyene: Charged residues at different
positions along the polyene of retinal-PSB could stabilize the resonating positive charge at
92-94

different locations along the polyene system that can lead to red or blue shift.

14

Different approaches (chemical, spectroscopic, computational, and mutagenesis) have been
utilized to elucidate the wavelength regulation mechanism in opsin on a molecular level. In the
early days, because of lack of genetic information about rhodopsins and poor development of

O

O

1-1 (440/500 nm:2700 cm-1)

1-6 (425/460 nm:1800 cm-1)

O

O

1-7 (392/420 nm:1700 cm-1)

1-8 (322/345 nm:2100 cm-1)

O

O

1-9 (270/315 nm:5300 cm-1)

1-10 (275/310 nm:4100 cm-1)

Figure 1-13.
11-cis-Retinal, dihydroretinal and tetrahydroretinal analogs with
absorption before and after incubation with rhodopsin and the opsin shift.
purification and expression of rhodopsins, model compounds and model systems played an
important role to elucidate the theories of wavelength regulation in rhodopsin. In 1967, Blatz et
al. proposed that the strength of interaction between the counterion and the PSB of the
chromophore could modulate the absorption of the bound retinal (Figure 1-12).

15

88

They

N+
H

nBu

1-11 (440 nm)

N+
H
NH+

1-12 (423 nm)

NH+

N+
N+
H

nBu

CO2H
1-14 (276 nma

1-13 (419 nm)

ClO4297 nmb)

N+
ClO4-

CO2H
1-15 (410 nma

nBu

420 nmb)

N+
ClO4-

CO2H
1-16 (502 nma

510 nmb)

Figure 1-14. Retinal analogs with different charged group along the polyene and the
a
b
corresponding UV-vis absorption ( in ethanol; treated with Et3N in ethanol).
-

-

-

-

measured the UV-vis spectrum of retinal PSB with different counterions (Cl , Br , I , ClO3 ,
-

CH3CO2 , etc) and showed that the size and charge of the counterion have a significant effect on
86

the absorption of retinal PSB.

Honig and Nakanishi suggested that there is an external point

charge to stabilize the resonance of retinal at different levels; thus affecting the chromophore’s
absorption (Figure 1-12).

92

They synthesized different dihydroretinals with a saturated bond at

different positions along the polyene (Figure 1-13). According to the point-charge theory, an
external charge close to the end of the polyene results in the largest bathochromic shift. The
result indicated that the largest red-shift was obtained when 11,12-dihydroretinal was incubated
with rhodopsin. Therefore, they proposed that there should be an external charged residue in
16

92

rhodopsin around C11-C12 of the bound 11-cis-retinal PSB.

The concept was further

elaborated by Sheves’s group, who synthesized a library of retinal analogs with different amines
(positive charge) or carboxylates (negative charge) placed at different position along the retinal
(Figure 1-14) and tested them in ethanol. The result (Figure 1-14) shows that negative charge at
84,95

different positions around the polyene could modulate the absorption of the chromophore.

In 1986, Nathans and co-workers isolated and sequenced DNAs encoding color
83

pigments.

Sequence identity between rhodopsin and color rhodopsins is around 40% while red

rhodopsin and green rhodopsin is almost the same (96% sequence identity).

83

With the sequence

identified, effort was placed into understanding the fundamentals of wavelength regulation in
human opsins using genetic methods. The first mutagenesis experiment designed to understand
color vision was carried out by Oprian in 1994. Oprian generated a chimeric protein using red
and green rhodopsin and was able to generate a mutant having absorption spectrum in between
red and green. Oprian et. al identified seven important mutations (S116Y, S180A, I230T,
A233S, Y277F, T285A and Y309F) that allowed them to convert the red pigment to green
96

pigment.

Interestingly, the mutations introduced were either switch from polar residue to

apolar residue or vice versa. Several research groups have studied the effect of individual
residues (within 5 Å away from the bound retinal in rod rhodopsin) on wavelength regulation by
replacing these residues in rod rhodopsin with the corresponding residues in green or red
rhodopsin based on the sequence alignment.

The results indicate that residues leading to

bathochromic shift in red or green pigments as compared to rod are either polar or non-polar but

17

Figure 1-15. Crystal structure of rhodopsin (11-cis-retinal in dark blue) and models of color rhodopsins (11-cis-retinal in
cyan for blue rhodopsin, 11-cis-retinal in green for green rhodopsin and 11-cis-retinal in red for red rhodopsin) with closed
residues in different colors.

18
18

97-101

not charged residues.

These results further dismiss the point-charged theory. In addition,

Sakmar and co-workers attempted to study the principles of hypsochromic shift in blue
rhodopsin relative to rod rhodopsin. They first studied the effect of individual segment in blue
rhodopsin on wavelength regulation by replacing each segment in blue pigment with the
corresponding sequence in rod rhodopsin. Additionally, the residues, within 5 Å away from the
bound retinal PSB in rhodopsin, were mutated. Their results show that replacing polar residues
with apolar residues (E122L, T124A) around the ionone ring and W265Y mutation around the
end of the chromophore are responsible for most of the hypsochromic shift in blue
97,102

rhodopsin.

It is interesting that after a decade of mutagenesis studies, most of the residues
96-102

identified are either polar or apolar residues rather than a fully charged residue.

In 2000,

there was a breakthrough in rhodopsin research, when Palczewski et al. were able to crystallize
the rhodopsin for the first time (Figure 1-5).

49

This development led the scientists in the field to

quickly revise the point-charge theory since from the crystal structure, it was evident that there is
no specific charged residue in the vicinity of the bound 11-cis-retinal in rhodopsin (Figure 115).

49,51

Nowadays, it is believed that it is the global electrostatic environment along the

polyene that plays an important role for wavelength regulation.

Although mutagenesis

experiments provide a vast amount of information, the progress towards understanding
wavelength regulation on a molecular level is still hampered by lack of crystal structures of the
other three colored rhodopsins. Computational studies have been used by many groups to further
investigate the phenomena of wavelength modulation. Since the sequence identity between
rhodopsin and color rhodopsins is around 40%,

83

19

Palczewski et al. employed homologous

modeling to build models for the three colored pigments based on the crystal structure of
rhodopsin (Figure 1-15).

51

Using the rhodopsin crystal structure and the color rhodopsin

models, we are able to analyze the information from the mutagenesis studies from a structural
basis. Based on the models, the seven sites responsible for engineering red rhodopsin into green
rhodopsin are located either close to

Rh 3 Å

the β-ionone ring region (S180A,

Rh 3.5 Å

I230T, A233S, Y277F and T285A)
or the PSB region (S116Y and

427 450

500 530

560 (nm)

Y309F) (Figure 1-15). Around the
ionone ring in red rhodopsin, there

Protein
Recover

are polar residues while green
rhodopsin

has

apolar

residues

Introduction of counterion

around the ionone ring. In addition,
we

found

that

blue

rhodopsin

11-cis-retinal
PSB with E113

11-cis-retinal
PSB

contains more polar residues (Ser87, Figure 1-16. Scheme shows the relative contributions
from the counterion E113, rhodopsin (Rh) and its
Ser292 and Ser289) close to the PSB residues (at certain distance around bound 11-cisretinal PSB) on modulating the energy of bound
region as compared with other chromophore based on Complete Active Space
Second-Order Perturbation (CASPT2). The blue
51
rhodopsins (Figure 1-15).
This arrow indicates the blue shift induced by the
counterion. The red arrow shows the red shift that is
caused by the protein around 3.0 to 3.5 Å from the
could be the reason for blue bound retinal PSB.
rhodopsin being more hypsochromically shifted than green and red rhodopsins. Kloppmann et
al. calculated the electrostatic potential around bound retinal in different microbial rhodopsins
103

and found that the electrostatic potential plays a vital role in wavelength regulation.

20

Recently, Andersen, et al. measured the retinal protonated Schiff base in vacuo (λmax ~ 610
104

nm).

This result provides a different perspective of wavelength regulation in rhodopsin. One

could suggest that rhodopsins modulate the wavelength of bound retinal PSB through blue
shifting the bound chromophore (starting at a maximum of 610 nm in vacuum). Tomasello and
co-workers conducted a calculation similar to Kloppmann’s with the human rhodopsins.

105

Based on their calculation, counterion Glu113 and the residues within 2.5 Å away from the
bound chromophore alone could blue shift the bound 11-cis-retinal PSB to around 410 nm from
600 nm, that is the λmax of retinal PSB in vacuum. However, their calculations also indicated
that the electrostatic potential induced by the residues between 3 Å and 3.5 Å away from the
bound 11-cis-retinal PSB in rhodopsin are able to recover part of the hypsochromic shift induced
by the residues 2.5 Å away from the bound retinal and the counterion Glu 113 (Figure 1-16).

105

These two computational studies provide new evidence that the electrostatic potential plays a
crucial role in wavelength modulation in rhodopsin proteins.
In addition, Blatz et al. proposed that the degree of planarity of the bound 11-cis-retinal
106

could modulate the absorption of the bound chromophore.

11-cis-Retinal suffers from steric

strains between the C5-methyl and the C8-hydrogen, and C13-methyl and C10-hydrogen. This
forces 11-cis-retinal to adopt a bent conformation (Figure 1-12) that minimizes the conjugation
between the double bond. It is further confirmed by the crystal structure of 11-cis-retinal that the
106

chromophore has twisted single bonds between C6-C7 and C12-C13.

Based on the solid state

NMR study and crystal structures of rhodopsin, it is believed that 11-cis-retinal PSB in
rhodopsin adopts a C5-C6 s-cis conformation (Figure 1-15).

21

90

It is also believed that different

color rhodopsins have a unique binding pocket that
O

could restrict the bound 11-cis-retinal PSB to adopt
only specific conformations, thus, modulating the

1-17 (440 nm

568 nm)

absorption. van der Steen et al. synthesized a variety
of ring-locked compounds (Figure 1-17) to study the
O

effect of retinal’s planarity and conformation around
the C5-C6 bond on wavelength regulation. The ring-

1-18 (465 nm

564 nm)

locked

analogs

were

incubated

with

bacteriorhodopsin; however, the study showed that the
O

planarity of the chromophore could only account for
-1

1300 cm
1-19 (485 nm

of the 5100 cm

-1

bathochromic shift

596 nm)
91

Figure 1-17.
C6, C7-Ring-lock observed in bacteriorhodopsin.
In addition, the
analog of all-trans-retinal with
corresponding λmax of their PSB and computational studies suggested that the induced
bound to bacteriorhodopsin.
planarity of bound 11-cis-retinal PSB cannot solely
-1

89

account for the 4870 cm shifted observed in red rhodopsin.

Apart from the three main hypotheses (Figure 1-12), other suggestions have been made by
different groups in their attempt to explain the wavelength regulation in rhodopsin. Beppu and
Kakitani suggested that aromatic residues induce important physical interactions with bound 11107-110

cis-retinal PSB and modulate the absorption.

They suggested that exciton coupling

between these aromatic groups and the chromophore in rhodopsin can manipulate the energy
level of the bound chromophore. Besides, the orientation of the aromatic residues relative to
retinal is an important factor for wavelength modulation because the induced electric field from

22

the aromatic ring interacts differently with the bound chromophore from different angles based
on their calculations. In addition, the mutagenesis experiments by Sakmar et al. showed that
Trp285 in the middle of the binding pocket is one of the key residues responsible for
bathochromic shifts of bound retinal PSB in rhodopsins (rod, red and green) (Figure 1-15).

97

1.1.5. Wavelength Regulations in Other Microbial Rhodopsins
As previously discussed, rhodopsin and color rhodopsins belong to a diverse family of
proteins across different organisms from vertebrates, invertebrates to microbes. All members of
this family of proteins shares a common 7-α-helical transmembrane motif and can covalently
bind to similar chromophores, either retinal or retinal like molecules, through the formation of a
49

protonated Schiff base with a lysine residue within the binding cavity.

The rhodopsin family is

mainly responsible for light sensing. Upon absorption of a photon, stereospecific isomerization
of retinal or the retinal like chromophore leads to a series of protein conformational changes that
result into the transduction of the signal from light to either chemical or electrical signal. In
vertebrates, the isomerization triggers a cascade of protein interactions that initiates signal
transduction through generation of an electrochemical signal across the membrane and the neural
signals are transmitted to the brain (Figure 1-9).

48,50,75,76

However, in bacteria,

bacteriorhodopsin is used as a proton pump for energy generation. Upon absorption of a photon
(~560 nm), the isomerization of all-trans-retinal to 13-cis-retinal occurs and a proton is
transferred across the membrane through the conformational changes that happen within the

23

Figure 1-18.
absorptions.

Stereoview of different microbial rhodopsins with the corresponding

24

protein, thus, generating an electrochemical gradient across the membrane. During this stage,
light (around 568 nm) is re-absorbed and triggers the isomerization of 13-cis-retinal back to alltrans-retinal. The energy stored from the electrochemical gradient across the membrane is used
111-113

to synthesize adenosine triphosphate for energy storage.

The crystal structure of

bacteriorhodopsin in dark-adapted state has been resolved by Grigorieff et al. in 1996 with a
h!

sRhII

sRhII

sRhII

sRhII

HtrII

HtrII

CheW

CheY

CheW

CheA
CheA
P
CheY

CheY

Flagella
Motor
Activation

Figure 1-19. Signal transduction of sensory rhodopsin. (sRhII: Sensory Rhodopsin II;
HtrII: Transducer; CheA: Histidine Kinase).

25

114

resolution at 3.5 Å.

The resolution of the crystal structure was improved to 1.55 Å by Luecke

and co-workers (Figure 1-18).

115

Halorhodopsin is another bacterial rhodopsin. In halorhodopsin, chloride is the counterion
for the retinal PSB (Figure 1-19).

Halorhodopsin’s function is similar to that of

bacteriorhodopsin. Upon absorption of a photon (570 nm), halorhodopsin acts as a chloride
pump to translocate the bound chloride ion across the membrane and generate an electrochemical
gradient that serves as energy source. Halorhodopsin has been crystallized and its structure was
determined by Kolbe et al. to 1.8 Å resolution (Figure 1-19).

116

A third type of microbial rhodopsin serves as a light sensor for phototaxis process in
117,118

halobacterium salinarium.

Sensory rhodopsins I and II are tightly coupled to tranducer

proteins HtrI and HtrII respectively.

These specific transducers modulate the function of

histidine kinases. During the isomerization of retinal in sensory rhodopsin, after absorption of
light, it activates the transducer protein HtrI and HtrII and downstream processes (Figure 1-20).
CheA is activated through interactions with HtrII with the assist of CheW. The complexes
phosphorylate CheY.

Phosphorylated CheY activates flagellar motor; thus, controlling the
118

flagellar motor of the microbe.

From the crystal structure (Figure 1-18), sensory rhodopsin I

shows a very similar structure as bacteriorhodopsin and halorhodopsin with all-trans-retinal as
its chromophore.

All-trans-retinal adopts a 6-s-trans conformation such as that found in
117

bacteriorhodopsin and halorhodopsin.

Different microbial rhodopsins display different absorptions (Figure 1-18) and several
hypotheses have been proposed to explain the wavelength modulation amongst microbial

26

rhodopsins.

The

crystal

structures of different microbial
rhodopsins

show

that

the

chromophore is surrounded by

G130
(S141)

tryptophan residues (Figure 118).

These tryptophans are

proven by mutagenesis studies

A131
(T142)

to be important for the observed
bathochromic

T204
(A215)

119-121

shift.

Figure 1-20.
Important residues around sensory
rhodopsin II (Blue) and bacteriorhodopsin (Green) with
Irving et al. also suggested that residues different between bacteriorhodopsin and sensory
rhodopsin II labeled (residues in parenthesis correspond
to rhodopsins is due to
the bathochromic shift in microbialbacteriorhodopsin). the electronic polarization from the
nearby aromatic residues. This hypothesis is based on the fact that dipole across the retinal PSB
is much larger in its excited state. When the retinal PSB is excited, the PSB is further polarized
leading to an increased the dipole moment of the bound chromophore. This in turn leads to the
polarization of aromatic residues around the excited chromophore. This induced polarization
could stabilize the excited dipole of retinal PSB, thus, lowering the HOMO-LUMO energy gap
122

that induces bathochromic shift.

This suggestion is further supported by the fact that retinal
123,124

PSB red shifts with addition of phenol or indole in ethanol.

It is interesting that the residues within 4 Å of the bound retinal PSB in sensory rhodopsin
II (λmax ~ 497 nm) and bacteriorhodopsin (λmax ~ 560 nm) are similar (Figure 1-20) with only
three amino acid differences. However, the absorption is shifted by 71 nm in bacteriorhodopsin.
Several studies have been conducted to understand the reason for such an absorption shift. There

27

are two hydroxyl amino acids near the β-ionone ring in bacteriorhodopsin (Ser141 and Thr142)
while sensory rhodopsin II replaces them with Gly130 and Ala131 (Figure 1-20). However,
mutagenesis studies carried by Shimono et. al showed that substitution of residues in sensory
rhodopsin II around the bound retinal with the corresponding residues present in
bacteriorhodopsin recovered only 25 nm of red-shift.

125

This indicated that the polarity changes

D75

D85
R72

R82

D201

D212

Figure 1-21. Crystal structure of bacteriorhodopsin (Green) and sensory rhodopsin II
(Cyan) with the water-mediated interaction between Arg82 (Bacteriorhodopsin) or
Arg72 (Sensory Rhodopsin II) and PSB of bound all-trans-retinal.
within the binding pocket cannot account for the entire 71 nm shift in bacteriorhodopsin.
Besides, Arg72 in sensory rhodopsin II takes the conformation which the guanidino group is 1.1
Å away from the PSB as compared with the guanidino group in Arg82 in bacteriorhodopsin
(Figure 1-21).

126

It was suspected that the displacement of Arg72 by 1.1 Å could alleviate the

interaction with counterion Asp212, thus strengthening the PSB-counterion interactions leading
to the blue shift in sensory rhodopsin II (Figure 1-21).

127

Although mutation of Arg82 did not

provide a clear conclusion about the function of this arginine, Arg72 in sensory rhodopsin II and
Arg82 in bacteriorhodopsin are suggested to be precisely positioned through the interactions with
128,129

the residues around them.

Professor Ullmann and co-worker identified 7 positions
28

(Gly130, Ala131, Thr204, Pro183, Asp193, Ala111 and Arg66) in sensory rhodopsin II that
could contribute different electrostatic potentials on the bound retinal as compared to
103

bacteriorhodopsin based on calculations.

It is interesting that four of the identified residues
-1 -

located more than 8 Å away from the bound retinal, are able to contribute about 0.5 kcal mol e
1

difference on the bound retinal in bacteriorhodopsin, as compared with sensory rhodopsin II,
103

and the collective effect is more significant.

These results suggested that a single amino acid

could not induce significant changes in absorption and it is a collaborative effort from different
residues that leads to significant changes on the absorption spectra of these rhodopsins.
1.1.6. Re-engineering a Protein Mimic to Probe the Wavelength Regulation in
Rhodopsin
After decades of studies, the theory behind wavelength regulation in rhodopsins at
molecular level has not been fully understood. Although gene sequences of different rhodopsins
and the crystal structure of bovine rhodopsin are known, the progress on dissecting the
mechanisms of wavelength modulation in rhodopsin proteins is still hampered by different
problems. Rhodopsins are membrane-bound proteins and protein expression and purification are
50,52

problematic.

In addition, when mutation is introduced, rhodopsin mutants are generally

poorly expressed compared to the wild-type protein so the characterization of mutants is
130

difficult. Although the crystal structure of bovine rhodopsin has been solved to 2.2 Å,

neither

color opsins, nor rhodopsin mutants have been crystalized yet so identification of the crucial
protein-chromophore interactions that modulate the energy level of the bound chromophore

29

become difficult. To tackle these problems, we initiate a program to design a protein mimic for
rhodopsin in order to study the phenomenon of wavelength regulation.
Protein engineering has been a central piece of scientific research in past decades. Since
131

the first example of site-directed mutagenesis in 1978,

protein engineering has been greatly

applied to different research areas (biochemical, biomedical, industrial, physiological, etc).
Rational mutagenesis has been widely used to answer the question of enzyme catalysis.

132-136

Rational design based on crystal structures of target proteins allow scientists to create enzymes
134,137,138

with new specificities

132,139-145

and reactivities.

Besides, through directed evolution,
133,139,144,146,147

researchers are able to generate proteins with novel secondary structures.

Also,

protein engineering has been utilized on medical research to generate biomaterial for healing or
probes for diagnosis.

148-152

We have initiated a de novo approach to the study of wavelength regulation in rhodopsin.
The plan entails the reengineering of a protein template as a rhodopsin mimic. With the protein
mimic, we can study the specific mechanisms involved in wavelength regulation. To be a good
protein mimic, this protein has to be able to bind retinal through the formation of PSB with
lysine, like rhodopsins.

Unlike rhodopsins, it has to be easily crystallized so we could

understand the specific interactions leading to wavelength regulation in different mutants. The
protein mimic should be easily expressed and purified. It should tolerate different mutations.
Based on these guildines, we chose the human cellular retinoic acid binding protein II
(CRABPII) as our first protein surrogate.
CRABPII (Figure 1-22) is a small cellular protein (16 KDa) found in all
153,154

vertebrates,

and belongs to intracellular lipid-binding proteins (iLBPs) family.
30

155

iLBPs

(Figure 1-23) are responsible for cytoplasmic transport of lipophilic molecules, such as long155

chain fatty acids and retinoids.

They share a common tertiary structure of 10-strand anti-

parallel β-barrel composed of two orthogonal five-stranded β-sheets and a helix-turn-helix
155

cap.

The β-barrel of CRABPII possesses a deep embedded binding cavity of about 600 Å

3

Figure 1-22. Stereoview of CRABPII (Green) with retinoic acid (Cyan)

(Figure 1-22). CRABPII is mostly expressed in skin,
159

plexus

160,161

and hematopoietic cells.

156

153,157

testis,

uterus ovary,

158

choroid

CRABPII helps to solubilize retinoic acid in the
154

cytoplasm and carries retinoic acid from the cytoplasm to the nucleus for gene regulation.
156

Retinoic acid is a biologically active metabolite of vitamin A (retinol).

31

It acts as a morphogen

during embryonic morphogenesis and is necessary for regulating vertebrate cell growth,
162-165

differentiation and homeostasis.
Also,

gene

expression

during

developmental stage is tightly regulated
by

the

retinoic
166,167

concentration.

acid

Upon binding

with retinoic acid, CRABP proteins
Figure 1-23. Crystal structure of fatty acid binding localize in the nucleus and channel the
protein (Green) with 1-anilinonaphthalene 8sulfonic acid (Blue).
bound retinoic acid to the retinoic acid receptor
(RAR).

168

O
1-17

RAR, a member of the steroid / thyroid

OH

hormone receptor family, is a transcription factor.

O

RAR interacts with retinoid ‘X’ receptor (RXR)
1-20

forming a heterodimer that recognizes RARresponse elements

in gene promoters for gene
OH

160,169-171

transcription.

Retinoic acid, therefore,

1-21

Figure 1-24. Structure of all-transretinal (1-17), all-trans-retinoic acid
From the crystal structures of CRABPII (1-20) and all-trans-retinol (1-21)

regulates the gene transcription in such a manner.

(Figure 1-25), we can see that the retinoic acid is bound through a salt bridge between Arg132
and the carboxylate of retinoic acid. Also, the retinoic acid interacts with Arg111 through a
172,173

water-mediated interaction and hydrogen bonds with Tyr134.

According to the

competition assay done on the R132M mutant by Yan’s group in 1997, the binding of the

32

mutant proteins with retinoic acid is 6 to 8

R132

times less than that of the wild-type
172,174

CRABPII.

We

have

Y134

also

demonstrated that mutation of Arg132 to
leucine

hampers the binding towards

retinoic acid 100 folds lower using
175

fluorescent quenching assay.

R111
Arg111

was also tested by the competitive binding

Figure 1-25. Crystal structure of CRABPII
with retinoic acid (Blue) with important
interactions shown in dash lines.
172

assay and showed a 40 to 45 times weaker binding with mutant R111M.

The effect of Tyr134

on retinoic acid binding is not tested because the mutation of Y134F leads to misfolded
175

protein.

However, an indirect comparison has been made. Mutant R132K:Y134F displays a
175

relatively less effective binding with retinoic acid as compared with single mutant R132K.

Based on the crystal structure (Figure 1-26),
Trp109

has

π-cation

interaction

with

Arg111, thus directing the conformation of
Arg111 that favors the interaction with
175

E73

retinoic acid.

R111
W109

Mutation of Trp109 to a

hydrophobic residue (leucine) weakens the
retinoic acid binding by ~1000 fold. This

Figure 1-26. Retinoic acid binding site in shows the importance of Trp109 in retinoic
CRABPII with interaction of Glu73 with
Trp109 and Arg111 shown.
175
acid binding.
Also, Glu73 provides a

33

tight hydrogen bonding with the indole nitrogen to constraint the conformation of Trp109
(Figure 1-26). Mutation of Glu73 to Ala reduces the retinoic acid binding by a factor of about 5.
That further supports that the π-cation interaction of Trp109 with Arg111 is important to orient

Lys132

NH2

Tyr134

109˚

H

O
3.73 Å

2.95 Å

3.1 Å

NH2

H

2.47 Å
3.4 Å

Glu121

O

Figure 1-27. Left: Model of R132K:R111L:L121E with all-trans-retinal; Right: Optimum
Bu
¨rgi-Dunitz angle at 109˚.
175

Arg111 in a proper conformation to interact with retinoic acid.

CRABPII fits the criteria to be a good protein template for a rhodopsin mimic. It can be
easily be expressed and the yield for the wild-type (WT) CRABPII expression under T7
176

promoter is 30 mg per liter.

As mentioned, a high-resolution (1.4 Å) crystal structure has

been obtained so that a rational redesign is possible. In addition, CRABPII has been mutated
extensively suggesting that the fold is resistant to structural changes by multiple mutations.
Finally, although CRABPII binds poorly to all-trans-retinal (Kd ~ 6 uM), it binds tightly to the
177

all-trans-retinoic acid (Figure 1-24) that has a similar structure to retinal.

Therefore, we

believed that we could easily enhance the binding towards retinal with proper mutations.

34

1.2. Probing the Wavelength Regulation using Re-engineered CRABPII as a Rhodopsin
Mimic
1.2.1. Re-engineering CRABPII into Rhodopsin Protein Surrogate
Recently, we have re-engineered CRABPII into a rhodopsin protein surrogate. Based on
the in silico mutagenesis on the crystal structure of CRABPII with retinoic acid, we identified
that R132K could have the best B¨ rgi-Dunitz trajectory for the ε-amino of lysine to attack the
u
carbonyl of retinal (Figure 1-27).

176

Based on the model, a counterion glutamate was installed at

121 to stabilize the bound retinal PSB.

During the course of engineering CRABPII into

rhodopsin mimic, we found that Arg111 hinders the formation of a bound all-trans-retinal PSB.
From the crystal structure of mutant R132K:Y134F, (mutant that cannot form SB with
engineered Lys132), the bound all-trans-retinal adopts two conformations, one of them which
interacts with Arg111 through a water-mediated interaction. The ordered water is held in place
through hydrogen bonding with T54 and R111. We believed that the ordered water restricts the
bound retinal from adopting a better conformation for nucleophilic attacked by Lys132 (Figure
1-27 and 1-28). In an attempt to restore the SB formation, R111L mutation is introduced to

35

178

Table 1-1. Kd and λmax for Retinal Bound to CRABPII Mutants.
Mutant

WT-CRABPII

6600 ± 360

No

379

280 ± 17

Yes

R132K:Y134F

404

120 ± 5

No

R132K:Y134F:L121E

381

160 ± 10

Yes

R132K:Y134F:R111L:L121E 446

200 ± 8

Yes

R132K:R111L

408

567 ± 36

Yes

R132K:R111L:L121E
b

377

R132K

a

Reductive Amination
λmax (nm) Kd (nM)
(all-trans-retinal) (+268 m/z)

449

1.36 ± 4.9

Yes

λmax of bound all-trans-retinal PSB
An adduct peak observed in mass spectrum after reductive amination
177,178

remove the ordered water.

Mutant R132K:Y134F:R111L is now able to bind all-trans-

176,178,179

In addition, we proposed that Tyr134 is important for SB formation

retinal as a PSB.

Lys132

Lys132
70˚

146˚
Arg111

Arg111
Thr54

Thr54

Figure 1-28. Crystal structure of CRABPII mutant R132K:R111L with important residues
and hydrogen bonding (~3 Å) shown.

36

0.12

base

PSB

Abs at 360 nm (a. u.)

Abs (a.u.)

0.08
0.06
0.04
0.02
0

pKa = 8.7

1.2

0.1

300

350 400 450 500
Wavelength (nm)

1
0.8
0.6
0.4
0.2

550

7.5

8.25
pH

9

9.75

Figure 1-29. Acid-base titration of mutant R132K:R111L:L121E with all-trans-retinal
monitored by UV-vis spectroscopy and the calculated pKa of the bound retinal PSB in
mutant R132K:R111L:L121E.

Lys132

Tyr134

Glu121

Leu111
Figure 1-30. Binding site of mutant R132K:R111L:L121E with bound all-trans-retinal
PSB stabilized by counterion Glu121.
based on the observation that single mutant R132K forms Schiff base with all-trans-retinal but
176,178,179

not the double mutant R132K:Y134F;

therefore, residue Tyr134 was maintained.

Finally, based on the model, a counterion glutamate was installed at 121 to stabilize the bound
retinal PSB. CRABPII mutant R132K:R111L:L121E was generated, which exhibits tight binding
with the all-trans-retinal with Kd around 2 nM (Table 1-1).

37

178

M+
15508.2

14000
O

H2N

Lys132

Intensity

+

12000
10000
8000
6000
4000
2000
0
14500 15000 15500 16000 16500 17000
Mass (Da)

Lys132

N+
H

M+
15508.5

NaB(CN)H3

Intensity

20000

M+
15776.3

15000
10000
5000
0
14500 15000 15500 16000 16500 17000
Mass (Da)
Lys132

N +
H2
m/z = Protein Mass + 268

Figure 1-31. Scheme for reductive amination and the mass spectrum of mutant
R132K:R111L:L121E and mutant R132K:R111L:L121E with all-trans-retinal after
178
reductive amination with an adduct peak shown (+ 268 m/z).

38

The triple mutant R132K:R111L:L121E binds
all-trans-retinal as a PSB that absorbs at 449

Lys296

nm with the pKa value around 8.7 based on
the acid-base titration followed by UV-vis
spectroscopy of the protein-chromophore
complex (Table 1-1) (Figure 1-29). Based on
Glu113

the reductive amination of the protein/

Figure 1-32. Binding site of rhodopsin with
chromophore complex analyzed by mass 11-cis-retinal showing a trans-imine.
spectroscopy, a [M + 268] adduct was observed (Figure 1-31), which is expected considering the
mass of retinal. The formation of Schiff base was further confirmed by crystallograpy.

178

A

crystal structure of mutant R132K:R111L:L121E with all-trans-retinal was obtained with a
resolution of 1.2 Å (Figure 1-30).

From the crystal structure, the retinal binds to

R132K:R111L:L121E through the formation of PSB with Lys 132. As expected, the retinal PSB
is interacts with counterion Glu121 (Figure 1-30). Interestingly, a cis-iminium bond is found in
triple mutant R132K:R111L:L121E–CRABPII (Figure 1-30) instead of the trans-iminium in
rhodopsin (Figure 1-32).
1.2.2. Effect of Electrostatic Potential on Wavelength Regulation in Rhodopsin
As mentioned in Section 1.1.4., residues responsible for wavelength regulation in
rhodopsin proteins are either polar or apolar residues rather than charged residues based on the
98,100-102,119

mutagenesis studies and the crystal structure of bovine rhodopsin.

Different amino

acids (charged, polar and apolar) induce different magnitude of electrostatic potential field.
However, in water environment with its high dielectric constant (εr = 78), the electrostatic field
39

induced by polar residues is negligible because water neutralize the electrostatic potential field
effectively by reorienting the dipole of the water molecule around the polar residues.

180-186

However, in vacuum environment with virtually 0 dielectric constant, the electrostatic potential
180-186

can have a huge and distal effect on the surrounding environment.

Several studies

indicate that the protein binding site has different dielectric constant as compared to bulk water
environment surrounding it. This is the results of the residues around the binding site acting as a
122,180,183,184,186

boundary to segregate the protein interior from the surrounding area.

Therefore, we hypothesized that the overall electrostatic potential from residues in different
rhodopsins (red, green, blue and rod) projected on the bound 11-cis-retinal should be different.
An electrostatic potential calculation was conducted on the bovine rhodopsin crystal structure
187-192

and the three color rhodopsin models using APBS package.

The overall electrostatic

potential projected on the bound 11-cis-retinal PSB from four different rhodopsin proteins are
shown in Figure 1-34. The calculation shows significant different among different rhodopsins
(blue, green, red and rod). Our calculations suggest that electrostatic potential can play an
important role on wavelength regulation that is similar to the calculations on bacteriorhodopsin
and human rhodopsins by Kloppmann and Tomasello, respectively as mentioned in Section
1.1.4. Based on these, we propose a formula for wavelength regulation of visual rhodopsins.
We divided the bound 11-cis-retinal into three regions: PSB (I), middle (II) and β-ionone ring
(III) (Figure 1-33). Different rhodopsins showed different electrostatic potential at each region.
Blue rhodopsin exhibits a significant amount of negative electrostatic potential around the PSB

40

-40

0

40
Blue (410 nm)

Rho (500 nm)

Green (530 nm)

Red (560 nm)

H
N+

H
H

Lys296
I

III

II

I

II

III

Rh

Blue/Rh

Blue

Red

Green

Rh

Less Polar

Blue/Green

Red

Red/Green

Non Polar

Polar

Figure 1-33. Electrostatic potential calculation (APBS suite) of blue, rod, green, and red
opsin (calculations of blue, green, and red are based on available homology models)
projected on the Van der Waals surface of retinal (the electrostatic potential of the
retinylidene chromophore was set to zero as to only illustrate the electrostatic
contribution of the protein). The chromophore is divided into three segments; the
qualitative average score for each segment represents the overall electrostatic potentials
that lead to the wavelength regulation of each opsin.
and as the rhodopsin proteins spectrum moves towards red, the negative electrostatic potential
around the PSB region diminishes. The PSB region shows a significant role in modulating the

41

wavelength as suggested by different groups.

86,88,98,102,103,127,128,193

In addition, the

electrostatic potentials around region II and around the β-ionone ring, take part in tuning the

Figure 1-34. Stereoview of R132K:R111L:L121E with residues that are 4 Å away from
the bound all-trans-retinal PSB in cyan color
absorption of rhodopsins and they work cooperatively. For example, region II of blue rhodopsin
shows a more negative dipole as compared with rhodopsin. In contrast, there is an increased
negative dipole around the β-ionone ring (region I) in rhodopsin as compared with blue
rhodopsin. The same trend is also observed in green and red rhodopsins, the red rhodopsin has
more negative electrostatic potential around the β-ionone as compared with green rhodopsin
(Figure 1-33). We would like to test this hypothesis in our rhodopsin protein surrogate. Based
on the projected electrostatic potential map on the bound retinal PSB in different rhodopsins, we
would like to reproduce similar electrostatic potential maps for the bound all-trans-retinal in our
protein surrogate R132K:R111L:L121E by installing polar or apolar residues around specific
regions (I, II and III). In addition, we will also study the relative importance of different region
(I, II and III) on wavelength regulation.

42

1.2.3. Probing the Wavelength Regulation in Rhodopsin using CRABPII–Triple Mutant
R132K:R111L:L121E
After CRABPII was re-engineered to bind all-trans-retinal through the formation of a
PSB with an engineered lysine, mutant R132K:R111L:L121E was used as a template for
studying specific interactions that can induce shifts in the absorption. Based on the crystal
structure, we identified all residues that were 4 Å away from the bound all-trans-retinal PSB
(Phe15, Leu28, Ala32, Ala36, Thr54, Thr56, Val58, Arg59, Asp77 and Met123) (Figure 1-34).
These residues would be mutated in order to study the different hypotheses behind wavelength
regulation. Based on calculations, the electrostatic potential is suggested to have a crucial role in
manipulating the energy between the excited state and the ground state of the bound
chromophore, thus, affecting its λmax (Figure 1-33). In order to obtain a maximum effect,
charged residues were installed. Phe15, Ala32, Ala36, Arg59, Thr56 and Val76 were replaced
with negatively charged residues to probe whether negative electrostatic potential along the
polyene could stabilize the excited state of bound retinal PSB, thus, modulating its absorption

43

Table 1-2. λmax and Kd of all-trans-retinal with R132K:R111L:L121E mutants having a
178

negatively charged residue installed at different positions.
a

Mutant

WT-CRABPII

377

6600 ± 360

No

R132K:R111L:L121E

449

1.36 ± 4.9

Yes

R132K:R111L:L121E:R59E

449

373 ± 16

N.D.

R132K:R111L:L121E:A32E

451

206 ± 36

Yes

R132K:R111L:L121E:A36E

387

206 ± 37

No

R132K:R111L:L121E:T56E

462

728 ± 84

N.D.

R132K:R111L:L121E:F15D

378

456 ± 46

N.D.

R132K:R111L:L121E:V76E
a

λmax (nm)

475

634 ± 64

N.D.

Kd (nM) with
all-trans-retinal

Reductive
b,c
Amination
(+268 m/z)

λmax of bound all-trans-retinal PSB

b

An adduct peak observed in mass spectrum after reductive amination

c

N.D. stand for not determined

(Figure

1-12).

From

Table

1-2,

mutations around the β-ionone ring

A32

region (A32E and R59E) (Figure 1-34)

T56

does not induced any significant protein
shift (λmax shifts from the bound retinal

R59

PSB in ethanol induced by CRABPII) as
compared

to

triple

V76

mutant

Figure 1-35. Position of Arg59, Ala32, Thr56 and
Val76 relative to the bound all-trans-retinal in
However, when glutamate was mutant R132K:R111L:L121E.

R132K:R111L:L121E.

44

installed at position 56 and 76, bathochromic shifts are observed (+13 nm and 26 nm
respectively). Based on the model, both V76E and T56E are close to the middle of the bound
chromophore (Figure 1-35). These results indicate that negatively charged residues barreled
inside the binding pocket can induce a significant effect on λmax of the bound retinal PSB.
However, both mutants R132K:R111L:L121E:V76E and R132K:R111L:L121E:T56E, do not
produce

a

completely

shifted

absorption

spectrum

like

the

456 nm

0.02

0.15

0.1

Retinal
Increase

0.05

Absorbance

Absorbance

UV-vis Spectrum of Retinal with KLE-T56R Absorption of Retinal with KLE-T56R
Deconvoluted by PeakFit(R)
0.2
0.025

0.015

377 nm

0.01
0.005

0
250 300 350 400 450 500 550 600
Wavelength (nm)

0
320 360 400 440 480 520 560 600
Wavelength (nm)

Figure 1-36. Left: UV-vis spectrum of all-trans-retinal titrated with CRABPII mutant
R132K:R111L:L121E:T56R; Right: Deconvolution of the UV-vis spectrum of all-transretinal with CRABPII mutant R132K:R111L:L121E:T56R.
R132K:R111L:L121E triple mutant (Figure 1-29). The absorption peak, therefore, is a mixture
of either unbound retinal (λmax = 377 nm) or retinal SB (λmax = 360 nm) with retinal PSB (λmax
> 430 nm) absorptions. To be able to uncouple the mixture of absorption peaks, we have
®

employed a program (PeakFit ) to deconvolute individual peaks (Figure 1-36). The most redshifted peak is the absorption of bound retinal PSB and it is verified by acid-base titration in

45

CRABPII triple mutant R132K:R111L:L121E. The absorption of bound retinal PSB is more
than 440 nm and can be converted to 360 nm, that is the transition of PSB to SB, by increasing
the pH of the buffer (Figure 1-30). In order to understand whether the small amount of retinal
PSB formed is due to the poor binding affinity with retinal, Kd of retinal with CRABPII mutants
were measured based on the quenching of the fluorescence from tryptophans in CRABPII by the
154,161

bound chromophore.

From the results (Table 1-2), all mutants showed weaker binding

with all-trans-retinal as compared to R132K:R111L:L121E. The binding becomes worse in
R132K:R111L:L121E:V76E, R132K:R111L:L121E:T56E and R132K:R111L:L121E:F15D.
This provided some evidence that the resolved peak around 370 nm in the UV-vis spectrum of
all-trans-retinal

with

R132K:R111L:L121E:V76E,

R132K:R111L:L121E:T56E

or

R132K:R111L:L121E:F15D, is unbound retinal. In addition, addition of acid does not red shift
the spectrum and this clarify that the observed peak at 360 nm is not SB.

Figure 1-37. Stereoview of CRABPII mutant R132K:R111L:L121E crystal structure with
Ala36, Ala32 and Phe15 shown (Cyan).
The three residues, Ala36, Ala32 and Phe15 are located on a helix-turn-helix motif (Figure
1-37). Mutations of a helix motif can result secondary structural change. Helix-turn-helix motifs
have been studied extensively for the last decade because of their interesting fold and

46

F15W

KL:A32E

WT

Figure 1-38. Crystal structure of CRABPII mutants F15W (Green), R132K:R111L:A32E
(Purple) and WT-CRAPBII.
function.

144,194-196

144,194-196

motif.

These motifs require specific sequence in order to fold into helix-turn-helix

The sensitivity of helix-turn-helix in CRABPII has been further confirmed by

the apo-structure of CRABPII mutants F15W and R132K:R111L:A32E. The crystal structure of
both apo-structures with mutations on the helix show a loss of density around the helical region
as compared with the apo-crystal structure of CRABPII and triple mutants R132K:R111L:L121E
(Figure 1-38). Since mutations on the helix-turn-helix could lead to drastic conformational
changes on the helix, it would be difficult to model the mutations precisely, therefore, further
investigations on Ala32, Ala36 and Phe15 were abandoned.
1.2.3.1.

Further Investigations on Wavelength Modulation and Retinal Binding in

CRABPII at Position 76 and 56

47

Table 1-3.
λmax and Kd of all-trans-retinal with R132K:
R111L:L121E mutants having V76 mutation.

Mutant

Kd (nM) with
all-trans-retinal

a

λmax (nm)

To

investigate

the role of position 76
on

wavelength

regulation in CRABPII,

R132K:R111L:L121E:V76

475

634 ± 64

R132K:R111L:L121E:V76F

449

166 ± 13

R132K:R111L:L121E:V76Y

458

523 ± 42

R132K:R111L:L121E:V76W

446

242 ± 28

(Table 1-3).

R132K:R111L:L121E:V76A

450

519 ± 66

spectra with a mixture

R132K:R111L:L121E:V76L

443

319 ± 36

of peaks were observed

R132K:R111L:L121E:V76S

455

634 ± 70

R132K:R111L:L121E:V76R

b

1.36 ± 4.9

R132K:R111L:L121E:V76E

a

449

439

622 ± 117

λmax of bound all-trans-retinal PSB.

An adduct peak observed in mass spectrum after reductive
amination.

a library of mutants
were created and tested
with

all-trans-retinal
Although

for most of the mutants,
deconvolution of the
data resolved the peaks.
The result indicated that

Figure 1-39. Stereoview of model mutant R132K:R111L:L121E with T56E and V76E
mutations shown.

48

mutations at position 76 could induce significant changes on absorption of the bound
chromophore. By comparing the resolved retinal PSB peak across different Val76 mutants, a
trend was observed. As mentioned, Glu at 76 leads to 26 nm red-shift. When a polar (His, Ser
and Tyr) but not charged residues was installed at 76, about 10 nm red-shift was observed. In
order to study the relative position of Glu 76 in the mutant R132K:R111L:L121E:V76E, a
minimized model was constructed. According to the model, the carboxylate of Glu76 is 5.5 Å
away from the C10 and C12 of the bound retinal (Figure 1-39). These results matched the model
84,95

study conducted by Sheves (Figure 1-14) that led to the point-charged theory.

Negative

electrostatic potential along the polyene further away from the PSB region can lead to red-shift.
This is because it can stabilize the resonating positive charge from the iminium nitrogen to the
Table 1-4.
λmax and Kd of all-trans-retinal
R132K:R111L:L121E mutants having T56 mutation.

with

Mutant

λmax (nm)

Kd (nM) with alltrans-retinal

R132K:R111L:L121E:T56

449

R132K:R111L:L121E:T56E

a

middle

of

the

chromophore.
However, the positive

1.36 ± 4.9

destabilizes

the

462

728 ± 84

propagation of charge

474

8451 ± 842

R132K:R111L:L121E:T56I

437

0.58 ± 4.2

R132K:R111L:L121E:T56Q

447

22.2 ± 9

R132K:R111L:L121E:T56W

449

250 ± 14

R132K:R111L:L121E:T56R

b

potential

R132K:R111L:L121E:T56D

a

electrostatic

456

341 ± 14

transfer from the PSB,
thus,

localizing

the

charge on the nitrogen
atom.

This prevents

bathochromic shift or

λmax of bound all-trans-retinal PSB.

An adduct peak observed in mass spectrum after reductive
amination.

causes blue shift (Table
1-3).

49

Mutant R132K:R111L:L121E:T56E also showed a significant protein shift with all-transretinal (Table 1-4).

Installation of Asp at position 56 leading to the mutant

R132K:R111L:L121E:T56D, yielded a protein with poor retinal binding. However, it showed a
larger red shift as compared with mutant R132K:R111L:L121E:T56E.

According to the

modeled protein (Figure 1-40), the favored conformations for both Asp and Glu at 56 are similar,
although the glutamate is further down in the pocket and closer to PSB. Therefore, Asp could
stabilize the resonating positive charge further away from the PSB as compared to Glu at 56.
This

would

lead

to

a

more

red

shifted

pigment

as

compared

to

mutant

R132K:R111L:L121E:T56E. As shown in Table 1-4, that is indeed observed. It is interesting
that when Arg was installed at position 56, it leads to a red-shift that is different from mutant
R132K:R111L:L121E:V76R which caused a blue shift. At present, we do not have a satisfying
reason for this observation.
Charged

T56E

T56D

residues at position 56
and 76 greatly inhibit
the binding of retinal
with CRABPII (Table

Figure 1-40. Model of CRABPII mutants R132K:R111L:L121E: 1-3 and 1-4). Based
T56D (left) and R132K:R111L:L121E:T56E (right).
on
the
sequence
alignment of fatty acid binding proteins, Val76 and Thr56 are highly conserved (Figure 1-41).
Position 56 is most often populated with Ser, Thr or Cys across the family of proteins while at
76, it is mainly Leu, Val or Ile. Therefore, replacement of these conserved residues with charged
residues could greatly affect the folding of the proteins as well as the function of the proteins. In

50

order to probe the later suggestion, we have generated several mutants with hydrophobic,
aromatic and polar mutations at position 76 and 56. From Tables 1-3 and 1-4, we can see that
the binding of these mutants with retinal is recovered as compared with mutants having charged
residues at positions 56 or 76. Also, more retinal PSB was formed in these mutants based on the
UV-vis spectroscopy study as compared with mutants with charged residues installed at position
56 or 76. This indicated that charged residues at those two conserved positions have a great
impact on the affinity towards all-trans-retinal.

hCRABPI
hCRABPII
rCRBPI
rCRBPII
hCRBPIII
hCRBPII
zCRBP
hCRBPIV
hAFABP
mFABPIV
hBFABP
hMFABP
rIFABP
hLFABP
zLFABP
hFABPI

Interestingly, we observed that large

ASKPHVEIRQDGDQFYIKTSTTVRTTEINFKVGEGFEEET--VDGRK
ASKPAVEIKQEGDTFYIKTSTTVRTTEINFKVGEEFEEQT--VDGRP
K--PDKEIVQDGDHMIIRTLSTFRNYIMDFQVGKEFEEDLTGIDDRK
K--PDKEIVQDGDHMIIRTLSTFRNYIMDFQVGKEFEEDLTGIDDRK
K--PDKEIEHQGNHMTVRTLSTFRNYTVQFDVGVEFEEDLRSVDGRK
T--QTKVIDQDGDNFKTKTTSTFRNYDVDFTVGVEFDEYTKSLDNRH
K--QTKVIVQNGDKFETKTLSTFRNYEVNFVIGEEFDEQTKGLDNRT
K--PQKVIEQNGDSFTIHTNSSLRNYFVKFKVGEEFDEDNRGLDNRK
K--PNMIISVNGDLVTIRSESTFKNTEISFKLGVEFDEIT--ADDRK
K--PNMIISVNGDLVTIRSESTFKNTEISFKLGVEFDEIT--ADDRK
K--PTVIISQEGDKVVIRTLSTFKNTEISFQLGEEFDETT--ADDRN
K--PTTIIEKNGDILTLKTHSTFKNTEISFKLGVEFDETT--ADDRK
---LKLTITQEGNKFTVKESSNFRNIDVVFELGVDFAYSL--ADGTE
K--PIVEIQQKGDDFVVTSKTPRQTVTNSFTLGKEADITT--MDGKK
K--PVTEIQQNGSDFTITSKTPGKTVTNSFTIGKEAEITT--MDGKK
K--GVSEIVQNGKHFKFTITAGSKVIQNEFTVGEECELET--MTGEK

Figure 1-41. Sequence alignment of CRABPII with proteins belong to Fatty Acid Binding
Protein (FABP); hCRABPI: human CRABPI; hCRABPII: human CRABPII; rCRBPI:
rat Cellular Retinol Binding Protein (CRBP) I; rCRBPII: rat CRBPII; hCRBPIII: human
CRBPIII; hCRBPII: human CRBPII; zCRBP: zebra CRBP; hCRBPIV: human CRBP
IV; hAFABP: human adipocyte FABP; mFABPIV: mouse FABP IV; hBFABP: human
Brain-type FABP; hMFABP: human muscle FABP; rIFABP: rat Intestinal FABP;
hLFABP: human Liver FABP; zLFABP: zebra Liver FABP; hFABPI: human FABP I.

51

hydrophobic residues (Leucine or Isoleucine) at position 56 and 76 could induce blue-shift of the
bound retinal PSB (Table 1-3 and 1-4).
1.2.3.2.
Mutation

Investigating the Effect of
at

position

59

on

Wavelength Regulation of CRABPII
As mentioned, mutations on the
helix might greatly destabilize the helixturn-helix

Arg59

motifs

in

CRABPII.

Therefore, further investigations on
Phe15, Leu28, Ala32 and Ala36 (all

Figure
1-42.
Crystal
structure
of
R132K:R111L:L121E with all-trans-retinal PSB
(Purple) and Arg59 shown.

although they are close to the bound retinal

R132K:R111L:L121E:R59E Titrated
with all trans-retinal
0.06

PSB (Figure 1-34 and 1-38).

From the

0.05

R132K:R111L:L121E

structure,

0.04

crystal

Arg59 is close to the β-ionone ring and on
the β-sheet (Figure 1-42).
previously,

As mentioned
mutant

Absorbance

presence on the α-helix) were not conducted

0.03
0.02
0.01

R132K:R111L:L121E:R59E does not induce

0
250 300 350 400 450 500 550 600
Wavelength (nm)

any change on the absorption of bound

retinal PSB (Table 1-2). Although mutant Figure
1-43.
UV-vis
spectra
of
R132K:R111L: L121E:R59E titrated with allR132K:R111L:L121E:R59E produced a trans-retinal.
fully-shifted UV-vis spectrum with retinal (Figure 1-43), charged residues could lead to

52

Table 1-5. λmax and Kd of all trans-retinal with R132K:R111L:L121E mutants
having R59 mutation.
a

Mutant

Reductive
Kd (nM) with
b,c
all-trans-retinal Amination
(+268 m/z)

R132K:R111L:L121E:R59

449

1.36 ± 4.9

Yes

R132K:R111L:L121E:R59E

450

373 ± 16

N.D.

R132K:R111L:L121E:R59Q

443

16 ± 7.5

N.D.

R132K:R111L:L121E:R59Y

444

133 ± 19

Yes

R132K:R111L:L121E:R59W

443

38 ± 8.2

N.D.

R132K:R111L:L121E:R59L

461

155 ± 9

Yes

R132K:R111L:L121E:R59D

452

N.D.

Yes

R132K:R111L:L121E:R59A
a

λmax
(nm)

448

92 ± 9

Yes

λmax of bound all-trans-retinal PSB.

b

An adduct peak observed in mass spectrum after reductive amination.

c

N.D. stands for not determined.

conformational changes on CRABPII as suggested by the model of R132K:R111L:L121E:V76E
(Figure 1-44). Moreover, liver fatty acid binding protein in chicken, which has a Gln at position
59 instead of Arg in CRABPII, showed a loop movement of more than 2 Å (Figure 1-45). That
suggests that mutations at position 59 could result in conformational change of the loop. To
fully study the effect of Arg59 mutation, a library of mutants with polar, aromatic and
hydrophobic residues installed at position 59 were created and tested with retinal (Table 1-5). In
general, all Arg59 mutants showed a fully shifted peak with retinal. However, the absorption of
bound

all-trans-retinal

PSB

is

not

sensitive

to

any

mutations

at

59

except

R132K:R111L:L121E:R59L. When leucine, a large hydrophobic residue, was introduced at
position 59, a 12 nm red-shift was induced. The 12 nm bathochromic shift of Arg59 mutant is
surprising and presently cannot be explained.
53

Figure 1-44.
Stereoview of overlay between apo-structure of CRABPII mutant
R132K:R111L:L121E (cyan) and model of CRABPII mutant R132K:R111L:L121E:V76E
(green).

Figure 1-45. Left: Overlay of holo-structure between CRABPII (Green) and chicken
LFABP (Cyan); Right: Overlay of apo-structure between CRABPII (Green) and chicken
LFABP (Cyan).
1.2.4. Probing the Wavelength Regulation in Rhodopsin using CRABPII and C15 aldehyde
As analyzed above, mutations on CRABPII do not induce a pronounced effect on the
absorption of the bound all-trans-retinal PSB (Table 1-2, 1-3, 1-4 and 1-5). That development
led us to further analyze the crystal structure of R132K:R111L:L121E with retinal in order to
understand the spectroscopic results. Based on the crystal structure, all-trans-retinal is not fully
embedded within CRABPII and the β-ionone ring is exposed to the solvent (Figure 1-46).

54

CRABPII
Surface

Arg59
Figure 1-46. Crystal structure of all-trans-retinal with
CRABPII mutant R132K:R111L:L121E with protein surface
(Black) and solvent exposed β-ionone ring (Cyan) shown.

Arg59

Arg59

Figure 1-47. Electrostatic potential calculation using APBS on CRABPII mutant
R132K:R111L:L121E. Left: Calculation based on vacuum environment; Right:
Calculation based on water environment (dielectric constant 78 was used).
Therefore, the effect of mutations close to the β-ionone ring, particularly position 59, is buffered
by the aqueous environment.

It has been well-documented that bulk water could greatly

55

neutralize the electrostatic potential generated by the
O polar or charged residues.180,182,185

Thus, the

1-17

electrostatic changes originating from charged or polar
residues installed at 59 could be neutralized by the

O
1-22

surrounding water. In order to probe this, an APBS
calculation was performed to study the electrostatic
O

potential projected on bound retinal chromophore by

1-23

the protein in the presence of water and in vacuum.
Figure 1-48. All-trans-retinal 1-17
and its short analogs (C17 aldehyde The calculations indicated that the water environment
1-22 and C15 aldehyde 1-23).

greatly neutralizes the electrostatic potential induced by

Arg59 (Figure 1-47). Because of this observation, our approach to study wavelength regulation
in CRABPII was revised.
A possible solution to fully
embed the chromophore within the
binding pocket would be to use a
shorter chromophore. Assuming that
the Schiff base with the new shorter
chromophore will form at a similar
position

as

that

of

retinal

in

CRABPII, a shorter chromophore
Figure 1-49. Overlay between holo-structure of
could move the β-ionone ring inside R132K:R111L:L121E with all-trans-retinal (Purple)
and
energy
minimized
models
of
the protein, thus, protecting the whole R132K:R111L:L121E with C aldehyde (Green) and
17
chromophore from the bulk water. C15 aldehyde (Blue) (protein surface in black
shadow).
56

Energy minimized models were built by incorporating shorter retinal analogs (Figure 1-48) with
CRABPII (Figure 1-49). By comparing the CRABPII models of C15 aldehyde (removing 5
carbon) and C17 aldehyde (removing 3 carbon) with the holo structure of CRABPII, one can
notice that C15 aldehyde was totally embedded within CRABPII (Figure 1-49).

1.2.4.1. Synthesis of C17 aldehyde and C15 aldehyde
To verify the later hypothesis, both C15 and C17 aldehydes were synthesized using a
standard protocol for all-trans-retinal synthesis (Scheme 1-1).

197

β-Ionone (1-24) was coupled

with diethyl cyanomethyl-phosphonate forming 3-methyl-5-(2,6,6-trimethylcyclhex-1-en-1-yl)
penta-2,4-dienenitrile (1-25) with a 3:1 E:Z ratio at C13. Nitrile 1-25 was reduced by DIBAL-H
at 0 ºC to yield C15 aldehyde (1-23) as a mixture of isomers. All-trans-C15 aldehyde was
purified by chromatography (Hexane:EtOAc / 9:1). C17 aldehyde (1-22) was synthesized in
similar manner. Briefly, continuing from C15 aldehyde (1-23), it was coupled with diethyl
cyanomethylphosphonate followed by nitrile reduction using DIBAL-H.

1.2.4.2. Binding of C17 aldehyde with R132K:R111L:L121E mutants
C15 Aldehyde and C17 aldehyde were tested with R132K:R111L: L121E.

Both

chromophores showed a UV-vis shift upon incubating with R132K:R111L: L121E (Figure 1-50).
C17 aldehyde showed a fully shifted spectrum while C15 aldehyde showed a predominant peak at

57

NaH, (EtO)2P(O)CH2CN

CN

O THF, rt, 90%, E:Z / 3:1
1-24

1-25

1) DIBAL-H in THF,
-78 ˚C
0 ˚C
2) Silica-H2O, 0 ˚C,
60 %, E:Z / 3:1

O
1-23

NaH, (EtO)2P(O)CH2CN
THF, rt, 90 %, E:Z / 3:1

CN

1-26
1) DIBAL-H, THF,
-78 ˚C
0 ˚C
2) Silica-H2O, 0 ˚C,
40 %, E:Z / 3:1
O
1-22
Scheme 1-1. Synthesis of C15 aldehyde 1-23 and C17 aldehyde 1-22.
337 nm that is either C15 aldehyde SB or unbound C15 aldehyde absorption (Figure 1-51). To
verify whether C17 aldehyde is short enough to interact effectively with CRABPII, C17 aldehyde
was tested with a series of Arg59 mutants, the results of which are shown in Table 1-6. Based on
the result, the mutations at position 59 do not significantly affect the absorption of the bound C17

58

UV-vis Spectrum of CRABPII
UV-vis Spectrum of CRABPII
Mutant KLE titrated with C17 aldehyde Mutant KLE titrated with C15 aldehyde
0.1
0.1

PSB formed

0.06
0.04

Absorbance

Absorbance

0.08

0.02

0.08
0.06

PSB formed

0.04
0.02

0

0
250 300 350 400 450 500 550 600
Wavelength (nm)

250 300 350 400 450 500 550 600
Wavelength (nm)

Figure 1-51. Absorption spectrum of R132K:R111L:L121E with C15 aldehyde 1-23
(Right) and C17 aldehyde 1-22 (Left) showing the PSB formation.
aldehyde, much like Arg59 mutants bound with all-trans-retinal (Table 1-6). However, mutant
R132K:R111L:L121E:R59E showed a blue shifted spectrum (λmax ~ -18 nm). The result agree
with the energy minimized model of C17 aldehyde with CRABPII (Figure 1-52) suggesting that
its β-ionone ring is still solvent exposed but the Glu59 is close to the middle of the bound C17
aldehyde in the model, therefore, leading to blue shift. Since C17 aldehyde is about 2.5 Å shorter
than all-trans-retinal, it was expected the β-ionone ring should be fully covered by the protein.
But, the model shows that the β-ionone ring of C17 aldehyde moves only 1.2 Å as compared with
the β-ionone ring of all-trans-retinal in the crystal structure. This is because Lys132 is more
relaxed and the PSB is situated further away from the binding pocket (1.3 Å) to compensate for
the shortened chromophore (Figure 1-52).

59

1.3 Å
1.3 Å

1.2 Å
1.2 Å

Figure 1-52. Overlay between crystal structure of R132K:R111L:L121E with all-transretinal (Purple) and model of R132K:R111L:L121E with C17 aldehyde (Green).
1.2.4.3. Binding of C15 aldehyde with CRABPII mutant R132K:R111L:L121E
We then tested C15 aldehyde with CRABPII mutants in order to see if it would be more
sensitive than C17 aldehyde and all-trans-retinal.

2.2 Å

As mentioned previously, it was expected that the
β-ionone ring would move 4.8 Å deeper inside the
protein. However, based on the model of C15

2.6 Å
2.6
2.2 Å
aldehyde with R132K:R111L:L121E, the βFigure 1-53. Overlay between crystal ionone ring of C15 aldehyde moves ~ 2.6 Å
structure of R132K:R111L:L121E with
all-trans-retinal (Purple) and model of deeper inside the binding pocket, however, is
R132K:R111L:L121E with C15 aldehyde
totally embedded in the protein. Similar to the
(Blue).

60

Table 1-7. λmax and Kd of all-trans-C15 with R132K:R111L:L121E mutants having R59
mutation.
a

Mutant

λmax
(nm)

Reductive
Kd (nM)
b
all-trans-retinal Amination
(+268 m/z)

R132K:R111L:L121E:R59

400

9030 ± 925

Yes

R132K:R111L:L121E:R59E

424

1 ± 4.3

Yes

R132K:R111L:L121E:R59D

414

261 ± 35

Yes

R132K:R111L:L121E:R59Q

416

583 ± 68

Yes

R132K:R111L:L121E:R59W

404

139 ± 25

Yes

R132K:R111L:L121E:R59Y

389

77 ± 16

Yes

410

N.D.

N.D.

R132K:R111L:L121E:R59A

405

635 ± 61

Yes

R132K:R111L:L121E:R59T

407

901 ± 65

Yes

R132K:R111L:L121E:R59L

a

c

λmax of bound C15 aldehyde PSB identified by deconvolution.

b

An adduct peak observed in mass spectrum after reductive amination.

c

N.D. stands for not determined.

C17 aldehyde model, the nitrogen of Lys132 moves away from the binding pocket by ~ 2.2 Å to
compensate for the total length changed (4.8 Å) from all-trans-retinal to C15 aldehyde (Figure 153). Bound C15 aldehyde showed a slight change in UV-vis spectrum upon incubation with
R132K:R111L:L121E. In order to verify the formation of C15 aldehyde SB, reductive amination
of the C15 aldehyde with R132K:R111L:L121E was conducted and an adduct peak [M + 216]

+

was observed in the mass spectrum that indicated the formation of the C15 aldehyde SB. In
addition, based on the fluorescence quenching experiment, C15 aldehyde binds poorly with

61

R132K:R111L:L121E (Table 1-7), helping to explain the slight shifted spectrum (Figure 1-51),
since the majority of C15 aldehyde is unbound (the λmax of the bound C15 aldehyde PSB was
identified to be about 400 nm based on the deconvolution, Table 1-7). The absorption is 20 nm
red-shifted as compared with the C15 aldehyde PSB in ethanol. The 20 nm bathochromic shift is
more than the retinal with R132K:R111L:L121E. This could be due to the fact that the C15
aldehyde is fully covered by the CRABPII binding pocket.
1.2.4.4.

Exploring the Effect of Mutations at Position 59 on the Wavelength

Modulation in CRABPII using C15 aldehyde
Assuming that the β-ionone
UV-vis Spectrum of CRABPII Mutant KLE:R59E
titrated with C15 aldehyde
0.8

1 equivalent

0.7

embedded

within

L121E-CRABPII,

0.6

Absorbance

ring of the C15 aldehyde is fully
R132K:R111L:
Arg59

0.5

were

0.4

R132K:R111L:L121E: R59E showed

0.3

424 nm

0.2

Mutant

a significant red-shifted spectrum (∆
-1

+44 nm, 2731 cm ) (Figure 1-54).

0.1
0
300

evaluated.

mutants

350
400
450
Wavelength (nm)

In addition, C15 aldehyde binds well

500

Figure 1-54.
UV-vis spectra of mutant
R132K:R111L:L121E:R59E titrated with C15
aldehyde.

with

mutant

R132K:R111L:

L121E:R59E and the majority forms
a protonated Schiff base (Kd ~ 1 nM)

62

(Table 1-7). To further probe the effect of Arg59 mutation on the absorption of bound C15
aldehyde PSB, a series of mutants at position 59 were measured. The results (Table 1-7) showed
that negative charge or dipole close to the ionone ring can induce red-shifting of the bound
chromophore. This result indicated that the revised point-charge theory, based on the change in
electrostatic potential at along the chromophore, can affect the absorption (Figure 1-12). In
addition,

by

comparing

mutants

R132K:R111L:L121E:R59E

(424

nm),

R132K:

R111L:L121E:R59Q (416 nm), R132K:R111L: L121E:R59L (410 nm) and R132K:R111L:
L121E:R59 (400 nm), we can conclude that the red-shifting is directly proportional to the
strength of the negative dipole (Table 1-7 and Figure 1-55). Moreover, mutant R132K:R111L:
L121E:R59D (414 nm) is less red-shifted thanR132K:R111L:L121E: R59E (424 nm) when
bound with C15 aldehyde. Aspartate, that
is one C-C bond shorter than glutamate, is
further away from the bound C15 aldehyde

electrostatic potential projected on the
bound chromophore by Asp59 is less as
compared with Glu59.
R132K:R111L:L121E:R59D
causes

less

red

Accordingly,
(414
shift

nm)
than

0.12

More electronegative
Normalized Absorbance

PSB. Therefore, the overall negative

0.1
0.08
0.06
0.04
0.02
0

R132K:R111L:L121E:R59E

(424

upon binding with C15 aldehyde.

KLE
KLE:R59E
KLE:R59Q
KLE:R59L

nm)

350

400
450
Wavelength (nm)

Figure 1-55.
Deconvoluted spectrum of
R132K:R111L:L121E with R59 mutations

63

The crystal structure of C15 aldehyde with R132K:R111L:L121E:R59W was solved by Dr.
Jia (Figure 1-56).

198

The overall holo-structure of R132K:R111L:L121E:R59W with C15

aldehyde is similar to holo-structure of R132K:R111L:L121E with all-trans-retinal. It contains
β-barrel, comprised of 10 anti-parallel β-sheets, with a helix-turn-helix cap (Figure 1-56). The
trajectory of the bound all-trans-retinal PSB in the mutant R132K:R111L:L121E holo-structure
is very similar to the bound C15 aldehyde PSB in mutant R132K:R111L:L121E:R59W (Figure 157). As predicted, the bound C15 aldehyde PSB in R132K:R111L:L121E:R59W is completely
embedded in the protein pocket.

As compared to the R132K:R111L:L121E with all3Å

C15

All transretinal

Figure 1-56. Stereoview of overlay between crystal structure of R132K:R111L:L121E with
all-trans-retinal (Cyan) and R132K:R111L:L121E:R59W with C15 aldehyde (C15) (Green)
trans-retinal structure, the β-ionone ring of the bound C15 aldehyde is about 4.8 Å deep inside
the binding pocket as compared to the β-ionone ring of bound all-trans-retinal in mutant
R132K:R111L:L121E. This further indicates that the enhanced sensitivity of the bound C15

64

aldehyde towards Arg59 mutations is
because the chromophore is completely

All-transretinal

C15 aldehyde

protected by the protein from the bulk
water surrounding the protein.
1.2.4.4.1. Investigating the Element for
Bound C15 aldehyde PSB Stabilization

As mentioned, bound C15 aldehyde
Figure 1-57.
Overlay of crystal structures
between all-trans-retinal in R132K:R111L:L121E
PSB with Arg59 mutants has a lower pKa.
and
C15
aldehyde
in
R132K:R111L:L121E:R59W.
value (pKa ~ 7.8) as compared to bound all-trans-retinal PSB in triple mutant
R132K:R111L:L121E

(pKa

~

8.9).

Based on the holo structure of mutant
R132K:R111L:L121E:R59W bound to

2.9 Å

C15 aldehyde, the Glu121 is 4.8 Å away

4.8 Å
2.8 Å

from the imine nitrogen (Figure 1-58).
However, the imine nitrogen in mutant

BTP
Glu121

R132K:R111L:L121E bound to all-

1-58.
Crystal
structure
of
trans-retinal is only 2.6 Å away from Figure
R132K:R111L:L121E:R59W with C15 aldehyde
Glu121. Thus, Glu121 is 2.2 Å further with Glu121 and BTP shown.
away from the iminium in the C15 bound retinal system as compared to protein bound to alltrans-retinal.

In addition, the crystal structure showed that there is 1,3-bistris(hydroxy-

methyl)methyl-amino)propane (BTP) molecule (from the crystallization buffer) next to the PSB

65

and the hydroxyl group from BTP hydrogen bonds (2.8 Å) with the imine nitrogen and Glu121
(2.9 Å). Therefore, Glu121 can interact with the bound PSB through a hydroxyl-mediated
interaction of BTP. Since there is no other negatively charged residue except Glu121 (4.8 Å
from the PSB) around the PSB of bound C15 aldehyde, the pKa value of the bound C15 aldehyde
PSB in mutant R132K:R111L:L121E:R59W (pKa ~ 7.7) is lower than the pKa value of bound
all-trans-retinal (pKa ~ 8.7).
The BTP molecule is also present in several apo-strcutures of CRABPII mutants and is
located at the same position as the holo-structure of mutant R132K:R111L:L121E:R59W (Figure
1-59).

198

However, in our UV-vis spectroscopic study, the phosphate saline buffer (PBS) we

used does not contain BTP; thus, the space occupied by BTP must be occupied by

Lys132

2.8 Å

4.8 Å
2.9 Å

Glu121

BTP

Figure 1-59. Overlay of R132K:R111L:L121E (Blue) and R132K:R111L:L121E:R59E
(Cyan) with holo-structure of R132K:R111L:L121E:R59W (Green). All three structures
contain BTP at the same position.
other molecules. Although we do not have a holo-crystal structure and the apo-structure of
R132K:R111L:L121E:R59W

without

R132K:R111L:L121E:Y134F

and

BTP

molecule,

an

apo-structure

R132K:R111L:L121E:Y134F:T54V

66

has

of
been

Y134 /
F134

T54

Y134 /
F134

T54 /
Glu121 V54

Glu121

Figure 1-60. Right: Overlay of R132K:Y134F:R111L:L121E apo-structure (Purple) with
R132K:R111L:L121E:R59W holo-structure (Green) with ordered water shown in purple
sphere; Left: Overlay of R132K:Y134F:R111L:L121E:T54V apo-structure (Cyan) with
R132K:R111L:L121E:R59W holo-structure (Green) with ordered waters shown in Cyan
spheres.
198,199

crystalized.
C15

aldehyde

By comparing the holo-structure of R132K:R111L:L121E:R59W bound to

with

the

apo-structures

of

mutant

R132K:R111L:L121E:Y134F

and

R132K:R111L:L121E:Y134F:T54V, an ordered water molecule in these apo-structures is
positioned where the hydroxyl group of BTP is in the C15 aldehyde holo-structure (Figure 1-60).
Interaction of the PSB with its putative counteranion through water-mediated hydrogen bonds is
a well-characterized system in bacteriorhodopsin and sensory rhodopsin (Figure 1-20).
Therefore, it is possible that Glu121 (counterion) also stabilizes the bound C15 aldehyde PSB in
R132K:R111L:L121E mutants through water-mediated interactions.
1.2.4.4.2. Studying the Effect of Electrostatic Potential on Wavelength Regulation based on
CRABPII R59 Mutant.
As discussed in Section 1.1.2 and 1.2.2, the electrostatic potential was suggested to play an
important role in wavelength regulation in rhodopsins.

With the crystal structure of C15

aldehyde with R132K:R111L:L121E:R59W in hand, we were able to conduct a more accurate
67

187-190

electrostatic calculation using the APBS package.
models

(R132K:R111L:L121E:R59

and

We have generated another two

R132K:R111L:L121E:R59E)

R132K:R111L:L121E:R59W holo-structure as a template.

using

the

The conformations of Arg59 in

R132K:R111L:L121E and Glu59 in R132K:R111L:L121E:R59E were chosen based on the holostructure

of

R132K:R111L:L121E

with

all-trans-retinal

and

apo-structure

of

R132K:R111L:L121E:R59E in order to create more accurate models. As we replaced Arg59
with Trp and then Glu, the electrostatic potential around the β-ionone ring becomes more
negative.

By

comparing

KLE:R59: 400 nm

R59

the

electrostatic

KLE:R59W: 408 nm

R59W

potential

and

the

absorption

KLE:R59E: 424 nm

of

KbT / ec
- 40

R59E
0

+40

Figure 1-61. Electrostatic potential around bound C15 aldehyde in different CRABPII
mutants (Left: R132K:R111L:L121E:R59; Middle: R132K:R111L:L121E:R59W; Right:
R132K:R111L:L121E:R59E) calculated by APBS package with the corresponding
absorption shown.
R132K:R111L:L121E:R59,

R132K:R111L:L121E:R59E

models

and

R132K:

R111L:L121E:R59W crystal structures around the bound C15 aldehyde (Figure 1-61), we
identified a pattern similar to human rhodospins (Figure 1-32). The calculations showed that the
λmax of the protein-chromophore complexes become more red-shifted as the negative
electrostatic potential around the β-ionone ring increases (Figure 1-61). These calculations with
68

the UV-vis spectroscopic data further supported the argument that the electrostatic potential in
proteins affects the absorptions of bound retinal. The stabilization of a resonating positive
charge originating from the PSB at different positions along the polyene is altered by the
localized electrostatic potential (Figure 1-32).
1.2.4.5. Probing the Effect of Negative Charged Residues on Wavelength Regulation
in CRABPII
As C15 aldehyde has showed to

Ala36

be more sensitive than C17 aldehyde and
Arg59

all-trans-retinal towards the mutations at
position 59, we further studied the effect
Ala32

Phe15

of

the

electrostatic

wavelength

Val76

placement

potential

regulation
of

negative

on

through
residues

at

Figure
1-62.
Crystal
structure
of
different positions (Val76, Thr56, Ala36,
R132K:R111L:L121E:R59W with C15 aldehyde.
Negative charges are installed at the position Ala32 and Phe15) along the polyene of
highlighted in blue.
bound C15 aldehyde (Figure 1-62). From Table 1-8, one can see that the negative charges at
different positions around the bound C15 aldehyde can induce different changes on C15 aldehyde
absorption. As the negative charge moves closer to the PSB, it induces a blue shift (Figure 1-62
and Table 1-8).

Based on the crystal structure of R132K:R111L:L121E: R59W with C15

aldehyde, Ala36 is pointing outside the binding pocket (Figure 1-62), therefore, mutation at this
position should not have any significant effect on the λmax of the bound chromophore in R132K:

69

R111L:L121E:A36E (402nm) as compared to R132K:R111L:L121E (Table 1-8).

R132K:

R111L:L121E:T56E did not induce any UV-vis spectrum change upon binding with C15
aldehyde.

According to the fluorescent quenching experiment, R132K:R111L:L121E:T56E

Table 1-8. λmax and Kd of all-trans-C15 with R132K:R111L:L121E mutants having
negative residues installed at different positions.
a

Mutant

1 ± 4.3

Yes

416

1256 ± 202

Yes

R132K:R111L:L121E:A36E

402

46 ± 6.7

Yes

R132K:R111L:L121E:T56E

c

334

N.D.

N.D.

R132K:R111L:L121E:F15D

397

1068 ± 108

Yes

R132K:R111L:L121E:V76E

c

424

R132K:R111L:L121E:A32E

b

Kd (nM) with Reductive
b
all-trans-retinal Amination
(+268 m/z)

R132K:R111L:L121E:R59E

a

λmax
(nm)

392

1808 ± 192

Yes

λmax of bound C15 aldehyde PSB identified by deconvolution
An adduct peak observed in mass spectrum after reductive amination.
N.D. stand for not determined.

showed a weak binding (µM) with C15 aldehyde. As mentioned in section 1.2.3.1., Thr56 is a
conserved residue, thus, installing a charged residues at this position could induce structural
changes that can affect the binding of the chromophore. In addition, introduction of glutamate at
position 56 can result in increased steric inside the protein, hindering the binding of C15
aldehyde.

70

1.2.4.6.

Exploring the Effect of

Mutation at Position 76 on Wavelength
Regulation in CRABPII using C15
aldehyde
6.0 Å

As discussed in section 1.2.3.1,
placement of charged residues at position

V76

76 leads to changes on the absorption
spectrum of bound retinal (Table 1-3).
The C15

Figure 1-63. Crystal structure of C15 aldehyde
aldehyde was also tested with a bound to mutant R132K:R111L:L121E:R59W.

series of Val76 mutants in order to investigate the effect of charged mutations on wavelength
regulation. The UV-vis spectra showed that mutations at position 76 generally showed blue shift
as compared with mutant R132K:R111L:L121E (Table 1-9). The mutations (polar, negative,

Table 1-9. λmax and Kd of all-trans-C15 with R132K:R111L:L121E mutants having
negative residues installed at different positions.
a

Mutant

400

9030 ± 925

Yes

R132K:R111L:L121E:V76E

392

169 ± 44

Yes

R132K:R111L:L121E:V76F

392

1808 ± 192

Yes

R132K:R111L:L121E:V76W

390

2371 ± 130

Yes

R132K:R111L:L121E:V76Y

388

1847 ± 288

Yes

R132K:R111L:L121E:V76R

b

Kd (nM) with Reductive
b
all-trans-retinal Amination
(+268 m/z)

R132K:R111L:L121E:V76

a

λmax
(nm)

394

809 ± 32

Yes

λmax of bound C15 aldehyde PSB identified by deconvolution
An adduct peak observed in mass spectrum after reductive amination

71

positive, aromatic) at position 76 do not induce any spectral changes among each other. Based
on the crystal structure of C15 aldehyde with mutant R132L:R111L:L121E:R59W (Figure 1-63),
Val76 is further away from the bound C15 aldehyde (~6.1 Å) as compared to the distance of
Val76 with bound all-trans-retinal in triple mutant (R132K:R111L:L121E).

Therefore, the

mutation should not induce a significant effect on the bound C15 aldehyde PSB. In addition, as
we have mentioned in section 1.2.3.1., Val 76 is a conserved residue. Thus, mutation at position
76 might result in conformational changes of the protein.

1.2.4.7. Study the Effect of Residues Around the Bound C15 aldehyde PSB
As mentioned in Section 1.1.4., residues around the PSB are believed to be crucial for
wavelength regulation in rhodopsin proteins. We observed changes on Kd, pKa and λmax of
bound all-trans-retinal PSB with mutants of Arg111 (Table 1-10).

However, mutant

R132K:L121E, where the positive charge was maintained at position 111, showed a 9 nm red
shift as compared with mutant R132K:R111L:L121E. The destabilization of the positive charge
on the iminium nitrogen by other positive charge bearing amino acid close by leads to the red
shift (Figure 1-12). When C15 aldehyde was incubated with mutant R132K:L121E, a 10 nm
redshift was also observed as compared with R132K:R111L:L121E (Table 1-10).

It is

interesting that although Arg111 is 9.8 Å away from the PSB, it can induce changes in the
absorption of the bound C15 aldehyde. By comparing the crystal structure of apo-structure
R132K:L121E with holo-structure of R132K:R111L:L121E:R59W with C15 aldehyde, Arg111
is hydrogen bonded with one of the Glu121 conformers (Figure 1-64).

72

There are two

conformations of the Glu121 (Glutamate A and B) and both of them have 50% occupancy which
means they both contribute equally. When Glu121 takes conformation B that hydrogen bonds
with Arg111, Glu121 would be 0.9 Å further away from the PSB as compared to original
Table 1-10. λmax and Kd of all-trans-C15 with R132K:R111L:L121E mutants having
mutation at Arg111.

179

a

Mutant

449

1.4 ± 4.9

Yes

8.7

R132K:R111:L121E

457

104 ± 11

Yes

7.8

R132K:R111V:L121E

449

31 ± 63

Yes

8.9

R132K:R111H:L121E

447

4.1 ± 4.7

Yes

7.8

R132K:R111K:L121E

b

Kd (nM) with Reductive
b
all-trans-retinal Amination pKa
(+268 m/z)

R132K:R111L:L121E

a

λmax
(nm)

460

236 ± 18

Yes

8.1

λmax of bound C15 aldehyde PSB identified by deconvolution
An adduct peak observed in mass spectrum after reductive amination

conformation, therefore, the overall negative electrostatic potential, induced by Glu121, around
the PSB region would be decreased, and that could lead to red shift.
As mentioned before, Glu121 makes a water mediated interaction with the PSB (Figure 161). We would like to investigate the importance of Glu121 in C15 aldehyde PSB formation and
stabilization. When all-trans-retinal binds to R132K:R111L:L121E, Glu121 is the counterion.
Therefore, placement of a non-charged residue at 121 has a big impact on the pKa of bound
retinal PSB (No PSB was formed at pH 7.3 with mutant R132K:R111L:L121)

178,179

but not on

the absorption. Several Leu121 mutants were tested with C15 aldehyde (Table 1-11). The pKa of
C15 aldehyde PSB in R132K:R111L:L121Q (pKa ~ 7.7) was higher than that of C15 aldehyde
73

The plane
of PSB

Figure 1-64. Left: Overlay between holo-structure of R132K:R111L:L121E:R59W with
C15 aldehyde (Cyan) and apo-structure of R132K:R111:L121E (Yellow) with Glu121
showed in two conformations (Conf.): Glu Conf. A overlapped with Glu121 in
R132K:R111L:L121E:R59W; Glu Conf. B hydrogen bonding with Arg 111 in
R132K:R111:L121E; Right: The crystal structure of R132K:R111LE:R59W with BTP
showed. The ordered water is not in the plane of PSB.
PSB in R132K:R111L: L121E (pKa ~ 6.7). Replacement of negative charged residue (Glu) with
a polar residue (Gln) weakens the hydrogen bonding with the ordered water. Therefore, the
ordered water could adopt a better geometry to interact with PSB. Based on the overlay model
between

C15

aldehyde-R132K:R111L:L121E:R59W

holo-structure

and

mutant

R132K:Y134F:R111L:L121E apo-structure, Glu121 interacts the PSB through water-mediated
interactions (Figure 1-64). L121Q mutation weakens the hydrogen bonding with the ordered
water, leading to a stronger interaction of the same water with the bound C15 aldehyde PSB, thus
enhancing its pKa. This is our hypothesis for the difference observed between sensory rhodopsin
II and bacteriorhodopsin. It is proposed that the different water network around PSB between
bacteriorhodopsin (pKa ~ 13.3) and sensory rhodopsin II (pKa ~ 12.0) leads to different pKa
value between these two rhodopsins. Modulating the interactions between both Asp202 and

74

water, that hydrogen bonds to PSB, greatly affects the pKa of the bound retinal PSB (Figure 120).

117,118,127,193

The important role of Glu121 in modulating the pKa of bound C15 aldehyde

PSB has also been demonstrated by another mutant R132K:R111L:L121D. The bound C15
aldehyde PSB in R132K:R111L:L121D showed slightly increased pKa of the bound C15
aldehyde PSB (Table 1-11). In addition, we found that L121 plays an important role for
wavelength modulation in CRABPII with C15 aldehyde, since the removal of negative charge at
position 121 enhances the red-shifting (Table 1-11). As mentioned before, electrostatic potential
around the chromophore is important for wavelength regulation in rhodospin. Electrostatic
potential of both mutants R132K:R111L:L121Q and R132K:R111L:L121D were calculated by
APBS. It showed that the negative electrostatic potential around the C15 aldehyde PSB for both
Table 1-11. λmax and Kd of all-trans-C15 with R132K:R111L:L121E mutants having
negatively charged residues installed at different positions.
Mutant

a

400

9030 ± 925

Yes

6.7

410

1845 ± 220

Yes

N.D.

R132K:R111L:L121D

418

147 ± 20

Yes

7.1

R132K:R111L:L121Q

413

208 ± 28

Yes

7.7

R132K:R111L:L121

c

pKa

R132K:R111:L121E

b

Kd (nM) with Reductive
b
all-trans-retinal Amination
(+268 m/z)

R132K:R111L:L121E

a

λmax
(nm)

398

106 ± 12

Yes

8.6

λmax of bound C15 aldehyde PSB identified by deconvolution
An adduct peak observed in mass spectrum after reductive amination
N.D. stands for not determined.

75

c

KLE

KLD

KLQ

KbT / ec
-40

Glu121

Asp121

Gln121
0

+40

Figure 1-65. Electrostatic potential around the bound C15 aldehyde in different CRABPII
mutants (Left: R132K:R111L:L121E; Middle: R132K:R111L:L121D; Right:
187-192
R132K:R111L:L121Q) calculated by the APBS package.
mutants is less than that in mutant R132K:R111L:L121E (Figure 1-65), therefore, promotes the
charge propagation along the polyene which leads to red-shifted.
These results further support the argument that reduction of the electrostatic potential
around the PSB enhances the red-shifting. Therefore, further increase in red-shifting should be
possible by complete removal of the polarity at position 121 (Figure 1-65). This should also
modulate the pKa of the bound C15 aldehyde PSB. Mutant R132K:R111L, having hydrophobic
leucine at 121, showed a 20 fold increase in pKa for the bound C15 aldehyde PSB (pKa of
R132K:R111L:L121E ~ 6.7; pKa of R132K:R111L ~ 8.6) (Table 1-11). However, the spectrum
showed the same λmax as R132K:R111L:L121E (400 nm) (Table 1-11). Further study with this
mutant will be discussed in the following sections.

76

1.2.4.8. Probing the Wavelength Regulation in CRABPII Mutant R132K:R111L
Since the mutant R132K:R111L (pKa ~ 8.6) showed a higher pKa value than mutant
R132K:R111L:L121E (pKa ~ 6.7) and its mutants (R132K:R111L:L121E:R59W (pKa ~ 7.9),
R132K:R111L:L121E:R59E (pKa ~ 7.4) and R132K:R111L:L121E:R59D (pKa ~ 7.7)), we
therefore installed glutamate at 59 to see whether a red-shifted spectrum could also be obtained.
Mutant R132K:R111L:R59E, however, does not show a similar spectrum as mutant
R132K:R111L:L121E:R59E (Table 1-12). The mutant R132K:R111L:R59E with C15 aldehyde
absorbed at 397 nm, which is the same absorption for C15 aldehyde bound R132K:R111L (Table
1-12).

To further investigate the effect of R59 mutations, we replaced arginine with a

Table 1-12. λmax and Kd of all-trans-C15 with R132K:R111L mutants.
Mutant

a

λmax
(nm)

Kd (nM) with Reductive
b
all-trans-retinal Amination
(+268 m/z)

pKa

R132K:R111L:L121

398

106 ± 12

Yes

8.6

R132K:R111L:R59E

397

81 ± 5.3

Yes

N.D.

R132K:R111L:R59L

406

203 ± 23

Yes

8.2

R132K:R111L:A32E

407

57.5 ± 8.4

Yes

8.6

R132K:R111L:A36E

397

37 ± 6.8

Yes

8.3

R132K:R111L:F15D

390

1062 ± 179

N.D.

8.3

R132K:R111L:F15Y

397

1.2 ± 5.9

N.D.

8.3

a
b
c

c

λmax of bound C15 aldehyde PSB identified by deconvolution.
An adduct peak observed in mass spectrum after reductive amination.
N.D. stands for not determined.

hydrophobic residues leucine. Mutant R132K:R111L:R59L showed a more red-shifted peak as
77

Figure 1-66. Stereoview of crystal structure of R132K:R111L:A32E with C15 aldehyde
(Green).
compared with R132K:R111L and R132K:R111L:R59E (Table 1-12) with a PSB pKa around
8.2, similar to that of C15 aldehyde PSB in R132K:R111L. In addition, mutations, that showed
effect on R132K:R111L:L121E, did not induce the same protein shift observed with
R132K:R111L mutants (Table 1-12). For example, mutant R132K:R111L:L121E:A32E (412
nm) exhibited a less red shift absorption than mutant R132K:R111L:L121E:R59E (424 nm) upon
binding with C15 aldehyde. However, mutant R132K:R111L:A32E (407 nm) showed a more
red-shifted spectrum as compared with mutant R132K:R111L:R59E (397 nm). These results
indicated that C15 aldehyde binds the R132K:R111L mutants in a different way than the
R132K:R111LE:L121E mutants. More studies on R132K:R111L would be discussed in the
following sections.

1.2.4.8.1. Discovery the Alternate Binding Site for C15 aldehyde in CRABPII

78

C15 in KLE:R59W

C15 in
KL:A32E

Figure 1-67. Stereoview of overlay holo-structure of R132K:R111L:L121E:R59W (Green)
and R132K:R111L:A32E (Cyan) with bound C15 aldehyde.
As mentioned, mutants R132K:R111L showed a 20 fold increase of pKa of bound C15
aldehyde without changing the λmax of the bound C15 aldehyde as compared with mutant
R132K:R111L:L121E (Table 1-12). We
were interested to see whether the
ordered water would stay at the same
position upon removal of polar residue at
121 that interacts with the PSB through a
water mediated interaction (Figure 1-60).
Also, we wanted to probe whether

Figure 1-68.
Overlay of holo-strcuture of
R132K:R111L:A32E (Cyan) and R132K:R111L:
L121E:R59W (Green) with bound C15 aldehyde
(C15) (in R132K:R111L:A32E: Orange; in
R132K:R111L:L121E:R59W: purple) and BTP
(Purple).

79

removal of a polar residue around the
PSB

would

induce

conformational

changes in the protein. Crystallography
of the mutants was pursued and crystal

126˚

structure

3.0 Å

of

mutants

R132K:R111L:A32E with C15 aldehyde
was solved (Figure 1-66). The crystal
structure showed a different binding
Figure 1-69. Bound C15 aldehyde PSB (Orange)
stabilized by the carbonyl group of Ser37.

mode as compared with all-trans-retinal

with R132K:R111L:L121E and C15 aldehyde with R132K:R111L: L121E-R59W (Figure 1-67).
The bound C15 aldehyde PSB in R132K:R111L:A32E is perpendicular to the bound
chromophore in R132K:R111L:L121E:R59W (Figure 1-67). This new conformation displaced
the BTP molecule in the holo structure of R132K:R111L:L121E:R59W (Figure 1-68).

In

addition, this conformation allows C15 aldehyde to be completely embedded within the protein
(Figure 1-66).

Unlike R132K:R111L:L121E: R59W, the PSB is stabilized by the main chain

carbonyl of Ser37 that is 3.0 Å
away from the PSB with CNO
angle of 126˚ (Figure 1-69).

N+
108˚

According to the calculation
conducted

H

-O

O

by Scheiner and

Figure 1-70.
The optimum CNO angle for PSB
Hillenbrand, the best angle for stabilization by counterion is 108˚.

80

stabilizing PSB is 108˚ (Figure 1-70).

200

Although the observed CNO angle is 18˚

F15
3.5 Å

away from the optimum, it can still
provide a good stabilization to the PSB.

7.0 Å

Therefore,

R59

the

pKa

of

bound

C15

aldehyde in R132K:R111L:A32E (pKa~
Figure 1-71. Relative position of Arg59 and
8.6) is above the pKa of the PSB in EtOH
Phe15
with
C15
aldehyde
in
R132K:R111L:L121E:A32E.
(Table 1-12). Based on the crystal structure, Arg59 is 7.0 Å away from the bound C15 aldehyde
(Figure 1-71), therefore, mutation at 59 does not have a significant effect on the bound C15
aldehyde (Table 1-12).

Mutant R132K:R111L: F15D induced a blue shift relative to

R132K:R111L. From the crystal structure (Figure 1-71), replacement of Phe15 with Asp could
increase the electrostatic potential around the PSB, thus, causing the 7 nm blue shift (Table 112). Results obtained from R132K:R111L:L121E mutants indicate that increasing the negative
electrostatic potential around the β-ionone ring causes red shift (Table 1-10). We would like to
apply a similar strategy in mutant R132K:R111L with the new binding docky by introducing
negative charged residues around the β-ionone ring, thus, mutant R132K:R111L:T54E was

81

generated.

However, mutant R132K:

R111L:T54E shows only a 4 nm red shift
as compared with mutant R132K:R111L.
We

were

able

to

crystalize

R132K:R111L:T54E with C15 aldehyde
illustrating that the carboxylate of Glu54

Ser12
Ala37

is around the middle of the chromophore
while in mutant R132K:R111L:L121E:
R59E, the negative charged Glu59 is

Figure
1-72.
PSB
stabilization
in
R132K:R111L:T54E
by
water-mediated
interactions with Ala36 and Ser12.

close to the β-ionone ring. Based on the electrostatic hypothesis we proposed in Section 1.2.2.,
mutant R132:R111L:T54E (+4 nm) does not induce the red shift as much as
R132K:R111L:L121E:R59E (+44 nm).

In addition, the PSB in R132K:R111L:T54E is

stabilized by two water-mediated interactions with Ser12 and Ala36 (Figure 1-72) but not the
main chain carbonyl of Ser37 in R132K:R111L:A36E (Figure 1-69). As previously mentioned,
PSB in R132K:R111L:L121E:R59W is stabilized through the water network that originates from
Glu121. Although both PSB in R132K:R111L:T54E and PSB in R132K:R111L:L121E:R59W
are stabilized in a similar manner, the pKa of PSB in R132K:R111L:T54E (pKa ~ 8.2) is about 5
fold higher than PSB in R132K:R111L:L121E:R59W (pKa ~ 7.7). The polar environment
around the PSB has been suggested to hugely affect the pKa of the PSB. The retinal PSB in
bacteriorhodopsin and rhodopsin are well protected by the protein (Figure 1-15, 18) from the
solvent

around

the

82

protein

PSB Barreled
inside the
protein
without water

PSB exposed to
empty space filled
with water

Figure 1-73. Comparison of PSB environment in R132K:R111L:L121E:R59W and
R132K:R111L:A32E.

and in both cases, the pKa values are high (13.3 and >16 respectively). On the other hand, the
pKa of retinal PSB in water or methanol is about 6.5. This is because the counterion for PSB in
water or methanol is chloride or trifluoroacetate most often, and is greatly solvated by the water,
therefore, the interaction between the PSB and counterion is weaken and pKa of retinal PSB
decreases. Through comparing the C15 aldehyde PSB in
both

R132K:R111L:L121E:R59W

R132K:R111L:T54E

crystal

structures,

and
PSB

in

R132K:R111L:T54E is protected from bulk water by the
protein

more

than

the

PSB

in

R132K:R111L:L121E:R59W (Figure 1-73). This might
Figure 1-74. Overlay of the bound
C15
aldehyde
in be the reason for the relatively high pKa in
R132K:R111L:T54E (Green) with
mutants
as
compared
to
bound
C15
aldehyde
in R132K:R111L
R132K:R111L:A32E (Orange).

83

R132K:R111L:L121E mutants. Bound C15 aldehyde in R132K:R111L:A32E is more red-shifted
(5 nm) as compared with R132K:R111L:T54E (Table 1-12). Glu32 is more than 6 Å away from
the bound chromophore so the electrostatic potential induced by Glu32 on the bound
chromophore should be negligible. According to the crystal structures (R132K:R111L:A32E
and R132K:R111L:T54E), bound C15 aldehyde PSB in mutant R132K:R111L:A32E adopts a 6s-trans conformation while in mutant R132K:R111L:T54E adopts a twist conformation (74˚)
around
C15

C5-C6

bond.

The

twisted

conformation decreases the conjugations that
contribute to blue shift (Figure 1-74). This

Rt

result is consistent with the proposed role of

C15

planarity

in

wavelength

89,90,106,122

regulations.

We have showed that C15 aldehyde
adopts two different conformations upon
binding with CRABPII mutants.

C15

Glu121

aldehyde
Leu121

in

L121E:R59W

mutant
showed

R132K:R111L:
a

similar

Figure 1-75. Overlay of all-trans-retinal (Rt) conformation as all-trans-retinal in mutant
in R132K:R111L:L121E (Blue), C15 aldehyde
R132K:R111L:L121E (Figure 1-56) while
(C15) in R132K:R111L:L121E:R59W (Purple)
and C15 in R132K:R111L:T54E (Green) with C
15 aldehyde in mutants R132K:R111L:
Glu 121 (Purple) in R132K:R111L:L121E and
R132K:R111L:L121E:R59W and Leu121
A32E and R132K:R111L:T54E adopted an
(Green) in R132K:R111L:T54E shown.

84

alternate

conformation

that

is

perpendicular

to

the

all-trans-retinal

in

mutants

R132K:R111L:L121E and C15 aldehyde in R132K:R111L:L121E:R59W (Figure 1-74). When
we overlayed these three structures (mutants R132K:R111L:L121E:R59W, R132K:R111L:A32E
and R132K:R111L: T54E), it is evident that Glu121 in mutant R132K:R111L:L121E:R59W
provides a hydrophilic surface that interacts poorly with the hydrophobic C15 aldehyde (Figure
1-75). That prevents the chromophore in mutant R132K:R111L:L121E:R59W from adopting a
conformation such as that in mutant R132K:R111L:A32E and R132K:R111L:T54E.

We,

therefore, suggested that C15 aldehyde in R132K:R111L: L121E mutants adopts a conformation
like all-trans-retinal in R132K:R111L: L121E and C15 aldehyde in R132K:R111L mutants
should adopt a conformation that is
perpendicular

to

all-trans-retinal

Fluorescence Quenching of KL:T54E and
KLE-R59W with C15 aldehyde
1

in

1.2.4.9.

Fluorescence as a Tool to

Identify the Relative Position of the
Bound C15 aldehyde
To verify the conformation of
bound

C15

aldehyde

R132K:R111L:L121E
R132K:R111L

mutants

mutants,

quenching was employed.

Relative Fluorescence

R132K:R111L:L121E.

0.9
0.8
0.7
0.6

in
and

0.5

0
4.333 10-7 8.667 10-7 1.3 10 -6
C15 aldehyde concentration (M-1)

fluorescent

Figure 1-76. Fluorescence quenching of C15
Tryptophans, aldehyde with R132K:R111L:L121E:R59W (top
curve) and R132K:R111L:T54E (bottom curve).

that fluoresce at 360 nm, are important

85

residues for binding constant measurement. Since the absorption of bound C15 aldehyde PSB
overlaps with the fluorescence of the tryptophan. Therefore, monitoring the relative amount of
fluorescence quenching can be used for binding affinity measurements. By plotting the amount
of fluorescence, from tryptophans, quenched by bound chromophore at different concentration of
chromophore, a binding constant can be calculated from the plot using non-linear curve fitting
(Figure 1-76). Based on the crystal structures, C15 aldehyde in R132K:R111L:L121E:R59W is
further away from tryptophans in CRABPII, particularly Trp109 (Figure 1-77) as compared with
mutants R132K:R111L:A32E and R132K:R111L: T54E (Figure 1-77). Therefore, the amount of
fluorescence quenched by the chromophore in R132K:R111L mutants (alternate binding site)
should be more than that in R132K:R111L:L121E (original trajectory) mutants. Thus, the
fluorescence at the saturation point (protein is fully bound with ligand) in R132K:R111L mutants
should be lower than that in R132K:R111L:L121E mutants. Data in Table 1-13 indicates that

Figure 1-77. Position of Trp (Cyan) in CRABPII relative to bound C15 aldehyde in
R132K:R111L:A32E (Green) and in R132K:R111L:L121E:R59W (Purple).
the quenched fluorescence ratio at saturation point for R132K:R111L mutants in general is about
20%

more

than

the

R132K:R111L:L121E
86

mutants

(Figure

1-76),

except

for

R132K:R111L:L121E:V76 mutants. Based on the fluorescence quenching, we find that the
overall amount of fluorescence quenched in all R132K:R111L mutants is more than
R132K:R111L:L121E mutants. This suggested that the bound C15 aldehyde in R132K:R111L
mutants should adopt the conformation much like the holo-structure of R132K:R111L:A32E and
R132K:R111L:T54E (Figure 1-77), which leads to the bound C15 aldehyde PSB more efficiently
quenching the tryptophan fluorescence as compared to R132K:R111L:L121E which binds C15
aldehyde PSB in a manner that places the chromophore is further away from the tryptophan
(Figure 1-77).

87

Table 1-13. Relative fluorescence at saturation for different CRABPII mutants.
c

Mutant

Quenced
a,b
Ratio
%

Mutant

Quenced
a,b
Ratio
%

Mutant

Quenced
a,b
Ratio
%

KLE

80

KLE:R59L

N.D.

KLE:V76Y

85

KL:L121D

94

KLE:T56E

N.D.

KLE:V76W

55

KL:L121Q

80

KLE:T56I

N.D.

KLE:V76R

50

KLE:R59E

81

KLE:T56R

N.D.

KL

70

KLE:R59D

87

KLE:T56W

91

KL:R59E

48

KLE:R59Q

79

KLE:F15D

79

KL:R59L

57

KLE:R59Y

83

KLE:A32E

85

KL:A32E

61

KLE:R59W

81

KLE:A36E

89

KL:F15Y

56

KLE:R59T

82

KLE:V76E

65

KL:F15D

69

KLE:R59A

82

KLE:V76F

76

KL:T54E

67

a

Quenched Ratio = Fluorescence at the beginning / Fluorescence at saturation

b

N.D. stands for not determined.

c

KLE stands for R132K:R111L:R121E.

1.2.4.10. Study the effect of BTP on C15 aldehyde binding
The crystal structure of mutant R132K:R111L:L121E:R59W with C15 aldehyde contains a
BTP molecule that is present in the buffer the crystals were grown (Figure 1-58). This BTP
molecule in the mutant R132K:R111L:L121E:R59W structure is located at the same position as
the C15 aldehyde in mutants R132K:R111L:T54E and R132K:R111L:A32E structure (Figure 172). Although fluorescence quenching could identify the relative position of the bound C15

88

aldehyde in CRABPII mutants (Section 1.2.4.9.), the actual position for the bound C15 aldehyde
in mutant R132K:R111L:L121E:R59W crystal structure could be modulated by the presence of
the BTP. For this reason, we investigated the effect of BTP on the conformation of the bound
C15 aldehyde.

Fluorescence quenching experiments using the same buffer used for

crystallography (100 mM BTP, pH 7.3) showed that the fluorescence level at the saturation point
does not have a significant change on both mutants R132K:R111L:L121E:R59W and
R132K:R111L:L121E:R59E (most red-shifted mutant with C15 aldehyde ~ 424 nm) as compared
with the value from the experiment using PBS buffer (buffer system used in the UV-vis
spectroscopic study) (Table 1-14). These results indicated that the BTP molecule does not affect
Table 1-14. Study on the Effect of BTP buffer.
Mutant

λmax
(nm)

Quenced
a
Ratio
%

pKa

R132K:R111L:L121E:R59E (in PBS Buffer)

1 ± 4.3

424

81

7.4

R132K:R111L:L121E:R59E (in BTP Buffer)

72 ± 30

429

86

6.9

R132K:R111L:L121E:R59W (in PBS Buffer)

139 ± 25 404

81

7.7

R132K:R111L:L121E:R59W (in BTP Buffer)
a

Kd
(nM)

252 ± 17 414

91

6.6

Quenched Ratio = Fluorescence at the beginning / Fluorescence at saturation

89

a)

b)

C15 aldehyde with R132K:R111L:L121E: C15 aldehyde with R132K:R111L:L121E:
R59E in PBS Buffer
R59E in BTP Buffer
0.1
0.05
0.08
Absorbance

Absorbance

0.04
0.03
0.02

0.06
0.04
0.02

0.01
0

350
400
450
Wavelength (nm)

0
300

500

350 400 450 500
Wavelength (nm)

550

d)

c)

C15 aldehyde with R132K:R111L:L121E: C15 aldehyde with R132K:R111L:L121E:
R59W in PBS Buffer
R59W in BTP Buffer
0.15
0.04

0.03
Absorbance

Absorbance

0.035

0.025
0.02
0.015

0.1

0.05

0.01
0.005
0

0
350
400
450
500
Wavelength (nm)

350
400
450
500
Wavelength (nm)

Figure 1-78. a) C15 aldehyde with R132K:R111L:L121E:R59E in BTP buffer at pH 7; b)
C15 aldehyde with R132K:R111L:L121E:R59E in PBS buffer at pH 7; c) C15 aldehyde
with R132K:R111L:L121E:R59W in BTP buffer at pH 7; d)
R132K:R111L:L121E:R59W in PBS buffer at pH 7.

C15 aldehyde with

the relative position of the bound C15 aldehyde. In addition, the binding towards the C15
90

aldehyde is weakened with BTP in the buffer. This might be due to the fact that the the binding
pocket is saturated by the BTP molecule at 100 mM concentration, therefore, hindering the
binding of C15 aldehyde that is present at nanomolar concentration.

In order to further

investigate whether the presence of BTP molecule in the binding pocket could affect the effect of
mutations on wavelength regulation in CRABPII, a UV-vis spectroscopic study was conducted.
Both mutants R132K:R111L:L121E:R59E and R132K:R111L:L121E:R59W do not show a
completely shifted peak in BTP buffer. Since both mutants show good binding (nM) with the
C15 aldehyde in BTP buffer, the incompletely shifted spectrum might be result of the lower pKa
of the bound C15 aldehyde PSB in the presence of the BTP molecule. Acid-base titration showed
that the pKa of the bound C15 aldehyde PSB in both mutants drops at least 0.5 unit in the
presence of BTP buffer as compared with the pKa of the bound C15 aldehyde in PBS buffer.
Therefore, in order to find the correct λmax for the bound PSB in both mutants, a BTP buffer at
pH 5 was used to get the full shifted UV-vis spectrum. The λmax for both mutants bound with
C15 aldehyde in BTP buffer showed the same trend as compared with the result in PBS buffer
(Table 1-14). That is, negatively charged residues close to the β-ionone ring induce red-shift
with the bound C15 aldehyde PSB (Figure 1-61). This result indicated that the bound C15
aldehyde PSB during the UV-vis study should adopt a conformation close to the holo-crystal
structure of mutant R132K: R111L:L121E:R59W (Figure 1-57). As we compared the UV-vis
spectra of C15 aldehyde bound to mutant R132K:R111L:L121E:R59E in PBS buffer and that in

91

UV-vis Spectrum of C15 aldehyde with UV-vis Spectrum of C15 aldehyde with
KLE:R59W Titrated with BTP
KLE:R59E titrated with BTP
0.2

0.1

0.18

Absorbance

0.09
0.08
Kd = 35 ± 2.7 mM

0.07
0.06

Absorbance

0.11

0.16
0.14

Kd = 202 ± 43 uM

0.12

0.05
0.1

0.04
0.03

0

0.05 0.1 0.15 0.2
Concentration (M)

0.25

0.08
0

0.005 0.01 0.015 0.02 0.025
Concentration (M)

Figure 1-79. Dissociation constant of BTP with R132K:R111L:L121E:R59W (Right)
and R132K:R111L:L121E:R59E (Left) calculated based on the absorbance change of
UV-vis spectrum of C15 aldehyde with CRABPII mutant titrated by BTP.
100 mM BTP buffer at pH 7, the ratio between the absorbance of 320 nm (bound C15 aldehyde
SB) and the absorbance of 425 nm (bound C15 aldehyde PSB) is different (Figure 1-78). It is
because the pKa of the bound C15 aldehyde PSB in R132K:R111L:L121E:R59E in PBS buffer
(pH = 7.4) is higher than that in BTP buffer. Since the addition of BTP molecule leads to
decrease of intensity at 425 nm, we can measure the binding constant of BTP with the C15
aldehyde-CRABPII mutant complex. BTP titration and Kd calculation show it binds poorly with
the complex (Kd ~ 202 µM) as compared to C15 aldehyde (from 1 nM to 9 µM) (Table 1-14)
(Figure 1-79). Thus, if the C15 aldehyde prefers to occupy the alternate binding pocket like BTP
in the R132K:R111L:L121E:R59E apo-structure (Figure 1-68), it seems too weak to compete

92

with C15 aldehyde in order to displace the bound C15 aldehyde such that it would adopt the
original trajectory like C15 aldehyde in R132K:R111L:L121E:R59W crystal structure (Figure 161).

1.2.4.11. Study the effect of Tyr134 on bound C15 aldehyde absorption and binding
The role of Tyr134 in CRABPII for all-trans-retinal binding has been fully
176,178,179

investigated.

In summary, Tyr134 has been suggested to activate the carbonyl of allTyr134

O

O
Lys132
H
HN H

Lys132

O

N

O

Glu121

H
H
H
O

H O

Tyr134

O

Glu121

O
H

O
H

Lys132
N

Tyr134

H

O

Glu121

O
H
H

O

Scheme 1-2. Proposed mechanism of Schiff base formation in R132K:R111L:L121E with
all-trans-retinal.
trans-retinal for nucleophilic attacked by Lys132 (Scheme 1-2). In addition, upon formation of
the PSB, Tyr134 forms hydrogen bonding with Glu121 (counterion for PSB). Introduction of
Y134F mutation hampers the ability PSB formation as we compared the result of mutant

93

Table 1-15. λmax and Kd of all-trans-C15 with Tyr134 mutants.
Mutant

a

λmax
(nm)

Kd (nM) with Reductive
b
all-trans-retinal Amination
(+268 m/z)

pKa

Quenched
c
Ratio
(%)

R132K:R111L:L121E

400

9030 ± 924

Yes

6.7

80

R132K:R111L:L121E:Y134F

402

844 ± 63

Yes

6.8

75

R132K:R111L:L121D (KLD)

418

147 ± 20

Yes

7.1

94

R132K:R111LD:Y134F

417

89 ± 13

Yes

7.1

83

a
b
c

λmax of bound C15 aldehyde PSB identified by deconvolution.
An adduct peak observed in mass spectrum after reductive amination.
Quenched Ratio = Fluorescence at the beginning / Fluorescence at saturation

R132K:Y134F:R111L:L121E (pKa ~ 6.5) and R132K:R111L:L121E (pKa ~ 8.7).

178,179

The

decrease of bound all-trans-retinal PSB pKa in mutant R132K:Y134F:R111L:L121E relative to
mutant R132K:R111L:L121E is because the loss of hydrogen bonding of Tyr134 with Glu121
176,178,179

hampers the ability of Glu121 as a counterion.

As C15 aldehyde was tested with

mutant R132K:Y134F:R111L:L121E, there is no effect on λmax and pKa of the bound C15
aldehyde PSB as compared to R132K:R111L:L121E (Table 1-15). Based on the crystal structure
of R132K:R111L:L121E:R59W with C15 aldehyde, unlike all-trans-retinal (Figure 1-30),
Glu121 interacts indirectly with the bound C15 aldehyde PSB through a water-mediated
interaction. Therefore, mutation of Tyr134 does not induce a significant effect on the bound C15
aldehyde PSB.
94

1.3. Conclusion
In short, we have demonstrated the power of protein engineering to study the fundamentals
theories of biological processes. We have successfully applied the de novo design principles to
reengineering of CRABPII into a rhodopsin mimic and we have accomplished the first step
towards studying wavelength regulation using our protein surrogate. Throughout the study of
all-trans-retinal with our first generation of rhodopsin mimics, we have pinpointed the
importance of having a fully embedded chromophore in order to study the actual effect from the
mutations. The ease of crystalizing the CRABPII and its mutants help us to tackle down the
problem arisen during the study. With the crystal structures, we provide detailed molecular
analysis on how electrostatic potential induced by residues could modulate the absorption of the
bound chromophore (Figure 1-32 and 64). Although the point-charge theory has been under
scrutiny since crystal structures of rhodopsin and bacteriorhodopsin has been published, recent
computational studies suggest that negative dipole along the chromophore could induce
wavelength modulation of the bound PSB. Using our simplistic system, we have demonstrated
that changing the strength of electrostatic potential and its relative position along the polyene
could modulate the wavelength of the bound chromophore at different degrees. This system has
been further applied to another protein to study the wavelength regulation in opsin proteins.
Human cellular retinol binding protein II (hCRBPII) is our second generation protein surrogate.
hCRBPII binds to both retinol and retinal. In addition, the bound retinal is fully embedded and
protected by the protein unlike with CRABPII. We have applied the design principle we learned
so far from CRABPII to reengineer hCRBPII to bind all-trans-retinal as a PSB.

95

1.4. Materials and Methods
i.

Protein Mutagenesis, Bacterial Expression and Purification

i.

Protein Mutation
Mutagenesis of all CRABPII proteins was performed on pET17b-CRABPII plasmid
®

according to Stratagene’s QuikChange
®

performed using Bio-rad

protocol. The polymerase chain reaction (PCR) was

iCycler Thermal Cycler with 96-well reaction module with the

specified PCR conditiions.
Table 1-16. PCR recipe for CRABPII
Reactant
DNA Template (10 ng/uL)
Forward Primer (11.2 pmol)
Backward Primer (11.2 pmol)
pfu ultra reaction buffer (10x)
dNTP (10mM)
d.d. H2O
pfu turbo polymerase

Amount (uL)
5
a
b
5
1
50-(12+a+b)
1

Table 1-17. PCR cycle for CRABPII
PCR Cycles
no. of cycles

Temperature (˚C)

Time

94 °C

4min

94 °C

1 min

55 °C

1 min

72 °C

6 min

1x

72 °C

15 min

1x

25 °C

10 min

1x

20x

96

Table 1-18. PCR Primers
Mutant

Prime
rs
name

Primers Sequence

Template

R132K:R11 bb179 5’1L:L121E:R
CCTCCACCACCGTGGAGACCACA
59E
G AG-3’

Seq.
No.

R132K:R11
1L:L121E

BB335

R132K:R11
1L:L121E

BB334

R132K:R11
1L:L121E

BB343

R132K:R11
1L:L121E

BB340

bb180 5’CTCTGTGGTCTCCACGGTGGTGGA
GG-3’
R132K:R11 bb181 5’1L:L121E:R
CCTCCACCACCGTGGCCACCACA
59A
G AG-3’
bb182 5’CTCTGTGGTGGCCACGGTGGTGG
A GG-3’
R132K:R11 bb183 5’1L:L121E:R
CCTCCACCACCGTGATGACCACA
59M
G AG-3’
bb184 5’CTCTGTGGTCATCACGGTGGTGG
A GG-3’
R132K:R11 bb185 5’1L:L121E:R
CCTCCACCACCGTGTGGACCACA
59W
GA G-3’
bb186 5’CTCTGTGGTCCACACGGTGGTGG
AG G-3’
R132K:R11 bb187 5’1L:L121E:R
CCTCCACCACCGTGTTGACCACA
59L
GA G-3’
bb188 5’CTCTGTGGTCAACACGGTGGTGG
AG G-3’

97

R132K:R11
1L:L121E

Table 1-17 (cont’d)

R132K:R11 bb189 5’1L:L121E:R
CCTCCACCACCGTGTACACCACA
59Y
GA G-3’

R132K:R11
1L:L121E

bb190 5’CTCTGTGGTGTACACGGTGGTGG
AG G-3’
R132K:R11 bb229 5’1L:L121E:R
CCTCCACCACCGTGCAGACCACA
59Q
GA G-3’

R132K:R11
1L:L121E

bb230 5’CTCTGTGGTCTGCACGGTGGTGG
AG G-3’
R132K:R11 bb253 5’1L:L121E:R
CCTCCACCACCGTGGACACCACA
59D
GA G-3’

R132K:R11
1L:L121E

BB438

R132K:R11
1L:L121E

BB367

R132K:R11
1L:L121E

BB370

R132K:R11
1L:L121E

BB371

bb254 5’CTCTGTGGTGTCCACGGTGGTGG
AG G-3’
R132K:R11 bb212 5’1L:L121E:V
GGAGCAGACTGAGGATGGGAGG76E
3’
bb213 5’-CCTCCCATCCTCAGTCTGCTCC3’
R132K:R11 bb214 5’1L:L121E:V
GGAGCAGACTCGGGATGGGAGG76R
3’
bb215 5’-CCTCCCATCCCGAGTCTGCTCC3’
R132K:R11 bb216 5’1L:L121E:V
GGAGCAGACTTGGGATGGGAGG76W
3’

98

Table 1-18 (cont’d)
bb224 5’-CCTCCCATCCCAAGTCTGCTCC3’
R132K:R11 bb222 5’1L:L121E:V
GGAGCAGACTCACGATGGGAGG76Y
3’

R132K:R11
1L:L121E

BB383

bb223 5’-CCTCCCATCGTAAGTCTGCTCC3’
R132K:R11 bb233 5’1L:L121E:V
GGAGCAGACTTTCGATGGGAGG76F
3’

R132K:R11
1L:L121E

bb234 5’CCTCCCATCGAAAGTCTGCTCC-3’
R132K:R11 bb374 5’-GGGCCTCCCATCCGCAGT
1L:L121E:V
CTGCTCCTCAAACTC-3’
76A
bb375 5’GAGTTTGAGGAGCAGACTGCGGA
TGGGAGGCCC-3’

R132K:R11
1L:L121E

BB721

R132K:R11 bb372 5’-GGGCCTCCCATCCGAAGT
1L:L121E:V
CTGCTCCTCAAACTC-3’
76S
bb373 5’GAGTTTGAGGAGCAGACTTCGGA
TGGGAGGCCC-3’

R132K:R11
1L:L121E

BB725

R132K:R11 bb376 5’R132K:R11
1L:L121E:V
GGGCCTCCCATCCAGAGTCTGCTC 1L:L121E
76L
CTCAAACTC-3’

BB719

bb377 5’GAGTTTGAGGAGCAGACTCTGGA
TGGGAGGCCC-3’
R132K:R11 bb206 5’-CCTCCGAGACCGTGCGCACC-3’ R132K:R11
1L:L121E:T
1L:L121E
bb207 5’-GGTGCGCACGGTCTCGGAGG-3’
56E

99

BB368

Table 1-18 (cont’d)
R132K:R11 bb208 5’-CCTCCCGCACCGTGCGCACC-3’
1L:L121E:T
bb209 5’-GGTGCGCACGGTGCGGGAGG56R
3’

R132K:R11
1L:L121E

BB385

R132K:R11 bb210 5’-CCTCCTGGACCGTGCGCACC-3’
1L:L121E:T
bb211 5’-GGTGCGCACGGTCCAGGAGG56W
3’

R132K:R11
1L:L121E

R132K:R11
1L:L121E:T
56D

R132K:R11
1L:L121E

BB709

R132K:R11
1L:L121E:F
15D

bb171 5’-CGGAAAACGACGAGGAATTGC- R132K:R11
3’
1L:L121E

BB329

bb172 5’-GCAATTCCTCGTCGTTTTCCG-3’

R132K:R11 bb151 5’1L:L121E:A
GCTGAGGAAGATTGAAGTGGCTG
32E
C-3’

R132K:R11
1L:L121E

BB276

R132K:R11 bb153 5’-GGCTGCAGAGTCCAAGCCAGC1L:L121E:A
3’
36E
bb154 5’-GCTGGCTTGGACTCTGCAGCC3’

R132K:R11
1L:L121E

BB278

R132K:R11
1L:R59E

R132K:R11
1L

BB434

bb152 5’GCAGCCACTTCAATCTTCCTCAGC
-3’

bb179 5’CCTCCACCACCGTGGAGACCACA
GA G-3’
bb180 5’CTCTGTGGTCTCCACGGTGGTGGA
G G-3’

100

Table 1-18(cont’d)
R132K:R11
1L:R59L

bb187 5’CCTCCACCACCGTGTTGACCACA
GA G-3’

R132K:R11
1L

BB427

R132K:R11
1L

BB275

R132K:R11
1L

BB277

bb149 5’-CGGAAAACTACGAGGAATTGC- R132K:R11
3’
1L

BB280

bb188 5’CTCTGTGGTCAACACGGTGGTGG
AG G-3’
R132K:R11
1L:A32E

bb151 5’GCTGAGGAAGATTGAAGTGGCTG
C-3’
bb152 5’GCAGCCACTTCAATCTTCCTCAGC
-3’

R132K:R11
1L:A36E

bb153 5’-GGCTGCAGAGTCCAAGCCAGC3’
bb154 5’-GCTGGCTTGGACTCTGCAGCC3’

R132K:R11
1L:F15Y

bb150 5’-GCAATTCCTCGTAGTTTTCCG3’
R132K:R11
1L:F15D

bb171 5’-CGGAAAACGACGAGGAATTGC- R132K:R11
3’
1L

BB328

bb172 5’-GCAATTCCTCGTCGTTTTCCG-3’
R132K:R11
1L:T54E

bb65

5’CTACATCAAAGAGTCCACCACCG
TGC G-3’

bb66

5’CGCACGGTGGTGGACTCTTTGAT
GTA G-3’

101

R132K:R11
1L

BB440

Melting Temperature Calculation for Primers:
Primers used for mutagenesis should have around 25 to 45 bases with a melting
temperature (Tm) ≥ 78 ˚C. The melting temperature is calculated according to the following
201

equation:

Tm (˚C) = 81.5 + 0.41 (%GC) – 675 / N – %mismatch
where:
GC stand for guanine and cytosine bases
N is total number of bases in primer
and value for %GC and %mismatch are whole number
iv. Dpn I Digestion of the Amplification Products
The PCR products were subjected to Dpn I digestion to destroy the template DNA. Dpn I
enzyme is specific for digesting methylated DNA. DNA purified from the E. coli is methylated
while DNA after PCR is not, therefore, Dpn I can specifically digest the template DNA that is
purified from E. coli but not the PCR product. Dpn I restriction enzyme (1 µL, 10 U/µl) is added
to the PCR products and the reaction mixture was incubated at 37 ˚C for 1 hour. The resulting
mixture is transfected to competent cell for storage and DNA amplification and purification.
v.

DNA Transformation of PCR product
®

The XL-1 Blue super competent cell from Novagen was de-thawed in ice and 10 µL of
PCR product was added after the cells were de-thawed. The cells were incubated in ice for 30
minutes. The cells were heat-shocked at 42 ˚C for 35 seconds and incubated on ice for another 2

102

minutes. Luria-Bertani (500 µL, LB) broth was added and incubated at 37 ˚C for 30 minutes.
The cells were centrifuged at 3000 rpm for 2 minutes and 500 µL of supernatant was discarded.
The cells were re-suspended gently and plated on LB agar plate with the corresponding antibiotic
(ampicilin and chloramphenicol) and incubated at 37 ˚C overnight.

vi.

DNA Purification
®

The DNA purfication followed the protocol of Qiagen Plasmid Maxi Kit. Briefly, a cell
colony with correct plasmid was innoculated in LB broth (5 mL) with specific antibiotic. The
cell culture was grown at 37 ˚C for 12 to 16 hours and the cell culture (1 mL) was inoculated
into LB broth (500 mL) with the corresponding antibiotic. The cell culture was then growed at
37 ˚C for another 12 to 16 hours. The cell culture was centrifuged at 4000 rpm for 15 min at 4
˚C. The supernatant was discarded and the cell pellet was resuspended in suspension buffered
followed by lysate buffer. The cells were then incubated at room temperature for no longer than
5 minutes. Cell lysate was neutralized by neutralizing buffer and was incubated on ice for
another 30 minutes. The lysate with precipitates was centrifuged at 7000 rpm at 4 ˚C for 30
®

minutes. The supernatant was then passed through the pre-equilibrated Qiagen -tip 500. It is
washed by washing buffer twice. The DNA was eluted with the elution buffer. The obtained
DNA was precipitated by isopropanol and further cleaned by 70% EtOH. The purified DNA was
dried and resuspended in deionized distilled water (d.d. water).
calculated by the following equation using UV-vis spectroscopy:

103

DNA concentration was

DNA concentration (ng/µL) = OD260 x 50 ng/µL x dilution facter (volume of DNA added /
total volume in UV curvette)
Purity of DNA = OD260/OD280 = 1.8 (pure)
>1.8 (RNA contamination)
< 1.8 (Protein contamination)
where OD260/280 stands for absorbance at 260 nm or 280 nm, respectively.

vii. Sample preparation for DNA sequencing
To a 500 µL sterilized eppendorf tube, 2000 ug of plasmid DNA and 30 pmol of
sequencing primer DNA were added and d.d. H2O was added to a total volume of 12 µl. The
primer DNA was synthesized by Research Technology Support Facility in Michigan State
University (primer design was based on the protocol in http://rtsf.msu.edu/custom-primers)
Briefly, the priming site for sequencing primer is 30 to 50 bases upstream from the target
sequence. The melting temperature should be around 55 ˚C to 65 ˚C based on the calculation of
201

the following formula:

Melting temperature (Tm) = 81.5 ˚C + 0.41 x (% GC) – 675 / primer length -% mismatch
(1)
Primer for Sequencing:
5’-GCTTCCTTTCGGGCTTTGTTAGCAGCC-3’

104

Table 1-19. Recipe for sequencing CRABPII:
Reactant
DNA Template (2000 ng)
Sequencing Primer (30 pmol)
d.d. H2O
Total volume

Amount (uL)
x
a
12-(x+a)
12

viii. Protein Expression of CRABPII Mutants
The plasmid was transformed into BL21(DE3) pLacI E. coli competent cell for protein
expression according to the procedure described in Materials and Methods Section v. and was
plated on LB/Amp/Clm plate according to the standard protocol. One colony was inoculated
with 5 mL of LB/Amp/Clm. The cell culture was incubated at 37 ˚C for 12 h at 220 rpm. Cell
culture (5 mL) was inoculated in 1.5 L of LB/ Amp (100 ug/mL)/ Clm (170 ug/mL). The 1.5 L
cell culture was incubated in a shaker at 37 ˚C and 220 rpm until OD600 was equal to 0.4 to 0.6.
IPTG (1 mM) was added into the cell culture to induce the expression and was incubated further
at 30 ˚C for another 6 to 12 hours.

ix. Protein Purification of CRABPII Mutants
Cells were harvested at 4 ˚C by centrifugation for 15 min at 7000 rpm. The supernatant
was discarded. The cells were resuspended in buffer A (10 mM Tris-HCl, pH 8). The cells were
lysed by sonication (Power 60%, 1 min x 3). The cell lysate was spun down at 4 ˚C for 30
minutes at 7000 rpm.

The supernatant was collected and purified by ion exchange
TM

chromatography using Q Sepharose

, Fast Flow resin (diameter: ~4 cm; height: ~ 10 cm;

binding capacity: 20 mg). The column was pre-washed with NaCl solution (2 M). The column

105

was pre-equilibrated with 5 column volumes of buffer A. Flow through was collected for gel
electrophoresis. The column was washed with 3 column volumes of buffer A. The protein was
eluted by 3 column volumes of buffer B (10 mM Tris-HCl, 100 mM NaCl, pH 8). All collected
fractions were checked by SDS-PAGE gel electrophoresis. The combined eluent was desalted
by ultrafiltration (250 mL ultrafiltration cell, Millipore, at 20 psi) through an YM-10
(NMWL=10000) membrane (Ultrafiltration Membranes, Material Regeneration Cellulose) with
two times with buffer A (200 mL each). The retentant was concentrated to 50 mL and further
purified by FPLC using anion exchange column.
x. FPLC Protocol:
Column: Source 15Q (quaternary ammonium strong anion exchanger)
Binding capacity: ~20 mg
Buffer A1: 50 mM Tris HCl
Buffer A2: 50 mM Tris
Buffer B1: H2O
Buffer B2: 2 M NaCl

106

Table 1-20. FPLC Method for CRABPII Purification:
Step Volume (mL) Description
1

0

Collect 3.0 mL fractions

2

0

Isocratic Flow pH = 8.1, 0% B, 20 mL, 5 mL/min

3

20

Quad tec auto zero

4

20

Load/Inject 50 mL, 4 mL/min

5

70

Isocratic Flow pH = 8.1, 0% B, 20 mL, 4 mL/min

6

90

Linear Gradient, pH = 8.1, 0-4% B, 30 mL, 4 mL/min

7

120

Isocratic Flow, pH = 8.1, 4% B, 20 mL, 4 mL/min

8

140

Linear Gradient, pH = 8.1, 4-8% B, 20 mL, 4 mL/min

9

160

10

180

Linear Gradient, pH = 8.1, 8-20% B, 20 mL, 4 mL/min
Linear Gradient, pH = 8.1, 20-75% B, 20 mL, 4 mL/min

11

200

Isocratic Flow, pH = 8.1, 100% B, 20 mL, 4 mL/min

12

220

Isocratic Flow, pH = 8.1, 0% B, 20 mL, 4 mL/min

13

230

End of protocol

The purified proteins were analyzed by SDS-PAGE electrophoresis.
B.

Reductive Amination Protocol
The CRABPII mutants (10 µM, 500µL) were incubated in dark with 2 to 5 equivalent of

all-trans-retinal (1 mM) or C15 aldehyde (2.4 mM) relative to the mutants for 2 hours to provide
sufficient time for Schiff base formations. About 10 equivalents of 1M NaCNBH3 were added
to the mixtures and were incubated in dark for 8 to 12 hours. The resulting mixtures were passed
TM

through ion exchange chromatography using Q Sepharose

, Fast Flow resin (diameter: ~1.5

cm; height: ~ 3 cm; binding capacity: 20 mg) and the proteins were purified using standard
procedure mentioned in Materials and Methods Section x. The fast Q column can remove excess
amount of unbound all-trans-retinal. The fractions were analyzed by UV-vis spectroscopy. The

107

fractions that contain protein absorption (OD280) were collected and concentrated.

The

concentrated sample was desalted by buffer exchange with purified H2O for at least 8 times
®

using Millipore Ultrafree

micro-centrifuge filter (pore size 10000 KDa). The samples were

concentrated to 30 µL final volume, ready for MALDI-TOF analysis. Protein sample (0.5 µL)
was mixed with matrix solution (0.5 µL) with the addition of lysozyme (0.1 µL, 1mM) as an
internal standard.
MALDI-TOF Experimental Parameters:
System: Shimadzu Biotech Axima CFR
Mode of operation: Linear
Acquisition mass range: 5000 – 20000 Da
Number of laser shots: 250 / spectrum
Nitrogen Laser: 337 nm
Laser intensity: 9 Hz
Calibration type: Internal standard (Lysozyme, MW = 14307 Da)
Matrix: Sinapinic acid (Fluka), Saturated solution (10 mg sinapinic acid, 670 µL purified
H2O, 0.3% trifluoroacetic acid (30 µL, 10% solution), 30% MeCN (300 µL)
Electrospray High-resolution Mass spectrometer parameter:
System: Water Q-Tof Ultima

TM

API

Cone Voltage: 35 V
Capillary Voltage: 3 kV
MCP: 1850
Source Temp: 90 ˚C
108

Desol. Temp: 200 ˚C
Mode: ES (+)
Scan time: 10 minutes
Scan range: 300 to 2500 m/z
Calibration type: Phosphoric acid
Table 1-21. Online Desalting Protocol using column (Beta-basic CN, 10 x 1 mm, 5 µm)
Step Time (min)
1
0-5

Description

2

5-8

20:80 (0.15% formic acid in H2O:acetonitrile)

3

8-10

Gradient: 20:80 to 95:5 (0.15% formic acid in H2O:acetonitrile)

Gradient: 95:5 to 20:80 (0.15% formic acid in H2O:acetonitrile)

C. Extinction Coefficient Determination
The extinction coefficients (ε) at OD280 of the mutants were measured according to the
38,39

method described by Gill and von Hippel.

The extinction coefficient (ε) of denatured protein

(the measurement of the UV-vis spectrum in 6M Guanidine HCl) is calculated based on the
following equation:
ε(den) at OD280 = a * ε(Trp) +x * ε(Tyr) + y * ε(Cys)
where a, x and y is the number of Trp, Tyr and Cys residues in the CRABPII mutants and
-1

-1

-1

-1

their ε has been determined previously (ε(Trp) = 5690 M cm , ε(Tyr) = 1280 M cm , ε(Cys) =
-1

120 M

-1

cm ) and WT-CRABPII contains 3 Trp, 2 Tyr and 1 Cys residues.

Therefore,

concentration of the protein can be determined in 6M Guanidine HCl solution based on the

109

Table 1-22. Extinction Coefficient of CRABPII mutants
-1

Mutants

-1

ε (M cm )
20215
18473
20408
19540
18653
26275
19655
21951
22161
19951
21807
20899
27194
37479
21014
21986
21807
24069
22163
25081
17784
20590
21954
20000
15726
20831
21350
24862
20572
21954

WT-CRABP
R132K:R111L:L121E:R59T
R132K:R111L:L121E:R59E
R132K:R111L:L121E:R59A
R132K:R111L:L121E:R59M
R132K:R111L:L121E:R59W
R132K:R111L:L121E:R59L
R132K:R111L:L121E:R59Y
R132K:R111L:L121E:R59Q
R132K:R111L:L121E:R59D
R132K:R111L:L121E:V76E
R132K:R111L:L121E:V76R
R132K:R111L:L121E:V76W
R132K:R111L:L121E:V76Y
R132K:R111L:L121E:V76F
R132K:R111L:L121E:T56E
R132K:R111L:L121E:T56R
R132K:R111L:L121E:T56W
R132K:R111L:L121E:F15Y
R132K:R111L:L121E:F15D
R132K:R111L:L121E:A32E
R132K:R111L:L121E:A36E
R132K:R111L:T54E
R132K:R111L:R59E
R132K:R111L:R59L
R132K:R111L:A32E
R132K:R111L:A36E
R132K:R111L:F15Y
R132K:R111L:F15D
R132K:R111L:T54E

corresponding ε(den). OD280 of the identical concentration of protein was measured in native
condition (PBS buffer, pH = 7.3). Based on Beer’s Law:
Abs(280) = ε(nat) * b * c;
ε = b*c / Abs (280)
where b is the cuvette path length, c is the concentration of the mutants
the ε of the native protein is calculated.
110

D. General Procedure for UV Binding Study

The Fitted

0.5

Spectrum
R2 > 0.99

344nm
The Experimental
Spectrum

Absorbance

0.4

0.3

414nm

0.2

0.1

0
300

350

400
450
Wavelength (nm)

500

Figure 1-80. Deconvoluted spectrum of C15 aldehyde with R132K:R111L:L121E:R59D.
-5

-6

The experiment was conducted in a 1 mL cuvette. CRABPII mutants (5x10 to 5x10 M)
-1

-1

in phosphate buffer were titrated with a solution of all-trans-C15 aldehyde (ε = 16200 M cm ,
λmax = 326 nm, 1 mM) or all-trans-retinal (ε = 42650 M

-1

-1

cm , λmax = 380 nm, 0.5 mM in

EtOH). UV-vis spectra from 200 nm to 600 nm were recorded upon every 0.1 or 0.2 equivalent
of C15 aldehyde or all-trans-retinal was added to the protein solution until 1 eq. The spectra
were measured at room temperature and the ethanol concentration was kept below 3%.

111

For most cases, the absorption spectrum is a combination of C15 Schiff base, C15 aldehyde
and C15 PSB absorption. In order to pick the correct λmax, deconvolution of the UV-vis spectra
®

were performed to separate each component. The deconvolution was performed with Peakfit
2

(Systat Software, Inc.) and for all cases an R of >0.995 was achieved (Figure 1-80).

E.

Fluorescence Titration
All samples were stored in glass containers as many plastic containers will leach

fluorescent impurities into the sample.

The buffer used was purified by passing through

sterilized 0.4 µm filter. Silanized glassware was used to avoid loss of protein and a change in
protein concentration. In addition, gelatin was used to avoid the loss of protein due to binding
with the surface of the cuvette. The fluorescence cuvette was allowed to sit with 3 mL of 0.01%
gelatin containing PBS buffer (4 mM NaH2PO4, 16 mM Na2HPO4, 150 mM NaCl, pH = 7.3) for
30~60 minutes. The buffer was discarded and the cuvette was rinsed with d.d. water. Protein
sample (0.5 uM, 3 mL in PBS buffer) was added and was stirred gently. The sample was excited
at 283 nm (excitation slit width = 1.0 nm) and the fluorescence was measured at the peak
maximum (340-350 nm, emission slit width = 12 nm). C15 aldehyde or all-trans-retinal was
added to the sample at varying concentration. The total volume of EtOH was kept under 2%.
Measurement at the peak maximum was taken after each portion of chromophore was added.
The concentration of C15 aldehyde was plot against the relative fluorescence intensity. The
titration was completed when there was no observable quenching of fluorescence upon addition
of chromophore.

112

Since retinal has been reported to quench the Trp fluorescence, therefore, a correction was
made prior to calculating the binding constant for CRABPII mutants with retinal or C15
aldehyde. A blank titration of all-trans-retinal or C15 aldehyde with 1.5 µM N-acetyltryptophan
was preformed and the quenched fluorescence was added back to the actual titration with
CRABPII mutants. The fluorescence of the N-acetyltryptophan peaks at 365 nm.
The correction of fluorescence titration required 4 calculations, upon completion of both
protein and blank titration with all-trans-retinal and C15 aldehyde.

!=

1.

Fmax – F
Fmax – F0

Determination of a value for α for every single point of the titration:
-1

where Fmax = fluorescence upon saturation cm
F0 = initial fluorescence
F = observed fluorescence
α = fraction of free binding sites
2.

Determine the free ligand concentration
R = R0 – nP0 (1– α)
R0 = ligand concentration

113

n = number of binding site
P0 = protein concentration

1

y = -18620x + 0.9947
R2 = R2=0.99
0.9955

0.9
0.8
0.7
0.6
0.5
0

3.

2E-06

4E-06

6E-06

8E-06 0.00001 1.2E-05

Find the fluorescence contribution of the free ligand, FR, to be deduced from
the blank (N-acetyltryptophanamide) titration.
where y is FR

4.

Subtract the fluorescence contribution of the free ligand from the actual readings and plot

the corrected fluorescence values vs. the ligand concentration.
(F – FR) vs. R0
5.

The Kd for each CRABPII mutant was determinded according to the method previously
174

described by Wang et al.,

using the fluorescence values calculated above.

The dissociation constant is calculated by Kaledaigraph® using non-linear square
regression analysis.

The corrected relative fluorescence intensity is plotted against the

114

concentration of all-trans-retinal or C15 aldehyde. Within the program, the protein concentration
and the final fluorescence intensity at saturation need to be solved in order to calculate the
dissociation constant:

𝐹
𝐹!
= 1 + 
×
𝐹!
𝐹!

𝑃! + 𝑅! + 𝐾! −

𝑃! + 𝑅! + 𝐾!
2𝑃!

!

− 4𝑃! 𝑅!

 

F = observed fluorescence  
Ff = fluorescence of free protein
Fb = fluorescence of bound protein
Pt = total protein concentration
Rt = total chromohore concentration
The equation put into the Kaleidagraph® :
y=cell(0,1)+cell(0,1)*((cell(#,1)/cell(0,1))-1)*((0.0000005+m0+m1-sqrt((0.0000005+m0+
m1)^ 2-4*0.0000005*m0))/(2*0.0000005)); m1=0.000000001
where
cell (0,1) = initial fluorescence intensity = 1
cell (#,1) = fluorescence intensity upon saturation
y = fluorescence value
m0 = chromophore concentration
m1 = Kd

115

F.

Molecular Modeling
Energy minimization and molecular dynamics were performed with Discover module in
®

InsightII obtained from Accelrys. The structures for computational studies are obtained from
the crystal structures available from protein database. The conformer of the mutated residues is
chosen based on the lowest energy conformation and the double bond is added on the
corresponding positions on bound retinal or C15 aldehyde before subjecting it to computational
study. The number of energy minimization steps is set to 5000 at 298 K for 1 femtosecond per
step using CVFF forcefield from InsightII before molecular dynamic calculations. After the
structure is minimized, molecular dynamic calculation is performed. The molecular dynamic is
equilibrated for 5000 steps for 1 femtosecond using CVFF forcefield, starting at 0 K. The
obtained structures are subject to the subsequent rounds of calculation, where the temperature of
each round is increased by 50 K until 298 K is reached.
Electrostatic potential calculation was performed using standard procedures described for
APBS package. The dielectric constant for the protein was set at 2 while the dielectric constant
for water was set for 78 before the start of calculations. The charge and radius of atoms on each
residues within the protein was added by a web-base program pdb2pqr using Amber94 forcefield
before subjecting it for electrostatic potential calculation.
All the protein figures were generated by PyMol Molecular Graphics Systems, version 1.2
educational, copyright by DeLano Scientific LLC.
G. Chemistry
General

116

All syntheses were carried out in a dry nitrogen atmosphere in the dark room with
minimum red lights unless otherwise specified. All the reaction vessels were flame-dried.
Tetrahydrofuran was distilled over sodium and benzophenone under dried N2 atmosphere.
Reactions were monitored by thin-layer chromatography (TLC, on Merck F254 silica gel 60
aluminum sheets, spots were either visible under light or UV-light (254 mm) or treated with an
oxidizing solution (KMnO4 stain) Column chromatography was performed with Silicycle silica
gel 60.
1

H-NMR spectra were recorded on a Varian Unity+500 spectrometer with deuterated
1

chloroform (CDCl3; ∂ = 7.24 ppm) as an internal standard. H noise-decoupled

13

C spectra were

recorded on a Varian Unity+500 spectrometer at 125 MHz with deuterated chloroform (CDCl3; ∂
= 77 ppm) as an internal standard.
Synthesis of (2E,4E)-3-methyl-5-(2,6,6-trimethylcyclohex-1-enyl)penta-2,4-dienenitrile (125)
1) NaH, THF, 0 ˚C
(EtO)2P(O)CH2CN

rt;

2)

CN
1-25

O
0 ˚C

1-24
rt.

2-(Diethylphosphono)ethanenitrile (886 mg, 5 mmol) was added to NaH (208 mg of 60%
dispersion in oil, 5 mmol) in distilled THF (25 mL) at 0 °C. The reaction mixture was stirred for
1 h at room temperature. To the anion of the phosphonate was added β-ionone (1-24) (480mg,

117

2.5 mmol). The mixture was stirred for another 6 h at room temperature. The reaction was
monitored by the TLC (Rf = 0.8 (diethyl ether:40 – 60 light petroleum ether / 20:80) and was
quenched by ice. The aqueous layer was extracted by diethyl ether three times and the combined
organic layers were dried by anhydrous sodium sulfate (anhyd. Na2SO4) and filtered. The
product was purified by chromatography (silica gel, diethyl ether:40 – 60 light petroleum /
20:80) to give 1-25 (434 mg, 2 mmol, 80% yield) of as a mixture of the 9-cis and all-trans
compounds (1:3).
1

H-NMR

(500MHz,

CDCl3)

of

(2E,4E)-3-methyl-5-(2,6,6-trimethylcyclohex-11

1

enyl)penta-2,4-dienenitrile (1-25): δ 6.53 (dm, JH-H = 16.0 Hz, 1H), 6.11 (d, JH-H, 16.0 Hz,
1

1

1

1H), 5.13 (d, JH-H = 0.8 Hz, 1H), 2.18 (d, JH-H, 0.8, 3H), 2.01 (t, JH-H = 7.2 Hz, 2H), 1.68 (m,
3H), 1.59 (m, 2H), 1.45 (m, 2H), 1.00 (s, 6H)
1

H-NMR

(500MHz,

CDCl3)

of

(2Z,4E)-3-methyl-5-(2,6,6-trimethylcyclohex-11

1

enyl)penta-2,4-dienenitrile (1-25): δ 6.57 (dm, JH-H = 16.0 Hz, 1H), 6.68 (d, JH-H, 16.0 Hz,
1

1

1H), 5.07 (m, 1H), 2.03 (d, JH-H = 1.4 Hz, 3H), 2.01 (t, JH-H = 7.2 Hz, 2H), 1.73 (m, 3H), 1.59
(m, 2H), 1.45 (m, 2H), 1.02 (s, 6H)
13

C-NMR

(125MHz,

CDCl3)

of

(2E,4E)-3-methyl-5-(2,6,6-trimethylcyclohex-1-

enyl)penta-2,4-dienenitrile (1-25): δ 157.3 (s), 136.6 (s), 135.6(s,), 132.7 (s), 130.1 (s), 117.2 (s),
94.8 (s), 39.4 (s), 34.2 (s), 33.1 (s), 28.8 (s), 21.7 (s), 19.0 (s), 16.5 (s)

118

13

C-NMR

(125MHz,

CDCl3)

of

(2Z,4E)-3-methyl-5-(2,6,6-trimethylcyclohex-1-

enyl)penta-2,4-dienenitrile (1-25): δ 156.8(s), 136.6 (s), 136.1 (s), 132.6 (s), 133.0 (s), 118.0 (s),
96.3 (s), 39.5 (s), 34.1 (s), 33.2 (s), 28.9 (s), 21.8 (s), 19.2 (s), 22.7 (s)
Synthesis of (2E,4E)-3-methyl-5-(2,6,6-trimethylcyclohex-1-enyl)penta-2,4-dienal (1-23)

CN
1-25

1) DIBAL-H, THF,
-78 ˚C
-20 ˚C, 12h;

O

2) Silica-H2O, 0 ˚C, 8h

1-23

A solution of 1-25 (434 mg, 2 mmol) in distilled THF (20 mL) was cooled to –78

˚C.

DIBAL-H (2.7 ml, 1.5 M, 4 mmol) was added via syringe. The reaction mixture was allowed to
warm to –40 °C in 2 h. A homogeneous mixture of water absorbed on silica (1.7 g, 0.3 g
water/g silica) was added and stirring was continued for another 2 h at 0 °C. All solids were
filtered off and thoroughly washed with diethyl ether. The organic layer was dried by anhyd.
Na2SO4 and the solvents were evaporated in vacuo.

The product was purified by

chromatography (silica gel, diethyl ether:40 – 60 light petroleum ether/5:95) to give 1-23 (248
mg, 1.14 mmol, 57% yield) as a mixture of the 9-cis and all-trans compounds.
1

H-NMR

(500MHz,

CDCl3)

of

(2E,4E)-3-methyl-5-(2,6,6-trimethylcyclohex-1-

1

1

enyl)penta-2,4-dienal (1-23): δ 10.1 (d, JH-H = 8 Hz, 1H), 6.71 (d, JH-H = 16.1 Hz, 1H), 6.18
1

1

1

1

(dm, JH-H, 16.0 Hz, 1H), 5.90 (dm, JH-H = 8.0 Hz, 1H), 2.28 (d, JH-H, 1 Hz, 3H), 2.01 (t, JHH

= 7.2 Hz, 2H), 1.76 (s, 3H), 1.59 (m, 2H), 1.45 (m, 2H), 1.00 (s, 6H)

119

1

H-NMR

(500MHz,

CDCl3)

of

(2Z,4E)-3-methyl-5-(2,6,6-trimethylcyclohex-1-

1

1

enyl)penta-2,4-dienal (1-23): δ 10.2 (d, JH-H = 8 Hz, 1H), 7.10 (d, JH-H = 16 Hz, 1H), 6.64
1

1

1

1

(dm, JH-H, 16.0 Hz, 1H), 5.88 (dm, JH-H = 8.0 Hz, 1H), 2.14 (d, JH-H, 1 Hz, 3H), 2.07 (t, JHH

= 7.2 Hz, 2H), 1.69 (s, 3H), 1.65 (m, 2H), 1.50 (m, 2H), 1.02 (s, 6H)
13

C-NMR

(125MHz,

CDCl3)

of

(2E,4E)-3-methyl-5-(2,6,6-trimethylcyclohex-1-

enyl)penta-2,4-dienal (1-23): δ 191.4 (s), 155.2 (s), 137.0 (s), 135.6(s), 135.6 (s), 132.7 (s), 128.7
(s), 39.4 (s), 34.2 (s), 33.2 (s,), 28.9 (s), 21.7 (s), 19.0 (s), 12.9 (s)
13

C-NMR

(125MHz,

CDCl3)

of

(2Z,4E)-3-methyl-5-(2,6,6-trimethylcyclohex-1-

enyl)penta-2,4-dienal (1-23): δ 190.2 (s), 155.3 (s), 137.2 (s), 136.5 (s), 132.6 (s), 127.8 (s),
127.7 (s), 39.4 (s), 34.2 (s), 33.2 (s), 28.9 (s), 21.7 (s), 19.0 (s), 14.1 (s)
Synthesis of (2E,4E,6E)-5-methyl-7-(2,6,6-trimethylcyclohex-1-enyl)hepta-2,4,6-trienenitrile
(1-26)

(EtO)2P(O)CH2CN

1) NaH, THF, 0 ˚C
2)

rt;

O

CN
1-26

1-23
0 ˚C

rt.

2-(Diethylphosphono)ethanenitrile (198 mg, 1.1 mmol) was added to NaH (46 mg of 60%
dispersion in oil, 1.1 mmol) in distilled THF (25 mL) at 0 °C. The reaction mixture was stirred
for 1 h at room temperature. To the anion of the phosphonate, aldehyde (1-23) (120 mg, 0.56
mmol) was added. The mixture was stirred for another 6 h at room temperature. The reaction

120

was monitored by the TLC (Rf = 0.8 (diethyl ether: 40 – 60 light petroleum ether / 20:80) and
was quenched by ice. The aqueous layer was extracted by diethyl ether three times and the
combined organic layers were dried by anhydrous sodium sulfate (anhyd. Na2SO4). The product
was purified by chromatography (silica gel, diethyl ether:40 – 60 light petroleum / 20:80) to give
of 1-26 as a mixture of the 9-cis and all-trans compounds (56 mg, 0.23 mmol, 41% yield,
E:Z/1:3).
1

H-NMR

(500MHz,

CDCl3)

of

(2E,4E,6E)-5-methyl-7-(2,6,6-trimethylcyclohex-11

2

enyl)hepta-2,4,6-trienenitrile (1-26): δ 7.39 (dd, JH-H = 12 Hz, JH-H = 15.8 Hz, 1H), 6.44 (d,
1

1

1

1

JH-H = 16 Hz, 1H), 6.10 (d, JH-H, 16.0 Hz, 1H), 6.08 (d, JH-H = 12 Hz, 1H), 5.26(d, JH-H =
1

1

15.8 Hz, 1H), 2.00 (d, JH-H, 1 Hz, 3H), 2.01 (t, JH-H = 7.2 Hz, 2H), 1.69 (m, 3H), 1.45 (m,
2H), 1.01 (s, 6H)
13

C-NMR (125MHz, CDCl3) of (2E,4E,6E)-5-methyl-7-(2,6,6-trimethylcyclohex-1-

enyl)hepta-2,4,6-trienal (1-26): δ 146.3 (s), 145.2 (s), 137.4 (s), 136.0 (s), 132.4 (s), 131.3 (s),
130.7 (s), 126.4 (s), 119.1 (s), 96.5 (s), 39.6 (s), 34.3 (s), 33.2 (s), 28.9 (s), 21.8 (s), 19.1 (s), 13.1
(s)
Synthesis

of

(2E,4E,6E)-5-methyl-7-(2,6,6-trimethylcyclohex-1-enyl)hepta-2,4,6-

trienal (1-22)
1) DIBAL-H, THF,
-20 ˚C, 12h;
CN -78 ˚C
1-26

2) Silica-H2O, 0 ˚C, 8h

121

O
1-22

A solution of 1-26 (56 mg, 0.23 mmol) in distilled THF (20 mL) was cooled to –78 °C.
DIBAL-H (0.18 mL, 1.5 M, 0.23 mmol) was added via syringe. The reaction mixture was
allowed to warm to –40 °C in 2 h. A homogeneous mixture of water absorbed on silica (0.3 g
water/ gram silica) was added and stirring was continued for another 2 h at 0 °C. All solids were
filtered off and thoroughly washed with diethyl ether. The organic layer was dried by anhyd.
Na2SO4 and the solvents were evaporated in vacuo. The all-trans product was purified by
chromatography (silica gel, diethyl ether:40 – 60 light petroleum ether/ 5:95) to give 1-22 (5 mg,
5 % yield).
1

H-NMR

(500MHz,

CDCl3)

of

(2E,4E,6E)-5-methyl-7-(2,6,6-trimethylcyclohex-1-

1

1

2

enyl)hepta-2,4,6-trienal (1-22): δ 9.59 (d, JH-H = 8 Hz, 1H), 7.51 (dd, JH-H = 11.8 Hz, JH-H =
1

1

1

15 Hz, 1H), 6.48 (d, JH-H, 16.1 Hz, 1H), 6.27 (d, JH-H, 11.8 Hz, 1H), 6.19 (d, JH-H = 16.1 Hz,
1

2

1

1

1H), 6.15 (d, JH-H = 15.0 Hz, JH-H = 8.0 Hz, 1H), 2.07 (d, JH-H, 1 Hz, 3H), 2.02 (t, JH-H =
7.2 Hz, 2H), 1.71 (s, 3H), 1.60 (m, 2H), 1.46 (m, 2H), 1.02 (s, 6H)
13

C-NMR (125MHz, CDCl3) of (2E,4E,6E)-5-methyl-7-(2,6,6-trimethylcyclohex-1-

enyl)hepta-2,4,6-trienal (1-22): δ 193.7 (s), 148.0 (s), 146.7 (s), 137.3 (s), 136.4 (s), 132.5 (s),
131.6 (s), 130.7 (s), 127.4 (s), 39.6 (s), 34.3 (s), 33.2 (s), 28.9 (s), 21.8 (s), 19.1 (s), 13.1 (s)

122

BIBLIOGRAPHY

123

Bibliography
(1)

Tate, C. G.; Schertler, G. F. X. Curr. Opin. Struct. Biol. 2009, 19, 386.

(2)
Hanson, M. A.; Cherezov, V.; Griffith, M. T.; Roth, C. B.; Jaakola, V. P.; Chien, E.
Y. T.; Velasquez, J.; Kuhn, P.; Stevens, R. C. Structure 2008, 16, 897.
(3)

Hanson, M. A.; Stevens, R. C. Structure 2009, 17, 8.

(4)

Mustafi, D.; Palczewski, K. Mol. Pharmacol. 2009, 75, 1.

(5)

Kobilka, B.; Schertler, G. F. X. Trends Pharmacol. Sci. 2008, 29, 79.

(6)

Baldwin, J. M.; Schertler, G. F. X.; Unger, V. M. J. Mol. Biol. 1997, 272, 144.

(7)

Fredriksson, R.; Schioth, H. B. Mol. Pharmacol. 2005, 67, 1414.

(8)
Madabushi, S.; Gross, A. K.; Philippi, A.; Meng, E. C.; Wensel, T. G.; Lichtarge,
O. J. Biol. Chem. 2004, 279, 8126.
(9)

Grant, S. M.; Langtry, H. D.; Brogden, R. N. Drugs 1989, 37, 801.

(10) Arkin, I. T.; Brunger, A. T.; Engelman, D. M. Proteins-Structure Function and
Genetics 1997, 28, 465.
(11)

Wang, T.; Duan, Y. Journal of the American Chemical Society 2007, 129, 6970.

(12)

Essen, L. O. Chembiochem 2001, 2, 513.

(13) Casado, V.; Cortes, A.; Mallol, J.; Perez-Capote, K.; Ferre, S.; Lluis, C.; Franco,
R.; Canela, E. I. Pharmacology & Therapeutics 2009, 124, 248.
(14) Rosenbaum, D. M.; Cherezov, V.; Hanson, M. A.; Rasmussen, S. G. F.; Thian, F.
S.; Kobilka, T. S.; Choi, H. J.; Yao, X. J.; Weis, W. I.; Stevens, R. C.; Kobilka, B. K.
Science 2007, 318, 1266.
(15) Jaakola, V. P.; Griffith, M. T.; Hanson, M. A.; Cherezov, V.; Chien, E. Y. T.; Lane,
J. R.; Ijzerman, A. P.; Stevens, R. C. Science 2008, 322, 1211.
(16)

Wade, N. J. A natural history of vision; The MIT Press: Cambridge, Mass, 1998.

(17)

Shorey, P. The Republic; Heinemann: London, 1935.

(18)

Ross, W. D. The Works of Aristotle; Clarendon: Oxford, 1931; Vol. 3.

124

(19)

Kepler, J. Ad Vitellionem Paralipomena; Marinium and Aubrii: Frankfurt, 1604.

(20) Newton, I. Opticks: or,a Treatise of the Reflections, Refractions, Inflections and
Colours of Light; Henry Colburn: London, 1704.
(21)

Putri, I.

(22)

Baumann, C. Vision Research 1977, 17, 1267.

(23)

Boll, F. Archiv der Anatomie und Physiologie 1877, 4.

(24)

Kuhne Vision Research 1977, 17, 1269.

(25) Kuehne, W. Ueber den Sehpurpur; Untersuchungen aus dem Physiologischen
Institut der Universitaet Heidelberg: Heidelberg, 1878.
(26)

Wald, G. Ber Zusammenkunft Dtsch Ophthalmol Ges 1965, 66, 8.

(27)

Wald, G. Science 1964, 145, 1007.

(28) Wald, G. Vitamins and Hormones-Advances in Research and Applications 1960,
18, 417.
(29)

Brown, P. K.; Wald, G. Science 1964, 144, 45.

(30)

Wald, G. Science 1968, 162, 230.

(31)

Crescite.F Vision Research 1968, 8, 333.

(32)

Dubois-Poulsen, A. Ann Ocul (Paris) 1968, 201, 257.

(33) Nakanishi, K.; Chen, A. H.; Derguini, F.; Franklin, P.; Hu, S. H.; Wang, J. Pure
and Applied Chemistry 1994, 66, 981.
(34)

Packer, L. Methods in Enzymology 1982, 88, R19.

(35) Stecher, H.; Palczewski, K. Vertebrate Phototransduction and the Visual Cycle,
Pt B 2000, 316, 330.
(36) Whitten, M. E.; Yokoyama, K.; Schieltz, D.; Ghomashchi, F.; Lam, D.; Yates, J.
R.; Palczewski, K.; Gelb, M. H. Vertebrate Phototransduction and the Visual Cycle, Pt B
2000, 316, 436.
(37) Wissinger, B.; Sharpe, L. T. American Journal of Human Genetics 1998, 63,
1257.
(38) Palczewski, K. Vertebrate Phototransduction and the Visual Cycle (Part A);
Academic Press: Burlington, 2000; Vol. 315.
125

(39) Palczewski, K. Vertebrate Phototransduction and the Visual Cycle (Part B);
Academic Press: Burlington, 2000; Vol. 316.
(40)

Shichi, H. Biochemistry of Vision; Academic Press: Burlington, 1983.

(41) Rattner, A.; Smallwood, P. M.; Williams, J.; Cooke, C.; Savchenko, A.;
Lyubarsky, A.; Pugh, E. N.; Nathans, J. Neuron 2001, 32, 775.
(42)

Boynton, R. Human Color Vision; Holt, Rinchart, & Winston: New York, 1979.

(43)

Mollon, J. D., Sharpe, L.T. Colour Vision; Academic Press: New York, 1983.

(44)

Bowmaker, J. K., Hunt, D. M. Current Biology 2006, 16, R484.

(45) Mustafi, D., Engel, A. H., Palczewski, K. Progress in Retinal and Eye Research
2009, 28, 289.
(46) Hoffmann, R. The Same and Not the Same; Columbia University Press: New
York, 1995.
(47)

Cohen, A. I. Journal of Cell Biology 1968, 37, 424.

(48) Luo, D. G.; Xue, T.; Yau, K. W. Proceedings of the National Academy of
Sciences of the United States of America 2008, 105, 9855.
(49) Palczewski, K.; Kumasaka, T.; Hori, T.; Behnke, C. A.; Motoshima, H.; Fox, B. A.;
Le Trong, I.; Teller, D. C.; Okada, T.; Stenkamp, R. E.; Yamamoto, M.; Miyano, M.
Science 2000, 289, 739.
(50)

Menon, S. T.; Han, M.; Sakmar, T. P. Physiological Reviews 2001, 81, 1659.

(51) Stenkamp, R. E.; Filipek, S.; Driessen, C.; Teller, D. C.; Palczewski, K. Biochim.
Biophys. Acta-Biomembr. 2002, 1565, 168.
(52) Filipek, S.; Stenkamp, R. E.; Teller, D. C.; Palczewski, K. Annual Review of
Physiology 2003, 65, 851.
(53) Scheerer, P.; Park, J. H.; Hildebrand, P. W.; Kim, Y. J.; Krauss, N.; Choe, H. W.;
Hofmann, K. P.; Ernst, O. P. Nature 2008, 455, 497.
(54) Benovic, J. L., Mayor, F. Jr., Somers, R. L., Caron, M. G., Lefkowitz, R. J. Nature
1986, 321, 869.
(55) Ovchinnikov, Y. A.; Abdulaev, N. G.; Feigina, M. Y.; Artamonov, I. D.; Zolotarev,
A. S.; Kostina, M. B.; Bogachuk, A. S.; Miroshnikov, A. I.; Martinov, V. I.; Kudelin, A. B.
Bioorg. Khim. 1982, 8, 1011.

126

(56)

Dratz, E. A.; Hargrave, P. A. Trends Biochem.Sci. 1983, 8, 128.

(57)

Bownds, D. Nature 1967, 216, 1178.

(58) Sakmar, T. P.; Franke, R. R.; Khorana, H. G. Proceedings of the National
Academy of Sciences of the United States of America 1989, 86, 8309.
(59)

Zhukovsky, E. A.; Oprian, D. D. Science 1989, 246, 928.

(60)

Nathans, J. Biochemistry 1990, 29, 9746.

(61)

Vogt, K. Facets of Vision; Springer - Verlag: Berlin, 1989.

(62) Seki, T.; Vogt, K. Comparative Biochemistry and Physiology B-Biochemistry &
Molecular Biology 1998, 119, 53.
(63) Yan, B.; Nakanishi, K.; Spudich, J. L. Proceedings of the National Academy of
Sciences of the United States of America 1991, 88, 9412.
(64)

Shieh, T.; Han, M.; Sakmar, T. P.; Smith, S. O. J. Mol. Biol. 1997, 269, 373.

(65)

Yoshizawa, T.; Kito, Y. Nature 1958, 182, 1604.

(66) Wang, Q.; Schoenlein, R. W.; Peteanu, L. A.; Mathies, R. A.; Shank, C. V.
Science 1994, 266, 422.
(67)

Cooper, A. Nature 1979, 282, 531.

(68)

Nakanishi, K. Chem. Pharm. Bull. 2000, 48, 1399.

(69)

Kliger, D. S.; Lewis, J. W. Israel Journal of Chemistry 1995, 35, 289.

(70) Degrip, W. J.; Gray, D.; Gillespie, J.; Bovee, P. H. M.; Vandenberg, E. M. M.;
Lugtenburg, J.; Rothschild, K. J. Photochemistry and Photobiology 1988, 48, 497.
(71)

Lambright, D. G.; Noel, J. P.; Hamm, H. E.; Sigler, P. B. Nature 1994, 369, 621.

(72) Lambright, D. G.; Sondek, J.; Bohm, A.; Skiba, N. P.; Hamm, H. E.; Sigler, P. B.
Nature 1996, 379, 311.
(73)

Noel, J. P.; Hamm, H. E.; Sigler, P. B. Nature 1993, 366, 654.

(74) Sondek, J.; Bohm, A.; Lambright, D. G.; Hamm, H. E.; Sigler, P. B. Nature 1996,
379, 369.
(75) Lamb, T. D. Proceedings of the National Academy of Sciences of the United
States of America 1996, 93, 566.

127

(76) Arshavsky, V. Y.; Lamb, T. D.; Pugh, E. N. Annual Review of Physiology 2002,
64, 153.
(77) Xu, J.; Dodd, R. L.; Makino, C. L.; Simon, M. I.; Baylor, D. A.; Chen, J. Nature
1997, 389, 505.
(78) Wilden, U.; Hall, S. W.; Kuhn, H. Proceedings of the National Academy of
Sciences of the United States of America 1986, 83, 1174.
(79) Kuhn, H.; Bennett, N.; Michelvillaz, M.; Chabre, M. Proceedings of the National
Academy of Sciences of the United States of America-Biological Sciences 1981, 78,
6873.
(80)
331.

Chabre, M. Annual Review of Biophysics and Biophysical Chemistry 1985, 14,

(81)

Dartnall, H. J. Vision Research 1968, 8, 339.

(82) Young, T. The akerian lecture: On the theory of light and colors London, 1802;
Vol. 92.
(83)

Nathans, J.; Thomas, D.; Hogness, D. S. Science 1986, 232, 193.

(84) Baasov, T.; Sheves, M. Journal of the American Chemical Society 1985, 107,
7524.
(85) Blatz, P. E.; Johnson, R. H.; Mohler, J. H.; Aldilaim.Sk; Dewhurst, S.; Erickson, J.
O. Photochemistry and Photobiology 1971, 13, 237.
(86)

Blatz, P. E.; Mohler, J. H. Biochemistry 1975, 14, 2304.

(87)

Blatz, P. E.; Mohler, J. H. Biochemistry 1972, 11, 3240.

(88)

Blatz, P. E.; Mohler, J. H.; Navangul, H. V. Biochemistry 1972, 11, 848.

(89) Wada, M.; Sakurai, M.; Inoue, Y.; Tamura, Y.; Watanabe, Y. Journal of the
American Chemical Society 1994, 116, 1537.
(90) Harbison, G. S.; Smith, S. O.; Pardoen, J. A.; Courtin, J. M. L.; Lugtenburg, J.;
Herzfeld, J.; Mathies, R. A.; Griffin, R. G. Biochemistry 1985, 24, 6955.
(91) Vandersteen, R.; Biesheuvel, P. L.; Mathies, R. A.; Lugtenburg, J. Journal of the
American Chemical Society 1986, 108, 6410.
(92) Arnaboldi, M.; Motto, M. G.; Tsujimoto, K.; Baloghnair, V.; Nakanishi, K. Journal
of the American Chemical Society 1979, 101, 7082.

128

(93) Honig, B.; Dinur, U.; Nakanishi, K.; Baloghnair, V.; Gawinowicz, M. A.; Arnaboldi,
M.; Motto, M. G. Journal of the American Chemical Society 1979, 101, 7084.
(94) Sheves, M.; Nakanishi, K.; Honig, B. Journal of the American Chemical Society
1979, 101, 7086.
(95) Sheves, M.; Nakanishi, K. Journal of the American Chemical Society 1983, 105,
4033.
(96)

Asenjo, A. B.; Rim, J.; Oprian, D. D. Neuron 1994, 12, 1131.

(97)

Beck, M.; Sakmar, T. P.; Siebert, F. Biochemistry 1998, 37, 7630.

(98)

Chan, T.; Lee, M.; Sakmar, T. P. J. Biol. Chem. 1992, 267, 9478.

(99)

Nathans, J. Brain / 1990, 55, 621.

(100) Nathans, J. Biochemistry 1990, 29, 937.
(101) Nakayama, T. A.; Khorana, H. G. J. Biol. Chem. 1991, 266, 4269.
(102) Lin, S. W.; Kochendoerfer, G. G.; Carroll, H. S.; Wang, D.; Mathies, R. A.;
Sakmar, T. P. J. Biol. Chem. 1998, 273, 24583.
(103) Kloppmann, E.; Becker, T.; Ullmann, G. M. Proteins-Structure Function and
Bioinformatics 2005, 61, 953.
(104) Andersen, L. H.; Nielsen, I. B.; Kristensen, M. B.; El Ghazaly, M. O. A.; Haacke,
S.; Nielsen, M. B.; Petersen, M. A. Journal of the American Chemical Society 2005, 127,
12347.
(105) Tomasello, G.; Olaso-Gonzalez, G.; Altoe, P.; Stenta, M.; Serrano-Andres, L.;
Merchan, M.; Orlandi, G.; Bottoni, A.; Garavelli, M. Journal of the American Chemical
Society 2009, 131, 5172.
(106) Blatz, P. E.; Liebman, P. A. Experimental Eye Research 1973, 17, 573.
(107) Beppu, Y. Journal of the Physical Society of Japan 1997, 66, 3303.
(108) Beppu, Y.; Kakitani, T. Photochemistry and Photobiology 1994, 59, 660.
(109) Kakitani, T.; Beppu, Y.; Yamada, A. Photochemistry and Photobiology 1999, 70,
686.
(110) Kakitani, H.; Kakitani, T.; Rodman, H.; Honig, B. Photochemistry and
Photobiology 1985, 41, 471.

129

(111) Jin, Y. D.; Honig, T.; Ron, I.; Friedman, N.; Sheves, M.; Cahen, D. Chemical
Society Reviews 2008, 37, 2422.
(112) Balashov, S. P.; Imasheva, E. S.; Govindjee, R.; Sheves, M.; Ebrey, T. G.
Biophysical Journal 1996, 71, 1973.
(113) Mitchell, P. Nature 1961, 191, 144.
(114) Grigorieff, N.; Ceska, T. A.; Downing, K. H.; Baldwin, J. M.; Henderson, R. J. Mol.
Biol. 1996, 259, 393.
(115) Luecke, H.; Schobert, B.; Richter, H. T.; Cartailler, J. P.; Lanyi, J. K. J. Mol. Biol.
1999, 291, 899.
(116) Kolbe, M.; Besir, H.; Essen, L. O.; Oesterhelt, D. Science 2000, 288, 1390.
(117) Vogeley, L.; Sineshchekov, O. A.; Trivedi, V. D.; Sasaki, J.; Spudich, J. L.;
Luecke, H. Science 2004, 306, 1390.
(118) Spudich, J. L.; Luecke, H. Curr. Opin. Struct. Biol. 2002, 12, 540.
(119) Greenhalgh, D. A.; Farrens, D. L.; Subramaniam, S.; Khorana, H. G. J. Biol.
Chem. 1993, 268, 20305.
(120) Hackett, N. R.; Stern, L. J.; Chao, B. H.; Kronis, K. A.; Khorana, H. G. J. Biol.
Chem. 1987, 262, 9277.
(121) Mogi, T.; Marti, T.; Khorana, H. G. J. Biol. Chem. 1989, 264, 14197.
(122) Houjou, H.; Inoue, Y.; Sakurai, M. Journal of the American Chemical Society
1998, 120, 4459.
(123) Kliger, D. S.; Milder, S. J.; Dratz, E. A. Photochemistry and Photobiology 1977,
25, 277.
(124) Milder, S. J.; Kliger, D. S. Photochemistry and Photobiology 1977, 25, 287.
(125) Shimono, K.; Ikeura, Y.; Sudo, Y.; Iwamoto, M.; Kamo, N. Biochim. Biophys.
Acta-Biomembr. 2001, 1515, 92.
(126) Shimono, K.; Iwamoto, M.; Sumi, M.; Kamo, N. Photochemistry and Photobiology
2000, 72, 141.
(127) Ikeura, Y.; Shimono, K.; Iwamoto, M.; Sudo, Y.; Kamo, N. Photochemistry and
Photobiology 2003, 77, 96.

130

(128) Hatanaka, M.; Sasaki, J.; Kandori, H.; Ebrey, T. G.; Needleman, R.; Lanyi, J. K.;
Maeda, A. Biochemistry 1996, 35, 6308.
(129) Morgan, J. E.; Gennis, R. B.; Maeda, A. Photochemistry and Photobiology 2008,
84, 1038.
(130) Okada, T.; Sugihara, M.; Bondar, A. N.; Elstner, M.; Entel, P.; Buss, V. J. Mol.
Biol. 2004, 342, 571.
(131) Hutchison, C. A.; Phillips, S.; Edgell, M. H.; Gillam, S.; Jahnke, P.; Smith, M. J.
Biol. Chem. 1978, 253, 6551.
(132) Allert, M.; Hellinga, H. W. Protein Science 2004, 13, 335.
(133) Allert, M.; Kjellstrand, M.; Broo, K.; Nilsson, A.; Baltzer, L. Journal of the
Chemical Society-Perkin Transactions 2 1998, 2271.
(134) Cheon, Y. H.; Park, H. S.; Kim, J. H.; Kim, Y.; Kim, H. S. Biochemistry 2004, 43,
7413.
(135) Hellinga, H. W. Nature Structural Biology 1998, 5, 525.
(136) Chung, H. H.; Benson, D. R.; Schultz, P. G. Science 1993, 259, 806.
(137) Davies, R. R.; Distefano, M. D. Journal of the American Chemical Society 1997,
119, 11643.
(138) Yang, W.; Jones, L. M.; Isley, L.; Ye, Y. M.; Lee, H. W.; Wilkins, A.; Liu, Z. R.;
Hellinga, H. W.; Malchow, R.; Ghazi, M.; Yang, J. J. Journal of the American Chemical
Society 2003, 125, 6165.
(139) Baltzer, L.; Nilsson, H.; Nilsson, J. Chemical Reviews 2001, 101, 3153.
(140) Dahiyat, B. I.; Mayo, S. L. Science 1997, 278, 82.
(141) Hecht, M. H.; Richardson, J. S.; Richardson, D. C.; Ogden, R. C. Science 1990,
249, 884.
(142) Allert, M.; Baltzer, L. Chemistry-a European Journal 2002, 8, 2549.
(143) Dwyer, M. A.; Looger, L. L.; Hellinga, H. W. Proceedings of the National
Academy of Sciences of the United States of America 2003, 100, 11255.
(144) Franklin, S. J.; Welch, J. T. Comments on Inorganic Chemistry 2005, 26, 127.
(145) Swift, J.; Wehbi, W. A.; Kelly, B. D.; Stowell, X. F.; Saven, J. G.; Dmochowski, I.
J. Journal of the American Chemical Society 2006, 128, 6611.

131

(146) Butterfoss, G. L.; Kuhlman, B. Annual Review of Biophysics and Biomolecular
Structure 2006, 35, 49.
(147) Graziano, J. J.; Liu, W. S.; Perera, R.; Geierstanger, B. H.; Lesley, S. A.; Schultz,
P. G. Journal of the American Chemical Society 2008, 130, 176.
(148) Berg, D. T.; Gerlitz, B.; Shang, J.; Smith, T.; Santa, P.; Richardson, M. A.; Kurz,
K. D.; Grinnell, B. W.; Mace, K.; Jones, B. E. Proceedings of the National Academy of
Sciences of the United States of America 2003, 100, 4423.
(149) Sanders, T.; Hawrot, E. Faseb Journal 2004, 18, C75.
(150) Brannigan, J. A.; Wilkinson, A. J. Nature Reviews Molecular Cell Biology 2002, 3,
964.
(151) Koren, E., Zuckerman, L. A., Mire-sluis, A. R. Current Pharmaceutical
Biotechnology 2002, 3.
(152) Binz, H. K.; Amstutz, P.; Pluckthun, A. Nature Biotechnology 2005, 23, 1257.
(153) Zheng, W. L.; Bucco, R. A.; Schmitt, M. C.; Wardlaw, S. A.; Ong, D. E.
Endocrinology 1996, 137, 5028.
(154) Jing, Y. K.; Waxman, S.; MirayLopez, R. Cancer Research 1997, 57, 1668.
(155) Banaszak, L.; Winter, N.; Xu, Z. H.; Bernlohr, D. A.; Cowan, S.; Jones, T. A.
Advances in Protein Chemistry, Vol 45 1994, 45, 89.
(156) Ong, D. E., Newcorner, M. E., Chytil, F. Cellular Retinoid-binding proteins. In The
Retinoids: Biology, Chemistry, and Medicine; 2nd ed.; Raven Press: New York, 1994.
(157) Ong, D. E.; Chytil, F. J. Biol. Chem. 1978, 253, 4551.
(158) Wardlaw, S. A.; Bucco, R. A.; Zheng, W. L.; Ong, D. E. Biology of Reproduction
1997, 56, 125.
(159) Yamamoto, M.; Drager, U. C.; Ong, D. E.; McCaffery, P. European Journal of
Biochemistry 1998, 257, 344.
(160) Kreutz, M.; Fritsche, J.; Andreesen, R.; Krause, S. W. Biochemical and
Biophysical Research Communications 1998, 248, 830.
(161) Delva, L.; Bastie, J. N.; Rochette-Egly, C.; Kraiba, R.; Balitrand, N.; Despouy, G.;
Chambon, P.; Chomienne, C. Molecular and Cellular Biology 1999, 19, 7158.

132

(162) Gudas, L. J., Sporn, M. B., Sporn, A. B. R. Cellular biology and biochemistry of
the retinoids. In the Retinoids: Biology, Chemistry, and Medicine; Raven Press: New
York, 1994.
(163) De Luca, L. M. Faseb Journal 1991, 5, 2924.
(164) Lotan, R. Biochimica Et Biophysica Acta 1980, 605, 33.
(165) Chambon, P. Faseb Journal 1996, 10, 940.
(166) Napoli, J. L. Biochimica Et Biophysica Acta-Molecular and Cell Biology of Lipids
1999, 1440, 139.
(167) Napoli, J. L. Faseb Journal 1996, 10, 993.
(168) Dong, D.; Ruuska, S. E.; Levinthal, D. J.; Noy, N. J. Biol. Chem. 1999, 274,
23695.
(169) Leroy, P., Krust, A., Kastner, P., Mendelsohn, C., Zelent, A., Chambon, P.
Retinoic Acid Receptors. In Retinoids in Normal Development and Teratogenesis;
Oxford University Press: Oxford, 1992.
(170) Mangelsdorf, D. J.; Thummel, C.; Beato, M.; Herrlich, P.; Schutz, G.; Umesono,
K.; Blumberg, B.; Kastner, P.; Mark, M.; Chambon, P.; Evans, R. M. Cell 1995, 83, 835.
(171) Giguere, V. Endocrine Reviews 1994, 15, 61.
(172) Chen, X.; Tordova, M.; Gilliland, G. L.; Wang, L. C.; Li, Y.; Yan, H. G.; Ji, X. H. J.
Mol. Biol. 1998, 278, 641.
(173) Vaezeslami, S.; Mathes, E.; Vasilelou, C.; Borhan, B.; Geiger, J. H. J. Mol. Biol.
2006, 363, 687.
(174) Wang, L. C.; Li, Y.; Yan, H. G. J. Biol. Chem. 1997, 272, 1541.
(175) Vasileiou, C.; Lee, K. S. S.; Crist, R. M.; Vaezeslami, S.; Goins, S. A.; Geiger, J.
H.; Borhan, B. Proteins-Structure Function and Bioinformatics 2009, 76, 281.
(176) Vasileiou, C., Michigan State University, 2005.
(177) Crist, R. M.; Vasileiou, C.; Rabago-Smith, M.; Geiger, J. H.; Borhan, B. Journal of
the American Chemical Society 2006, 128, 4522.
(178) Vasileiou, C.; Vaezeslami, S.; Crist, R. M.; Rabago-Smith, M.; Geiger, J. H.;
Borhan, B. Journal of the American Chemical Society 2007, 129, 6140.

133

(179) Vasileiou, C.; Wang, W. J.; Jia, X. F.; Lee, K. S. S.; Watson, C. T.; Geiger, J. H.;
Borhan, B. Proteins-Structure Function and Bioinformatics 2009, 77, 812.
(180) Simonson, T.; Brooks, C. L. Journal of the American Chemical Society 1996,
118, 8452.
(181) Warshel, A.; Russell, S. T. Q. Rev. Biophys. 1984, 17, 283.
(182) Warshel, A.; Papazyan, A. Curr. Opin. Struct. Biol. 1998, 8, 211.
(183) King, G.; Lee, F. S.; Warshel, A. Journal of Chemical Physics 1991, 95, 4366.
(184) Schutz, C. N.; Warshel, A. Proteins-Structure Function and Genetics 2001, 44,
400.
(185) Nandi, N.; Bagchi, B. Journal of Physical Chemistry B 1997, 101, 10954.
(186) Lazaridis, T.; Karplus, M. Proteins-Structure Function and Genetics 1999, 35,
133.
(187) Baker, N. A.; Sept, D.; Joseph, S.; Holst, M. J.; McCammon, J. A. Proc. Natl.
Acad. Sci. U. S. A. 2001, 98, 10037.
(188) Baker, N.; Holst, M.; Wang, F. J. Comput. Chem. 2000, 21, 1343.
(189) Dolinsky, T. J.; Nielsen, J. E.; McCammon, J. A.; Baker, N. A. Nucleic Acids
Research 2004, 32, W665.
(190) Holst, M.; Baker, N.; Wang, F. J. Comput. Chem. 2000, 21, 1319.
(191) Holst, M. J.; Saied, F. J. Comput. Chem. 1995, 16, 337.
(192) Holst, M.; Saied, F. J. Comput. Chem. 1993, 14, 105.
(193) Iwamoto, M.; Furutani, Y.; Sudo, Y.; Shimono, K.; Kandori, H.; Kamo, N.
Biophysical Journal 2002, 83, 1130.
(194) Aravind, L.; Anantharaman, V.; Balaji, S.; Babu, M. M.; Iyer, L. M. Fems
Microbiology Reviews 2005, 29, 231.
(195) Harrison, S. C.; Aggarwal, A. K. Annual Review of Biochemistry 1990, 59, 933.
(196) Santos, C. L.; Tavares, F.; Thioulouse, J.; Normand, P. Fems Microbiology
Reviews 2009, 33, 411.
(197) Creemers, A. F. L.; Lugtenburg, J. Journal of the American Chemical Society
2002, 124, 6324.

134

(198) Jia, X. F., Michigan State University, 2008.
(199) Vaezeslami, S.; Jia, X. F.; Vasileiou, C.; Borhan, B.; Geiger, J. H. Acta
Crystallographica Section D-Biological Crystallography 2008, 64, 1228.
(200) Scheiner, S.; Hillenbrand, E. A. Proceedings of the National Academy of
Sciences of the United States of America 1985, 82, 2741.
(201) Sambrook, J., Fritsch, E. F., Maniatis, T. Molecular Cloning: A Laboratory
Manual; Cold Spring Harbor Laboratory Press: Cold Spring Harbor, NY., 1989.

135

Chapter 2
Development of in vivo Color Screening System for Probing the Long Range Interactions
on Wavelength Regulation of Rhodopsin in CRBPII
2.1. Introduction
Residues further away from the binding pocket or active site of proteins can be important
1

for protein reactivity and specificity.

Many examples of long range interactions in tuning
2-11

enzyme activity have been identified through random mutagenesis experiments.

Wada et al.

improved the reactivity of N-acetylneuraminic acid aldolase, (natural substrates are N-acetyl-Dmannosamine and D-arabinose) for substrates (N-acetyl-L-mannosamine and L-arabinose)
through random mutagenesis. The N-acetylneuraminic acid aldolase mutants selected were able
to produce the unnatural sugar that is enantiomeric to its wild-type product in vitro. The
mutations identified were F115L, V251I and T98H. However, none of the mutations are in close

Figure 2-1. Crystal structure (Right) and anaglyph stereo picture (Right) of Nacetylneuraminate lyase (Left) showing the active site lysine (Purple) and mutated
residues (Blue).

136

proximity to the active site and the closest mutation, V251I, is at least 13.9 Å away from the
6

active residue, Lys165 (Figure 2-1). In addition, Jackson and Fersht in 1993, identified surface
charged residues of subtilisin BPN’ that stabilize the transition state of amide hydrolysis. The
-1

contribution was about 0.6 kcal mol and was based on the long-range electrostatic interactions
5

(>15 Å) between the surface charged residues and transition state within the protein.

In Chapter 1, we have proposed
that

electrostatic

potential

could

modulate the absorption of bound retinal
PSB.

In 2002, Kloppmann and co-

workers conducted a computational
study on bacteriorhodopsin (λmax = 560
nm) and sensory rhodopsin II (λmax =
Figure
2-2.
Crystal
structure
of
bacteriorhodopsin with hightlighted residues 497 nm). The crystal structures of
around the bound retinal PSB (Residues <5 Å
and
sensory
away are shown in purple while residues >5 Å bacteriorhodopsin
away are shown in cyan).
rhodopsin II indicate that the residues within 5 Å away from the bound retinal PSB are very
similar in both proteins.

Substitution of those residues in sensory rhodopsin with the

corresponding residues in the bacteriorhodopsin did not afford a protein with absorption close to
12,13

bacteriorhodopsin.

Therefore, residues that are not close to the retinal binding site, must be

the origin of the difference in absorption between these two rhodopsins. From the calculations,
they identified 4 residues that are more then 5 Å away from the bound retinal PSB, and are able
to contribute a significant electrostatic potential on retinal collectively (Figure 2-2).

137

14

This result

indicated that long-range interactions (>5 Å away from the binding site), referred to second shell
interactions, are also contributing to the wavelength regulation in rhodopsins. Apart from the
electrostatic potential, conformational changes away from the binding site also affect the
wavelength regulation in microbial rhodopsin. In bacteriorhodopsin, Arg82 is pointing towards
retinal PSB while in sensory rhodopsin, Arg72 faces away from the retinal PSB (Figure 2-3).
This specific Arg is about 9 Å away from the retinal PSB but was shown to have a large effect on
the corresponding absorption in microbial rhodopsin (Figure 2-3).

15-18

In addition, Yoshitsugu

et al. showed that A178R mutation in proteorhodopsin, that is 25 Å away from the retinal
binding site, induces a 20 nm red shift on the bound retinal PSB.

19

This further proves the

importance of second shell interactions in wavelength regulations in rhodopsins.
Directed evolution has been a popular research subject for the last decade since it was first
20

introduced in 1967.

Directed evolution incorporates the Darwinian principles of mutation and

D75

D85
R72

R82

D212

D201

Figure 2-3. Crystal structure of bacteriorhodopsin (Green) and sensory rhodopsin II
(Cyan) with the water-mediated interaction between Arg82 (Bacteriorhodopsin) or
Arg72 (Sensory Rhodopsin II) and PSB of bound all-trans-retinal.

138

Parental gene
Error-Prone PCR

DNA Purification

Bacteria with
Different
Mutated Gene

Selection

Figure 2-4. Scheme for directed evolution.
21-23

selection into experiments for improving biocatalysts and cellular processes.

It generally
6

includes two processes, creating of a library followed by selection. A library of mutants (>10 )
is generated through random mutagenesis and is followed by selecting mutants based on the
corresponding phenotype. These processes would be repeated until the best mutant is created
(Figure 2-4). Directed evolution has been widely used in biomedical, industrial, physiological
22,24-32

and chemical research.

There are numerous successful examples for using directed
6,10,23-26,28,33

evolution to improve enzyme activity, specificity, selectivity and stability.

Lipases

are one such example. Lipases are carboxylesterases that catalyze the hydrolysis of long-chain
acylglycerols stereospecifically and regio-specifically (Figure 2-5).

139

34

They are easy to

O
O
R

R

O

O

Triacylglyrerol
Lipase

O

O
R

O

+

O

R

O

R

O

R

OH

OH

O
R = long-chain fatty acid
Figure 2-5. Human pancreatic lipase hydrolyzed triacylglycerol to diacylglycerol and
fatty acid regiospecifically.

handle and are readily available in large quantities; therefore, they have been widely used in
chemical synthesis and industry. Because of their attractiveness in industrial and academic
research, many research programs have focused on improving lipases stability or creating new
substrate specificity for lipases through directed evolution.

31

In 1997, Reetz et al. were able to

improve the enantioselectivity of lipase from Pseudomonas aeruginosa by 40 folds in one single
25

experiment.

In addition, Tsien, 2008 Nobel Prize winner in chemistry, and his colleagues were

able to apply directed evolution to evolve monomeric red fluorescent protein (mRFP) from
30

Discosoma sp. into different color (yellow to red) fluorescent protein.

However, these mRFP

mutants were prone to aggregate so these mRFP mutants were subjected to another round of
30

directed evolution to improve their stability by breaking down the aggregation.

The mutants

obtained from the second round of directed evolution could be used as alternatives for green
fluorescent protein as protein fusion tags to visualize biological processes in vivo.
As mentioned before, directed evolution involves two steps. The target genes are first
mutated randomly. Then, the gene products from random mutagenesis are subject to selection or
screening based on the mutants’ phenotype.

There are numerous methods for introducing

140

random mutations into genes, with new techniques for random mutagenesis still being
developed. Two well-known and well-applied techniques are error-prone polymerase chain
reaction (PCR) and DNA shuffling. Polymerase chain reaction is a technique for amplifying the
target genes in vitro. During PCR, one tries to avoid errors so that pure gene products can be
obtained. The error can be avoided by using high fidelity DNA polymerases under optimal
35

conditions.

However, during error prone PCR, errors are introduce purposely during the PCR

process (Figure 2-4).
mixture.

36

2+

DNA polymerases are sensitive to Mg

concentrations in the reaction

Through increasing the MgCl2 concentration, the fidelity of DNA polymerases

decrease and the enzymes are more prone to introduce errors during PCR.

36,37

A library of

mutants with random mutations is then generated. The other well-known method for random
mutagenesis is called DNA shuffling that is developed by Stemmer.
40

diverse library of mutants than error-prone PCR.
involves shuffling of the genes.

38-40

It can generate a more

DNA shuffling as the name suggests

During DNA shuffling, a library of mutated genes or a

collection of genes that belong to same family are randomly digested by DNAseI.

The

randomized fragments of genes with different size are re-ligated through normal PCR protocol
(Figure 2-6).

39,40

When a library of mutants is generated, mutants are subjected to screening and

selection. However, to identify the right targets from a library that has more than a million
mutants, a powerful screening or selection method has to be developed. A selection or screening
method is described as a method to distinguish target proteins from others based on the mutants
phenotype. While selection identifies qualified mutants based on a single readout, screening is
described as a method to identify qualified mutants based on activity levels that are set by

141

DNAaseI

Re-ligation

Selection

Figure 2-6. Scheme for DNA shuffling.
researchers. Each enzyme or protein requires specific selection or screening based on the
phenotype.

For example, Minshull’s group in 1999 used DNA shuffling to improve

commercially available subtilisin’s properties including thermostability, organic solvent
41

tolerance and acid-base stability.

Subtilisins are important serine endopeptidases with broad

substrate specificity that hydrolyze peptides bonds. To be able to screen subtilisin mutants,
Minshull and his colleagues used a specific substrate that only fluoresces after a specific amide
bond is cleaved by subtilisin (Figure 2-7).

142

Subtilisin BPN'

F
OH

B
F

B
F

caesin

F

O

NH

Fluorescence at 560 nm

O

+ H2N
Figure 2-7. Scheme describing a screening system for subtilisin BPN’.

caesin

2.2. Study the Second Shell Interactions on Wavelength Regulation using Engineered
Rhodopsin Mimics
In Chapter 1, we have demonstrated the importance of having a fully embedded
chromophore in order to study wavelength regulation in our protein surrogate, CRABPII (Figure
2-8 and Table 2-1). Based on the crystal structures of C15 aldehyde with CRABPII mutants
(R132K:R111L:L121E:R59W)
retinal

with

(R132K:R111L:L121E),

and

CRABPII
C15

all-transmutants

aldehyde

is

completely covered within the binding pocket of
CRABPII while the β-ionone ring of all-transretinal is solvent exposed (Figure 2-8). Based on
Figure 2-8. Overlay of C15 aldehyde
CRABPII
mutants
UV-vis spectroscopy, C15 aldehyde is more with
(R132K:R111L: L121E:R59W) (Blue)
and all-trans-retinal with CRABPII
mutants
(R132K:R111L:
L121E)
(Green).
143

sensitive to the mutations in CRABPII than all-transretinal.

The mutants that induce different shift in

absorption with bound C15 aldehyde PSB, do notproduce
any changes with bound all-trans-retinal PSB absorption
(Table 2-1). In order to use the full-length retinal to study
the wavelength regulation in protein surrogates, we have
Figure 2-9.
Overlay crystal switched our protein mimic from CRABPII to human
structure of CRBPII (Green)
with
all-trans-retinol
and cellular retinol binding protein II (CRBPII). We suggested
CRABPII (Blue) with all-transthat the bound retinal in CRBPII should be fully
retinal.
embedded by the protein. The suggestion is based on the holo-structure of CRBPII (Figure 29).

42

The retinol is 5 Å deeper inside CRBPII as compared to the retinoic acid bound to

CRABPII.
Table 2-1. Absorption of CRABPII R59 mutants with C15
aldehyde and all-trans-retinal.
small
Mutants

λmax (nm)
(C15-aldehyde)

R132K:R111L:L121E:R59 400
R132K:R111L:L121E:R59E 424
R132K:R111L:L121E:R59Q413
R132K:R111L:L121E:R59L 388
R132K:R111L:L121E:R59D414

λmax (nm)
(all-trans-retinal)
449
450
443
441
446

144

cytoplasmic

CRBPII is a
monomeric
protein

of

approximately 15.5 KDa.
It is mainly responsible to
transfer and to protect the

all-trans-retinol
43-47

within cells.

(Figure

2-11)

Retinoids play a

crucial role in physiology of
vertebrates, particularly in cell
growth

and

embryonic

differentiation,

development

and

45,46,48

CRBPII belongs to
Figure 2-10. Crystal structure of CRBPII with all- vision.
trans-retinol (Blue) and the anaglyph picture on the
right.
the family of intracellular lipid42

binding proteins like CRABPII and shares a similar structural topology.

It is comprised of a

β-barrel formed by 10 antiparallel β-sheets and capped by a helix-turn-helix motif (Figure 2-10).
CRBPII binds retinol as a natural substrate with a 10 nM Kd and also binds retinal with Kd equal
to 50 nM. Evident from the crystal structure of CRBPII with retinol, the retinol is bound through
hydrogen bonding with Gln108 and is completely embedded within binding pocket (Figure 210).

Recently, we have crystallized the holo-

structure of CRBPII with retinal (Figure 2-12). As

OH
all-trans-retinol

predicted, the retinal is fully covered by CRBPII and
occupies the same place within the binding pocket as
retinol (Figure 2-12). As mentioned in Chapter one,

O
all-trans-retinal

protein-chromophore

inter-actions

are

greatly

buffered by the solvent molecules around the solvent

Figure 2-11. Structure of all-trans- exposed chromophore. Therefore, it is beneficial to
retinol and all-trans-retinal.
re-engineer CRBPII into the second generation

145

3.0
K132
T51
Figure 2-12. Crystal structures of all-trans-retinal in CRBPII (Green, left) and all-transretinal PSB in CRABPII mutant (R132K:R111L:L121E, cyan, right).
rhodopsin mimic for studying wavelength regulation. The binding of retinal is mainly through
the hydrophobic contacts between the polyene and CRBPII and hydrogen bonding with Thr51
(Figure 2-12).

There is a major difference between retinal in CRABPII mutant

(R132K:R111L:L121E) and retinal with CRBPII. The bound retinal in CRBPII adopts a C6-C7
s-trans conformation while in CRABPII, retinal adopts as C6-C7 s-cis conformation (Figure 212). We have successfully applied the knowledge we learned from re-engineering CRABPII into
generating a rhodopsin mimic with CRBPII. Wang et al. engineered CRBPII to bind all-transretinal as a PSB through engineered Lys108 (Figure 2-13).

49

In addition, CRBPII mutants are

capable to modulate the absorption of the bound retinal through site-directed mutagenesis of
nearby residues (<5 Å) (Table
2-2).

49

Table 2-2. Absorption of CRBPII Mutants bound to allAs mentioned before, trans-retinal.

Ulrich computational study

Mutants

λmax (nm)

showed that the electrostatic

Q108K:K40L

506

potential

Q108K:K40L:R58W

519

Q108K:K40L:R58W:T51V

565

Q108K:K40L:R58W:T51V:T53C

585

induced

by

individual residues, that are

146

more than 5 Å from the retinal binding site, are small.

This second generation of rhodopsin

mimic appears to be more sensitive to mutations as predicted (Table 2-2). Therefore, we believe
our second generation rhodopsin mimic are
CRBP (Q108K:K40L) Titrated
with all-trans-retinal
0.5
506 nm

suitable to study long range interaction that
are small in strength.

0.4

study, Ulrich et al. were able to identify
key residues 5 Å away from the retinal
binding site that affect the electrostatic
potential around bound retinal. In order to

Absorbance

During an elegant computational

Every 0.2 eq.
all trans-retinal

0.3
0.2
0.1

further probe the role of second shell

0
250 300 350 400 450 500 550 600 650
Wavelength (nm)
interactions in wavelength regulation, Figure 2-13. Titration of CRBPII mutants
Q108K:K40L with all-trans-retinal.
50-54
APBS package
was used to calculate

the electrostatic potential map of different color rhodopsins based on 1) the complete structure of
the protein; 2) the residues 5 Å away; and 3) the residues within 5 Å from the binding site of
bound retinal (Figure 2-14). Based on the calculations, the electrostatic potential, projected from
the residues within 5 Å from the retinal binding site are similar between Blue and Rod as well as
between Red and Green. The calculation suggested that the residues that are more than 5 Å
away from the bound retinal are responsible for the differences observed between the latter two
groups of proteins. For example, the rod and blue rhodopsin both have an intense negative
electrostatic potential close to the PBS region, however, the distal residues (>5 Å away from the
bound retinal PSB) of Rod induce a positive electrostatic field around the PBS region of the

147

>5 Å

<5 Å

-40
0
40
Blue (410 nm) Rho (500 nm) Green (530 nm) Red (560 nm)
II
III
I
H+
N
I

II

Rh
Lys296

Blue/Rh

Blue

Polar

Red

Green

Rh

Less Polar

III

Blue/
Red/
Red
Non Polar
Green
Green
Figure 2-14. Electrostatic potential projected on 11-cis-retinal in rod rhodopsin and color
50-52,93
rhodopsins using APBS Package.
Top: Electrostatic potential calculation from the
residues >5 Å away from the retinal binding site; Middle: residues <5 Å away from the
retinal binding site; Bottom: Electrostatic potential calculation of whole protein.
retinal. This helps to neutralize the negative electrostatic induced by the close resideus (<5 Å),
therefore, overall electrostatic potential around the PSB region for Rod is less negative as
compared to Blue rhodopsin leading to the red shift of Rods. A similar pattern also occurs
between Green and Red rhodopsin (Figure 2-14). The calculations show that although the effect
on the electrostatic potential from the distal residues, more than 5 Å away from the bound

148

chromophore (second shell interactions), is not as significant as that of the residues directly in
contact with the chromophore, the second shell interactions can still affect the electrostatic
potential projected on the bound 11-cis-retinal directly, thus, affecting absorption of the
chromophore. In addition, these second shell residues could also affect the absorption of the
bound chromophore indirectly through modulating the polarity of the residues around the
chromophore. Therefore, we propose that surface residues in CRBPII could act collectively
changing the electrostatic potential around the retinal PSB. In Figure 2-15, we identified four
residues, that are either negatively or polar on the surface of the protein (Asp61, Asp63, Thr120
and Glu118) directly above or below the PSB region of the bound retinal. They were mutated to
Arg. Based on APBS calculations, the electro-static potential map around the PSB region is less
negative after the substitution of specific surface residues. However, it is difficult to identify
specific residues far away from the binding site (>5 Å) that have significant effect on the
chromophore’s absorption. From the crystal structures, there are 23 residues within 5 Å away
from the retinal PSB. If we would like to study the effect of distal residues (110 residues)
110

individually, 4

of mutations have to be made

D61

based on four different properties of amino acid
(charged, polar, apolar and aromatic).

D63

D61R

In

addition, as demonstrated by Ulrich’s calculation,

E118
T120

the distal residues usually do not have significant

T120R
effect by themselves.

D63R

14

E118R

Therefore, one would Figure 2-15. Projection of electrostatic
potential by CRBPII mutants on allhave to generate mutants with different trans-retinal (Left: Q108K:K40L; Right:
Q108K:K40L:D61R:D63R:T120R:E118
combinations of different mutations thus R).

149

increasing the potential pool of mutants. Use of directed evolution strategies would be practical
alternative in order to study the second shell interactions on the wavelength regulation.
2.2.1. Design of Color Screening Method for CRBPII
To be able to apply directed evolution to study second shell interactions on wavelength
regulation using the second generation of rhodopsin mimic (CRBPII), we need to develop an
efficient selection or screening method to read the results of a library of mutants from random
mutagenesis. When CRBPII mutants bind to all-trans-retinal, the result is a variety of different
absorptions that correspond to different colors (Figure 2-16).

55

The color is the phenotype for

retinal-CRBPII complex formation and could be used for screening. We propose that all-transretinal-synthesizing bacteria that express a CRBPII mutant will bind the all-trans-retinal
produced by the bacteria. The formation of a CRBP-retinal complex will stain the cells with
colors corresponding to the UV-vis absorption (Figure 2-16).

A similar system has been

developed by Jung’s group in 2008. Jung’s group used a retinal producing system to screen
proteorhodopsin mutants based on the color in order to identify residues that modulate
wavelength in proteorhodopsin.
58,59

proteobacteria.

56,57

Proteorhodopsins are rhodopsins found in marine

They bind all-trans-retinal as a chromophore and show absorption around

490 nm to 525 nm (depend ing on the type of proteorhodopsin). Like bacteriorhodopsin,
proteorhodopsins are light-driven proton pumps and are necessary for energy generation in
bacteria using light.

60-62

The all-trans-retinal-synthesizing bacteria are a bacterial strain having plasmids that
contain β-carotene (Figure 2-16) synthesizing gene cluster and the β-carotene dioxygenase gene

150

!-carotene
mouse !-carotene dioxygenase
(mono-oxygenase)/ BCDOX
O + O
all trans-retinal

NH2

H2N

NH+

+HN

Figure 2-16. Proposed in vivo color screening method for CRBPII mutants.
63,64

for cleavage of β-carotene into all-trans-retinal.

β-Carotene is an essential photoprotecting
65-69

agent for the bacteria, fungi, algae and higher plants.

and the β-carotene synthesizing gene

cluster consists of multiple genes encoding enzymes that are necessary for β-carotene
63,70-72

biosynthesis (Figure 2-17) and is obtained from Pantoea ananatis.

The precursor for

carotenoid biosynthesis is farnesyl diphosphate (FPP) that is bio-synthesized from the 2-Cmethyl-D-erythri-tol 4-phosphate (MEP) pathway in E. coli. This 2-C-methyl-D-erythritol 4phosphate (MEP) pathway in E. coli is responsible to produce isopentenyl diphosphate (IPP) and
dimethylallyl diphosphate (DMAPP) (Figure 2-18). DMAPP is further modified by enzymes
including farnesyl diphosphate (FPP) synthase into FPP (Figure 2-18).
151

70,73

FPP is elongated by

O
P

O

O
P

O

O-

FPP

O-

OIPP
CrtE
O
O

O-

GGPP
GGPP

P

O
O

P

O-

O-

CrtB

Phytoene

CrtI

Lycopene
CrtY

!-carotene
Figure 2-17. Biosynthesis of β-carotene in vivo.
GGPP (geranylgeranyl-pyrophosphate) synthase to GGPP. Two GGPP molecules are coupled

152

together

by

phytoene synthase
to phytoene that is

HO
HO

O
O
P
P
O - O - OO
O

OPO32OH

further
dehydrogenated by
phytoene
desaturase

to

lycopene.
Lycopene

is

FPP
Figure 2-18. MEP Pathway in E. coli.

O
O
P
P
O - O - OO
O

finally cyclized by lycopene cyclase and β-carotene is obtained (Figure 2-17).

63,70-72

β-Carotene synthesized by enzymes transcribed from the β-carotene gene cluster is
subjected to β-carotene 15,15’-dioxygenase (BCDOX) from Mus musculus, generating all-transretinal (Figure 2-19). β-Carotene 15,15’-dioxygenase (BCDOX) as the name suggested cleaves

!-carotene

mouse !-carotene
dioxygenase (BCODX)

O2

O

+

O

all trans-retinal
Figure 2-19. Proposed mechanism of β-carotene cleavage into all-trans-retinal.

153

β-carotene to all-trans-retinal (Figure 2-19). BCDOX is responsible for β-carotene metabolism
74

and vitamin A formation as initially described in 1930 by Moore.

The first enzymatic activity

of β-carotene cleavage was reported 35 years later in 1965 by Olson and Hayaishi in rat liver and
75

rat intestine.

Since then, many research groups have tried to obtain pure BCDOX or to study
76-82

the mechanism of β-carotene cleavage in crude lysate.

It was found that the cleavage

required molecular oxygen and the cleavage occurred at the central position to generate two
83

molecules of all-trans-retinal.

Early studies by Olson’s group and Goodman’s group showed
2+

that the enzyme required iron (Fe ) as a co-factor for β-carotene cleavage since iron chelating
75,84

agents completely shut down the reaction.

However, it is suggested to be a non-heme

dependant oxygenase because the partially purified enzyme does not have heme characteristic
absorption (lacking the soret band at 415 nm).

85

In addition, sulfhydryl alkylating reagents such
75-77

as: p-chloromercuribenzoate and N-ethylmaleimide inhibit the enzyme activity.

This

indicated that cysteines in the enzyme play an important role for catalysis, probably as a result of
2+

Fe

coordination. Based on amino acid sequence alignment, it is believed that BCDOX belongs
85

to the family of non-heme iron containing dioxygenases.

In 2001, Leuenberger and co-

workers were able to purified BCDOX to homogeneity and demonstrated that BCDOX cleaves
86

β-carotene through a mono-oxygenase mechanism using labeled oxygen.
incubated with β-carotene in the presence of

17

18

BCDOX was

O2 and H2 O. Through analysis by the high

154

!-carotene
H218O

17O

+

17O

mouse !-carotene dioxygenase
(mono-oxygenase)/ BCDOX

2

18O

+

all trans-retinal

Figure 2-20. Elucidation of the mechanism of β-carotene cleavage using labeled oxygen.
resolution mass spectrometry (HRMS), the isolated retinals contained 50% enrichment of each
labeled oxygen (Figure 2-20). The result indicated that the cleavage follows a monooxygenase
mechanism and that only one oxygen atom is transferred from molecular oxygen during the βcarotene cleavage. Therefore, the name of β-carotene dioxygenase was changed to β-carotene
mono-oxygenase, although in order to lessen the confusion, we will continue to use the name
BCDOX in our disccusion.
2.2.2. Testing the in vivo Activity of BCDOX and Biosynthesis of β-carotene
To test whether the system could be used to screen mutants of CRBPII, bacteria were
transfected with the plasmid pOrange that contains the gene cluster of β-carotene biosynthesis
(Figure 2-11).

63,64

The plasmid was tested in different bacterial strains (XL-1 Blue and JM109).

BL21(DE3)pLysS cells that have a well-developed and control system for protein expression,
however, was not tested since both BL21(DE3)pLysS and pOrange contain the same antibiotic

155

p1

5o

ri

pOrange

E,

B,

Plac

I,Y

JM 109

C

rt

mR

C

XL-1 Blue

JM 109/
pOrange
JM 109

XL-1 Blue/
pOrange

XL-1 Blue

Figure 2-21. in vivo test of pOrange in different bacteria strain.
(chloramphenicol) resistance gene. This gene encodes the chloramphenicol acetyltransferase
enzyme that inactivates chloramphenicol by acetylating it. Since BL21(DE3)pLysS already has
the chloramphenicol resistance gene, the bacteria would not maintain pOrange,, that has the same
antibiotic resistance gene, in the cell because maintaining the extra plasmid in the cell requires
additional energy. The redundancy of antibiotic resistance causes pOrange to be expelled by the
bacteria. Both transfected bacteria (XL-1 Blue and JM109), after growing for 24 hour in dark,
showed orange coloration compared with bacteria without pOrange (Figure 2-21). pOrange
transfected XL-1 Blue cells showed more intense color compared with transfected JM 109
(Figure 2-21). To verify β-carotene is produced inside the cell, the cells were lyzed after 24 hour

156

by sonication and the chromophores were

a)

!-carotene

extracted.

HPLC analysis of the extracts

verified the presence of β-carotene (Figure 222).
The β-carotene synthesizing bacteria
(XL-1 Blue/ pOrange) were then transfected

b)
all-trans-retinal

with the BCDOX containing gene. Dr. RabagoSmith in our lab had demonstrated the use of
87

mouse BCDOX to cleave β-carotene in vivo.

However, BCDOX was cloned under the
Figure 2-22.
HPLC trace for
chromophore a) Extracted from XL-1
Blue transformed with pOrange; b)
Extracted from XL-1 Blue transformed
with pOrange and pBAD-BCDOXCRBPII (KS).

control of a T5 promoter which is a relatively
87

weak.

In the orginal design, the CRABPII

was installed after BCDOX gene (1.8kb).

However, the expression of CRABPII was not observed after cloned it after BCDOX under T5
87

promoter because CRABPII gene was too far away from the weak promoter.

In order to tackle

the latter problem, we have redesigned the plasmid for the screening system. For the new
plasmid, both BCDOX and CRBPII would be under different promoters as to prevent the
sluggish expression for the second protein. In addition, this new plasmid could allow better
control of the protein expression individually. Although BCDOX expression is problematic,
Vogt

et

al.

and

Cunningham,

et

al.

157

were

able

to

express

mammalian

O2
AraC

Pc
O1 O2

PBAD

AraC
I1

araB

I2

araA

araD

L-Arabinose

Pc
O1 O2

AraC AraC
I1

I2

PBAD
araB

araA

araD

Figure 2-23. Transcription mechanism of PBAD promoter.
63,88

BCDOXs in E. coli. with the PBAD promoter.
which induce with L-arabinose (Figure 2-23).

89

pBAD is characterized as a strong promoter

The L-arabinose operon under control of PBAD

comprises of three genes (araB, araA and araD) that encode for enzymes responsible for
converting L-arabinose to xyulose-5-phosphate that is used for energy production.

PBAD

promoter is under the control of the repressor protein AraC which is transcribed from the gene
90

araC.

The regulation of PBAD promoter has been well-studied. araC was transcribed from the

promoter (PC) to produce AraC. AraC is a repressor protein that binds operator sites, O2 and I2.
The binding between AraC and operators leads to a looped conformation of DNA which prevents
transcrip-tion of protein downsteam (Figure 2-23).

91

In the presence of L-arabinose, L-arabinose

binds AraC and leads to conformational changes of AraC. AraC can no longer bind to O2 site,

158

PBAD

5

aC

or

ar

i

pBAD-BCDOX

pOrange

Plac

pBR

322

Y

,I,

,B

Cm

tE

R

Cr

Ap R

p1

With L-Arabinose

ori

Without L-Arabinose

Figure 2-24. Cell pellet of XL-Blue transformed with pOrange + pBAD-BCDOX (Left:
with addition of L-arabinose; Right: without addition of L-arabinose).
the loop is released; therefore, transcription is allowed (Figure 2-23).
BCDOX into the pBAD-Topo

®

92

The plan was to clone

vector that contains the PBAD promoter. However, BCDOX

contains restriction sites that are also present on the multiple cloning sites (MCS) in pBAD®

Topo . Therefore, the enzyme corresponding to MCS cannot be used because the enzymes also
cleave the BCDOX gene. To solve the problem, we have installed Bgl2 restriction site in pBAD®

®

Topo . BCDOX was cloned into pBAD-Topo between Bgl2 and Pme1 restriction site and the
pBAD-BCDOX plasmid was obtained (Figure 2-24). pBAD-BCDOX was then transformed into
XL-1 Blue/pOrange. The resulting XL-1 Blue/pOrange/pBAD-BCDOX was tested to check

159

PT7

Ptac
Q

R

cI
La

BP

R

C

Ap

pCWori(+)

Ap R

pET17b-CRBP

pBR

322

ori

pBR
322

Ptac

IQ

BP
R

C

c
La

ori

pBR

Ap R

pCWori(+)-CRBP

322

ori

Figure 2-25. Cloning strategy for CRBPII.
whether BCDOX would be expressed as a functional protein to cleave β-carotene. The XL-1
Blue/pOrange/ pBAD-BCDOX bacteria were grown and expression was induced with 0.2% of Larabinose. After 24 hours, the bacteria with L-arabinose showed pale yellow colonies (Figure 224) while the bacteria without addition of L-arabinose were orange (Figure 2-24). The result
indicated that BCDOX is expressed as a functional enzyme in vivo.
2.2.3. Cloning of CRBPII into pCWori (+) Vector
With the retinal producing bacteria (XL-1 Blue/pOrange/pBAD-BCDOX) on hand,
CRBPII mutants could be screened. Since XL-1 Blue cells do not contain T7 polymerase, pET17b-CRBPII, in which CRBPII is under the control of T7 promoter, cannot be used directly. In
160

addition, pET17b vector contains the ampicillin resistant gene (Figure 2-25). Therefore, it is
impossible to co-transfect both pET17b and pBAD-TOPO (Figure 2-25) because they contain the
same ampicillin resistant gene. To be able to create a screening system, CRBPII was cloned in
pCWori(+) under the Ptac promoter. Ptac promoter originates from the Plac promoter, although it
is stronger than Plac for protein expression.

89

This is because Ptac contains an extra

promoter that facilitates the binding of RNA polymerase.
repressor protein called LacIq.
89

transcription.

LACUV5

Ptac is also tightly controlled by a

LacIq binds the lac operon of Ptac promoter and blocks

Isopropyl-β-D-1-thiogalactopyranoside (IPTG) binds LacIq leading to
89

conformational change of LacIq and it subsequent neutralization that enables transcription.

2.2.4.

Cloning of Gene Cassette (LacIq-Ptac-CRBP / LacIq-Ptac-CRBPII (KS)

(Q108K:K40S)

/

LacIq-Ptac-CRABPII

/

LacIq-Ptac-CRABPII

(R132K:R111L:L121E)) into pBAD-BCDOX Plasmid and Evaluation

(KLE)

of the Color

Screening Methods
The gene cassette that includes LacIq, Ptac and CRBPII was cloned into pBAD-BCDOX.
It is necessary to include LacIq to ensure enough repressor protein production in vivo for Ptac
promoter regulation. The gene cassette was designed to clone between transcription terminator
gene and the ampicillin resistance gene (Figure 2-26). In order to clone the gene cassette,
restriction sites were introduced. Based on the restriction enzyme analysis for pBAD-BCDOX
and LacIq-Ptac-CRBPII, XhoI and NotI were chosen because there were no corresponding cut
site between the two genes. After the restriction sites were introduced upstream of the ampicillin

161

Ptac

BP

R

ar

D

O

X

pBAD-BCDOX

pCWori(+)-CRBP

Ap R

Ap R

pBR
322

BC

aC

C

Q
I
ac
L

PBAD

pBR

322

ori

ori

PBAD

ara

C

BC

D

O

pBR322 ori

pBAD-BCDOXCRBP

Ap R

P

CRB

LacI Q

X

Ptac

Figure 2-26. Cloning strategy for plasmid containing BCDOX and CRBPII (pBADBCDOX-CRBP).
resistant gene, the bacteria were able to grow. This indicated that the mutations before the
ampicillin resistance gene do not inhibit the transcription of the resistance gene. The gene
cassette (LacIq-Ptac-CRBPII) was cloned into pBAD-BCDOX and pBAD-BCDOX-CRBPII was
obtained (Figure 2-26). This final plasmid was transfected to XL-1 Blue and CRBPII and
BCDOX were selectively expressed with IPTG and L-arabinose, respectively. Based on sodium
dodecyl sulfate polyacrylamide gel electrophoresis gel (SDS-PAGE) analysis, CRBPII was
expressed; however, BCDOX expression cannot be identified from the protein gel because the

162

BCDOX band (about 60KDa) overlaps with
native proteins produced by XL-1 Blue. Based
on

color

analysis

(Figure

2-24),

a

XL-1

1 2 3 4 5

97.4KDa
66.2KDa
45KDa

Blue/pBAD-BCDOX can produce pale yellow or

31KDa

white colonies, therefore, BCDOX should be

21.5KDa
active and expressed in vivo. Leaky expression

14.4KDa

of CRBPII was observed based on SDS-PAGE
analysis (Figure 2-27).

Although LacIq is

6.5 KDa

expressed, it does not completely inhibit CRBPII
expression before IPTG addition. This might be b
a problem because CRBPII is expressed before

the exponential growth phase of E. coli that is the Figure 2-27. a) SDS-PAGE analysis of
protein expression of BCDOX and
optimum growth phase for protein expression. CRBPII in XL-1 Blue (Lane 1: XL-1 Blue;
Lane 2: XL-1 Blue expressed BCDOX
This mighy lead to low expression for CRBPII (with addition of L-arabinose); Lane 3:
XL-1 Blue expressed CRBPII (with
because of the metabolic burden for the addition of IPTG); Lane 4: bio-rad SDSPAGE size marker (Broad-range)(From
Bottom: 6.5 KDa, 14.4 KDa, 21.5 KDa, 31
uncontrolled early expression.
KDa, 45 KDa, 66.2 KDa, 97.4 KDa, 116.3
To fully evaluate the screening method, KDa, 200 KDa); Lane 5: Pure CRABPII;
b) expand image of SDS-PAGE around 15
another three gene (CRBPII-KS, CRABPII and KDa; pointer indicated CRBPII protein.
CRABPII-KLE) were cloned into the pBADBCDOX via gene cassette (LacIq-Ptac-target gene) (Figure 2-26). As mentioned previously,
CRBPII mutants showed a wide range of absorption maxima. However, the pKa of bound retinal
PSB is generally around 7.5, which is a problem since the pH in the cytoplasm of E. coli. is
56

generally more than 8.0.

The high pH in vivo would reduce the level of PSB (the red-shifted
163

component) and consequently lead to a
weak

color

intensity

(Figure

2-28).

bound retinal PSB has a higher pKa value
(about 8.0) so it should have enough
retinal PSB in vivo.

CRABPII and

CRABPII-KLE were also cloned as a
control.

XL-1

Blue/pOrange

was

Absorbance

CRBPII-KS was chosen because the

Acid Titration of Retinal with CRBPII
(Q108K:K40L)
0.5
pH 8.23
Retinal SB
pH 7.15
0.4
Retinal PSB
0.3
0.2
0.1
0

transfected with the four different newly

300

400

500

600

700

Wavelength (nm)

cloned plasmids (pBAD-BCDOX-CRBPII Figure 2-28.
Absorption of CRBPII
(Q108K:K40L) at pH 7.15 and pH 8.23.
or CRBPII (KS) or CRABPII or CRABPII
(KLE)) (Figure 2-26). L-Arabinose and IPTG were added to the transfected bacteria (XL-1
Blue/pOrange/pBADBCDOX-target gene) at
OD600 around 0.8.

The

bacteria were grown for 24
hours. The color of spinned
down colonies of each
transfected bacteria was
pOrange/
pOrange/
pBAD-BCDOX- pBAD-BCDOXcompared. The bacteria,
CRBP
CRBP
(with IPTG)
(with L-arabinose)
Figure 2-29. Cell pellets of XL-1 Blue transfected with grown in the presence of
pOrange and pBAD-BCDOX-CRBPII (Left: without inducer;
only L-arabinose, showed a
Middle: Addition of IPTG; Right: Addition of L-arabinose).

pOrange/
pBAD-BCDOXCRBP

164

very a pale yellow cell pellet (Figure 2-29), which indicated that BCDOX was expressed as a
functional enzyme to cleave β-carotene.

Based on HPLC analysis, all-trans-retinal was

produced in significant amounts compared with cell without BCDOX (Figure 2-22). An orange
cell pellet was observed for bacteria with the addition of IPTG. This showed that BCDOX was
tightly regulated and no over expression is occurred without addition of L-arabinose. Orange cell
pellets were obtained from both CRBPII (KS) (λmax = 505 nm) and CRABPII (KLE) (λmax =

pOrange/
pOrange/
pOrange/
pOrange/
pBAD-BCDOX- pBAD-BCDOX- pBAD-BCDOX- pBAD-BCDOXCRBP II (KS)
CRBP II
CRABP II
CRABP II (KLE)
Figure 2-30. Cell pellet of XL-1 Blue/pOrange with CRPBII (KS), WT-CRBPII, WTCRABPII or CRABPII (KLE) expressed together with BCDOX.
449 nm) containing bacteria (Figure 2-30). However, CRBPII (KS) that absorbed at 505 nm
should show a greenish cell pellet. As mentioned before, the pKa of bound retinal PSB in
CRBPII (KS) was around 8.0; therefore, in vivo, the color may reflect a mixture of retinal PSB
55

and SB absorption (Figure 2-31).

In addition, the orange cell pellet might be due to poor

CRBPII (KS) expression. The unbound retinal might be solvated by the cell membrane and
could not diffuse back to the cytoplasm when there is enough expressed CRBPII (KS). SDSPAGE electrophoresis gel analysis was conducted. The result showed that the CRBPII and
mutants are not expressed well. In addition, a large amount of inclusion bodies were observed in
55

the crude lysate. CRBPII mutants’ expressions are known to be difficult at high temperature.

165

Retinal with CRBPII
(Q108K:K40S) at pH 7.73
0.1

Absorbance

When CRBPII mutants were expressed at
temperature higher than 30 ˚C, significant

Retinal SB

0.08

However, they are expressed well at 24 ˚C.

Retinal PSB

55

amount of inclusion bodies were observed.

0.06

This is a problem for our system because
0.04

BCDOX is known to be a temperature
0.02

sensitive enzyme.

Based on the study

conducted by various groups, 37

0
320

400
480
560
Wavelength (nm)

Figure 2-31. Absorption of all-trans-retinal
with CRBPII mutant (Q108K:K40S) at pH
7.73.

˚C

is

optimum for BCDOX activities, a dramatic
decrease is observed when the temperature is
higher or lower than 37˚C. BCDOX activity

79,87,88

almost completely shuts down at 24 ˚C.
2.4. Conclusion

Based on the HPLC analysis, biosynthesis of β-carotene and all-trans-retinal in vivo have
been demonstrated and XL-1 Blue was identified as a better host for CRBPII screening. In
addition, a plasmid with two different promoters (PBAD and Ptac) for protein expression was
created. Each promoter works separately from each other and two proteins under different
promoter can be expressed individually with the addition of the corresponding inducers (IPTG or
L-arabinose).

Although the screening system suffers from poor protein expression and leaky

expression, the feasibility of using retinal producing bacteria as hosts for CRBPII screening has
been demonstrated. To better control the protein expression of CRBPII, T7 promoter should be

166

employed. As mentioned previously, the reason that we cannot use T7 promoter is because
pOrange and BL21(DE3)pLysS, host for T7 protein expression has the same antibiotic
resistance. To solve this problem, we could switch the chloramphenicol resistance gene with
kanamycin resistance gene so T7 protein expression system could be employed for our color
screening system. Then, we could clone the T7-promoter-CRBPII gene cassette from pET17bCRBPII (Figure 2-25) into pBAD-BCDOX, and thus express CRBPII under the control of T7
promoter. In addition, we could also tried to express our host XL-1 Blue/ pOrange/ pBADBCDOX-CRBPII at 24 ˚C. When enough CRBPII mutants and BCDOX were obtained, we
could elevate the temperature to 30 ˚C for generating all-trans-retinal from BCDOX. Once the
better color screening system are created, we can generate a random library and study the
wavelength regulation in rhodopsins contributed from the long-range interactions. As mentioned
in the introduction, there are several method to generate a library of mutants for directed
evolution. In site-saturation mutagenesis, only residues at selected positions are randomized.
This method has its own advantages over the other random mutagenesis techniques. Since the
high resolution crystal structure of CRBPII with bound retinal has been resolved, we know the
positions of different residues relative to the bound retinal. Therefore, we can selectively mutate
the surface residues rather than randomize the whole protein. This method allows us to filter the
false positives that contain the mutations on residues within 5 Å of the bound chromophore. In
addition, we can also selectively randomize the surface residues at specific region around the
bound all-trans-retinal PSB (PSB, middle and β-ionone ring regions) (Figure 2-14), therefore,
we can study the relative importance of second shell interactions at different region of the bound
retinal PSB.

167

2.5. Materials and Methods
A. Plasmid Manipulation
i. Plasmid Purification
Plasmid purification has been detailed in Chapter 1. The plasmid purification follows
®

protocol in Qiagen Plasmid Purification Maxi Kit . Briefly, a colony from the LB agar plate
with corresponding antibiotic is inoculated in 500 mL LB with correct antibiotic. The cell
culture is grown overnight about 12 hours to 16 hours. Cells are spinned down by centrifugation
(7000 rpm, 15 mins, 4 ˚C). Supernatant is discarded. The resulting cell pellet is subjected to
Qiagen Plasmid Purification Kit using Qiagen anion exchange column. The purified DNA is
dried and is re-dissolved in sterilized H2O. The concentration and the quality of DNA is
analyzed by UV-vis spectroscopy following the equation below:
DNA concentration (ng/µL) = OD260 x 50 ng/µL x dilution facter (volume of DNA added /
total volume in UV curvette)
Purity of DNA = OD260 / OD280 = 1.8 (pure)
>1.8 (RNA contamination)
< 1.8 (Protein contamination)
ii. Sample preparation for DNA sequencing:
To a 500 µL sterilized eppendorf tube, plasmid DNA (2000 ng) and sequencing primer
DNA (30 pmol) are added and sterilized H2O is added to a total volume of 12 µL. The primer
DNA is synthesized by Research Technology Support Facility in Michigan State University and

168

the primer design is based on the protocol in http://rtsf.msu.edu/custom-primers. The priming site for
sequencing primer is 30 to 50 bases upstream from the target sequence. The melting temperature
should be around 55 ˚C to 65 ˚C based on the calculation of the following formula (REF):
Melting temperature (Tm) = 81.5 ˚C + 0.41 x (% GC) – 675 / primer length -% mismatch
Primer for Sequencing:
®

For sequencing PBAD promoter of pBAD-TOPO :
5’- TCT ATA ATC ACG GCA GAA AAG TCC-3‘

BB437
®

For sequencing gene after PBAD promoter of pBAD-TOPO :
5’- ATG CCA TAG CAT TTT TAT CCA TAA GAT TAG-3‘

BB459

For sequencing BCDOX:
5’-GATCTCGATCCCGCGAAATTAATACGAC-3‘

BB044

5’-CCAGACCCTAGAGACCTTGGAGAAGG-3‘

BB045

5’-CGAGGAGAAGTCCAGGCTGACC-3‘

BB046

5’-GATCGATCTCGATCCCGCG-3‘

BB055

5’-GCAGACTGGAATGCAGTGAAGC-3‘

BB052

For sequencing pBAD-before ampicillin resistance gene:
5’-CCTTTCGTTTTATCTGTTGTTTGTC-3‘

BB459

For sequencing gene after Ptac promoter:
5’-GGTTCTGGCAAATATTCTGAAATGAGCTGTTGAC-3‘

169

BB533

Table 2-3. Recipe for Sequencing for Gene in Chapter 2

Reactant
DNA Template (2000ng)
Sequencing Primer (30 pmol)
sterilized H2O
Total volume

Amount (µL)
x
a
12-(x+a)
12

iii. Polymerase Chain Reaction (PCR) Conditions
™

A 50 µl PCR reaction is set up based on the Phusion

DNA polymerase protocol from

FINNZYMES. The PCR reaction follows the standard recipe of Phusion™DNA polymerase:
Table 2-4. PCR Recipe for Phusion
Component
5X Phusion HF Buffer
Template DNA (100 ng/µL)
dNTPs (10 mM)
Forward Primer
Reverse Primer
Phusion DNA polymerase
sterilized H2O
Total Volume

TM

DNA Polymerase

Volume (µL)
10
1
1
n
m
0.5
add to total volume of 50 µL
50 µL

Concentration
1X
2 ng/µL
200 uM
0.5 uM
0.5 uM
0.02 Unit/µL
N/A
N/A

iv. Primer Design for PCR and Mutagenesis:
The primers for both gene amplification and site-directed mutagenesis should have a
length of at least 20 nucleotides. Primers design for gene amplification should have melting
temperature (Tm) above 60 ˚C and below 69 ˚C while for site-directed mutagenesis, the primers
should fall-between 65 ˚C and 72 ˚C. The primers for site-directed mutagenesis should have the
mutated codon in the middle of the primers. To calculate Tm for the primers, nearest-neighbor
method is employed, because the primer Tm calculations can change significantly if different

170

methods are used.

Tm for primers can be calculated from the web-based calculator in

Finnzymes’ website (http://www.finnzymes.com/tm_determination.html).
Table 2-5. Primers for gene amplification with restriction site (underline) installed:
Template
DNA

Primer Sequence
Forward Primer
Install-Not1
pBAD
Reverse Primer
Install-Not1
pBAD
Forward Primer
Install-Xho1
pBAD
Reverse Primer
Install-Xho1
pBAD
Forward Primer
Install-Bgl2
pBAD
Reverse Primer
Install-Bgl2
pBAD

for
to5’-GCGAAGCAGCGGCCGCGAGGGTGG
for
to5’-CCACCCTCGCGGCCGCTGCTTCGC
for
to5’-CGAGGGTCTCGAGCAGGACGC
for
to5’-GCGTCCTGCTCGAGACCCTCG

pBADTOPO®
pBADTOPO®
pBADTOPO®
pBADTOPO®

for
5’-CTTTAAGAAGGAGATATACAGATCT
to
ATGGGCTCTGGATCCG

pBADTOPO®

for
5’-CGGATCCAGAGCCCATAGATCT
to
GTATATCTCCTTCTTAAAG

pBADTOPO®

Table2-6. Primers for site-directed mutagenesis with mutated codon underline:
Primer Sequence
Forward Primer for
BCDOX (Bgl2)
Reverse Primer for
BCDOX (Pme1)
Forward Primer for
CRBPII (EcoR1)
Reverse Primer for
CRBPII (Hind3)
Forward Primer for
CRABPII (EcoR1)
Reverse Primer for
CRABPII (Hind3)
Reverse Primer for
CRABP-KLE (Hind3)

5’-CTAAGATCTATGGAGATAATATTTGG
CCAG
5’-GAGTTTAAACTCATGGTGCTGTCGGA
TCTGT GG
5’-CTAGAATTCATGGCGAGGGACCAGA
ATGG AACC
5’-CTAAAGCTTTCACTTCTTTTTGAAC
AC TTGACG
5’-CTAGAATTCATGCCAAACTTCTCTGG
CAAC
5’-CTAAAGCTTTCACTCTCGGACGTAG
AC CCTGGT
5’-CTAAAGCTTTCACTCTCGGACGTAG
AC CTTGGT

171

Template
DNA
pET29bBCDOX
pET29bBCDOX
pET17bCRBPII
pET17bCRBPII
pET17bCRABPII
pET17bCRABPII
pET17bCRABPII

Forward Primer for
pCWori(+) (Not1)
Reverse Primer for
pCWori(+) (Xho1)

5’-GAAGCGGCCGCTCACTGCCCGC

pCWori(+)

5’-GATCTCGAGTTTGTAGAAACGCAAA
AAGGC

pCWori(+)

Table 2-7. PCR Conditions for Site-directed Mutagenesis
Number of Step

Cycle(s)

1

1X

2

30X

3
4

1X
1X

Temp. ( ˚C)
98
98
Tm + 3
72
72
4

Time (s)
30
10
30s
30 s/1kb
10 min
hold

v. DNA Purification using Agarose Gel:
Genes amplified from PCR were purified by QIAquick Gel Extraction Kit. PCR products
were subjected to electrophoresis using 1% agarose gel (0.3 g agarose/30 mL sterilized H2O/8
µL Ethidium bromide (10 mg/mL)). PCR products is mixed with endostop in a 4 to 1 ratio.
Both PCR products and standard size marker were loaded on to agarose gel. The agarose gel
was ran at 5V/cm. The genes were detected by long wavelength UV. The products with the
correct size were cut. The weight of gel slices with amplified gene were measured. Three
volume of buffer QG (gel solubilizing buffer) is added to 1 volume of gel (300 µL of buffer QG
for 100 mg of gel). The mixtures were incubated at 50 ˚C for 10 minutes with gentle mixing
until the gel had dissolved. One volume of isopropanol is added into the mixture and the
resulting mixture was applied to QIAquick column followed by centrifugation for 1 minute at
13000 rpm. The QIAquick column is wash with Buffer PE and spinned down for 1 minute at

172

13000 rpm. The flow through is discarded and an additional 1 minute centrifugation is applied at
13000 rpm. The DNA was eluted from the column using 50 µL sterilized H2O.

B. Cloning
i) Cloning (CRABPII, CRABPII mutants (R132K:R111L:L121E), CRBPII and CRBPII
mutant (Q108K:K40S)) into pCWori(+)
1) Gene Digestion:
Sequential gene digestion was needed for inserting PCR product (CRABPII, CRABPII
mutants (R132K:R111L:L121E), CRBPII and CRBPII mutant (Q108K:K40S)) into pCWori(+):
Table 2-8. PCR Recipe for Cloning into pCWori(+)
CRBPII/
CRABPII
Volume (µL)

pCWori(+)
Volume (µL)

CRBPII /
CRABPII

48

0

pCWori(+) (800 ng)

0

8

EcoRI

6

6

EcorRI Buffer (X10)

6

3

sterilized H2O

0

13

The mixtures were incubated at 37 ˚C for 3 hours.

The digestion was stopped by

incubating the mixtures at 70 ˚C for 15 minutes. The gene products were purified using
QIAquick gel purification kit, ready for the second digestion.

173

Table 2-9. Gene Digestion for CRABPII/CRBPII and pCWori
CRBPII/
CRABPII
Volume (µL)

pCWori(+)
Volume (µL)

CRBPII /
CRABPII

48

0

pCWori(+) (800 ng)

0

48

HindIII

6

6

NEB Buffer 2 (X10)

6

6

The mixtures were incubated at 37 ˚C for 3 hours.

The digestion was stopped by

incubating the mixtures at 70 ˚C for 15 minutes.
2) Calf Alkaline Phosphatase (CIP)
The vector pCWori(+) is dephosphorylated using Calf Alkaline Phosphatase to prevent religation of vector.

Table 2-10. CIP Recipe for pCWori(+)
pCWori(+)
Volume (µL)
pCWori(+) (800ng)

60

CIP

2

The reaction was incubated at 37 ˚C for 2 hours. The gene products were purified using
QIAquick gel purification kit. The concentration of isolated digestion products were measured

174

by comparing the intensity of the gene band against different concentration standard on agarose
gel.
3) Gene Ligation:
Ligation of the purified insert (CRABPII, CRABPII mutants (R132K:R111L:L121E),
CRBPII and CRBPII mutant (Q108K:K40S)) and vector pCWori(+) were performed via
standard protocol with T4 ligase. The amount of insert required is calculated based on the
following equation:
Amount of insert (µL) = [100 ng * ratio (Insert / vector) * mass ratio (insert size
(kb) / vector size (kb))] ÷ insert concentration (ng/µL)
Table 2-11. Ligation Recipe for Cloning into pCWori(+)
Different Ligations (Insert : vector / 10 : 1)
CRABPII
(R132K:R111L:
L121E)
Volume (µL)

CRBPII
CRBPII
(Q108K:K40S)
Volume (µL)
Volume (µL)

CRABPII
Volume (µL)

1.25 (100 ng) 1.25 (100 ng)

1.25 (100 ng) 1.25 (100 ng)

CRBPII (5 ng/µL)

25

0

0

0

CRBPII
(Q108K:K40S)
(5 ng/µL)

0

25

0

0

CRABPII
(5 ng/µL)

0

0

25

0

CRABPII
(R132K:R111L:L1
0
21E)
(5 ng/µL)

0

0

25

Ligase Buffer
(X 10)

2.9

2.9

2.9

pCWori(+)
ng/µL)

(80

2.9

175

The mixtures were incubated at 16 ˚C for 16 hours. The reaction were quenched at 70 ˚C
for 25 minutes.

The ligased products (pCWori(+)-CRBP or pCWori(+)-CRABPII) were

transformed into XL-1 Blue competent cell according to the procedure described in Chapter 1.
Five colonies were selected from the LB plate containing tetracycline and ampicillin and the
plasmid from each colonies was purified via the QIAGEN Mini Plasmid Purification Kit
protocol. The purified plasmids were subjected to the DNA sequencing and the plasmid with the
correct insert sequence was used.
®

ii. Cloning of β-carotene monooxygenases (BCDOX) into pBAD-TOPO
1) Gene Digestion:

Sequential gene digestion is needed for inserting PCR product (BCDOX) into pBAD®

TOPO :
®

Table 2-12. Gene Digestion Recipe for Cloning into pBAD-TOPO 1
BCDOX
Volume (µL)
48

BCDOX
pBAD-TOPO
ng/µL)

®

pBAD-TOPO
Volume (µL)
0

(810 0

8

Bgl2

6

6

NEB Buffer 3 (X10)

6

3

sterilized H2O

0

13

176

®

The mixtures were incubated at 37 ˚C for 8 hours.

The digestion was stopped by

incubating the mixtures at 70 ˚C for 15 minutes. The gene products were purified using
QIAquick gel purification kit, ready for second digestion:
®

Table 2-13. Gene Digestion Recipe for Cloning into pBAD-TOPO 2
BCDOX
Volume (µL)

pBAD-TOPO
Volume (µL)

48

0

0

48

Pme1

6

6

NEB Buffer 4 (X10)

6

®

6

BCDOX
pBAD-TOPO

®

The mixtures were incubated at 37 ˚C for 12 hours. The digestion was stopped by
incubating the mixtures at 70 ˚C for 15 minutes.
4) Calf Alkaline Phosphatase (CIP)
The vector pBAD-TOPO

®

was dephosphorylated using Calf Alkaline Phosphatase to

prevent re-ligation of vector.
®

Table 2-14. CIP Recipe for Cloning into pBAD-TOPO

pBAD-TOPO
Volume (µL)
pBAD-TOPO
CIP

®

60
2

177

®

The reaction were incubated at 37 ˚C for 2 hours. The gene products were purified using
QIAquick gel purification kit.

The concentration of isolated digestion products were

measured by comparing the intensity of the gene band against different concentration standard
on agarose gel.
b) Gene Ligation:
Ligation of the purified insert (CRABPII, CRABPII mutants (R132K:R111L:L121E),
®

CRBPII and CRBPII mutant (Q108K:K40S)) and vector pBAD-TOPO were performed via the
standard protocol with T4 ligase. The amount of insert required was calculated based on the
following equation:
Amount of insert (µL) = [100 ng * ratio (Insert/vector) * mass ratio (insert size
(kb)/vector size (kb))] ÷ insert concentration (ng/µL)
®

Table 2-15. Gene Ligation Recipe for Cloning into pBAD-TOPO

Different Ligations (Insert :
vector / 10 : 1)
Volume (µL)
pBAD-TOPO
ng/µL)

®

(100 1 (100 ng)

BCDOX (40 ng/µL)

10.7

Ligase Buffer
(X 10)

1.3

The mixtures were incubated at 16 ˚C for 16 hours. The reaction were quenched at 70 ˚C
for 25 minutes.

The ligased products (pBAD-BCDOX) were transformed into XL-1 Blue

competent cell according to the procedure described in Chapter 1. Five colonies were selected

178

on the LB plate containing tetracycline and ampicillin and the plasmid from each colonies were
purified based on the QIAGEN Mini Plasmid Purification Kit protocol. The purified plasmids
were subjected to the DNA sequencing and the plasmid with the correct insert sequence were
used.

iii.

Cloning of LacIq-Ptac-CRBPII, LacIq-Ptac-CRBPII (Q108K:K40S), LacIq-Ptac-

CRABPII and LacIq-Ptac-CRABPII (R132K:R111L:L121E) into pBAD-BCDOX
1) Gene Digestion:
Double gene digestion was needed for inserting PCR product (LacIq-Ptac-CRBPII, LacIqPtac-CRBPII

(Q108K:K40S),

LacIq-Ptac-CRABPII

and

LacIq-Ptac-CRABPII

(R132K:R111L:L121E)) into pBAD-BCDOX:
Table 2-16. Digestion Recipe for Cloning into pBAD-BCDOX

Insert:
Vector:
LacIq-Ptac-(CRBPII / pBAD-BCDOX
CRABPII)
Volume (µL)
Volume (µL)
Insert:
LacIq-Ptac- 48
(CRBPII / CRABPII)

0

Vector:
BCDOX
ng/µL)

pBAD(1200 0

8

Not1

6

6

Xho1

6

6

NEB Buffer 3 (X10)

7

3

sterilized H2O

3

7

179

The mixtures were incubated at 37 ˚C for 3 hours. The double digestion was stopped by
incubating the mixtures at 70 ˚C for 15 minutes. The gene products of the inserts were purified
using QIAquick gel purification kit.
2) Calf Alkaline Phosphatase (CIP)
The digested vector pBAD-BCDOX is dephosphorylated using Calf Alkaline Phosphatase
to prevent re-ligation of vector.
Table 2-17. CIP Recipe for Cloning into pBAD-BCDOX

pBAD-BCDOX
Volume (µL)
pBAD-BCDOX

30

CIP

2

The reaction was incubated at 37 ˚C for 2 hours. The gene products were purified using
QIAquick gel purification kit. The concentration of isolated digestion products were measured
by comparing the intensity of the gene band against different concentration standard on agarose
gel.
3) Gene Ligation:
Ligation of the purified digested inserts (LacIq-Ptac-CRBPII, LacIq-Ptac-CRBPII
(Q108K:K40S), LacIq-Ptac-CRABPII and LacIq-Ptac-CRABPII (R132K:R111L:L121E)) and
vector pBAD-BCDOX were performed the standard protocol on T4 ligase. The amount of
inserts required was calculated based on the following equation:
Amount of insert (µL) = [100 ng * ratio (Insert/vector) * mass ratio (insert size
(kb)/vector size (kb))] ÷ insert concentration (ng/µL)
180

Table 2-18. Gene Ligation Recipe for Cloning into pBAD-BCDOX

Different Ligations (Insert : vector / 10 : 1)
LacIq-PtacCRBPII
Volume
(µL)

LacIq-PtacCRBPII
(Q108K:K40S)
Volume (µL)

LacIq-PtacLacIq-PtacCRABPII
CRABPII
(R132K:R111L:
Volume (µL) L121E)
Volume (µL)

1 (100 ng)

1 (100 ng)

1 (100 ng)

1 (100 ng)

LacIq-Ptac-CRBPII
(50 ng/µL)

5.5

0

0

0

LacIq-Ptac-CRBPII
(Q108K:K40S)
(40 ng/µL)

0

6.9

0

0

LacIq-PtacCRABPII
(50 ng/µL)

0

0

5.5

0

LacIq-PtacCRABPII
(R132K:R111L:L1
21E)
(60 ng/µL)

0

0

0

4.7

Ligase Buffer
(X 10)

1

1

1

1

sterilized H2O

2.5

1.1

2.5

3.3

pCWori(+)
ng/µL)

(100

The mixtures were incubated at 16 ˚C for 16 hours. The reaction was quenched at 70 ˚C
for 25 minutes.

The ligased products pBAD-BCDOX-CRBPII, pBAD-BCDOX-CRBPII

(Q108K:K40S), or pBAD-BCDOX-CRABPII and pBAD-BCDOX-CRABPII were transformed
into XL-1 Blue competent cell according to the procedure described in Chapter 1. Five colonies
181

were selected on the LB plate containing tetracycline and ampicillin and the plasmid from each
colony was purified with the QIAGEN Mini Plasmid Purification Kit protocol. The purified
plasmids were subjected to the DNA sequencing and the plasmid with the correct insert sequence
were used.
C. Protein Expression Analysis
The cells with target plasmids were inoculated in LB (5 mL) with the corresponding
antibiotic based on the cells phenotype and antibiotic resistance gene on the plasmids. The cell
cultures were grown at 37 ˚C for 12 to 16 hours under dim red light. Cell cultures (500 μL) was
inoculated into LB (10 mL) with the corresponding antibiotic. The fresh cell cultures were
growed at 37 ˚C under dim red light until the OD600 was about 0.8 to 1.0. Either or both
inducers (100 µL of 20% L-arabinose for BCDOX expression and 50 µL of 2mM IPTG) were
added and the cell cultures were further grown for an additional 20 hours at 20 ˚C. The cells
were collected by centrifugation in 15 mL sterilized centrifugation tubes (4000 rpm, 10 minutes,
4 ˚C, under dim red light). Supernatants were discarded and the cell pellets were resuspended in
Tris-buffer (5mL, 25 mM, pH 8.0). The cells were then lyzed by sonication (Power 60%, 1 min
x 3) at 4 ˚C. The cell lysate were centrifuged (4000rpm, 10 minutes, 4 ˚C, under dim red light).
The supernatant were analyzed by SDS-PAGE.
D. HPLC analysis of in vivo Biosynthesis of β-carotene and all-trans-retinal
The chromophore of cell lysate from the above Section v were extracted with hexane (5
mL), three times. Briefly, hexane (5 mL) is added into cell lysate and mixed vigorously. The

182

mixtures were centrifuged (5000 rpm, 20 minutes, 4 ˚C). The top layer (organic portion) was
removed and collected. The combined organic portion was blown dried using nitrogen. The
extracts were resolubilized by hexane (10 μL) and the resulting solution was subjected to HPLC
analysis (1 mL/minute, 1% EtOAc in hexane, 2 hours).
E. Color Screening of CRBPII, CRABPII and both mutants
The cells with target plasmids were inoculated in 5mL of LB with the corresponding
antibiotic based on the cells phenotype and antibiotic resistance gene on the plasmids. The cell
cultures were grown at 37 ˚C for 12 to 16 hours under dim red light. The cell cultures (500 μL)
were inoculated into 10mL of LB with the corresponding antibiotic. The fresh cell cultures were
growed at 37 ˚C under dim red light until the OD600 was about 0.8 to 1.0. Either or both
inducers (100 µL of 20% L-arabinose for BCDOX expression and 50 µL of 2mM IPTG) were
added and the cell cultures were further grown for an additional 20 hours at 20 ˚C. The cells
were collected by centrifugation in 15mL sterilized centrifugation tubes (4000rpm, 10 minutes, 4

˚C, under dim red light).

The cell pellets were analyzed based on the color appearance of the

cell pellet.

183

BIBLIOGRAPHY

184

Bibliography
(1)

Morley, K. L.; Kazlauskas, R. J. Trends in Biotechnology 2005, 23, 231.

(2)
Howes, B. D.; RodriguezLopez, J. N.; Smith, A. T.; Smulevich, G. Biochemistry
1997, 36, 1532.
(3)
Raag, R.; Martinis, S. A.; Sligar, S. G.; Poulos, T. L. Biochemistry 1991, 30,
11420.
(4)
Rod, T. H.; Radkiewicz, J. L.; Brooks, C. L. Proceedings of the National Academy
of Sciences of the United States of America 2003, 100, 6980.
(5)

Jackson, S. E.; Fersht, A. R. Biochemistry 1993, 32, 13909.

(6)Wada, M.; Hsu, C. C.; Franke, D.; Mitchell, M.; Heine, A.; Wilson, I.; Wong, C. H.
Bioorganic & Medicinal Chemistry 2003, 11, 2091.
(7)
de Kreij, A.; van den Burg, B.; Venema, G.; Vriend, G.; Eijsink, V. G. H.; Nielsen,
J. E. Journal of Biological Chemistry 2002, 277, 277 (18) 15432.
(8)

Noivirt-Brik, O.; Unger, R.; Horovitz, A. Bmc Structural Biology 2009, 9.

(9)

Chin, J. K.; Klinman, J. P. Biochemistry 2000, 39, 1278.

(10) Fong, S.; Machajewski, T. D.; Mak, C. C.; Wong, C. H. Chemistry & Biology
2000, 7, 873.
(11) Iffland, A.; Tafelmeyer, P.; Saudan, C.; Johnsson, K. Biochemistry 2000, 39,
10790.
(12) Shimono, K.; Ikeura, Y.; Sudo, Y.; Iwamoto, M.; Kamo, N. Biochim. Biophys.
Acta-Biomembr. 2001, 1515, 92.
(13)
141.

Shimono, K.; Iwamoto, M.; Sumi, M.; Kamo, N. Photochem. Photobiol. 2000, 72,

(14)

Kloppmann, E.; Becker, T.; Ullmann, G. M. Proteins 2005, 61, 953.

(15) Hoffmann, M.; Wanko, M.; Strodel, P.; Konig, P. H.; Frauenheim, T.; Schulten, K.;
Thiel, W.; Tajkhorshid, E.; Elstner, M. Journal of the American Chemical Society 2006,
128, 10808.
(16) Luecke, H.; Schobert, B.; Lanyi, J. K.; Spudich, E. N.; Spudich, J. L. Science
2001, 293, 1499.

185

(17) Ren, L.; Martin, C. H.; Wise, K. J.; Gillespie, N. B.; Luecke, H.; Lanyi, J. K.;
Spudich, J. L.; Birge, R. R. Biochemistry 2001, 40, 13906.
(18) Shimono, K.; Hayashi, T.; Ikeura, Y.; Sudo, Y.; Iwamoto, M.; Kamo, N. Journal of
Biological Chemistry 2003, 278, 23882.
(19) Yoshitsugu, M.; Shibata, M.; Ikeda, D.; Furutani, Y.; Kandori, H. Angewandte
Chemie-International Edition 2008, 47, 3923.
(20) Mills, D. R.; Peterson, R. L.; Spiegelm.S Proceedings of the National Academy of
Sciences of the United States of America 1967, 58, 217.
(21)

Arnold, F. H. Accounts of Chemical Research 1998, 31, 125.

(22)

Arnold, F. H.; Volkov, A. A. Current Opinion in Chemical Biology 1999, 3, 54.

(23)

Kuchner, O.; Arnold, F. H. Trends in Biotechnology 1997, 15, 523.

(24) Marti, S.; Andres, J.; Moliner, V.; Silla, E.; Tunon, I.; Bertran, J. Chemical Society
Reviews 2008, 37, 2634.
(25) Funke, S. A.; Eipper, A.; Reetz, M. T.; Otte, N.; Thiel, W.; Van Pouderoyen, G.;
Dijkstra, B. W.; Jaeger, K. E.; Eggert, T. Biocatalysis and Biotransformation 2003, 21,
67.
(26)

Koeller, K. M.; Wong, C. H. Nature 2001, 409, 232.

(27) Schmid, A.; Dordick, J. S.; Hauer, B.; Kiener, A.; Wubbolts, M.; Witholt, B. Nature
2001, 409, 258.
(28) Farinas, E. T.; Bulter, T.; Arnold, F. H. Current Opinion in Biotechnology 2001,
12, 545.
(29)

Koeller, K. M.; Wong, C. H. Nature Biotechnology 2000, 18, 835.

(30) Shaner, N. C.; Campbell, R. E.; Steinbach, P. A.; Giepmans, B. N. G.; Palmer, A.
E.; Tsien, R. Y. Nature Biotechnology 2004, 22, 1567.
(31)

Jaeger, K. E.; Eggert, T. Current Opinion in Biotechnology 2002, 13, 390.

(32) Park, J. B.; Buehler, B.; Habicher, T.; Hauer, B.; Panke, S.; Witholt, B.; Schmid,
A. Biotechnology and Bioengineering 2006, 95, 501.
(33)
964.

Brannigan, J. A.; Wilkinson, A. J. Nature Reviews Molecular Cell Biology 2002, 3,

(34)

Chapus, C.; Rovery, M.; Sarda, L.; Verger, R. Biochimie 1988, 70, 1223.

186

(35)

Bartlett, J. M. S.; Stirling, D. Methods Mol Biol 2003, 226, 3.

(36) Pritchard, L.; Corne, D.; Kell, D.; Rowland, J.; Winson, M. Journal of Theoretical
Biology 2005, 234, 497.
(37)

Leung, D., Chen, E., Goeddel, D. Technique 1989, 1, 11.

(38)

Stemmer, W. P. C. Nature 1994, 370, 389.

(39) Stemmer, W. P. C. Proceedings of the National Academy of Sciences of the
United States of America 1994, 91, 10747.
(40)
288.

Crameri, A.; Raillard, S. A.; Bermudez, E.; Stemmer, W. P. C. Nature 1998, 391,

(41) Ness, J. E.; Welch, M.; Giver, L.; Bueno, M.; Cherry, J. R.; Borchert, T. V.;
Stemmer, W. P. C.; Minshull, J. Nature Biotechnology 1999, 17, 893.
(42) Tarter, M.; Capaldi, S.; Carrizo, M. E.; Ambrosi, E.; Perduca, M.; Monaco, H. L.
Proteins 2008, 70, 1626.
(43) Banaszak, L.; Winter, N.; Xu, Z. H.; Bernlohr, D. A.; Cowan, S.; Jones, T. A.
Advances in Protein Chemistry, Vol 45 1994, 45, 89.
(44)

Wang, L. C.; Li, Y.; Yan, H. G. J. Biol. Chem. 1997, 272, 1541.

(45)

Li, E.; Norris, A. W. Annual Review of Nutrition 1996, 16, 205.

(46) Newcomer, M. E.; Jamison, R. S.; Ong, D. E. Subcellular Biochemistry, Vol 30
1998, 30, 53.
(47) Yost, R. W.; Harrison, E. H.; Ross, A. C. Journal of Biological Chemistry 1988,
263, 18693.
(48)

Li, E. Molecular and Cellular Biochemistry 1999, 192, 105.

(49)

Wong, W.; Michigan State University: East Lansing, 2008.

(50)

Holst, M.; Baker, N.; Wang, F. J. Comput. Chem. 2000, 21, 1319.

(51)

Baker, N.; Holst, M.; Wang, F. J. Comput. Chem. 2000, 21, 1343.

(52) Baker, N. A.; Sept, D.; Joseph, S.; Holst, M. J.; McCammon, J. A. Proc. Natl.
Acad. Sci. U. S. A. 2001, 98, 10037.
(53)

Holst, M.; Saied, F. J. Comput. Chem. 1993, 14, 105.

(54)

Holst, M. J.; Saied, F. J. Comput. Chem. 1995, 16, 337.
187

(55)

Wang, W.; Michigan State University: East Lansing, 2008.

(56) Kim, S. Y.; Waschuk, S. A.; Brown, L. S.; Jung, K. H. Biochim. Biophys. ActaBioenerg. 2008, 1777, 504.
(57) Jung, J. Y.; Choi, A. R.; Lee, Y. K.; Lee, H. K.; Jung, K. H. Febs Letters 2008,
582, 1679.
(58) Beja, O.; Spudich, E. N.; Spudich, J. L.; Leclerc, M.; DeLong, E. F. Nature 2001,
411, 786.
(59) Beja, O.; Aravind, L.; Koonin, E. V.; Suzuki, M. T.; Hadd, A.; Nguyen, L. P.;
Jovanovich, S.; Gates, C. M.; Feldman, R. A.; Spudich, J. L.; Spudich, E. N.; DeLong, E.
F. Science 2000, 289, 1902.
(60) Miranda, M. R. M.; Choi, A. R.; Shi, L. C.; Bezerra, A. G.; Jung, K. H.; Brown, L.
S. Biophysical Journal 2009, 96, 1471.
(61) Dioumaev, A. K.; Brown, L. S.; Shih, J.; Spudich, E. N.; Spudich, J. L.; Lanyi, J.
K. Biochemistry 2002, 41, 5348.
(62) Varo, G.; Brown, L. S.; Lakatos, M.; Lanyi, J. K. Biophysical Journal 2003, 84,
1202.
(63)

von Lintig, J.; Vogt, K. Journal of Biological Chemistry 2000, 275, 11915.

(64) Kim, Y. S.; Kim, N. H.; Kim, H. J.; Lee, J. K.; Kim, S. W.; Oh, D. K. Applied
Microbiology and Biotechnology 2007, 76, 1339.
(65) Goodwin, T. W. The biochemistry of the carotenoids; Chapman and Hall: London,
1980.
(66)

Koyama, Y. Journal of Photochemistry and Photobiology B-Biology 1991, 9, 265.

(67) Mathews-Roth, M. M.; Pathak, M. A.; Fitzpatrick, T. B.; Harber, L. C.; Kass, E. H.
New England Journal of Medicine 1970, 282, 1231.
(68)

Paerl, H. W. Oecologia 1984, 61, 143.

(69)

Mathewsroth, M. M. Federation Proceedings 1987, 46, 1890.

(70) Yoon, S. H.; Kim, J. E.; Lee, S. H.; Park, H. M.; Choi, M. S.; Kim, J. Y.; Shin, Y.
C.; Keasling, J. D.; Kim, S. W. Applied Microbiology and Biotechnology 2007, 74, 131.
(71) Nishizaki, T.; Tsuge, K.; Itaya, M.; Doi, N.; Yanagawa, H. Applied and
Environmental Microbiology 2007, 73, 1355.

188

(72)
750.

Schmidt-Dannert, C.; Umeno, D.; Arnold, F. H. Nature Biotechnology 2000, 18,

(73)

Wang, C.; Bammler, T. K.; Eaton, D. L. Toxicological Sciences 2002, 70, 20.

(74)

Moore, T. Biochemical Journal 1930, 24, 692.

(75) Olson, J. A.; Hayaishi, O. Proceedings of the National Academy of Sciences of
the United States of America 1965, 54, 1364.
(76)

Fidge, N. H.; Smith, F. R.; Goodman, D. S. Biochemical Journal 1969, 114, 689.

(77)

Goodman, D. S.; Huang, H. S. Science 1965, 149, 879.

(78)

Harashima, K. Biochimica Et Biophysica Acta 1964, 90, 211.

(79)

Sklan, D. British Journal of Nutrition 1983, 50, 417.

(80) Britton, G., Liaaen-Jensen, S., Pfander, H. Carotenoids; Birhåuser: Basel, 1998;
Vol. 3.
(81)

Britton, G. Faseb Journal 1995, 9, 1551.

(82) Olson, J. A. Moern Nutrition in Health andDisease; Williams and Wilkins:
Baltimore, 1999.
(83) Nagao, A.; During, A.; Hoshino, C.; Terao, J.; Olson, J. A. Archives of
Biochemistry and Biophysics 1996, 328, 57.
(84)

Olson, J. A. American Journal of Clinical Nutrition 1969, 22, 953.

(85)

Lange, S. J.; Que, L. Current Opinion in Chemical Biology 1998, 2, 159.

(86) Leuenberger, M. G.; Engeloch-Jarret, C.; Woggon, W. D. Angewandte ChemieInternational Edition 2001, 40, 2614.
(87)

Rabago-Smith, M., Michigan State University, 2006.

(88) Redmond, T. M.; Gentleman, S.; Duncan, T.; Yu, S.; Wiggert, B.; Gantt, E.;
Cunningham, F. X. J. Biol. Chem. 2001, 276, 6560.
(89) Cantrell, S. Vectors for the Expression of Recombinant Proteins in E. coli;
Humana Press: Totowa, New Jersey, 2003; Vol. 235.
(90) Wilcox, G.; Boulter, J.; Lee, N. Proceedings of the National Academy of Sciences
of the United States of America 1974, 71, 3635.
(91)

; Corp., I., Ed.; Invitrogen Corp.: Carlsbad, CA; Vol. Version G.
189

(92)

Casadaban, M. J. J. Mol. Biol. 1976, 104, 557.

(93) Dolinsky, T. J.; Nielsen, J. E.; McCammon, J. A.; Baker, N. A. Nucleic Acids
Research 2004, 32, W665.

Chapter Three
Engineering CRABPII and CRBPII into a Retinal isomerase
3.1. Introduction
3.1.1. Isomerization Triggers Biological Processes

Steric
R1

Isomerization

R1

R2

R2
Far Away
Figure 3-1. Isomerization changes the geometrical relationship with surrounding
area.
Isomerization of double bonds requires a large geometrical rearrangement (Figure 3-1) that
results in change of the spatial relationship around the double bond. Nature has taken advantage
1-8

of this process to trigger a variety of biological processes.

The isomerization in vivo is always

initiated by an external stimulus, such as light. In addition, the isomerization of a molecule
allows for information to be transformed and transferred across an interface triggering
downstream processes. These biological processes are important for organisms to survive and to
adopt to their environments. One example is the cis-trans isomerization of fatty acids. Fatty
acids are key components of cell membranes. Cell membranes in prokaryotes are mainly made
of phospholipids while in eukaryotes, they comprise of a variety of different components that
include phospholipids, transmembrane proteins, glycolipids, polysaccharides, surface peptides
190

and glycopeptides (Figure 3-2). Since a large proportion of cell membranes is phospholipids, the
geometry of phospholipids displayed on the cell membrane affects their properties (fluidity and
permeability). Generally, fatty acids in cell membrane are a mixture of saturated and unsaturated

Glycoprotein
Surface Protein

Transmembrane
Protein

Phospholipids

Figure 3-2. Schematic diagram of cell membrance.
hydrocarbons. In nature, unsaturated fatty acids adopt a cis geometry that is tightly controlled by
9-11

the regiosepecific and stereoselective desaturase enzymes.

Despite several decades of

research, it was not until the 1990s that
the first trans-configured unsaturated
fatty acids were reported in some
12,13

prokaryotes.

These

trans

Isomerization

unsaturated fatty acids were later found
to be presented only under certain
environmental stress, such as: heat,

Better Packing

presence of toxic chemicals, oxidative
conditions,

8,14-18

etc.

Figure 3-3. Isomerization leads to a better pack cell
The membrane.

191

isomerization from cis fatty acids to trans fatty acids was suggested as an adaptive mechanism
that allows bacteria to survive in dangerous environments based on the findings from
13,17,18

Pseudomonas putida P8 and Vibrio sp. studies.

It was found that both Pseudomonas

putida P8 and Vibrio sp. contain cis-trans isomerase (Cti) for unsaturated fatty acids, which is
induced when cells are exposed to elevated temperatures or high concentration of toxic
8,14,16,19,20

phenol.

The isomerization is beneficial to bacteria to survive because trans

unsaturated fatty acids can be packed better than cis isomers (Figure 3-3).

21-23

This

transformation decreases the permeability for toxic compounds into the cell and also decreases
the fluidity of the cell membrane, preventing it from dissociating because of the elevated
temperature. The cis-trans isomerization of unsaturated fatty acids also occurs in eukaryotes.
Under oxidative stress, for example, in the presence of thiyl radicals, the trans unsaturated fatty
2,11,24-27

acids are formed.

However, eukaryotes do not contain Cti. It is proposed that the

isomerization of fatty acids inside the eukaryotic cell membranes goes through a radical
2,11,25,26

mechanism.

Isomerization leads to a better packing of the cell membrane and protects

the cell from oxidative stress.

192

3.1.2. Biological Processes Triggered by
Light Induced Isomerization

O

3.1.2.1. Phytochromes
Besides

chemically

induced

N
H

or

3-1

enzymatic isomerization, light is also a
major inducer for cis-trans isomerization of
double bonds.

11 13

Absorption of photon by

14

12

light sensing proteins leads to isomerization
of the chromophore.

N

NH

O

N
H

N
H

3-2

The isomerization

results in protein conformational changes

O

O

that trigger biological processes. There are

S

3-3
three well-studied photoreceptor protein

Figure 3-4. Chromophores in different light
sensing proteins. 3-1. phytochromobilin in
families in Nature that can initiate a lightphytochromes where R1 is the repeating unit of
Retinal in
induced signal transduction through the shown dipyrrole; 3-2.
rhodopsins (isomerization occurs at C11-C12 in
isomerization of a double bond. They are vertebrates and invertebrates rhodopsin and at
C13-C14 in microbial rhodopsin; 3-3.
coumaric acid in xanthopsins.
rhodopsins, xanthopsins and
phytochromes.

These

phytochromobilin

three

ATP

ADP

families are characterized based

PAS

GAF

P4

HKD

on their protein sequence and the
chromophore
proteins
Rhodopsins
xanthopsins

bound
(Figure
use
employ

to

each

Pi

3-4).

Rcp1

P
Rcp1

retinal, Figure 3-5. Schematic diagram of the domain included
in bacterial phytochrome Cph1 and its response
coumaric regulator Rcp1. GAF: Chromophore binding domain;
HKD: histidine kinase domain.
193

acid while phytochromes incorporate phytochromobilin (Figure 3-4). Phytochromes were first
discovered as plant photoreceptors that can respond to red or far-red light. They also exist as a
28

dimer in eukaryotic cytoplasm and translocate to the nucleus upon absorption of far-red light.

Phytochrome comprises of four different domains that are PAS, GAF (chromophore binding
domain), P4 and a histidine kinase domain (Figure 3-5).

Phytochromes bind to

phytochromobilin for light sensing. Phytochromobilin is a linear tetrapyrole that exists in cis
conformation in its inactive form (Figure 3-4). The chromophore binds to phytochromes via a
thioether linkage (Figure 3-6).

29,30

Upon absorption of red-light, a cis-trans isomerization is

triggered that leads to protein conformational change and phytochromes translocate to the
28

nucleus.

The conformational changes

activate the histidine kinase domain that
can phosphorylate proteins downstream
and activate the kinase cascade which

Thiolinkage

6,31-38

can control the gene transcription.
Consequently,
responsible

for

phytochromes

are

light-induced

gene
Figure
3-6.
Crystal
structure
of
bacteriophytochrome bound to the chromophore
(Cyan) through thio-linkage.

transcription.
3.1.2.2.

Xanthopsins (Photoactive

Yellow Protein)
Xanthopsins refer to a photoreceptor family of protein that contain trans-p-coumaric acid
39

through a thiol-ester linkage.

The trans-p-coumaric acid functions as the light sensor for the

194

protein.

From all xanthopsins, photoactive yellow protein (PYP) is the most well-studied

Coumaric
Acid
Figure 3-7. Stereoview of photoactive yellow protein with trans-p-coumaric acid
shown.
xanthopsin and its crystal structure is available (Figure 3-7).

40,41

PYP is a small water soluble

protein (14 KDa) that is responsible for negative phototaxis of blue light.
43

1985 from archeobacteria Ectothiorhodospira halohila.

42

It was isolated in

It also belongs to the Per-Arnt-Sim

(PAS) family that is signaling proteins which Per, Arnt and Sim stands for three different
44,45

proteins that are first identified having PAS domain.

PYP contains a PAS domain that

adopts a α/β fold comprising of five strand of β-sheet and a α-helix on either side of the βsheets.

The ground state PYP absorbs at 446 nm.

Upon absorption of blue light, the

chromophore isomerizes to the strained cis-configuration and reaches intermediate state pR, that
is followed by subsequent relaxation and proton transfer from Glu46 to form signaling state pB.
During the transformation from pR to pB, the portions around the binding site unfold and a

195

signal is generated (Figure 3-8).

46-49

This process allow the bacteria to move away from the

hazardous ultraviolet and blue light.
3.1.2.3. Rhodopsin
A large family of photoreceptor proteins are the rhodopsins (see Chapter 1). Certain
microbes contain sensory rhodopsin or bacteriorhodopsin.
3,50,51

suggested is for bacteria to sense the light
purposes.

52-54

Sensory rhodopsin as the name

while bacteriorhodopsin is for energy generation

In brief, upon absorption of light by microbial rhodopsin at a specific wavelength

(bacteriorhodopsin ~ 560 nm, sensory rhodopsin I ~ 587 nm and sensory rhodopsin II ~ 497 nm),
the C13-C14 double bond of all-trans-retinal (Figure 3-4) isomerizes and leads to subsequent
protein conformational changes. For sensory rhodopsin, the conformational changes trigger
50,51

downstream processes and the bacteria moves.

196

However, the conformational changes in

Tyr42
O
H
Thr50

O

O-

O
Cys69

H
Glu46

S

O

pG

O

pB

HO

Cys69

S
O
pR
Tyr42
O
H
OH
Glu46

S

O
O

O
Cys69

Figure 3-8. Schematic diagram for photoactive yellow protein light cycle.

197

bacteriorhodopsin lead to proton transfer across the membrane that creates energy in the form of
52,54

proton gradient across the membrane.

Details of both microbial rhodopsins have been

discussed in Section 1.1.5.
In both vertebrates and invertebrates, 11-cis-retinal is the chromophore that is incorporated
in rhodopsin for light sensing.

Upon absorption of light by 11-cis-retinal in rhodopsin,

isomerization to all-trans-retinal initiates protein conformational changes. The changes trigger
the signal tranductions cascade downstream as discussed in Chapter 1 (Figure 1-9, see Section
1.1.3.). This isomerization allows light to transform into electrical signals that are transmitted to
the brain to be decoded into an image.
3.1.2.4. Isomerization for Recycling 11-cis-Retinal in Vertebrates and Invertebrates
3.1.2.4.1. Vertebrates
Recycling of 11-cis-retinal from all-trans-retinal
follows a set of biochemical transformation, not
requiring light.

This recycling happens in retinal

pigment epithelium (RPE) (Figure 3-9).

55-59

Retinal
Pigment
Epithelium
Rod Cell
Outer Segment

Disk

The all-

Cytoplasm

trans-retinal is reduced to all-trans-retinol by a
60

membrane-associated alcohol dehydrogenases.

All-

Rod Cell
Inner Segment

Mitochodria

trans-retinol diffuse to the retinal pigment epithelium

Nucleus
that contains several enzymes for 11-cis-retinal
regeneration.

The diffusion of all-trans-retinol is

suggested

be

to

catalyzed

by

lecithin

Light

Synapse
Rod Cell

retinol
Figure 3-9. Structure of rod cell.

198

O

Opsin

Retinol
Dehydrogenase

+HN

OH

Rhodopsin
!max = 500 nm
hv

O

scALDH
RPE65

N

OH
metarhodopsin II
!max = 380 nm
LRAT

OC(O)R

Signal Transduction
Cascade

Figure 3-10. 11-cis-Retinal cycle in retina. (scALDH stands for short-chain alcohol
dehydrogenase/reductase; LRAT stands for lecithin retinol acyl transferase.)

199

acyltransferase (LRAT) that converts all-trans-retinol to all-trans-retinyl ester and facilitates the
transfer of the retinol across the membrane.

61,62

All-trans-retinyl ester is hydrolyzed and

isomerize by isomerohydrolase in retinal pigment epithelium to all-trans-retinol (Figure 359,63,64

10).

These processes are suggested to be overall thermodynamically favorable. The
-1

-1

energy from the hydrolysis of ester (-5 to -10 kcal mol ) is coupled with 4 kcal mol
65

required to isomerize to all-trans-retinal to 11-cis-retinal.

energy

The 11-cis-retinol is finally oxidized

by retinol dehydrogenase to 11-cis-retinal that is bound with opsin to regenerate rhodopsin
(Figure 3-10).
In 1993, Hamel et. al identified a protein, RPE65, that is expressed predominantly in the
66

RPE.

RPE65 is a membrane-associated protein and its molecular mass is 61,961 Da based on
67

the MALDI.

The observed mass is higher than the calculated value (60944 Da) based on its
67,68

the amino acid sequence.

69

This indicated that RPE65 is post-translational modified.

The

role of RPE65 is further confirmed by knock-down experiments. It was found that mouse that
has its RPE65 gene knocked down, does not have 11-cis-retinal in the retina and rhodopsin
70

regeneration is impaired.
71,72

eyes.

Recently, RPE65 has been purified to homogeneity from bovine

The purified RPE65 binds all trans-retinyl ester as a substrate and catalyzes its
63,73

isomerization that produces 11-cis-retinol.

Since then, effort has focused on elucidation of

mechanism of isomerization. There are two proposed mechanisms for isomerization without
light.

The first proposal follows an addition-elimination mechanism (Figure 3-11).

It is

suggested that a nucleophilic residue from the enzyme attacks the polyene (presumably at C11 or

200

Enzyme

Nu

Enzyme

Nu

O
O

C15H31
3-5

3-4

Enzyme

3-7

Enzyme

Nu

OH

Nu

3-6

O
H

H
Figure 3-11. Proposed addition-elimination mechanism for isomerization in RPE65.

carbons closer to the β-ionone), followed by subsequent double bond migration and elimination
of the acyl group to generate the enzyme linked intermediate (3-5). Conformational change of
C11-C12 to s-cis leads to 3-6. This intermediate (3-6) is hydrated by the water, which leads to
74

double bond migration and elimination of the nucleophile generating 11-cis-retinal.

The other

proposed mechanism involves the formation of a carbocation intermediate (Figure 3-12). It is
suggested that the RPE65 binds to all trans-retinyl ester. The binding is followed by deacylation
of the ester to produce a carbocation intermediate (3-8).

75,76

Conformational change of C11-C12

carbocation intermediate and hydration forms 11-cis-retinol (Figure 3-12).

11-cis-Retinol

produced by RPE65 is oxidized by the retinol dehydrogenase to generate 11-cis-retinal. 11-cis201

LA
O
C15H31

O
LA
O
O

C15H31
3-8

3-4

3-7

OH

3-9
O
H

H

Figure 3-12. Second proposed mechanism for isomerization in RPE65.
Retinal is transferred by cellular retinal binding
protein (CRALBP) back to photoreceptor cell to
regenerates rhodopsin (Figure 3-10).
3-10

77-79

Since the

NH2

second proposed mechanism suggests RPE65
NH2
3-11

generates a cationic intermediate during the
catalysis, Golczak and co-workers have synthesized

Figure 3-13. Amine analogs of retinal several amine containing analogs as transition state
as inhibitors for RPE65.

202

Figure 3-14. Stereoview of RPE65 colored according to secondary structure (Helix:
Cyan; β-sheet: Purple; Loop: Pink). The hydrophobic tunnel is located in the middle of
the structure.
mimics (Figure 3-13). The amines are protonated under physiological pH. They found that
72

these amines are able to inhibit the isomerization activity in RPE65.
80

structure of RPE65 was resolved to 2.14 Å resolution.

Recently, the crystal

It contains a seven-bladed β-propeller

(Figure 3-14). The β-propellers are covered by a helical cap that protects the active site. There
is an iron(II) cofactor located near the top face of the propeller axis with four His residues
coordinated to the metal (Figure 3-15). The iron is located within the hydrophobic tunnel that is
proposed to be both the entry for the
starting material and the exit tunnel for
the product.
structure,

they

Based on the crystal
proposed

a

more

detailed mechanism (Figure 3-14).
Upon binding of all trans-retinyl ester,
the iron polarizes the acyl group and
catalyzes the deacylation of the ester
Figure 3-15. Iron binding site in RPE65 with
forming the carbocation intermediate 3- histidine coordinated to iron shown (Red)

203

Fe2+
O
O

Fe2+

C15H31

O

3-4

C15H31

O

3-8

Fe2+

O
O

3-9

C15H31

H
O
H

3-7

OH

Figure 3-16. Proposed mechanism involves iron activated deacylation followed by
isomerization and hydration of the cationic intermediate.
8. The de-acylation is followed by isomerization and hydration, producing 11-cis-retinol (Figure
3-16). In order to prove that the oxygen of the 11-cis-retinol is originates from water, they
conducted an experiment using oxygen-18 labeled water.

204

Based on the mass spectrum of the

reaction mixture, an oxygen labelled retinol was generated. This result confirms that the oxygen
of retinol is coming from the water.
3.1.2.4.2. Invertebrates
Invertebrates use a different pathway to recycle 11-cis-retinal. One of the well-studied
examples is the light-activated process in cephalopods. Cephalopods such as octopus and squid,
also use rhodopsin for light sensing. While vertebrates requires subsequent enzymatic reactions
to recycle 11-cis-retinal (see Sections 4.1.2.4.1.), cephalopods use retinochrome to regenerate
11-cis-retinal from all-trans-retinal in one step.

Retinochrome is a photosensitive pigment

found in the photoreceptor cells of cephalopods. While rhodopsin is located in the rhabdomal
membrane of outer segment of photoreceptor cell, retinochrome is stored in the somatic layer
81

(inner segment) of the photorecptor cell (Figure 3-17).

It was found that retinochrome binds both all-trans-retinal and 11-cis-retinal through the
formation of a PSB with Lys275 and yields two different absorptions (all-trans-retinal ~495 nm;
11-cis-retinal ~470 nm). Based on the sequence alignment with rhodopsin, it is believed that
retinochrome belongs to the same 7-α-helical transmembrane protein family as rhodopsins.
However, counterion Glu113 in rhodopsin for PSB stabilization is replaced by Met in
82,83

retinochrome.

In 2000, Terakita et al. conducted a mutagenesis experiment in retinochrome

to replace Glu and Asp with Gln and Asn respectively. This mutagenesis experiment suggested
82,83

that Glu181 is the counterion for the bound all-trans-retinal PSB in retinochrome.

It was

found that when retinochrome is irradiated by light (> 500 nm), the bound all-trans-retinal
isomerizes to 11-cis-retinal. The experiment conducted by Hara and co-worker showed that 80%
84-86

of pure all-trans-retinal is isomerized to 11-cis-retinal.
205

Interestingly, incubation of the

complex in dark for 24 hours after light irradiation,
87

recovered 86% of all-trans-retinal.

Light

This indicated

Microvilli
that light is necessary for isomerization of all-trans- (Rhabdomal
membrane)
retinal in retinochrome. It is proposed that upon

Cytoplasm

absorption of light by rhodopsin, 11-cis-retinal is
isomerized to all-trans-retinal.

All-trans-retinal is

then transferred to the inner segment of photoreceptors
88

by cellular retinaldehyde binding protein.

The all-

trans-retinal is then bound to retinochrome (λmax ~495
nm).

Irradiation of retinochrome-all-trans-retinal

complex

with

light

(>495

nm)

triggers

Somatic
layer

Mitochodria
Nucleus
Synapse
Squid
Photoreceptor

the

Figure 3-17. Schematic diagram of
isomerization of all-trans-retinal PSB to 11-cis-retinal squid photoreceptor.
PSB (λmax ~440 nm). The 11-cis-retinal is diffused out of the retinochrome and binds to cellular
88

retinaldehyde binding protein.

The 11-cis-retinal is finally transferred to the rhabdomal

membrane by retinaldehyde binding protein and is bound to rhodopsin to regenerate the pigment
(Figure 3-18). In 1992, Schwemer and co-workers also found that honey-bee adopts a similar
89-92

vision system for recycling retinal in retina.

3.2. Engineering CRABPII into Retinal Isomerases
3.2.1. Design Principles
We would like to engineer CRABPII into a retinal isomerase that specificically uses light
as an energy source, much like rhodopsin and retinochrome (Figure 3-10 and 3-18). Upon
206

Inner Segment

Outer Segment

+HN

Retinochrome

+HN

O

Retinal
Binding
Protein

Rhodopsin
hv = 480 nm

hv > 495 nm

H+
N

O

Figure 3-18. Schematic diagram of squid vision system.
207
207

H+
N

irradiation of rhodopsin with appropriate wavelength light, the specific isomerization of the 11cis-retinal to all-trans-retinal is initiated. All-trans-retinal has a poor affinity for the rhodopsin,
therefore, is expelled out of the binding site of rhodopsin, which allows for 11-cis-retinal
93

molecule to reload. The process is very efficient (quantum yield about 67%)
as compared to the photoisomerization in buffer.

94

and stereospecific

As a proof of principle, we would like to

generate an enzyme for a downhill process (11-cis-retinal to all-trans-retinal is favored by 4 kcal
-1

mol ) like rhodopsin, as a first generation retinal isomerase. In order to design a 11-cis-retinal
isomerase, the protein has to satisfy into the following criteria (Figure 3-19):
1.

It has to bind well with 11-cis-retinal.

2.

It has to bind 11-cis-retinal as a PSB so the absorption for the bound

chromophore (>440 nm) is different from that of the free chromophore (λmax ~380 nm).
Therefore, light that isomerizes the bound chromophore, does not isomerize the unbound
chromophore.
3.

The light source used should only isomerize the bound chromophore.

4.

In order to propagate the catalytic cycle, the protein should bind poorly

with all-trans-retinal.
After the first generation of retinal isomerase is created, we will target the uphill process
that is the isomerization of all-trans-retinal to 11-cis-retinal specifically by applying the same
principle. To be an all-trans-retinal isomerase, the protein has to bind well with all-trans-retinal
as PSB and binds poorly with 11-cis-retinal.

208

H+
N

hv > 450 nm

!max > 450 nm
Kd ~ µM

+HN

!max> 450 nm
Kd ~ nM

O
NH2

+
O
!max = 378 nm

Figure 3-19. Design principle for reengineering CRABPII to all-trans-retinal isomerase.

209

3.2.2. Synthesis of 11-cis-retinal
In order to test the design principle of retinal
isomerase,

11-cis-retinal

was

synthesized

previously described (Scheme 3-1).

95,96

11-cis retinal (99%)

as
The

synthesis begins with the coupling of β-ionone with
all trans retinal (1%)

TMS-protected yilde through the HWE reaction
forming alkyne 3-11 in 65% yield with E to Z ratio
of 4 to 1.

After desilylation of TMS-protected

alkyne 3-11, unprotected alkyne 3-12 was formed in
85% yield. The alkyne 3-12 was then coupled with

Figure 3-20. HPLC trace of purified
11-cis-retinal.

vinyl iodide using Sonogashira coupling followed by deprotection with TBAF, forming alcohol
3-14 in 81% overall yield after two steps.

Alkynol 3-14 was selectively hydrogenated by

activated Zn forming 11-cis-retinol. Oxidation of retinol forms 11-cis-retinal (3-16) and it is
further purified by HPLC to 99% purity (Figure 3-20). A 5% overall yield was obtained after
hydrogenation, oxidation and HPLC purification.

The yield was low because of the Zn-

activation was problematic. The Zn activation requires a completely inert atmosphere. Slight
amount of oxygen can deactivate the activated Zn.
3.2.3. Identification of CRABPII Candidate for 11-cis-Retinal Isomerase
3.2.3.1. UV-vis study
As discussed in Section 3.2.1, the first generation of retinal isomerase has to bind 11-cisretinal through the formation of a PSB. Therefore, several CRABPII mutants with Lys132 at the
active site were tested with 11-cis-retinal (Table 3-1).

210

97,98

The result showed a similar binding

TMS
O
EtO P
EtO

TMS

H
1. nBuLi, 0 ˚C to RT, THF;

TBAF, THF,
3-11 0 ˚C, 85% yield
(E:Z / 3.3:1)

2.
O

3-12

3-10
0 ˚C to RT, overnight,
65% yield, (E:Z / 4:1)
OH

OTBS
1. Pd(PPh3)4 (0.1 mol%),
rt, i-PrNH2
OTBS
2. CuI (0.1 mol %)
3.

I

TBAF, THF
0 ˚C, 90% yield
(E:Z / 3.5:1)

H
3-13

3-12
89% yield (E:Z / 3.3:1)

Activated Zn,

MnO2 / CH2Cl2

i-PrOH:H2O/2:1

5% overall yield
after 2 steps and
HPLC Purification

3-14

3-15

OH

Scheme 3-1. Synthesis of 11-cis-retinal.
211
211

3-16

O

Table 3-1. 11-cis-Retinal and all-trans-retinal tested with different CRABPII mutants
97,98,117,124
with engineered R132K.
11-cis-retinal
Entry Mutants

λmax
(nm)

all-trans-retinal

Reductive λ
Reductive
b, max
b,c
Amination (nm) Amination
c

1

R132K:R111L:L121E

451

N.D.
a

449

Yes

2
3
4
5
6
7
8

R132K:L121E
463
R132K:R111L:F15Y
352
R132K:R111L:T54E
357
R132K
370
R132K:Y134F
380
R132K:R134F:R111L:L121E:T54V 368
R132K:R134F:R111L:L121E
386

N.D.
Yes
Yes
Yes
No
Yes
N.D.

457
375
380
370
380
447
449

Yes
Yes
N.D.
Yes
No
Yes
Yes

9

WT-CRABPII

No

380

No

a
b
c

383

λmax of bound 11-cis-retinal or all-trans-retinal PSB identified by deconvolution
+

An adduct peak [M + 268] observed in mass spectrum after reductive amination
N.D. stands for not determined

pattern to that of all-trans-retinal. Through the preliminary screening, we found that Tyr134
inhibits the formation of 11-cis-retinal PSB. As we tested mutants R132K and R132K:Y134F
with 11-cis-retinal, we found that mutant R132K can form Schiff base with 11-cis-retinal while
mutant R132K:Y134F cannot give an adduct peak under reductive amination. This result is
similar to the study with all-trans-retinal (Table 3-1). Based on previous observations with alltrans-retinal, the formation of all-trans-retinal Schiff base is recovered when the Glu121 and
Val54 were installed (R132K:Y134F:R111L:L121E:T54V). Comparison of binding affinities
for mutants R132K:Y134F:R111L:L121E and R132K:Y134F:R111L:L121E:T54V with
R132K:R111L:L121E indicates that Y134F mutation inhibits the formation of 11-cis-retinal PSB
also. Thus, Y134F is necessary for the formation of 11-cis-retinal PSB. As mentioned in
Chapter 1, Tyr134 helps to orient the bound all-trans-retinal into the conformation that favors

212

a)

Acid Base Titration of 11zRt
with KLE
0.1
Absorbance at 455 nm

0.02

Absorbance

0.08
0.06
0.04
0.02
0

b)

350

400 450 500
Wavelength (nm)

0.015

0.01

0.005

0
6.5 7 7.5 8 8.5 9 9.5 10 10.5
pH

550

Acid Base Titration of 11zRt
with KE
0.2
Absorbance at 455 nm

0.095

0.15
Absorbance

pKa = 8.8

0.1

0.05

0

350

400 450 500
Wavelength (nm)

550

0.09

pKa = 7.2

0.085
0.08
0.075
0.07
0.065
0.06
0.055
6.5

7

7.5

8

pH

Figure 3-21. Acid base titration of 11-cis-retinal (11zRt) with CRABPII a) mutant
R132K:R111L:L121E (KLE) and b) R132K:R121E (KE).
nucleophilic attack by the engineered Lys132. Thus, Tyr134 might serve the same role for 11cis-retinal

PSB

formation

in

CRABPII.

In

addition,

the

results

with

mutants

R132K:R111L:F15Y and R132K:R111L:L121E indicate that glutamate is necessary for PSB
213

formation (Table 3-1). To further probe the effect of Glu121 on stabilization of PSB, the pKa
value for mutant R132K:L121E and R132K:R111L:L121E were measured (Figure 3-21). As
discussed in Section 1.2.4.3., Glu121 in mutant R132K:L121E adopts two different
conformations, one of which is further away from the bound all-trans-retinal PSB based on the
apo-crystal structure (Figure 1-68). Thus, the bound all-trans-retinal PSB in R132K:L121E (pKa
~ 7.2) is different from the PSB in R132K:R111L:L121E (pKa ~ 8.7). We found the same
changes on pKa with 11-cis-retinal (Figure 3-21). These results further prove the importance of
Glu121 in the stabilization of 11-cis-retinal PSB in CRABPII mutants.
3.2.3.2. Prediction of 11-cis-retinal Binding Position in CRABPII
Based on the initial screening, Tyr134 is important for SB formation with 11-cis-retinal
and Glu121 is important for stabilization of 11-cis-retinal PSB. These results are similar to the
study with all-trans-retinal. In order to probe the binding conformation for 11-cis-retinal in
CRABPII, 11-cis-retinal was docked in apo-R132K:R111L:L121E structure using Autodock 4.0

Figure 3-22. Stereoview of overlay of crystal structure of R132K:R111L:T54E with C15
aldehyde (C15) and 11-cis-retinal (11zRt) docked model.
214

99

package.

Autodock is the program that can find the optimized binding conformations of a

ligand with a protein based on the calculated binding energy. The binding energy between the
ligand and the protein is calculated based on the Van der Waals forces, electrostatic interactions
and hydrophobic interactions. Autodock 4.0 identified one possible binding mode. The docked
11-cis-retinal adopts a conformation much like the C15 aldehyde (Figure 3-22) in R132K:R111L
mutants (the alternate binding site). 11-Cis-Retinal is 1.4 Å shorter in overall (Figure 3-23). The
overall dimension of the alternate binding pocket in
O
3-2

CRABPII is about 15 Å (Figure 3-24). Fitting of alltrans-retinal into the alternate binding pocket (Figure
3-22) lead to severe steric clash. However, binding of

14.4 Å
11-cis-retinal in the alternate binding pocket did not
lead to offensive interactions and provided more
O
3-16

hydrophobic

contacts

with

protein

pocket

as

compared with the original conformation observed
with

13.0 Å

all-trans-retinal

in

CRABPII

mutant

R132K:R111L:L121E.

O
3-17

3.2.3.3. Comparison of the Binding Constant with
Both 11-cis-retinal and all-trans-Retinal across

9.8 Å
Different CRABPII Mutants
Figure 3-23.
Structure and
corresponding length of all-transThe design principle for creating 11-cis-retinal
retinal (3-2), 11-cis-retinal (3-16) and
C15 aldehyde (3-17).
isomerase requires the protein to bind 11-cis-retinal
and form a PSB. In addition, the protein should bind poorly to all-trans-retinal so that it can be
215

15.8 Å

13.5 Å
Figure 3-24. The distance between Lys132 and the Phe65 and Ile63 (Purple) in crystal
structure R132K:R111L:T54E with C15 aldehyde.
expelled. The unbound all-trans-retinal absorbs at a lower wavelength (380 nm) as compared
with bound 11-cis-retinal PSB (>440 nm) (Figure 3-19). Therefore, the ideal candidate should
have a good binding constant for 11-cis-retinal and poor binding with all-trans-retinal. The
binding constant of different CRABPII were measured.
Based on the result (Table 3-2) CRABPII mutant R132K:R111L:F15Y is a good candidate.
Mutant R132K: R111L:F15Y binds well with 11-cis-retinal (Kd ~ 47±8 nM) but binds poorly to
all-trans-retinal (Kd ~ 2455±240 nM) (Table 3-2), therefore, when the product (all-trans-retinal)
is formed, the all-trans-retinal should be replaced by another 11-cis-retinal molecule. This
allows the propagation of catalytic cycle (Figure 3-19).

However, CRABPII mutant

R132K:R111L:L121E binds relatively poorly with 11-cis-retinal (Kd ~ 351 nM) as compared
with all-trans-retinal (Kd ~ 1 nM) despite it forms a PSB with both 11-cis-retinal and all-transretinal (Table 3-2).

As mentioned in Section 3.2.1.3., 11-cis-retinal might bind into the

alternative pocket (Figure 3-22) as suggested by Autodock 4.0 model. As discussed in Section

216

Table 3-2. The binding constant of different CRABPII mutants with 11-cis-retinal and
97,98,117,124
all-trans-retinal.
11-cis-retinal
all-trans-retinal
a

Entry Mutants

a

Kd (nM)

Kd (nM)

351 ± 29.8

1 ± 4.9

N.D.

104 ± 11

1
2

R132K:R111L:L121E

3

R132K:R111L:F15Y

47 ± 8

2455 ± 240

4

R132K:R111L:T54E

436 ± 37

5069 ± 159

5

R132K

50 ± 16

280 ± 17

6

R132K:Y134F

630 ± 53

120 ± 4.9

7

R132K:R134F:R111L:L121E:T54V

210 ± 18

2.7 ± 7.0

8

R132K:R134F:R111L:L121E

9

WT-CRABP

c

R132K:L121E

c

N.D.

200 ± 8.3

644 ± 59

6000

N.D. stands for not determined

1.2.4.8.1., the presence of hydrophilic Glu121 in mutant R132K:R111L:L121E creates an
unfavorable interaction with the hydrophobic chromophore in the alternate binding site (Figure
1-74) and diminishes the binding affinity to 11-cis-retinal. Therefore, the enhanced binding
affinity to 11-cis-retinal from mutant R132K:R111L:F15Y might also indicate that the 11-cisretinal is bound to an alternate bindng pocket (Figure 3-22). A similar observation is made with
another R132K:R111L mutant, R132K:R111L:T54E. Mutant R132K:R111L:T54E also binds
tightly with 11-cis-retinal (Kd ~ 436 nM), that is about 10-folds tighter than all-trans-retinal (Kd
~ 5069 nM) (Table 3-1). As a control experiment, the binding of WT-CRABPII, that does not
have engineered K132, with 11-cis-retinal has been measured and it indicated that WT-CRABPII
binds poorly with 11-cis-retinal (Kd ~ 644±59 nM), however, it has higher affinity with 11-cis-

217

retinal relative to all-trans-retinal (Kd ~ 6000 nM). Further studies on whether formation of PSB
with R132K can enhance the binding of 11-cis-retinal will be discussed in detail later.
3.2.3. Discovery of Isomerization Activity in CRABPII
11-Cis-retinal is sensitive to acid, oxygen, light, heat

11-cis-retinal
a)

100,101

all-trans-retinal etc.

To ensure the 11-cis-retinal is maintained during the

UV-vis and fluorescent spectroscopic studies of 11-cis-retinal

b)

with CRABPII mutants, the chromophore was extracted after
each spectroscopic study. The chromophore from each sample

c)

was extracted with hexane and the extracted samples were
analyzed by HPLC (see Materials and Methods). During the
preliminary screening, the extracts after the UV-vis study and

Figure 3-25. HPLC trace of
a) 11-cis-retinal; b) 11-cisretinal
with
R132K:
R111L:L121E after 30 mins
of incubation in dark at rt;
c)
11-cis-retinal
with
R132K: R111L:F15Y after
30 mins of incubation in
dark at rt.

fluorescent study, showed substantial isomerization (Figure 325). This result is the first reported 11-cis-retinal isomerization
activity for CRABPII. Fearful that this observation was a result
of experimental error, two control experiments were performed.
11-Cis-retinal and 11-cis-retinal SB were incubated in PBS

buffer (pH = 7.3) for 30 minutes. However, upon extraction and HPLC analysis showed no signs
of isomerization (Table 3-3, Entry 1 and 2).
necessary for 11-cis-retinal isomerization.

This confirmed that CRABPII mutants are

It is also possible that the retinal isomerization

218

a, b

Figure 3-3. Isomerization activity of different CRABPII mutant.
Entry

Mutants

13-cisretinal
%

1

11-cis-retinal

0

99.1

0

0.9

30

2

11-cis-retinal SB

0

98.8

0

1.1

30

3

WT-CRABPII

8.5

55.1

0

36.4

30

4

R132K:R111L:L121E

7.7

60.7

0

31.5

30

5

R132K:R111L:F15Y

11.8

42.9

0

45.3

30

6

R132K:R111L:L121E

5.0

0

0

95

30

7

BSA

0

99

0

1.0

30

8

CRALBP

0

97

0

3.0

30

c

11-cisretinal
%

9-cisretinal
%

all-transretinal
%

Time
(min)

a

The percentage is normalized based on the molar extinction coefficient of different retinal
isomers.
b
0.2 equivalent of 11-cis-retinal was incubated with corresponding mutants for 30 minutes in
dark unless specified
c
0.2 equivalent of all-trans-retinal was incubated with corresponding mutants for 30 minutes in
dark.
happened during the extraction when the protein is denatured or the isomerization might have
happened on the protein surface. In order to probe these two possibilities, another two control
experiments were set up. 11-Cis-retinal was incubated with cellular retinaldehyde (retinal)
binding protein (CRALBP). CRALBP is a transfer protein for 11-cis-retinal and 11-cis-retinol in
photoreceptors of eyes.

102

It is proposed that CRALBP helps to transfer 11-cis-retinol from

RPE65, that isomerizes and hydrolates all-trans-retinyl ester to 11-cis-retinol, to alcohol
dehydrogenase.
102

regenration.

Also, it helps to transfer the 11-cis-retinal to rhodopsin for pigment
It binds 11-cis-retinal tightly with dissociation constant around 21 nM and does
103,104

not isomerize the bound 11-cis-retinal.

The extracted chromophore after the incubation

219

with CRALBP showed only 3% of all-trans-retinal (Table 3-3, entry 8). This suggested that the
isomerization does not happen during the chromophore extraction.

The second control

experiment was the incubation between bovine serum albumin (BSA) protein and 11-cis-retinal.
BSA is a protein that does not bind to 11-cis-retinal. It has 60 lysines that might form SB or PSB
with 11-cis-retinal. Incubation of BSA with 11-cis-retinal showed no sign of isomerization
(Table 3-3, entry 7). This further suggested that isomerization does not occur non-specifically
with interaction with a protein. Since it seems that the isomerization approaches an equilibrium
ratio of 11-cis-retinal and all-trans-retinal, incubation of mutant R132K:R111L:L121E with alltrans-retinal was examined to probe if any isomerization is evident. The chromophore extracted
from R132K:R111L:L121E after 30 minutes incubation with all-trans-retinal showed a small
amount of 13-cis-retinal (5%) formation with no sign of 11-cis-retinal formation (Table 3-3,
Entry 6).
3.2.4. Attempt to Dissect the Mechanism of 11-cis-retinal Isomerization
3.2.4.1. Is a Small Fraction of the Protein Responsible for the Isomerization?
Once the polypeptides are synthesized in vivo, they are folded into different tertiary
structures. During the folding processes, majority of the protein population folds into the correct
105,106

and functional tertiary structure, however, small portions are misfolded.
107-109

misfolded protein obeys the law of normal distrubution.

The ratio of

In order to investigate whether

the isomerase activity is the result of a small portion of misfolded “functional” proteins, we
incubated the protein with different equivalents of 11-cis-retinal. Assuming the reaction is not
catalytic, we hypothesized that if the isomerization is induced by a small fraction of “misfolded”
protein, these small populations of misfolded protein would be eventually saturated with

220

Table 3-4. Isomerization study of R132K:R111L:L121E with different equivalent of 11a, b
cis-retinal.
Equivalent Relative to
R132K:R111L:L121E

13-cis-retinal
%

11-cis-retinal
%

9-cis-retinal
%

all-trans-retinal
%

0.1 eq of 11-cis-retinal

5.7

51.4

0

42.9

0.2 eq of 11-cis-retinal

3.6

63.7

0

32.7

0.4 eq of 11-cis-retinal

3.0

59.5

0

37.5

0.6 eq of 11-cis-retinal

3.6

62.4

0

34.0

0.8 eq of 11-cis-retinal

2.8

66.4

0

30.8

1.0 eq of 11-cis-retinal

3.1

63.8

0

33.1

a

The percentage is normalized based on the molar extinction coefficient of differet retinal
isomers.
b
Incubation of different equivalent of 11-cis-retinal with R132K:R111L:L121E at dark for 2 h.
increasing amount of 11-cis-retinal; therefore, the rest of the 11-cis-retinal would bind to the
normal protein (Kd = 351 nM) that was assumed not to isomerize 11-cis-retinal to all-transretinal. This should lead to an increase in the ratio of 11-cis-retinal to all-trans-retinal. The
result (Table 3-4) showed that there is no change of the overall ratio as more 11-cis-retinal is
used. Therefore, it suggested that the isomerization is either not induced by a small portion of
possibly misfolded proteins or it is catalytic.
3.2.4.2. Investigating the Relationship between Isomerization and PSB or SB Formation
In order to investigate whether PSB or SB formation is necessary for the isomerization
activity overved with CRABPII mutants (R132K:R111:L121E and R132K:R111L:F15Y), WTCRABPII and mutants without nucleophilic Lys132 (R132L:R111L:L121E and R132L:R111L)
were tested. Both WT-CRABPII, mutants R132L:R132L:L121E and R132L:R111L isomerize
11-cis-retinal to its isomers (Table 3-5). This indicates that PSB or SB formation are not the

221

required for isomerization activity in CRABPII. WT-CRABPII maintains similar reactivity like
mutants R132K:R111L:L121E and R132K:R111L. However, we found that R132L mutation
can hamper the isomerization (-14% and -52% conversion as compare to R132K:R111L)
activity.

Both mutants R132L:R111L:L121E (15.4% conversion) and R132L:R121L

Table 3-5. Study the Effect of NaCl on Isomerization in R132K:R111L:L121E with 11-cisa, b
retinal.
NaCl (M)

13-cis-retinal
%

11-cis-retinal
%

9-cis-retinal
%

all-trans-retinal
%

100

1.7

88.9

0

9.4

0

0.9

90.2

0

8.9

a

The percentage is normalized based on the molar extinction coefficient of different retinal
isomers.
b
Incubation of 0.4 equivalent of 11-cis-retinal with R132K:R111L:L121E in dark for 30 min.
(5.5% conversion) show less conversion in 30 minutes as compare to the corresponding mutants
with R132K mutation R132K:R111L:L121E (39.3% conversion) and R132K:R111L:F15Y
(57.1% conversion) (Table 3-5). Such activity differences may be due to poor binding affinity to
11-cis-retinal by both R132L mutants.

However, based on the fluorescence quenching

experiment, both mutants R132L:R111L:L121E (Kd ~ 228 nM) and R132L:R111L (Kd ~ 111
nM) bound much better as compared to WT-CRABP with binding constant around 644 nM
(Table 3-5). Therefore, binding does not have significant impact on isomerization activity.
Based on the docking model of 11-cis-retinal with mutant R132K:R111L:L121E, Lys132 is
close to carbonyl of the bound 11-cis-retinal. Thus, Lys132 in mutants R132K:R111L:L121E
and R132K:R111L:F15Y and Arg132 in WT-CRABP can enhance the isomerization activity
through activation of the carbonyl of bound 11-cis-retinal.

222

Thus, we proposed that the

mechanism of bound 11-cis-retinal isomerization involves 1,4 conjugated addition-rotationelimination of a nucleophile (Figure 3-26).
Nu

Nu
B
O

H

OH

B
Nu
O

O

H

Figure 3-26. Proposed mechanism of isomerization in CRABPII.
3.2.4.3. Probing the Possible Michael Addition-Isomerization Mechanism
Since the retinal isomerization occurs without light irradiation, possibility of a Michael
addition-isomerization-elimination mechanism similar to the proposed mechanism in RPE65 as
described in Section 3.1.4.1. (Figure 3-26) mechanism similar to the proposed mechanism in
RPE65 as described in section 3.1.4.1 (Figure 3-11) was considered. External nucleophile from
the media was examined. Phosphate is a rarely seen as a nucleophile and chloride ion is a decent
nucleophile. We compared the rate of isomerization in PBS buffer and phosphate buffer without
sodium chloride. The result indicated that the present of chloride does not enhance the rate of

223

isomerization, therefore, chloride was eliminated as a possible player in the proposed mechanism
(Table 3-5).
Recently, Liu and co-workers studied the
role of hydrogen bonds on thermal isomerization
110

in rhodopsin using deuterated water.

It was

found that 11-cis-retinal in rhodopsin undergoes
thermal isomerization to all-trans-retinal at 65
111

˚C.

According to the crystal structure of

Figure 3-27. Hydrogen bonding network
rhodopsin, there is a hydrogen bonding network at around the bound 11-cis-retinal in
rhodopsin.
the retinal binding site that involves order water
molecules (Figure 3-27).

112,113

To probe the role of hydrogen bonding on thermal isomerization

in rhodopsin, the rate of thermal decay in deuterated solvent was measured. Their results
indicated that deuterated water can slow
down

the

thermal

110

isomerization,

since deuterium of deuterated water can
form stronger hydrogen bonds than
hydrogen in water. They then proposed
that the hydrogen bonding is important
for stabilizing the tertiary structure of
rhodopsin

that

affects

the

thermal

Figure 3-28. Ordered water molecules within the
110
retinal binding pocket in R132K:R111L:L121E.
isomerization activity in rhodopsin.
Based on the crystal structure of R132K:R111L:L121E, there is an extended ordered water
224

network in the binding pocket around the bound retinal (Figure 3-28). The ordered water
network is thought to be important for maintaining the structure of CRABPII,

114-117

therefore,

the same experiment was carried out in order to study the role of ordered water for retinal
isomerization in CRABPII. The result showed that deuterated water had a large effect on the
isomerization

activity

in

CRABPII.

The

rate

for

deuterated

PBS

buffer

a, b

Table 3-6. Effect of deuterated solvent on isomerization in R132K:R111L:L121E.

D2O

13-cis-retinal
%

11-cis-retinal
%

9-cis-retinal
%

all-trans-retinal
%

No

2.3

82.6

0

15.1

Yes

1.0

91.0

0

8.0

a

The percentage is normalized based on the molar extinction coefficient of different retinal
isomers.
b
Incubation of R132K:R111L:L121E with 0.2 equivalent of 11-cis-retinal at dark for 30 min.

is almost two fold lower than the non-deuterated solvent (Table 3-6). This suggested that the
ordered waters have a crucial role for retinal isomerization in CRABPII. To further probe the
effect of ordered waters, the half-life and decay constant of mutant R132K:R111L:F15Y for
retinal isomerization was measured in deuterated and non-deuterated water. The results indicate
that the rate is hampered by the deuterated solvent (Figure 3-29). There are several ways that
deuterated water could affect the 11-cis-retinal isomerizaton in CRABPII: 1) Deuterated water
can further stabilize the tertiary structure of CRABPII, therefore, slow down the conformational
changes that could lead to isomerization in the solution; 2) Deuterated water could lower the rate
of SB or PSB formation in CRABPII, thus, affecting the rate of isomerization; 3) Deuterated
118-120

waters are suggested to be less nucleophilic than water,

225

therefore, the isomerization in

Rate of 11-cis-retinal Isomerization
Rate of 11-cis-retinal Isomerization
in KL-F15Y Incubated
in KL-F15Y Incubated
with Deuteriated PBS buffer (pH = 7.3)
with PBS buffer (pH = 7.3)
90

80

Decay Constant (!):

70

0.00608

60

t1/2 = 114 s

50
40

% of 11-cis-retinal

100

90
% of 11-cis-retinal

100

80

Decay Constant (!):

70

0.00404

60

t1/2 = 172 s

50
40
30

30

20

20
0

1000 2000 3000 4000 5000 6000
Time (second)

0

1000 2000 3000 4000 5000 6000
Time (second)

Figure 3-29. Rate of isomerization of R132K:R111L:F15Y in PBS buffer and deuteriated
PBS buffer.
deuterated buffer could be slower if water is the nucleophile for isomerization based on our
suggested mechanism (Figure 3-26). These suggestion will be further discussed later.
Although we have shown the fact that formation of SB and PSB is not related to the
isomerization activity, the rate for the SB formation is measured in deuterated buffer. The rate of
SB formation was not affected by the use of deuterated solvent which indicates that the SB
formation is not involved in retinal isomerization for both R132K:R111L:F15Y and
R132K:R111L:L121E (Figure 3-30).
In addition, we found that the isomerization activity can be slowed by low temperature
(Table 3-7). Protein in solution is constantly changing the conformations, through folding and
unfolding processes, and low temperature can slow down conformational changes. It is possible
that one of the intermediate conformations triggers the isomerization. As mentioned previously,

226

Rate of SB formation in
Deuterated Buffer

Rate of SB Formation in
PBS Buffer
0.014

0.016

Absorbance at 400 nm

Absorbance at 400 nm

0.018

0.014

t1/2 = 43.9 s

0.012
0.01
0.008
0.006
0.004

0.012
0.01
0.008
0.006
0.004

0

5

10
15
Time (min)

20

t1/2 = 35.7 s

25

0

5

10
15
Time (min)

20

25

Figure 3-30. Rate of SB formation in deuterated PBS buffer and PBS buffer.
the ordered water network within CRABPII is important for maintaining its tertiary structure.
Hydrogen bonding between deuterated water is stronger than that in water. Therefore, the use of
deuterated water can strengthen the ordered water network within CRABPII which can slow
down the protein conformational change, thus, slow down the isomerization activity. In order to
test whether deuterated buffer slow down the isomerization activity through decreasing the rate
of conformational change, the incubation of 11-cis-retinal with CRABPII mutants was carried
out in buffers containing different amounts of glycerol. Glycerol is used as a co-solvent to
increase the viscosity of the buffer so it can slow down protein conformational changes. This
study (Table 3-7) showed that protein conformational changes are not the main cause for retinal

227

a, b

Table 3-7. Study of the effect of glycerol on retinal isomerization in CRABPII.
Glycerol (%)

13-cis-retinal
%

11-cis-retinal
%

9-cis-retinal
%

all-trans-retinal
%

0

1.9

81.0

0

17.1

0

c

0.6

94.4

0

5

10

1.8

81.0

0

17.2

15

1.6

81.7

0

16.7

5

1.1

84.5

0

14.4

a

a

The percentage is normalized based on the molar extinction coefficient of different retinal
isomers.
b
Incubation of R132K:R111L:L121E with 0.2 equivalent of 11-cis-retinal at dark for 30 min.
c

Incubation at 4 ˚C.

isomerization in CRABPII as no significant changes were observed (Table 3-7). That indicates
that ordered waters affect the rate of isomerization in different ways.
The possibility that ordered water could act as nucleophile for the proposed mechanism
was

considered

Although

the

(Figure

crystal

3-26).

structure

of

CRABPII mutants with 11-cis-retinal
has not been resolved yet, holo-crystal
structures

of

R132K:R111L:L121E

with bound all-trans-retinal (Figure 328)

and

R132K:R111L:T54E

with

bound C15 aldehyde (Figure 3-31)
showed that there are many ordered

Figure 3-31. Ordered water network in CRABPII
mutant R132K:R111L:T54E.

228

a

Table 3-8. Study the pH effect on isomerization in CRABPII.

pH

Entry Mutants

13-cisretinal
%

11-cisretinal
%

9-cisretinal
%

all-transretinal
%

1

11-cis-retinal SB

b

7.3

0

98.2

0

1.8

2

11-cis-retinal SB

b

9.5

0

97

0

3.0

3

R132K:R111L:L121E

c

5.9

0.3

91.9

0

7.8

4

R132K:R111L:L121E

c

7.3

0.8

87.0

0

12.1

5

R132K:R111L:L121E

c

8.3

1.2

80.6

0

18.2

6

R132K:R111L:L121E

c

8.9

1.5

78.0

0

20.5

7

R132K:R111L:F15Y

d

5.7

1.4

97.3

0

2.7

8

R132K:R111L:F15Y

d

7.3

3.4

83.0

0

15.6

9

R132K:R111L:F15Y

d

9.1

4

69.0

0

27.6

10

WT-CRABP

c

5.8

0

88.3

0

11.7

11

WT-CRABP

c

7.3

1.9

75.7

0

22.4

12

WT-CRABP

c

8.5

5.5

52.1

0

42.4

a

The percentage is normalized based on the molar extinction coefficient of different retinal
isomers.
b
Incubation at dark for 30 min.
c

Incubation with 0.2 equivalent of 11-cis-retinal at dark for 15 min.

d

Incubation with 0.4 equivalent of 11-cis-retinal at dark for 3 min.

waters in CRABPII crystal structure. These ordered waters are in close proximity to the bound
chromophore. If water can act as the nucleophile, increase of the pH value of the buffer can
increase the amount of nucleophilic water, in the form of hydroxide, and therefore, should
increase the rate of isomerization. The isomerization activity was enhanced under alkaline
conditions with different mutants (Table 3-8), suggesting that ordered water might be involved in

229

the isomerization, possibly as a nucleophile. In addition, a control experiment of 11-cis-retinal
SB in buffer at different pH showed that alkaline condition does not enhance the rate of
isomerization without CRABPII (Table 3-8). Based on these results, the most likely nucleophile
for isomerization is either a hydroxide ion or an activated water molecule (Figure 3-26).
3.2.5. Catalytic Isomerization in CRABPII
In our original plan to engineer CRABPII into a retinal isomerase, we required a CRABPII
mutant that binds tightly to 11-cis-retinal as a PSB. In addition, it has to bind poorly to all-transretinal. Upon irradiation of light with specific wavelength (>440 nm), 11-cis-retinal should
isomerize to all-trans-retinal. Theoretically, if the binding affinity of 11-cis-retinal is much
better than all-trans-retinal, product displacement with substrate would lead to a catalytic cycle
(Figure 3-19). However, we found that our CRABPII mutants already has the isomerization
activity in dark and mutant R132K:R111L:F15Y binds well with 11-cis-retinal (Kd ~ 47 nM) and
binds poorly with all-trans-retinal (Kd ~ 2455 nM). We incubated mutant R132K:R111L:F15Y
with different equivalence of 11-cis-retinal.

From the result, we found that mutant

R132K:R111L:F15Y has a turnover of ~1.3 with 5 equivalence of 11-cis-retinal (Table 3-9,
entry 6 and 7). The turnover number increases to 2 when 10 equivalent of 11-cis-retinal was
used (Table 3-9, entry 8 and 9). This suggests that product release is a problem and koff for
bound all-trans-retinal should be optimized. The same experiment was also conducted on
mutant R132K:R111L:L121E and WT-CRABPII.

Mutant R132K:R111L:L121E show no

turnover with 5 equivalent reaction (Table 3-10, entry 3 and 4) and become catalytic when 10
equivalents of 11-cis-retinal were used (Table 3-10, entry 5 and 6). This result is due to the fact

230

Table 3-9.
Study the
a, b
R132K:R111L:F15Y.

turnover

Time
(h)

ability

for

11-cis-retinal

isomerization

Turnover 13-cis11-cisall-transno.
retinal (%) retinal (%) retinal (%)

Entry Mutants

eq.b

1

11-cis-retinal c

N/A 4

N/A

0

98.6

1.4

2

11-cis-retinal with
all-trans-retinal c

N/A 0

N/A

0

24.4

75.6

3

11-cis-retinal with
all-trans-retinal c

N/A 4

N/A

0

27.8

72.2

4

R132K:R111L:F15Y 0.2

2

N/A

2.8

60.3

36.9

5

R132K:R111L:F15Y 0.2

4

N/A

2.7

62.9

34.3

6

R132K:R111L:F15Y 5

2

1.3

2.3

74.5

23.2

7

R132K:R111L:F15Y 5

4

1.3

1.8

72.5

25.7

8

R132K:R111L:F15Y 10

2

1.3

0.9

87.3

11.8

9
a

in

R132K:R111L:F15Y 10

4

2.0

1.6

80.5

17.9

Incubation in dark.

b

The percentage is normalized based on the molar extinction coefficient of differet retinal
isomers.
c
N/A means not applicable.
that R132K:R111L:L121E binds all-trans-retinal better than 11-cis-retinal, therefore,
diminishing the catalytic activity. This result indicates that the binding affinity of 11-cis-retinal
over all-trans-retinal is important for our retinal isomerase design. WT-CRABPII does not show
any catalytic activity probably because WT-CRABPII has poor binding affinity to 11-cis-retinal
(Table 3-10, entry 9 to 11). In addition, a control experiment of incubating a mixture of alltrans-retinal and 11-cis-retinal was conducted. The ratio between the 11-cis-retinal and all-

231

trans-retinal does not change after incubation for 4 hours without protein (Table 3-9, Entry 2 and
3). This result indicates that the isomerization activity observed is not due to the different
stability of 11-cis-retinal and all-trans-retinal in the reaction media.
Table 3-10.
Study the turnover ability
a, b
R132K:R111L:L121E and WT-CRABPII.
Entry Mutants

for

Equivalent Time
b
(h)

11-cis-retinal

isomerization

13-cisretinal (%)

11-cisretinal (%)

all-transretinal (%)

1

R132K:R111L:L121E 0.2

2

3.6

62.5

33.8

2

R132K:R111L:L121E 0.2

4

3.6

58.1

38.1

3

R132K:R111L:L121E 5

2

3.7

81.9

16.6

4

R132K:R111L:L121E 5

4

1.5

83.0

15.5

5

R132K:R111L:L121E 10

2

1.5

86.8

11.7

6

R132K:R111L:L121E 10

4

1.2

83.5

15.3

7

WT-CRABPII

0.2

2

3.1

65.8

31.1

8

WT-CRABPII

0.2

4

3.0

66.1

30.8

9

WT-CRABPII

5

2

1.1

89.2

9.8

10

WT-CRABPII

5

4

1.1

86.5

12.4

11
a

in

WT-CRABPII

10

4

0.9

90.1

9.0

Incubation in Dark

b

The percentage is normalized based on the molar extinction coefficient of different retinal
isomers
3.2.6. Attempted to Decrease the Isomerization Activity in CRABPII
Previously, we suggested that the isomerization in CRABPII involves the Michael addition
of hydroxide to the conjugated polyene followed by rotation of a single bond and retro-Michael
addition to form all-trans-retinal (Figure 3-26).

This is because the rate of isomerization

increases as the pH value of the buffer increases. Since the Michael addition can be activated

232

through polarization of the carbonyl of 11-cis-retinal, if the suggested mechanism is valid,
minimizing the polar interaction between protein and the aldehyde oxygen, such as: R132
mutants, should lower the isomerization activity. Our isomerization results agree with this

Figure 3-32. Stereoview of 11-cis-retinal (Green) model showing Thr54 and Thr56
(Purple) closed to C11-C12 of the bound 11-cis-retinal.
hypothesis since mutants R132L:R111L:L121E and R132L:R111L, which have leucine at 132
position, have a much lower isomerization activity (16.4% and 5.5% conversion, respectively) as
compared

with

the

Lys132

mutants

R132K:R111L:L121E

(39.3%

conversion),

R132K:R111L:F15Y (51.1% conversion) and WT-CRABP (44.9% conversion) (Table 3-3 and
3-4). In addition, we suggest that polarity around the C11-C12 of the bound 11-cis-retinal could
also affect the strength of the nucleophile (suggested to be the ordered water molecule). Polar
residues could further polarize the ordered water through hydrogen bonding, therefore,
enhancing its nucleophilicity.

Based on the AutoDock model of 11-cis-retinal with

R132K:R111L:L121E in the alternate binding pocket (Figure 3-22), Thr56 and Thr54 are close
to the C11-C12 of the bound 11-cis-retinal (Figure 3-32). Thus, mutants R132K:R111L:T54V
and R132K:R111L:L121E:T56V were generated. The isomerization assay, using these mutants,
indicated that both hydrophobic mutations can slow down the isomerization (Table 3-11, Entries
1, 2, 5 and 6). However, when the incubation of R132K:R111L:L121E:T56V with 11-cis-retinal

233

was followed for 2 hours, the amount of 11-cis-retinal isomerized was the same as
R132K:R111L:L121E (~50% conversion) (Table 3-11, Entries 1 and 3). These results indicate
that the isomerization activity can be slowed down by hydrophobic mutation but the overall
amount of isomerized 11-cis-retinal does not change (Table 3-11).
a, b

Table 3-11. Study the effect of T56V and T54V mutation on isomerization activity.
Entry Mutants

13-cisretinal
%

11-cisretinal
%

all-trans9-cis-retinal
retinal
%
%

1

R132K:R111L:L121E

8.7

47.9

0

42.7

2

R132K:R111L:L121E:T56V

2.2

73.2

0

24.6

3

R132K:R111L:L121E:T56V

4.7

51.1

0

44.2

4

R132K:R111L:F15Y

12.2

31.0

0

56.2

5

R132K:R111L

7

39

0

52.4

6

R132K:R111L:T54V

2.4

62.8

0

34.8

c

a

The percentage is normalized based on the molar extinction coefficient of different retinal
isomers.
b
Incubation with 0.2 equivalent of 11-cis-retinal in dark for 30 min unless specified.
c

Incubation for 2 h.

3.3. Searching for Another Counterion for Bound 11-cis-retinal PSB in CRABPII
As mentioned previously, our strategy entails generating the mutants that bind 11-cisretinal as a PSB but bind poorly with all-trans-retinal (Figure 3-19).

Although

R132K:R111L:L121E binds 11-cis-retinal as a PSB, the binding affinity for 11-cis-retinal (Kd ~
185 nM) is worse than that for all-trans-retinal (Kd ~ 1 nM). Therefore, product inhibition
occurs with R132K:R111L:L121E. Since R132K:R111L:F15Y mutant binds 11-cis-retinal as a

234

SB and the autodock model suggests
that the bound 11-cis-retinal adopts an
alternate conformation, we considered
alternate counterions, other than Glu121,
for the 11-cis-retinal SB. As suggested
by the model, Ser12 and Ser37 are close

S12

S37

to the bound 11-cis-retinal PSB (Figure
3-33). Thus, aspartate residues were
Figure 3-33. Model of 11-cis-retinal with R132K:
R111L:L121E with Ser12 and Ser37 showed installed at these two positions to act as
(Purple).
a counterion for the bound 11-cis-retinal
SB.

The UV-vis spectra as shown in

Figure

3-34,

indicated

that

11zRt with KL:F15Y:S12D and
KL:F15Y:S37D

mutant
and

0.035

R132K:R111L:F15Y:S12D

cannot

0.03

produce

a

fully

shifted

spectrum.

However, when the spectra of 11-cisretinal with R132K:R111L: F15Y:S37D
and R132K: R111L:F15Y:S12D were
overlapped, a shoulder is observed for
mutant

R132K:R111L:F15Y:S12D

(Figure 3-34) indicating a small amount

Relative Absorbance

R132K:R111L:F15Y:S37D

0.025
0.02
0.015
0.01

KL:F15Y:S12D
KL:F15Y:S37D

0.005
0
320 360 400 440 480 520 560 600 640
Wavelength (nm)

Figure 3-34. UV-vis spectra of R132K:R111L:
of PSB formation.
No noticeable F15Y:S12D and R132K:R111L:F15Y:S37D with
11-cis-retinal.
shoulder is observed when all-trans-

235

retinal is bound. These results suggested that mutant R132K:R111L:F15Y:S12D could be a
good template for further optimization to provide a better counterion for bound 11-cis-retinal.
3.4. human Cellular Retinol Binding Protein II as a Retinal Isomerase
As discuss in Chapter 2, we have
successfully

engineered

human

cellular

retinol binding protein II (CRBPII) into a
121

second generation of rhodopsin mimic.

In

addition, the crystal structure of all-transretinal with wild-type CRBPII (WT-CRBPII)
has been obtained and shows that retinal is
Figure 3-35. Crystal structure of retinal
with CRBPII with the ordered water inside
the binding site shown.

completely embedded by CRBPII. Based on
the crystal structure, there are less ordered

water molecules within the CRBPII binding pocket
as compared with CRABPII (Figure 3-28 and 3-35).
11-Cis-retinal was first docked in CRBPII by
Autodock 4.0. Based on the autodock model, 11-cisretinal binds CRBPII like all-trans-retinal in CRBPII
(Figure 3-36), unlike the two different binding modes
in CRABPII (Figure 3-22). Both WT-CRBPII and
mutant R108K:K40L exhibit isomerization activity
However, similar to what was Figure 3-36. Crystal structure of
CRBPII with bound all-trans-retinal
previously described for CRABPII (WT-CRABPII, (tRt) (Blue) and docked 11-cis-retinal
(11zRt) (Green).
(Table 3-13).

236

Table 3-12. Study of 11-cis-retinal isomerization activity using CRBPII.

a, b

Mutants

13-cisretinal
%

11-cisretinal
%

all-trans9-cis-retinal
retinal
%
%

WT-CRBP

0.8

85.8

0

13.4

Q108K:K40L (R132K:R111L:)

4.1

75.6

0

20.3

10.6

45.3

0

46.1

c

R132K:R111L::T51V:T53C:R58W
a

The percentage is normalized based on the molar extinction coefficient of differet retinal
isomers.
b
Incubation with 0.2 equivalent of 11-cis-retinal in dark for 30 min unless specificied.
c

Incubation for 2 h.

mutants R132L:R111L:L121E and R132L:R111L), formation of SB is not necessary for 11-cisretinal isomerization. In CRBPII, Gln108 is closed to the carbonyl of the bound retinal and
mutant Q108K can form a PSB with all-trans-retinal. Comparing the isomerization activity of
WT-CRBPII with mutant Q108K:K40L indicates that positive charged residue (Lys108) close to
retinal oxygen enhances the isomerization activity as suggested with CRABPII with mutants
R132L:R111L:L121E and R132L:R111L and their corresponding R132K mutants (Table 3-2
and 3-3).
Recently, we have obtained a series of CRBPII mutants that upon binding all-trans-retinal
121

absorb at a wide range of wavelength (475 nm to 620 nm).

Since the docked model shows

that 11-cis-retinal can bind CRBPII similar to all-trans-retinal (Figure 3-36), bound 11-cisretinal might show similar absorption shift. UV-vis spectra indicate differences between the
bound 11-cis-retinal and bound all-trans-retinal with different mutants (Table 3-13).

WT-

CRBPII shows a red-shifted peak (390 nm) with all-trans-retinal while bound 11-cis-retinal did

237

Table 3-13. UV-vis study of 11-cis-retinal with CRBPII and its
mutants.

λmax with
λmax with
all-trans-retinal 11-cis-retinal

Mutants

not

(375

nm).

However, upon binding
with

mutant

Q108K:K40L,
bound

the

11-cis-retinal

WT-CRBP

390

375

Q108K:K40L

508

522

R132K:R111L:T51V

530

534

shifted peak (522 nm)

R132K:R111L:T51V:T53C

537

537

as

R132K:R111L:L119Q

522

522

bound all-trans-retinal

R132K:R111L:T51V:T53C:L119Q 539

535

(508 nm) (Table 3-13).

R132K:R111L:T51V:T53C:R58W 589

556

showed a more red-

compared

with

With the removal of
polar residues Thr51

close to the PSB region
of

bound

all-trans-

KL

KL:T51V

retinal, both bound alltrans-retinal (530 nm)
and 11-cis-retinal (534
nm) showed significant
red-shifted spectra as
compared with mutant
Q108K:K40L (508 nm
and

522

respectively).

nm
The

Figure 3-37. Electrostatic potential calculation on CRBPII
mutants Q108K:K40L (Left) and Q108K:K40L:T51V:T53C
(Right) using APBS package.

238

induced red-shift in R132K:R111L:T51V is due to the removal of a negative electrostatic around
the PSB (Figure 3-37) that allows for the positive charge propagation from the iminium nitrogen
along the polyene (see Section 1.3).

When Thr53 is replaced by Cys, mutant

R132K:R111L:T51V:T53C shows a 7 nm red-shift (537 nm) when bound to all-trans-retinal and
a 3 nm red-shifted when bound to 11-cis-retinal (537 nm) (Table 3-13). Both T51V and T53C
mutations induce less red-shifting with 11-cis-retinal as compared with all-trans-retinal. A
similar trend is observed as we compare the mutants Q108K:K40L:L119Q with Q108K:K40L:
L119Q:T51V:T53C (Table 3-13). Q108K:K40L:L119Q:T51V:T53C is about 17 nm more redshifted (539 nm) as compared with Q108K: K40L:L119Q (522 nm). These results suggests that
11-cis-retinal bind CRBPII similar
to

all-trans-retinal

further

away

L1119Q.

from

but

possibly

T53C

and

Q108K:K40L:T51V:

T53C:R59W (589 nm) yields a large
red-shift (52 nm) when bound to alltrans-retinal

as

compared

Q108K:K40L:T51V:T53C

with
(537

Q108K:K40L: Figure 3-38. Overlay model of docked 11-cis-retinal
in CRBPII with holo-structure of all-trans-retinal
T51V:T53C:R58W (556 nm) shows with CRBPII.
nm).

only

However,

a

20

nm

red-shifted

when

bound

to

11-cis-retinal

as

compared

with

Q108K:K40L:T51V:T53C (537 nm) (Table 3-13). From the model, 11-cis-retinal is deeper in
the binding pocket, thus, the β-ionone ring is further away from Arg58 (Figure 3-38). This can
explain the relatively attenuated red shift observed with 11-cis-retinal. The UV-vis studies

239

suggest that 11-cis-retinal binds at a similar
11-cis-retinal with
KL:T51V:T53C:R58W

position as the bound all-trans-retinal
0.015

because both T51V and R58W induces red-

589 nm

shift on 11-cis-retinal much like all-transHowever, the precise orientation

might be different. This is suggested by the
smaller red-shift obtained from mutations

Absorbance

retinal.

556 nm

0.01

0.005

T53C, Q119L and R58W (Table 3-13).
Besides, as we previously discussed,
the

Q108K:K40L:T51V:

induces

more

red-shifting

0
300

T53C:R58W
(33

nm

differences) on bound all-trans-retinal (589

400
500
600
Wavelength (nm)

700

Figure 3-39. UV-vis spectra of 11-cis-retinal
with R132K:R111L:T51V:T53C:R58W.

nm) than 11-cis-retinal (556 nm). Since the λmax difference is so large, we can observe the
retinal isomerization by UV-vis by the following the decrease of the 556 nm peak (or increase of
the 589 nm peak) in real time (Figure 3-39).
3.5. Synthesis of Ring Locked 11-cis-retinal Analog for Crystallography
The exact position of 11-cis-retinal within the CRABPII binding pocket has been
speculated to change depending on whether position 121 is occupied by Leu or Glu in CRABPII.
Since 11-cis-retinal isomerizes to mixture of isomers with CRABPII, crystallographic studies are
not possible.

In order to probe the binding mode of 11-cis-retinal in CRABPII, a non-

isomerazble 11-cis-retinal ring-locked analog 3-18 (Scheme 3-2) was synthesized for
crystallography.

240

3.6.1. Synthesis of Ring-locked 11-cis-retinal Analog
The synthesis of ring locked analog 3-18 started by the condensation of 3-methoxy-1cyclohexenone (3-20) and -ionone affording the aldol product 3-21 in 52% yield. The ketone
3-21 was reduced by lithium aluminum hydride followed by dehydration and rearrangement with
1M H2SO4 to afford 3-22 in quantitative yield. Enone 3-22 was acetylated and after the
elimination, a mixture of E and Z ketone 3-24 was formed in 60% yield. The mixture of ketone
3-24 was submitted to HWE reaction with diethyl methylacetylphosphonate. The resulting ester
was reduced by DIBAL-H followed by MnO2 oxidation to afford 3-18 in 50% overall yield.
The final product was further purified by HPLC. However, the HPLC condition used
122

before for this molecule did not work well for us.

So optimization for the purification was

carried out prior to the actual purification. Optimized conditions (2% EtOAc in hexane and
normal phase on silica column) were determined and used to purify the 11-cis ring locked retinal.
The recovered yield from the HPLC purification was only ~40%, probably as a result of
decomposition on silica gel. Further optimization is needed in the future for recovering more
material. For example, the column could be equilibrated with 0.05% of triethylamine and
solvent should be pretreated with 0.05% of triethylamine in order to remove trace amount of acid
in the solvent and also the silica column.

241

1) LDA, THF
2)
1) 1.5 eq.
(CH3O)3CH,
MeOH

O

O

2) 0.2 ml 37% HCl,
57% yield

3-19

3-20

H O

HO

O

O
3) AcOH, 0 ˚C
OMe 4) pH 7.3 buffer,
52% yield

1) LiAlH4, 0 ˚C / ether

HO

2) H2SO4 / H2O (1:7),
OMe 50% yield

O

3-21

3-22
(EtO)2P(O)CH2CO2Et,
NaH, THF, 75% yield

1) Ac2O, DMAP, 24 h,
Toluene
2) DBN, reflux 2h,
60% yield

O
3-23
1) DIBAL-H, THF
2) wet silica
O
OEt

3-24

3) MnO2, CH2Cl2
50% overall
yield in 3 steps

3-18

CHO

Scheme 3-2. Synthesis of ring-locked 11-cis-retinal analog 3-18.
3.5.2. Comparison of UV-vis Spectroscopic and Fluorescent Spectroscopic Study between
11-cis-retinal and its Ring-locked Analog 3-18 with CRABPII Mutants
In order to investigate whether ring-locked analog 3-18 behaves similarly with 11-cisretinal, UV-vis spectroscopic studies were conducted. From the UV-vis, analog 3-18 exhibits

242

11zRt Ring-Lock 42 with
R132K:R111L:L121E

11zRt Ring-Lock 42 with KL:F15Y

0.08
Absorbance

0.3

Formation of
SB

0.25

0.06
0.4 equivalent

0.04

0.2

0.8 equivalent
0.6 equivalent

0.15

0.4 equivalent

0.1

0.2 equivalent

0.2 equivalent

0.02
0

Formation of
PSB

0.6 equivalent
Absorbance

0.1

0.05
300

400
500
600
Wavelength (nm)

0

700

300

400
500
600
Wavelength (nm)

700

Figure 3-40. UV-vis spectra of ring-lock analog 3-18 with mutant R132K:R111L:F15Y
and mutant R132K:R111L:L121E.
similar spectroscopic changes observed with 11-cis-retinal.

Glu121 is necessary for PSB

formation with analog 3-18 as we compared mutants R132K:R111L:F15Y (without Glu121) and
R132K:R111L:L121E

(Figure

3-40).

As

in

the

case

of

11-cis-retinal,

mutant

R132K:R111L:L121E produces a more red-shifted complex as compared to mutant
R132K:R111L upon binding with 3-18 (Table 3-14). In order to further probe the similarity
between 11-cis-retinal and ring-locked analog 3-18, the pKa of the PSB for both 11-cis-retinal
and analog 3-18 in R132K:L121E and R132K:R111L:L121E have been measured and compared.
The pKa of bound 3-18 PSB in mutant R132K:L121E (pKa ~ 8.7) is lower than the bound 3-18
PSB with R132K:R111L:L121E (pKa ~ 9.7) by about 1 unit (Figure 3-41). The change of pKa is
similar to that observed for bound 11-cis-retinal PSB in CRABPII mutants (Figure 3-21).
To further compare the ring-locked analogs 3-18 with 11-cis-retinal, the binding with
CRABPII mutants has been measured. Like 11-cis-retinal, the ring-locked analog 3-18 binds
243

Table 3-14. UV-vis study of 11-cis-retinal ring-locked analog with CRABPII mutants.
11-cis-retinal

analog 3-18

all-trans-retinal

λmax (nm)

λmax (nm)

λmax ( nm)

R132K:R111L:L121E

462

467

449

R132K:L121E

463

470

457

R132K:R111L:F15Y

360

372

373

R132K:R111L:T54E

358

N.D.

378

Mutants

R132K:R111L:F15Y (Kd ~ 290 nM) as tight as all-trans-retinal (Kd ~ 241 nM). In addition,
R132K:R111L:L121E has a higher affinity with all-trans-retinal than both 11-cis-retinal and the
analog 3-18 (Table 3-15). However, we found that R132K: R111L:F15Y have less affinity to
the ring locked analog as compared with 11-cis-retinal. We have to be aware that 11-cis-retinal
Acid Base Titration of 42
with KE

Acid-base Titration of 42
with KLE
0.15

Absorbance

0.08
0.06

pKa = 8.7
Acidified

0.04

Absorbance

0.1

pKa = 9.7
0.1

Acidified
0.05

0.02
0
250 300 350 400 450 500 550 600
Wavelength (nm)

Figure 3-41.
Acid-base
R132K:R111L:L121E.

titration

of

0
250 300 350 400 450 500 550 600 650
Wavelength (nm)

11-cis-retinal

244

with

R132K:L121E

and

isomerizes in the binding pocket. As discussed in Section 3.2.3.2., 11-cis-retinal is suspected to
bind at different positions within CRABPII depending on the mutant discussed (Figure 3-22).
The extra two methylene in the analog might lead to analog 3-18 bind differently as compared to
11-cis-retinal, thus, leading to different binding constant.

However, based on the UV-vis

spectroscopic study, we believe that the ring-locked analog 3-18 behaves the same way as 11cis-retinal. Analog 3-18 bound with several CRABPII mutants is undergoing crystallographic
screens in an attempt to answer this question unequivocally.

Table 3-15.
mutants.

Binding Study of 11-cis-retinal ring-locked analog 3-18 with CRABPII

11-cis-retinal

analog 42

all-trans-retinal

Kd (nM)

Kd (nM)

Kd (nM)

R132K:R111L:L121E

351 ± 29.8

241 ± 28

1 ± 4.9

R132K:R111L:F15Y

47.0 ± 8.1

290 ± 30

2455 ± 240

Mutants

3.6. Conclusion
Isomerization is an process that triggers a variety of biological processes such as visions,
phototaxis, energy generation, etc. that help organisms survive. Although the actual mechanisms
of isomerization in some of the described processes is solved yet, we would like to apply the de
novo design principle to engineer CRABPII as a 11-cis-retinal isomerase. The criteria to be a
good retinal isomerase candidate included: 1. binds 11-cis-retinal; 2. binds poorly with alltrans-retinal; 3. binds 11-cis-retinal as PSB; 4. shows a distinct absorption spectrum upon
forming protein-chromophore complex, so that the light source used can only isomerize the
bound chromophore but not the free one.

We successfully identified the mutant
245

R132K:R111L:F15Y that binds well with 11-cis-retinal but does not bind well with all-transretinal (Table 3-1 and 3-4). Mutant R132K:R111L:F15Y, although does not form PSB with 11cis-retinal, shows isomerization activity (Table 3-2 and 3-3). This was a steering surprise, our
investigation to uncover the source of this process.

In order to understand the possible

mechanism for such isomerization, we have conducted experiments including the use of
deuterated buffer and modulating the incubation buffer pH for incubation.

Based on our

preliminary results, we propose that an ordered water molecule undergoes Michael addition of
the bound 11-cis-retinal. Isomerization of C11-C12 (via rotation of the single bond) and a retroMicheal results in all-trans-retinal (Figure 3-26).

We have identified that mutant

R132K:R111L:F15Y posses catalytic activity due to the different binding affinity towards 11cis-retinal (Kd ~ 47 nM) and all-trans-retinal (Kd ~ 2 uM). In addition, we also found the
isomerization activity in CRBPII.

In addition for the first time, we are able to see the

isomerization occuring on real time with UV-vis spectroscopy.
Since CRABPII and its mutants already posses isomerization activity in dark, we would
like to move to other template to engineer a retinal isomerase that isomerize the bound retinal
only when specific light irradiates on it. As mentioned in Section 3.2.3., CRALBP binds
specifically with 11-cis-retinal and we have demonstrated that the bound retinal does not
isomerize (Table 3-3). Recently, CRALBP has been crystallized by He and co-workers to 2.14
Å (Figure 3-42).

123

Consequently, CRALBP could be a good candidate for retinal isomerase.

Based on the crystal structure, there are two residues, Phe161 and Met223 that are close to the
bound 11-cis-retinal (Figure 3-42). Replacing theses residue with lysine will result in BürgiDunitz trajectories of 96˚ and 125˚, respectively.

246

Figure 3-42. Crystal structure of cellular retinal binding protein (left) and the 11-cisretinal binding site (right).
3.7. Material and Methods
A) Biology
ii.

Protein Mutagenesis, Bacterial Expression and Purification

i. Protein Mutation on pET-17b vector
The PCR conditions and recipes have been described in detail in Chapter 1 (Material and
Methods).

247

Table 3-16. Primers Sequence for CRABPII Mutation in Chapter 3
Mutant

R132K:R111L:
F15Y:S12D

R132K:R111L:
:F15Y:S37D

R132K:R111L:
L121E:T56V

R132K:R111L:
L121E:T54V

Primers
name

Primers Sequence

BBB478

5’-CACTGCTGGCTTGTCCGC
TGCAGCCACAG-3’

BBB479

5’-CTGTGGCTGCAGCGGAC
AAGCCAGCAGTG-3’

BBB507

5’-CTCGTAGTTTTCATCTCG
GATGATTTTCCAG-3’

BBB508

5’-CTGGAAAATCATCCGAG
ATGAAAACTACGAG-3’

BBB480
BBB481
BB67
BB68

Template

Seq.
No.

R132K:R111L BB
:F15Y
918

R132K:R111L BB
:F15Y
919

5’-CTACATCAAAACCTCCG
TTACCGTGC GCACCACAG-3’ R132K:R111L BB
:F15Y
920
5’-CTGTGGTGCGCACGGTA
ACGGAGGT TTTGATGTAG-3’
5’-CTACATCAAAGTCTCCAC
CACCGTGC G-3’

R132K:R111L
5’-CGCACGGTGGTGGAGACT :L121E
TTGATGT AG-3’

ii. DNA Transformation and Purification
Procedures for DNA transformation, and purification of pET-17b plasmids, that contains
CRABPII mutants, using QIAGEN plasmid purification kit are detailed in Chapter 1 (Materials
and Methods).
DNA sample preparation for DNA sequencing has been detailed in Chapter 1 (Materials
and Methods).
iii. Protein Expression and Purifications
Protein expression and purification of CRABPII mutants (pET-17b vector) are described in
detail in Chapter 1 (Materials and Methods).

248

B. Protein Characterization and Binding Assays
i. Calculation Extinction Coefficients of CRABPII mutants
The absorption extinction coefficients (ε) for CRABPII mutants are determined according
to the method described by Gill and von Hippel. The details have been described in Chapter 1
(Materials and Methods).
ii. Reductive Amination
The procedures for reductive amination and the electrospray mass spectrometry setting
have been fully described in Chapter 1 (Materials and Methods)
iii. UV-vis and Fluorescent Spectroscopic Study
The procedures and experimental setup for UV-vis binding study and fluorescent
quenching experiment has been mentioned in detail in Chapter 1 (Material and Methods)
iv. HPLC study
All studies are performed in dark room with dim red lights unless otherwise specified.
1) Isomerization Study (A single point measurement)
All isomerization study was performed as described here unless specified. In general,
CRABPII mutants (5 µM, 500 µL) in 1 mL microcentrifugal tube were incubated at rt for 30
min. 11-Cis-retinal (0.5 mΜ, 2 μL, 0.2 eq) was added and the reaction mixtures were incubated
at rt for 30 min. The reactions were quenched by addition of hexane (200 µL) and mixed
vigorously by vortex. The mixtures were centrifuged (4000 rpm, 2 min). The top layer (organic)
was collected. The reaction mixtures were extracted with hexane (200 μL) for two times. The
combined organics were dried under nitrogen. The extracts were re-dissolved in hexane (10 μL)

249

and the solutions were analyzed by HPLC with silica column (2% EtOAc in Hexane with 0.05%
Et3N, 1 ml/min, Rainin microsorb-MV

TM

, 4.6 x 250 mm, 5 µm bead size). Absorbance at 360

nm was used for detection.
ii.

Sample Preparation for D2O Experiment
The 500 µL samples of CRABPII mutants at ~50 µM were concentrated to about 50 µL

using Millipore filters. The OD280 of concentrated samples was measured using UV-vis and the
concentration of the sample was calculated based on the absorption extinction coefficient of the
mutants. The concentrated samples were diluted to 5 µM by deuterated PBS buffer (40 mM
NaH2PO4, pH = 7.3) and incubated at rt for 30 min before the experiment.
iv.

Determining the Rate of Isomerization
CRABPII mutants (5.5 mL, 50 µM) were incubated at rt for 30 min. 10 200 µL of hexane

in 1 mL centrifugal tubes were prepared and incubated on ice. 11-cis-retinal (0.5 mM, 11 μL)
was added into the CRABPII sample at rt and mixed gently. Reaction mixture (500 μL) was
removed at t = 30s, 60s, 120s, 300s, 600s, 900s, 1800s, 3000s, 4800, 7200 and mixed with the
prepared hexane vigorously.

The resulting mixtures were incubated at 0

˚C.

Once the

experiment was finished the chromophores of each samples were extracted with hexane and
analyzed according to the procedures mentioned in isomerization study in this Chapter (Materials
and Methods).
vi.

Determining the Rate of SB formation
CRABPII mutants (5 µM, 1 mL) sample in 1 mL quartz cuvette was incubated at rt for 30

min. 11-Cis-retinal (0.5 mM, 2 μL) was added and the measurement was started (UV-vis kinetic

250

study conditions: Temperature = 22 ˚C, total scan time = 25 min, scan rate = every 20 s at 400
nm, integration time = 0.1 s)
B. Chemistry:
General:
All syntheses were carried out in a dry nitrogen atmosphere in the dark room with
minimum red lights unless otherwise specified. All the reaction vessels were flame-dried.
Tetrahydrofuran was distilled over sodium and benzophenone under dried N2 atmosphere.
Reactions were monitored by thin-layer chromatography (TLC, on Merck F254 silica gel 60
aluminum sheets, spots were either visible under light or UV-light (254 mm) or treated with an
oxidizing solution (KMnO4 stain). Column chromatography was performed with Silicycle silica
gel 60.
1

H-NMR spectra were recorded on a Varian Unity+ 500 spectrometer with deuterated

chloroform (CDCl3; δ = 7.24 ppm) or deuterated DMSB (DMSO-d6; δ = 2.5 ppm, br) as an
1

internal standard. H noise-decoupled

13

C spectra were recorded on a Varian Unity+ 500

spectrometer at 125 MHz with deuterated chloroform (CDCl3; δ = 77 ppm) or deuterated DMSΟ
(DMSO-d6; δ= 39.5 ppm) as an internal standard.

251

Synthesis of diethyl 3-(trimethylsilyl)prop-2-ynylphosphonate
1) NaHMDS, 0 ˚C
HP(O)(OEt)2

rt, THF

2) TMS
0 ˚C

Br
rt, THF

O
EtO P
EtO

TMS

To a stirred solution of diethyl phosphonate (1.99 g, 14.4 mmol) in THF (100 mL),
NaHDMS (14.4 mL, 14.4 mmol, 1M in toluene) was added dropwise at 0 ˚C. The reaction was
stirred at rt for 30 min. The mixture was cooled down to 0 ˚C and acetylene (2.5 g, 13.1 mmol)
was added dropwise. The mixture was further stirred at rt for 6h. Reaction was quenched by
addition of saturated ammonium chloride solution and organic layer was separated. The aqueous
layer was extracted with ethyl acetate for three times. The organics were washed with brine and
dried over anhydrous MgSO4. The crude was evaporated under vacuo yielding 80% of final
ylide 2 (2.6 g, 10.5 mmol).
1

H-NMR (CDCl3, 500 MHz): δ = 0.13 (s, 9H), 1.33 (td, J = 7 Hz, 0.5 Hz, 6H), 2.78 (d, J =

22.5 Hz, 2H), 4.17 (m, 6H)
13

C HNMR (CDCl3, 125 MHz): δ = 95.7 (d, J = 13.9 Hz), 87.8 (d, J = 8.8 Hz), 63.1 (d, J =

6.8 Hz), 19.8 (d, J = 19.1 Hz), 16.3 (d, J = 7.5 Hz), -0.22
31

P NMR (CDCl3, MHz): δ = 98.6

252

Synthesis of trimethyl((5E)-4-methyl-6-(2,6,6-trimethylcyclohex-1-enyl) hexa-3,5-dien-1ynyl)silane (3-11)

O
EtO P
EtO

TMS

1. nBuLi, 0 ˚C

TMS
RT, THF;

2.

3-11
O
3-10
0 ˚C

rt, overnight

To a stirred solution of ylide (5.2 g, 20.9 mmol) in THF (50 mL), nBuLi (9.5 mL, 20.9
mmol, 2.2 M) was added dropwise at 0 ˚C. The mixture was stirred at rt for 1h. β-ionone (2.68
g, 14.0 mmol) dissolved in THF (10 mL) was added dropwise at 0 ˚C. The reaction was stirred
at rt for 6 h. The reaction was quenched by addition of saturated ammonium chloride. The
organic layer was separated and the aqueous layer was extracted by ethyl acetate for three times.
The organics were washed with brine and was dried over anhydrous MgSO4. The final product
3-14 was purified by column chromatography (20% EtOAc: Hex: 0.05% Et3N) affording 65%
yield (2.6 g, 9.1 mmol) in a mixture of isomer (cis: trans/ 1:4).
trimethyl((5E)-4-methyl-6-(2,6,6-trimethylcyclohex-1-enyl) hexa-3,5-dien-1-ynyl)silane in
E/Z mixture:
1

H-NMR (CDCl3, 500 MHz): δ = 0.17 (s, 9H), 0.19 (s, 9H), 0.98 (s, 6H), 1.06 (s, 6H) 0.45

(m, 2H), 1.53 (s, 3H), 1.60 (m, 2H), 1.66 (d, J = 0.5 Hz, 3H), 1.91 (d, J = 1Hz, 3H), 1.98 (t, J = 6
Hz, 2H), 2.04 (s, J = 1 Hz, 3H), 2.04 (m, 2H), 5.34 (s, 1H), 5.41 (d, J = 0.5 Hz, 1H), 6.05 (d, J =
16 Hz, 1H), 6.24 (d, J = 16 Hz, 1H), 6.34 (d, J = 16 Hz, 1H), 6.88 (d, J = 16 Hz, 1H)

253

13

C HNMR (CDCl3, 125 MHz): δ = 148.8, 148.1, 137.4, 137.1, 135.4, 131.7, 131.1, 130.4,

130.1, 108.4, 106.9, 104.0, 103.3, 101.2, 44.0, 39.5, 34.2, 33.5, 33.0, 29.0, 28.9, 21.8, 21.6, 19.2,
19.1, 15.1, 0.1
Synthesis of 1,3,3-trimethyl-2-((1E)-3-methylhexa-1,3-dien-5-ynyl)cyclohex- 1-ene

TMS

3-11

H
TBAF, THF,
0 ˚C

3-12

TMS-Acetylene 3-11 (2.2 g, 7.7 mmol) in THF (50 mL) was treated with TBAF (1M, 30.8
mmol) and stirred at rt for 2h. When all the starting material was consumed, the reaction was
quenched by addition of sat. NH4Cl. The organics were isolated and the aqueous layer was
extracted with diethyl ether three times. The combined organics was concentrated under vacuo.
The product was further purified by column chromatography with hexane, yielding 85% of
acetylene 3-12 (1.4 g, 6.5 mmol, cis:trans/ 1:4)
E:Z mixture of 3-12:
1

H-NMR (CDCl3, 500 MHz): δ = 0.99 (s, 6H), 1.02 (s, 6H), 1.43-1.47 (m, 2H), 1.56-1.63

(m, 2H), 1.67 (m, 3H), 1.73 (s, 3H), 1.93 (m, 3), 1.98 (t, J = 6 Hz, 2H), 2.01 (t, J = 6Hz, 2H),
2.06(m, 3H), 3.14 (d, J = 2 Hz, 1H), 3.26 (d, J= 3Hz, 1H), 5.31 (s, 1H), 5.38 (m, 1H), 6.06 (d, J
= 14 Hz, 1H), 6.26 (d, J = 16 Hz, 1H), 6.32 (d, J = 16 Hz, 1H), 6.79 (d, J = 16 Hz, 1H)
13

C HNMR (CDCl3, 125 MHz): δ = 148.8, 148.1, 137.4, 137.1, 135.4, 131.7, 131.1, 130.4,

130.1, 108.4, 106.9, 104.0, 103.3, 101.2, 44.0, 39.5, 34.2, 33.5, 33.0, 29.0, 28.9, 21.8, 21.6, 19.2,
19.1, 15.1, 0.1

254

Synthesis of (E)-3-iodobut-2-en-1-ol
OH

1. Cp2TiCl2, i-BuMgBr, Et2O, 0 ˚C

OH
I

2. I2

i-BuMgCl (68.6 mmol, 2.4 eq, 1.25M, 54.8 mL) were added to 5% mol of CpTiCl2 (1.43
mmol, 349 mg) in THF (50 mL) at 0 ˚C under Argon. The reaction was stirred for 1h at 0 ˚C.
Alcohol 23 (2.0 g, 28.6 mmol) was added dropwisely at 0 ˚C. And the reaction was warmed up
to rt and stirred for an addition of 6h. The reaction was then quenched with 1.65 equivalent of I2
(11.96 g, 47 mmol) in 200 mL of THF at 0 ˚C affording 54% of vinyl iodide 22 (3.05 g, 37
mmol)
1

H-NMR (CDCl3, 500 MHz): δ = 2.44 (m, 3H), 2.02 (s, 1H), 4.07 (t, J= 6 Hz, 2H), 6.39 (t,

J = 6.5 Hz, 1H)
Synthesis of (E)-tert-butyl(3-iodobut-2-enyloxy)dimethylsilane

OH
I

TBSCl, Imidazole,
0 ˚C, DMF

OTBS
I
˚

The alcohol (2.8 g, 34 mmol) was dissolved in dried DMF (25 mL) and stirred at 0 C.
TBSCl (7.65 g, 51 mmol) and Imidazole (4.63 g, 68 mmol) were added. The reaction was stirred
at rt overnight. The reaction was quenched by addition of water and diethyl ether. The organics
were isolated and the aqueous layer was extracted with diethyl ether for 3 times. The combined

255

organics was washed with sat. NaCl and concentrated under vacuo. The crude was further
purified by column chromatography (EtOAc:Hex/ 1:99 with 0.05% Et3N)
1

H-NMR (CDCl3, 500 MHz): δ = 6.25 (t, J = 6.04 Hz, 1H), 4.08 (d, J = 6.59 Hz, 2H),

2.37 (s, 3H), 0.86 (s, 9H), 0.032 (s, 6H)
13

C HNMR (CDCl3, 125 MHz): δ = 140.32, 95.72, 60.40, 27.86, 25.65, 18.06, -5.42

Synthesis

of

tert-butyl(((2E,8E)-3,7-dimethyl-9-(2,6,6-trimethylcyclohex-1-en-1-yl)nona-

2,6,8-trien-4-yn-1-yl)oxy)dimethylsilane (3-13)
1. Pd(PPh3)4 (0.1 mol%),
rt, i-PrNH2
OTBS
I

2. CuI (0.1 mol %)
3.

OTBS
H

3-13

3-12

To a solution of vinyl iodide (268 mg, 0.86 mmol) in iPrNH2 (3 mL),
tetrakis(triphenylphosphine)palladium (16.4 mg, 0.014 mmol) was then added. The solution was
stirred at room temperature under Ar for 5 min. CuI (2.8 mg, 0.014 mmol) was added and stirred
for another of 5 min, followed by the addition of acetylene 3-12 (168 mg, 0.71 mmol) in iPrNH2
(1 mL). The reaction mixture was stirred for an addition of 5 h until TLC showed that the
starting material was consumed. The reaction was quenched by the removal of iPrNH2 under
vacuo. The residue was dissolved in Et2O and extracted with aqueous NH4Cl. The Organics
were washed with saturated NaCl, and dried over anhydrous Na2SO4 and the solvent was

256

removed by rotary evaporation.

The product was purified by column chromatography

(EtOAc/Hexane, 1:99) yielding 3-13 (250 mg, 0.63 mmol, 89%, E:Z/ 3.27:1)
E:Z mixture:
1

H-NMR (CDCl3, 500 MHz): δ = 0.057 (s, 6H), 0.062 (s, 6H), 0.885 (s, 9H), 0.889 (s,

9H), 0.995 (s, 6H), 1.05 (s, 6H), 1.45 (m, 2H), 1.59 (m, 2H), 1.67 (s, 3H), 1.76 (s, 3H), 1.81 (s,
3H), 1.82 (d, J = 1 Hz, 3H), 1.93 (d, J = 1.5Hz, 3H), 1.99 (t, J = Hz, 2H), 2.03 (s, 3H), 4.1 (d, J
= 6.5 Hz, 2H), 4.26 (d, J = 6.5 Hz, 2H), 5.43 (s, 1H), 5.50 (s, 1H), 5.89 (dt, J = 1.5 Hz, 6.5 Hz,
1H), 5.90 (dt, J = 1Hz, 6.5 Hz, 1H), 6.08 (d, J = 15.5 Hz, 1H), 6.22 (d, J = 16Hz, 1H), 6.30 (d, J
= Hz, 1H), 6.82 (d, J = 16.5 Hz, 1H)
13

C HNMR (CDCl3, 125 MHz): δ = -5.23, -5.17, -5.15, -5.11, 15.01, 17.65, 17.70, 18.34,

19.13, 19.19, 19.21, 21.64, 21.81, 25.86, 25.92, 25.98, 28.06, 28.90, 29.01, 33.37, 34.17, 34.21,
39.57, 39.91, 60.10, 60.66, 86.13, 98.81, 107.15, 108.64, 109.75, 119.34, 129.25, 129.93, 129.99,
131.04, 131.45, 135.55, 135.94, 135.98, 137.32, 140.65, 146.01, 147.04
Synthesis

of

trimethyl((5E)-4-methyl-6-(2,6,6-trimethylcyclohex-1-enyl)hexa-3,5-dien-1-

ynyl)silane (3-14)

OTBS

OH

TBAF,
3-13

THF, 0 ˚C

3-14

TBS protected acetylene 3-13 (150 mg, 0.38 mmol) was dissolved in distilled THF (20
mL), followed by an addition of TBAF (1 M, 1.9 mL, 1.9 mmol) dropwise. The reaction mixture
was stirred for 2 h at 0 ˚C and monitored by TLC (Hex/ EtOAc, 9:1). The reaction was

257

quenched by addition of water and extracted by Et2O for three times. The organics were
combined and dried over anhydrous Na2SO4. The crude was concentrated under vacuo and
further purified by column chromatography (Hex/EtOAc, 8:2 with 0.05% Et3N) yielding
acetylene 3-14 (98 mg, 0.30 mmol, 90% yield, E:Z/ 3.45:1). E-acetylene 3-14 was purified by
HPLC using semi-prep Zorbox Column (22.5mm x 0.943 mm, Hex/EtOAc, 9:1 with 0.05%
Et3N), recovering 90% of overall acetylene.
A Zorbox semi-prep column was used (22.5mm x 0.943mm).

A mobile phase

(Hex/EtOAc, 90:10 with 0.05% Et3N) was employed with a flow rate of 2 mL/min. The peak
was detected at 317 nm. A solution of 0.5 mg/ul was prepared and 5 μl was injected for each
run. The fraction at 23 min (Z-isomer) and 25 min (E-isomer) was collected and was
concentrated under vacuum and further analyzed by H-NMR.
1

H-NMR (CDCl3, 500 MHz): δ = 0.994 (s, 6H), 1.047 (s, 6H), 1.25 (br, 1H), 1.45 (m, 2H),

1.59 (m, 2H), 1.67 (d, J = 1Hz, 3H), 1.75 (d, J = 1Hz, 3H), 1.85 (t, J = 1Hz, 3H), 1.87 (t, J =
1Hz, 3H), 1.93 (d, J = 1.5Hz, 3H), 1.99 (t, J = 6Hz, 2H), 2.03 (d, J = 1Hz, 3H), 4.23 (d, J = 5.5
Hz, 2H), 4.24 (d, J = 5.5Hz, 2H), 5.43 (s, 1H), 5.50 (s, 1H), 5.94 (td, J = 6.5Hz, 1.5Hz, 1H), 5.99
(td, J = 6.5Hz, 1.5Hz, 1H), 6.08 (d, J = 16Hz, 1H), 6.23 (d, J = 16.5Hz, 1H), 6.32 (d, J = 16.5Hz,
1H), 6.79 (d, J = 16.5Hz, 1H)
E-acetylene 3-14:
H-NMR (CDCl3, 500 MHz): δ = 1.047 (s, 6H), 1.25 (br, 1H), 1.45 (m, 2H), 1.59 (m, 2H),
1.85 (t, J = 1Hz, 3H), 1.87 (t, J = 1Hz, 3H), 1.99 (t, J = 6Hz, 2H), 2.03 (d, J = 1Hz, 3H), 4.23 (d,

258

J = 5.5 Hz, 2H), 5.50 (s, 1H), 5.99 (td, J = 6.5Hz, 1.5Hz, 1H), 6.23 (d, J = 16.5Hz, 1H), 6.32 (d,
J = 16.5Hz, 1H)
13

C HNMR (CDCl3, 125 MHz): δ = 15.05, 17.67, 19.20, 21.65, 28.90, 33.04, 34.22, 39.57,

59, 27, 86.95, 98.39, 108.40, 121.60, 129.54, 130.08, 134.33, 135.45, 137.45, 147.45
Synthesis of 11-cis-retinol

OH
Activated Zn,
3-14

i-PrOH:H2O / 2:1

3-15

OH

To a stirred solution of activated Zn (2 g) in MeOH/H2O (4 mL, 1:1), acetylene 3-14 (12
mg, 0.04 mmol) in 1 mL MeOH was added. The reaction mixture was stirred under Ar for 16 h.
The Zn was filtered by a pad of celite. The filtrate was extracted with Et2O two times. The
organics were dried over anhydrous Na2SO4 and concentrated under vacuo yielding (7 mg, 0.024
mmol, 61%). The crude was used in the next reaction without further purification.
1

H-NMR (CDCl3, 500 MHz): δ = 0.99 (s, 6H), 1.02 (s, 6H), 1.43-1.47 (m, 2H), 1.56-1.63

(m, 2H), 1.67 (m, 3H), 1.73 (s, 3H), 1.93 (m, 3), 1.98 (t, J = 6 Hz, 2H), 2.01 (t, J = 6Hz, 2H),
2.06(m, 3H), 3.14 (d, J = 2 Hz, 1H), 3.26 (d, J= 3Hz, 1H), 5.31 (s, 1H), 5.38 (m, 1H), 6.06 (d, J
= 14 Hz, 1H), 6.26 (d, J = 16 Hz, 1H), 6.32 (d, J = 16 Hz, 1H), 6.79 (d, J = 16 Hz, 1H)

259

Synthesis of 11-cis-retinal

MnO2, CH2Cl2

3-15

3-16

OH

O

The crude product from previous reaction was dissolved in CH2Cl2 (10 mL) under N2 at 0

˚C. The reaction mixture was stirred for 2h and was monitored by TLC.

MnO2 was filtered by a

pad of celite. The filtrate was dried over anhydrous Na2SO4 and concentrated under vacuo. The
crude was passed through a short silica column, eluting with Hex/EtOAc, 9:1 with 0.05% Et3N.
The product was further purified by HPLC using a semi-prep HPLC column (Zorbax, 0.92 mm x
22.5 mm, eluting with Hex/EtOAC, 95:5 with 0.05% Et3N) yielding 0.5 mg final product (7.7%
yield)
A Zorbox semi-prep column was used (22.5 mm x 0.943 mm). A mobile phase (Hex/
EtOAc, 97:3 with 0.05% Et3N) was employed with a flow rate of 2 mL/min. The column was
pre-treated with the mobile phase for 5 column volume. The peak was detected at 380 nm. All
the procedure were camed out under dim red light. A solution of 0.5 mg/ul was prepared and 5
μl was injected for each run. The fraction at 18 min (by-product) and 24 min (11-cis retinal) was
collected and was immediately put it under 0 ˚C. The solution was dried under vacuo on ice.
The reaction was analyzed by H-NMR (C6D6). It was found that the product would be best kept

260

it at -78 ˚C in benzene. Therefore, the product was dissolved in benzene and a solution of 3.24 x
-3

10 M was prepared and was kept it at -78 ˚C.
Pure 11-cis-retinal:
1

H-NMR (d6-benzene, 500 MHz): δ = 1.07 (s, 6H), 1.44 (m, 2H), 1.56 (m, 2H), 1.68 (s,

3H), 1.74 (d, J = 1Hz, 3H), 1.76 (d, J = 1.5Hz, 3H), 1.91 (t, J = 6.5Hz, 2H), 5.58 (d, J = 11Hz,
1H), 6.10 (d, J = 7Hz, 1H), 6.22 (d, J = 16Hz, 1H), 6.34 (d, J = 15.5Hz, 1H), 6.38 (dd, J = 12Hz,
1H), 6.57 (d, J = 12.5Hz, 1H), 9.90 (d, J = 8 Hz, 1H)
The synthesis of 3-methoxy-2-cyclohexen-1-one (3-20)

OMe

O
MeO
O

O

OMe

HCl, MeOH

3-19

OMe
3-20

Trimethyl orthoformate (5.3 g, 50 mmol) and a catalytic quantity of 37% HCl (0.1 mL)
were added successively to a well-stirred solution of 1,3-cyclohexanedione (3.75 g, 34.5 mmol)
in methanol (50 mL), in which the air had been completely replaced by nitrogen. The yellowish
reaction mixture was stirred at rt for 12 h. The resulting orange-red solution was concentrated
under vacuo. The final product was purified by vacuo distillation (b.p. 80˚C at 50 mTorr) in
57% yield (2.5 g, 19.8 mmol).
1

H NMR (CDCl3, 500 MHz): δ = 5.34 (s, 1 H), 3.66 (s, 3 H), 2.77 (t, J = 6.5 Hz, 2 H), 2.32

(t, J = 6.5 Hz, 2H), 1.96 (m, 2H)

261

Synthesis

of

(E)-6-(2-hydroxy-4-(2,6,6-trimethylcyclohex-1-enyl)but-3-en-2-yl)-3-

methoxycyclohex-2-enone (3-21)

O

O

1. LDA, -78 ˚C, THF
OH
OMe

3-20

2.

OMe
3-21

O

n-BuLi (6.8 mL, 2 M in hexane, 12 mmol) was added gradually to a stirred solution of
(1.84 mL, 12 mmol) of diisopropylamine distilled THF (10 mL) at –78 ˚C. The reaction mixture
was stirred at rt for 30 min. 3-Methoxy-2-cyclohexen-1-one (1.5 g, 12 mmol) distilled THF (10
mL) was added dropwisely to the stirred reaction mixture at –78 ˚C. The reaction mixture was
stirred at rt for an additional 30 min. β-Ionone (2.3 g, 12 mmol) in distilled THF (10 mL) was
added dropwise and the reaction was stirred at 0 ˚C for 3h. The reaction was then cooled down
to –60 ˚C. The reaction was quenched by addition of acetic acid (0.792 g, 13.2 mmol). The
mixture was warmed to rt and 20 mL of PBS buffer at pH 7.3 was added. The organic layer was
separated and the aqueous layer was extracted with ether (20 mL), three times. The combined
organics were washed with brine and dried over anhydrous MgSO4. The crude was concentrated
under vacuo.

The crude aldol was purified by column chromatography (50% Et2O/pet

Ether/0.05% Et3N as eluent) yielding pure 3-21 (2 g, 6 mmol) as a mixture of isomers.

262

1

H NMR (CDCl3, 500 MHz): δ = 6.11 (d, J = 16 Hz, 1H), 5.66 (s, 1H), 5.46 (d, J = 16 Hz,

1H), 5.36 (s, 1H), 3.69 (s, 3H), 2.3-2.6 (m, 3H), 2.0-2.1(m, 1H), 1.9-2.0 (m, 2H), 1.66 (s, 3H),
1.5-1.7 (m, 3H), 1.4-1.5 (m, 2H), 1.33 (s, 3H), 0.98 (s, 6H)
Synthesis

of

(E)-6-(2-hydroxy-4-(2,6,6-trimethylcyclohex-1-enyl)but-3-en-2-yl)-3-

methoxycyclohex-2-enol (3-26)

O

OH

LiAlH4, 0 ˚C

OH

OH

THF
OMe

OMe
3-26

3-21

To a stirred solution of LiAlH4 (0.428 g, 10.62 mmol) in dried ether (10 mL), methoxy
enone 3 (1.68 g, 5.31 mmol) dissolved in dried ether (10 mL) was added dropwise at 0 ˚C. The
reaction mixture was gradually heated to rt for 1 h. The reaction was further stirred for 1 h at rt.
The mixture was then cooled down to 0 ˚C and 2 mL of H2SO4/ice-water (1:7) was added
dropwise. The organic layer was separated and the aqueous layer was extracted with ether (10
mL) for three times. The organics were washed with brine and dried over anhydrous MgSO4.
The crude was concentrated under vacuo and the product was purified by column
chromatography (50% ether/pet. ether/ 0.05% Et3N) affording 0.5 g crude diol 3 in a mixture of
isomers

263

Synthesis of (E)-4-(2-hydroxy-4-(2,6,6-trimethylcyclohex-1-enyl)but-3-en-2-yl)cyclohex-2enone (3-22)

OH
H2SO4/ H2O (1:7)
OH
3-26

OH
OMe Et2O

3-22

O

Crude Diol 3-26 was dissolved in ether (10 mL) and 2 mL of H2SO4/H2O (1:7). The
mixture was stirred for 2 h at 0 ˚C. The organic layer was separated and the aqueous layer was
extracted with Et2O (5 mL), three times. The combined organics were washed with brine and
was dried over anhydrous MgSO4. The crude was concentrated under vacuo and purified by
column chromotagraphy (50% Et2O/ pet. ether/ 0.05% Et3N as an eluent) yielding of enone 3-22
(0.3 g, 1.1 mmol) as a mixture of isomer (1:1.63)
1

H NMR (CDCl3, 500 MHz): δ = 7.14 (dt, J = 10.5 Hz, 2 Hz, 1H), 7.10 (dt, J = 10.5 Hz, 2

Hz, 1H), 6.14 (dd, J = 16 Hz, 1 Hz, 1H), 6.11 (dd, J = 16 Hz, 1 Hz, 1H), 6.06 (m, 1H), 6.03 (m,
1H), 5.51 (d, J = 16 Hz, 1H), 5.46 (d, J = 16 Hz, 1H), 2.4-2.6 (m, 3H), 1.7-2.0 (m, 4H), 1.64 (s,
3H), 1.60 (s, 3H), 1.5-1.7 (m, 3H), 1.4-1.5 (m, 2H), 1.41 (s, 3H), 1.36 (s, 3H), 0.94(s, 6H)
Synthesis

of

(E)-4-((E)-4-(2,6,6-trimethylcyclohex-1-enyl)but-3-en-2-ylidene)cyclohex-2-

enone (3-23)

1. Ac2O, DMAP, toluene
OH

2. DBN, reflux

O

O

3-23

3-22

264

To a solution of enone 3-22 (0.6 g, 2.1 mmol) in dried toluene (50 mL), Ac2O (0.428 g, 4.2
mmol) and 4-(dimethylamino)pyridine (0.31 g, 2.6 mmol) were added. The reaction was stirred
under N2 for 2 days at rt. After 2 days, 1,5-diazabicyclo[4.3.0]non-5-ene (0.525 g, 4.2 mmol)
was added and the solution was refluxed for an additional 2 h. The mixture was then cooled
down to rt and 20 mL water was added. The organic layer was separated. The aqueous layer
was further extracted with 20 mL ether, three times. The organics were washed with brine and
dried over anhydrous MgSO4. The crude was concentrated under vacuo and was further purified
by column chromatography (30% ether/ pet. ether/ 0.05% Et3N) to yield tetraenone 3-23 (300
mg, 1.1 mmol) as a mixture of isomer.
1

H NMR (CDCl3, 500 MHz): δ = 7.70 (d, J = 10 Hz, 1H), 7.61 (d, J = 10 Hz, 1H), 6.68 (d,

J = 16 Hz, 1H) 6.62 (d, J = 16 Hz, 1H), 6.49 (d, J = 16 Hz, 1H), 6.36 (d, J = 16 Hz, 1H), 5.93 (d,
J = 10 Hz, 1H), 5.89 (d, J = 10 Hz, 1H), 2.88 (t, J = 7 Hz, 2H), 2.85 (t, J = 7 Hz, 2H), 2.54 (t, J =
7 Hz, 2H), 2.52 (t, J = 7 Hz, 2H), 2.08 (s, 3H), 2.05 (s, 3H), 2.04 (t, J = 6 Hz, 2H), 1.74 (s, 2H),
1.74 (s, 3H), 1.64 (m, 2H), 1.48 (m, 2H), 1.05 (s, 6H), 1.04 (s, 6H)
Synthesis

of

ethyl

2-((E)-4-((E)-4-(2,6,6-trimethylcyclohex-1-enyl)but-3-en-2-ylidene)

cyclohex-2-enylidene)acetate (3-24)
1. (EtO)2P(O)CH2CO2Et,
NaH, THF
O
3-24

3-23

265

CO2Et

To a solution of stirred NaH (62 mg, 1.5 mmol, 60% in mineral oil) in THF (30 mL), ethyl
2-(diethoxyphosphoryl)acetate (332 mg, 1.5 mmol) was added dropwise at 0 ˚C. The mixture
was then stirred at rt for 30 min. Tetraenone 3-23 (200 mg, 0.75 mmol) dissolved in THF (10
mL) was added dropwise to the mixture at 0 ˚C. The mixture was stirred at rt for an additional 6
h. The reaction was quenched by the addition of ice water. The organic layer was separated.
The aqueous layer was extracted with ether, three times. The organics were washed with brine
and dried over anhydrous MgSO4. The crude was concentrated under vacuo and purified by
column chromatography (20% ether/ pet. ether/ 0.05% Et3N) giving ester 3-24 (180 mg, 0.53
mmol) as a mixture of isomer.
1

H NMR (CDCl3, 500 MHz): δ = 7.70 (d, J = 10 Hz, 1H), 7.61 (d, J = 10 Hz, 1H), 6.68 (d,

J = 16 Hz, 1H) 6.62 (d, J = 16 Hz, 1H), 6.49 (d, J = 16 Hz, 1H), 6.36 (d, J = 16 Hz, 1H), 5.93 (d,
J = 10 Hz, 1H), 5.89 (d, J = 10 Hz, 1H), 2.88 (t, J = 7 Hz, 2H), 2.85 (t, J = 7 Hz, 2H), 2.54 (t, J =
7 Hz, 2H), 2.52 (t, J = 7 Hz, 2H), 2.08 (s, 3H), 2.05 (s, 3H), 2.04 (t, J = 6 Hz, 2H), 1.74 (s, 2H),
1.74 (s, 3H), 1.64 (m, 2H), 1.48 (m, 2H), 1.05 (s, 6H), 1.04 (s, 6H)
Synthesis of 2-((E)-4-((E)-4-(2,6,6-trimethylcyclohex-1-enyl)but-3-en-2-ylidene)cyclohex-2enylidene)ethanol (3-27)
DIBAL-H, 0 ˚C,
THF

3-24

3-27

CO2Et

266

OH

Ester 6 (100 mg, 0.29 mmol) was dissolved in THF (20 mL) and was stirred at 0 ˚C.
DIBAL-H (0.235 mL, 1M in toluene) was added dropwise. The mixture was stirred at 0 ˚C for 2
h. The reaction was quenched by addition of sat. NH4Cl and ether. The organic layer was
separated. The aqueous layer was extracted with ether, three times. The organics were washed
with brine and dried over anhydrous MgSO4. The crude was concentrated under vacuo. The
final product was purified by column chromatography (30% Et2O/ pet. ether/ 0.05% Et3N)
affording of final product 3-27 (80 mg, 0.27 mmol) as a mixture of isomer.
1

H NMR (300MHz, CDCl3): δ = 1.00 (s, 6H), 1.4-1.7(m, 4H), 1.71(s, 3H), 2.0(m, 5H),

2.4-2.6(m, 4H), 4.1(m, 2H), 5.4-5.6(t, 1H), 6.0-6.4(m, 2H), 6.5-6.8(m, 2H).
Synthesis of 2-((E)-4-((E)-4-(2,6,6-trimethylcyclohex-1-en-1-yl)but-3-en-2-ylidene)cyclohex2-en-1-ylidene)acetaldehyde

MnO2, CH2Cl2,
0˚C

3-27

3-25

OH

O

To a stirred solution of 11-cis ring locked retinol 3-27 (60 mg, 0.2 mmol), MnO2 was
added at 0 ˚C. The reaction mixture was stirred for 4 h. The mixture was filtered through a celite
pad and the crude was concentrated under vacuo.

The crude was purified by column

chromatography yielding product 3-25 (50 mg, 0.169 mmol) as a mixture of isomer. The desired
isomer was obtained by HPLC purification (2% Et2O/ hexane as eluent) on silica column.

267

A Zorbox semi-prep column was used (22.5 mm x 0.943 mm). A mobile phase (Hex/
EtOAc, 97:3 with 0.05% Et3N) was employed with a flow rate of 2 mL/min. The column was
pre-equilibrated with the mobile phase for 5 column volume. The peak was detected at 380 nm.
All the procedure was carried under dim red light. A solution of 0.5 mg/ul was prepared and 5
μL was injected at each run. The fraction at 23 min (11-cis ring locked retinal analog) was
collected and stored on ice. The fractions were concentrated under vacuum and further analyzed
˚

by H-NMR (C6D6). It was found that the product would be best kept it at –78 C in benzene.
-3

Therefore, the product was dissolved in benzene and a solution of 3.24 x 10 M was prepared
and kept at –78 ˚C.
1

H NMR (CDCl3, 500 MHz): δ = 10. 07 (d, J = 8 Hz, 1H), 7.04 (d, J = 9 Hz, 1H), 6.62 (d,

J = 16 Hz, 1H), 6.42 (d, J = 16 Hz, 1H), 6.25(d, J = 10 Hz, 1H), 5.89 (d, J = 8 Hz), 2.98 (t, J = 6
Hz, 2H), 2.70 (t, J = 7 Hz, 2H), 2.19 (s, 3H), 2.06 (t, J = 6 Hz, 1H), 2.04 (s, 3H), 1.76 (s, 3H),
1.63 (m, 2H), 1.58 (s, 3H), 1.50 (m, 2H), 1.06 (s, 3H)
13

C NMR (CDCl3, 500 MHz): δ = 190.6, 156.1, 138.2, 135.6, 134.6, 131.4, 130.9, 130.4,

130.1, 128.6, 125.4, 39.6, 34.3, 33.2, 29.7, 29.0, 24.8, 24.6, 21.9, 19.2, 13.7

268

BIBLIOGRAPHY

269

Bibliography
(1)
Dugave, C. cis-trans Isomerization in Biochemistry; Wiley-VCH Verlag GmbG &
Co. KGaA: Weinheim, 2006.
(2)
Ferreri, C.; Kratzsch, S.; Landi, L.; Brede, O. Cellular and Molecular Life
Sciences 2005, 62, 834.
(3)
Hoff, W. D.; Jung, K. H.; Spudich, J. L. Annual Review of Biophysics and
Biomolecular Structure 1997, 26, 223.
(4)

Keweloh, H.; Heipieper, H. J. Lipids 1996, 31, 129.

(5)
Kort, R.; Hoff, W. D.; VanWest, M.; Kroon, A. R.; Hoffer, S. M.; Vlieg, K. H.;
Crielaard, W.; VanBeeumen, J. J.; Hellingwerf, K. J. Embo Journal 1996, 15, 3209.
(6)
Quail, P. H.; Boylan, M. T.; Parks, B. M.; Short, T. W.; Xu, Y.; Wagner, D.
Science 1995, 268, 675.
(7)
Van der Horst, M. A.; Hellingwerf, K. J. Accounts of Chemical Research 2004,
37, 13.
(8)

Weber, F. J.; Isken, S.; Debont, J. A. M. Microbiology-Uk 1994, 140, 2013.

(9)

Lange, S. J.; Que, L. Current Opinion in Chemical Biology 1998, 2, 159.

(10)
421.

Fox, B. G.; Lyle, K. S.; Rogge, C. E. Accounts of Chemical Research 2004, 37,

(11)
19.

Ferreri, C.; Panagiotaki, M.; Chatgilialoglu, C. Molecular Biotechnology 2007, 37,

(12) Guckert, J. B.; Ringelberg, D. B.; White, D. C. Canadian Journal of Microbiology
1987, 33, 748.
(13) Guckert, J. B.; Hood, M. A.; White, D. C. Applied and Environmental Microbiology
1986, 52, 794.
(14) Diefenbach, R.; Heipieper, H. J.; Keweloh, H. Applied Microbiology and
Biotechnology 1992, 38, 382.
(15) Heipieper, H. J.; Diefenbach, R.; Keweloh, H. Applied and Environmental
Microbiology 1992, 58, 1847.
(16) Heipieper, H. J.; Loffeld, B.; Keweloh, H.; Debont, J. A. M. Chemosphere 1995,
30, 1041.

270

(17) Okuyama, H.; Sasaki, S.; Higashi, S.; Murata, N. Journal of Bacteriology 1990,
172, 3515.
(18) Okuyama, H.; Okajima, N.; Sasaki, S.; Higashi, S.; Murata, N. Biochimica Et
Biophysica Acta 1991, 1084, 13.
(19) Heipieper, H. J.; Weber, F. J.; Sikkema, J.; Keweloh, H.; Debont, J. A. M. Trends
in Biotechnology 1994, 12, 409.
(20) Heipieper, H. J.; Meulenbeld, G.; VanOirschot, Q.; DeBont, J. A. M. Applied and
Environmental Microbiology 1996, 62, 2773.
(21)

Seelig, J.; Waespesarcevic, N. Biochemistry 1978, 17, 3310.

(22) Heipieper, H. J.; de Waard, P.; van der Meer, P.; Killian, J. A.; Isken, S.; de Bont,
J. A. M.; Eggink, G.; de Wolf, F. A. Applied Microbiology and Biotechnology 2001, 57,
541.
(23)
1.

Heipieper, H. J.; Meinhardt, F.; Segura, A. Fems Microbiology Letters 2003, 229,

(24) Ferreri, C.; Costantino, C.; Perrotta, L.; Landi, L.; Mulazzani, Q. G.;
Chatgilialoglu, C. Journal of the American Chemical Society 2001, 123, 4459.
(25) Chatgilialoglu, C.; Zambonin, L.; Altieri, A.; Ferreri, C.; Mulazzani, Q. G.; Landi, L.
Free Radical Biology and Medicine 2002, 33, 1681.
(26) Ferreri, C.; Samadi, A.; Sassatelli, F.; Landi, L.; Chatgilialoglu, C. Journal of the
American Chemical Society 2004, 126, 1063.
(27) Chatgilialoglu, C.; Ferreri, C.; Ballestri, M.; Mulazzani, Q. G.; Landi, L. Journal of
the American Chemical Society 2000, 122, 4593.
(28)

Sakamoto, K.; Nagatani, A. Plant Journal 1996, 10, 859.

(29)

Manabe, K.; Nakazawa, M. Journal of Plant Research 1997, 110, 109.

(30) Wagner, J. R.; Zhang, J. R.; Brunzelle, J. S.; Vierstra, R. D.; Forest, K. T. Journal
of Biological Chemistry 2007, 282, 12298.
(31) Fankhauser, C.; Yeh, K. C.; Lagarias, J. C.; Zhang, H.; Elich, T. D.; Chory, J.
Science 1999, 284, 1539.
(32)

Hetmann, A.; Kowalczyk, S. Postepy Biologii Komorki 2004, 31, 155.

(33) Krall, L.; Reed, J. W. Proceedings of the National Academy of Sciences of the
United States of America 2000, 97, 8169.

271

(34)

Lillo, C. Biochemical Journal 2008, 415, 11.

(35) Muller, R.; Fernandez, A. P.; Hiltbrunner, A.; Schafer, E.; Kretsch, T. Plant
Physiology 2009, 150, 1297.
(36)

Quail, P. H. Annual Review of Genetics 1991, 25, 389.

(37)

Sharma, R. Current Science 2001, 80, 178.

(38) Shen, Y. P.; Zhou, Z. Z.; Feng, S. H.; Li, J. G.; Tan-Wilson, A.; Qu, L. J.; Wang,
H. Y.; Deng, X. W. Plant Cell 2009, 21, 494.
(39) Baca, M.; Borgstahl, G. E. O.; Boissinot, M.; Burke, P. M.; Williams, D. R.; Slater,
K. A.; Getzoff, E. D. Biochemistry 1994, 33, 14369.
(40)

Borgstahl, G. E. O.; Williams, D. R.; Getzoff, E. D. Biochemistry 1995, 34, 6278.

(41) Genick, U. K.; Soltis, S. M.; Kuhn, P.; Canestrelli, I. L.; Getzoff, E. D. Nature
1998, 392, 206.
(42) Sprenger, W. W.; Hoff, W. D.; Armitage, J. P.; Hellingwerf, K. J. Journal of
Bacteriology 1993, 175, 3096.
(43) Tomioka, H.; Takahashi, T.; Kamo, N.; Kobatake, Y. Biochem. Biophys. Res.
Commun. 1986, 139, 389.
(44) Hefti, M. H.; Francoijs, K. J.; de Vries, S. C.; Dixon, R.; Vervoort, J. European
Journal of Biochemistry 2004, 271, 1198.
(45)

Ponting, C. P.; Aravind, L. Current Biology 1997, 7, R674.

(46) Ihee, H.; Rajagopal, S.; Srajer, V.; Pahl, R.; Anderson, S.; Schmidt, M.; Schotte,
F.; Anfinrud, P. A.; Wulff, M.; Moffat, K. Proceedings of the National Academy of
Sciences of the United States of America 2005, 102, 7145.
(47) Hoff, W. D.; Vanstokkum, I. H. M.; Vanramesdonk, H. J.; Vanbrederode, M. E.;
Brouwer, A. M.; Fitch, J. C.; Meyer, T. E.; Vangrondelle, R.; Hellingwerf, K. J.
Biophysical Journal 1994, 67, 1691.
(48) Meyer, T. E.; Tollin, G.; Hazzard, J. H.; Cusanovich, M. A. Biophysical Journal
1989, 56, 559.
(49)
418.

Meyer, T. E.; Yakali, E.; Cusanovich, M. A.; Tollin, G. Biochemistry 1987, 26,

(50) Yan, B.; Nakanishi, K.; Spudich, J. L. Proceedings of the National Academy of
Sciences of the United States of America 1991, 88, 9412.

272

(51)

Spudich, J. L.; Luecke, H. Curr. Opin. Struct. Biol. 2002, 12, 540.

(52) Jin, Y. D.; Honig, T.; Ron, I.; Friedman, N.; Sheves, M.; Cahen, D. Chemical
Society Reviews 2008, 37, 2422.
(53) Balashov, S. P.; Imasheva, E. S.; Govindjee, R.; Sheves, M.; Ebrey, T. G.
Biophysical Journal 1996, 71, 1973.
(54)

Mitchell, P. Nature 1961, 191, 144.

(55) Timmers, A. M.; Vangroningenluyben, D.; Degrip, W. J. Experimental Eye
Research 1991, 52, 129.
(56) McBee, J. K.; Kuksa, V.; Alvarez, R.; de Lera, A. R.; Prezhdo, O.; Haeseleer, F.;
Sokal, I.; Palczewski, K. Biochemistry 2000, 39, 11370.
(57) McBee, J. K.; Van Hooser, J. P.; Jang, G. F.; Palczewski, K. Journal of Biological
Chemistry 2001, 276, 48483.
(58) Lamb, T. D. Proceedings of the National Academy of Sciences of the United
States of America 1996, 93, 566.
(59) Bernstein, P. S.; Law, W. C.; Rando, R. R. Journal of Biological Chemistry 1987,
262, 16848.
(60) Rattner, A.; Smallwood, P. M.; Nathans, J. Journal of Biological Chemistry 2000,
275, 11034.
(61)

Trehan, A.; Canada, F. J.; Rando, R. R. Biochemistry 1990, 29, 309.

(62) Batten, M. L.; Imanishi, Y.; Maeda, T.; Tu, D. C.; Moise, A. R.; Bronson, D.;
Possin, D.; Van Gelder, R. N.; Baehr, W.; Palczewski, K. Journal of Biological Chemistry
2004, 279, 10422.
(63) Moiseyev, G.; Crouch, R. K.; Goletz, P.; Oatis, J.; Redmond, T. M.; Ma, J. X.
Biochemistry 2003, 42, 2229.
(64)

Gollapalli, D. R.; Rando, R. R. Biochemistry 2003, 42, 5809.

(65)

Deigner, P. S.; Law, W. C.; Canada, F. J.; Rando, R. R. Science 1989, 244, 968.

(66) Hamel, C. P.; Tsilou, E.; Pfeffer, B. A.; Hooks, J. J.; Detrick, B.; Redmond, T. M.
Journal of Biological Chemistry 1993, 268, 15751.
(67) Ma, J. X.; Zhang, J. Z.; Othersen, K. L.; Moiseyev, G.; Ablonczy, Z.; Redmond, T.
M.; Chen, Y. M.; Crouch, R. K. Investigative Ophthalmology & Visual Science 2001, 42,
1429.

273

(68)

Jin, M. H.; Li, S. H.; Moghrabi, W. N.; Sun, H.; Travis, G. H. Cell 2005, 122, 449.

(69) Takahashi, Y.; Moiseyev, G.; Ablonczy, Z.; Chen, Y.; Crouch, R. K.; Ma, J. X.
Journal of Biological Chemistry 2009, 284, 3211.
(70) Redmond, T. M.; Yu, S.; Lee, E.; Bok, D.; Hamasaki, D.; Chen, N.; Goletz, P.;
Ma, J. X.; Crouch, R. K.; Pfeifer, K. Nature Genet. 1998, 20, 344.
(71) Kiser, P. D.; Golczak, M.; Lodowski, D. T.; Chance, M. R.; Palczewski, K.
Proceedings of the National Academy of Sciences of the United States of America
2009, 106, 17325.
(72) Golczak, M.; Kuksa, V.; Maeda, T.; Moise, A. R.; Palczewski, K. Proceedings of
the National Academy of Sciences of the United States of America 2005, 102, 8162.
(73)

Maiti, P.; Gollapalli, D.; Rando, R. R. Biochemistry 2005, 44, 14463.

(74) Kuksa, V.; Imanishi, Y.; Batten, M.; Palczewski, K.; Moise, A. R. Vision Research
2003, 43, 2959.
(75) Stecher, H.; Gelb, M. H.; Saari, J. C.; Palczewski, K. Journal of Biological
Chemistry 1999, 274, 8577.
(76) Moiseyev, G.; Chen, Y.; Takahashi, Y.; Wu, B. X.; Ma, J. X. Proceedings of the
National Academy of Sciences of the United States of America 2005, 102, 12413.
(77) Gudas, L. J., Sporn, M. B., Sporn, A. B. R. Cellular biology and biochemistry of
the retinoids. In the Retinoids: Biology, Chemistry, and Medicine; Raven Press: New
York, 1994.
(78)

Lamb, T. D.; Pugh, E. N. Prog. Retin. Eye Res. 2004, 23, 307.

(79) Napoli, J. L. Biochimica Et Biophysica Acta-Molecular and Cell Biology of Lipids
1999, 1440, 139.
(80) Palczewski, K.; Kumasaka, T.; Hori, T.; Behnke, C. A.; Motoshima, H.; Fox, B. A.;
Le Trong, I.; Teller, D. C.; Okada, T.; Stenkamp, R. E.; Yamamoto, M.; Miyano, M.
Science 2000, 289, 739.
(81)

Hara, T.; Hara, R. Journal of General Physiology 1976, 67, 791.

(82) Terakita, A.; Yamashita, T.; Shichida, Y. Proceedings of the National Academy of
Sciences of the United States of America 2000, 97, 14263.
(83) Terakita, A.; Koyanagi, M.; Tsukamoto, H.; Yamashita, T.; Miyata, T.; Shichida,
Y. Nature Structural & Molecular Biology 2004, 11, 284.

274

(84) Hara, R.; Hara, T.; Tokunaga, F.; Yoshizawa, T. Photochemistry and
Photobiology 1981, 33, 883.
(85)

Hara, T.; Hara, R. Methods in Enzymology 1982, 81, 190.

(86)

Hara, T.; Hara, R. Nature-New Biology 1973, 242, 39.

(87)

Ozaki, K.; Hara, R.; Hara, T.; Kakitani, T. Biophysical Journal 1983, 44, 127.

(88)

Terakita, A.; Hara, R.; Hara, T. Vision Research 1989, 29, 639.

(89)

Smith, W. C.; Goldsmith, T. H. Visual Neuroscience 1991, 7, 237.

(90) Pepe, I. M.; Cugnoli, C. Journal of Photochemistry and Photobiology B-Biology
1992, 13, 5.
(91)

Smith, W. C.; Goldsmith, T. H. Journal of General Physiology 1991, 97, 143.

(92)

Pepe, I. M.; Cugnoli, C. Photochemistry and Photobiology 1998, 68, 257.

(93)

Dartnall, H. J. Vision Research 1968, 8, 339.

(94)

Birge, R. R.; Schulten, K.; Karplus, M. Chemical Physics Letters 1975, 31, 451.

(95) Borhan, B.; Souto, M. L.; Um, J. M.; Zhou, B. S.; Nakanishi, K. Chemistry-a
European Journal 1999, 5, 1172.
(96) Borhan, B.; Souto, M. L.; Imai, H.; Shichida, Y.; Nakanishi, K. Science 2000, 288,
2209.
(97)

Crist, R. M.; Michigan State University: East Lansing, 2005.

(98)

Vasileiou, C.; Michigan State University: East Lansing, 2006.

(99) Morris, G. M.; Goodsell, D. S.; Halliday, R. S.; Huey, R.; Hart, W. E.; Belew, R.
K.; Olson, A. J. Journal of Computational Chemistry 1998, 19, 1639.
(100) Birge, R. R.; Sullivan, M. J.; Kohler, B. E. Journal of the American Chemical
Society 1976, 98, 358.
(101) Futterma.A; Futterma.S Biochimica Et Biophysica Acta 1974, 337, 390.
(102) Saari, J. C. The Retinoids; Raven Press: New York, 1994.
(103) Wu, Z. P.; Hasan, A.; Liu, T. Y.; Teller, D. C.; Crabb, J. W. Journal of Biological
Chemistry 2004, 279, 27357.

275

(104) Wu, Z. P.; Yang, Y. W.; Shaw, N.; Bhattacharya, S.; Yan, L.; West, K.; Roth, K.;
Noy, N.; Qin, J.; Crabb, J. W. Journal of Biological Chemistry 2003, 278, 12390.
(105) Fitzkee, N. C.; Fleming, P. J.; Gong, H. P.; Panasik, N.; Street, T. O.; Rose, G. D.
Trends in Biochemical Sciences 2005, 30, 73.
(106) Dill, K. A.; Ozkan, S. B.; Weikl, T. R.; Chodera, J. D.; Voelz, V. A. Current
Opinion in Structural Biology 2007, 17, 342.
(107) Krishna, M. M. G.; Lin, Y.; Englander, S. W. J. Mol. Biol. 2004, 343, 1095.
(108) Rose, G. D.; Fleming, P. J.; Banavar, J. R.; Maritan, A. Proceedings of the
National Academy of Sciences of the United States of America 2006, 103, 16623.
(109) Mendoza-Espinosa, P.; Garcia-Gonzalez, V.; Moreno, A.; Castillo, R.; Mas-Oliva,
J. Mol. Cell. Biochem. 2009, 330, 105.
(110) Liu, J.; Liu, M. Y.; Nguyen, J. B.; Bhagat, A.; Mooney, V.; Yan, E. C. Y. Journal of
the American Chemical Society 2009, 131, 8750.
(111) Del Valle, L. J.; Ramon, E.; Bosch, L.; Manyosa, J.; Garriga, P. Cell. Mol. Life
Sci. 2003, 60, 2532.
(112) Okada, T.; Fujiyoshi, Y.; Silow, M.; Navarro, J.; Landau, E. M.; Shichida, Y.
Proceedings of the National Academy of Sciences of the United States of America
2002, 99, 5982.
(113) Janz, J. M.; Farrens, D. L. J. Biol. Chem. 2004, 279, 55886.
(114) Burns, L. L.; Dalessio, P. M.; Ropson, I. J. Proteins-Structure Function and
Bioinformatics 1998, 33, 107.
(115) Lalonde, J. M.; Bernlohr, D. A.; Banaszak, L. J. Biochemistry 1994, 33, 4885.
(116) Vaezeslami, S.; Jia, X. F.; Vasileiou, C.; Borhan, B.; Geiger, J. H. Acta
Crystallographica Section D-Biological Crystallography 2008, 64, 1228.
(117) Vasileiou, C.; Lee, K. S. S.; Crist, R. M.; Vaezeslami, S.; Goins, S. A.; Geiger, J.
H.; Borhan, B. Proteins-Structure Function and Bioinformatics 2009, 76, 281.
(118) Zhao, X. G.; Tucker, S. C.; Truhlar, D. G. Journal of the American Chemical
Society 1991, 113, 826.
(119) Lee, I.; Lee, W. H.; Lee, H. W. Journal of Physical Organic Chemistry 1993, 6,
361.

276

(120) Arnett, E. M., Mckelvey, D. K. Solute-Solvent Interaction; Marcel Dekker: New
York, 1969.
(121) Wang, W.; Michigan State University: East Lansing, 2008.
(122) Fan, G. B.; Siebert, F.; Sheves, M.; Vogel, R. Journal of Biological Chemistry
2002, 277, 40229.
(123) He, X. Q.; Lobsiger, J.; Stocker, A. Proceedings of the National Academy of
Sciences of the United States of America 2009, 106, 18545.
(124) Crist, R. M.; Vasileiou, C.; Rabago-Smith, M.; Geiger, J. H.; Borhan, B. Journal of
the American Chemical Society 2006, 128, 4522.

277

Chapter 4
Engineering CRABPII as a Protein Fusion Tag
4.1. Introduction
DNA recombination is a technique that allows researchers to create non-natural DNA
through fusion of different sequences together. Since the first example of recombinant DNA
1,2

technology published in 1972 by Paul Berg,

gene fusion has been greatly applied in biological

studies, particularly for proteomics to study protein structure and function. The technique has
3-9

10-19

been widely used for gene regulation and expression
20-23

localization

as well as protein purification

through the production of fusion proteins.

and

By incorporating certain tags

(sequences) with the target proteins, the proteins can be purified into homogenity easily using
combination of specific tag,
17,18

such as: His-tag,
12,13,24

Ni-NTA
GST

O

25-27

tag,

S-tag,

calmodulin-binding

O
OO-

peptide

Ni2+

N
28,29

tag,

etc.

and

their

O
corresponding resins. His-tag

Resin

O-

N

NH

O

N

NH

HN

N
H

R

His-His

R

is one of the best-known

O
examples and has been widely Figure 4-1. Schematic diagram of His-tag bound with
Nickel (II) nitrilotriacetic acid (Ni-NTA).
applied for protein purification. His-tag is a hexahistidine peptide attached at either the C- or the
N-terminus of the target protein. Its use is based on the fact that nickel (II)-nitriloacetic acid (Ni-

278

NTA) has a high affinity for adjacent histidine residues (Figure 4-1). Ni-NTA complex is
immobilized on resin and is used as chromatographic matrix. The protein with His-tag fused on
either the N-terminus or C-terminus is expressed and purified using the Ni-NTA resin column.
30-39

In addition, fusion proteins have been employed for studying protein-protein interactions
40-43

well as solubilizing insoluble proteins.

as

The technique has also been used to assist

crystallography. For example, in 2006, Cherezov et al. created a hybrid of lysozyme and β2adrenergic receptor; the chimeric structure crystallized at high resolution (Figure 4-2) while pure
44

protein cannot be crystallized at that time.

In Nature, there are a lot of important
biological processes, such as redox chemistry,
signal transductions, apoptosis, transcription
and

translation,
45-57

modifications, etc.

post-translational
These processes are !2AR

highly regulated and any disorder of these
systems would lead to various diseases, such
as cancer, etc. Thus, better understanding of
these

processes

can

provide

valuable

information to develop better drugs and

T4L

therapy for different diseases. Therefore, it is
necessary for scientists to develop probes

58-63

to visualize and study these processes in vivo.

Figure 4-2.
Crystal structure of
chimeric protein β2-Adrenergic G
protein-coupled
receptor
(β2AR)
(Green) and lysosome (T4L) (Cyan).

279

Such probes can also be used for diagnosis

1.0

508nm

Relative Absorbance/
Fluorescence

395nm

purposes.

39,47,64-69

Moreover, these visual

probes help us to develop high-throughput
screening for drugs and enzymes.

475nm
0.5

70-73

Protein

tags is one of the techniques developed for
tracking the biological processes in vivo.
From all-the commercially available tags,
green fluorescent protein (GFP) was broadly

0
400

500

employed as a protein fusion tag for such

Wavelength (nm)

purposes.
GFP was first discovered by
Figure 4-3. Schematic spectra of green
fluorescence protein. (Blue: absorption
74
Shimomura in 1962. GFP was isolated from
spectrum; Green: fluorescent spectrum)
Aequorea jellyfish as a by-product protein. In his footnote during the isolation, he wrote “a
protein giving solutions that look slightly greenish in sunlight, only yellowish under tungsten
lights, and exhibiting a bright, greenish fluorescence in the ultraviolet of a Mineralite, has also
75

been isolated from squeezates.”
minor peak at 475 nm.

75

GFP contains 238 amino acids that absorbs at 395 nm with a

When excited at 395 nm, GFP fluoresces at 508 nm. While when

excited at 475 nm, it emits at 503 nm (Figure 4-3).
77

a quantum yield around 78%.

76

GFP is an efficient fluorescent protein with

It was later found that the ratio between the two absorption

peaks is sensitive to the pH of the buffer.

76,78,79

Shimomura proposed that the chromophore in

GFP is 4-(p-hydroxybenzylidene)-imidazolidin-5-one, based on the analysis of the peptide, that
retains

the

visible

absorbance,
280

after

proteolysis

of

Fluorophore

Figure

4-4.

Stereoview

of

green

fluorescent

protein

with

fluorophore

4-(p-

hydroxybenzylidene)-imidazolidin-5-one shown in cyan.
80

the denatured GFP.

Its structure was further confirmed by the crystal structure of GFP

resolved by Ormo and co-workers (Figure 4-4). GFP comprises of 11 β-strands forming a β¨
barrel and it is connected to an α-helix bearing a chromophore in the middle of the β-barrel
(Figure 4-4).

81

Based on the crystal structure and the amino acid sequence, the fluorophore is formed by
cyclization

of

Thr65,

Tyr66

and

Gly67

(Figure

281

4-5).

76,82-84

The

currently

T65
HO
N
H

Y66

G67
R

G67

O
N
H

O

H
N

N
H

O

O

Foldiing

HO

t1/2 ~ 10 min

HN

O

O

H
R

N
H
HO

OH
Y66

T65
4-2

4-1

Cyclization

O
N
HO

HN

O
OH

N
H

R
OH

4-3
t1/2 ~ 3 mins for cyclization
+ dehydration

O

H2O2

N
HO

N
N
H
4-5

O

aerial oxidation

O

N

O2

N

HO

R
OH

t1/2 ~ 19 - 83 min

N
H
4-4

O
R
OH

Figure 4-5. Mechanism of fluorophore 4-5 formation in green fluorescent protein.

282

B

accepted mechanism starts with the nucleophilic attack of

O

amide nitrogen of Gly67 to the carbonyl of Thr65 forming

N
N

HO

O

intermediate 4-3 after GFP has been folded into its native

R

N
H

structure.

It is followed by dehydration forming

OH

intermediate 4-4. Oxidative dehydrogenation of the α-β

4-6

bond of residue 66 by molecular oxygen forms
O

imdazolinone 4-5 (Figure 4-5).
N

As mentioned above, the WT-GFP exhibits two

N

-O

O

different absorption peaks that are believed to be the

R

N
H

neutral phenol and the phenoxide ion forms of the

OH

4-7
fluorophore by the S65 mutation of the wild type protein
Figure 4-6.
Two forms of
fluorophore in green fluorescent
85,86
(Figure 4-6).
GFP has been engineered to absorb at
protein.

475 nm by maintaining the pure phenoxide ion
85

form of the fluorophore.

Tsien’s group and

Wlodawer’s group suggested that mutation

Tyr66
Ser65

S65T causes the Glu222 to remain neutral
within

the

hydrophobic

pocket.

This

eliminates the electrostatic repulsion between

Glu222

the phenoxide (Tyr65) of the fluorophore and
Glu222 (Figure 4-7).

Removal of the

Figure 4-7.
Fluorophore in green
fluorescent protein showing Glu222 closed
electrostatic repulsion promotes the phenoxide
to Ser65 and Tyr66.
formation of the fluorophore. In 1999, Remington’s group showed that the Glu222 in this
87

mutant is neutral at physiological pH.

Later, Thastrup et al. introduced the F64L mutation to
283

improve the folding efficiency of GFP at 37 ˚C that is later referred as an enhanced GFP (EGFP),
88

a versatile tool for mammalian cell research.

In 1994, Chalfie et. al published the first paper that described the expression of GFP in
Caenorhabditis elegans under the control of the mec-7 promoter. The result showed that the
GFP’s fluorescence was observed at the location where Mec-7 is expressed. This suggested that
89

GFP can be used as a genetic marker for protein expression.

This was the first example of

GFP used as a gene marker. The pioneered work on GFP bought Chalfie, Shimomura and Tsien
a Noble prize in 2008. Since then, GFP has been widely applied for biomedical research for
23,38,62,90-96

monitoring protein localization, protein expression and gene regulation in vivo.
has also been used as a physiological indicator.

GFP

Kneen and co-workers demonstrated that

EGFP’s fluorescence is sensitive to different pHs; therefore, EGFP could be used as
97

physiological pH indicator.

In addition, GFP is mostly used as a fusion protein tag. Most of

the techniques developed for fluorescent spectroscopy can be applied to GFP. Forster resonance
¨
energy transfer (FRET) is one of these techniques. FRET is a phenomenon that occurs when two
fluorophores are closed to each other and the emission spectrum from one fluorophore
(fluorescent donor) overlaps with the excitation spectrum of the other fluorophore (fluorescent
acceptor) (Figure 4-8b). The energy is transferred non-radiativiely through dipolar coupling
between the chromophores (Figure 4-8).

98

6

Since the FRET efficency shows 1/r dependence

where r stands for the intermolecular distance between fluorescent donor and acceptor, it can be
98,99

used as a molecular ruler (Figure 4-8c).

Since the GFP can be fused to other proteins easily,

GFP

been

and

GFP

variants

have

widely

284

used

as

FRET

reporters

a)

Energy
Transfer
non-Radiantly

hv

-

-

+

+

Fluorescence of
Fluorescent
Donor (Green)
Absorption
Spectrum for
Fluorescent
Acceptor (Red)

Red
Fluorescence
c)

Relative Red
Fluorescent
Intensity

b)

Relative Absorbance/
Fluorescence

r

1/r2

r (Å)

Wavelength (nm)

Figure 4-8. a) Schematic diagram for FRET. Left: Light irradiation of fluorescent
donor; Right) Fluorescent donor is excited and the energy is transferred to fluorescent
acceptor non-radiantly through excited dipole-dipole interaction between fluorescent
donor and acceptor. Fluorescence of fluorescent donor disappears and fluorescent
acceptor is excited and emits red fluorescence. b) Sketch spectra shows the fluorescent
spectrum from fluorescent donor overlapping with the absorption spectrum of fluorescent
acceptor. c) Sketch graph shows the fluorescent intensity of fluorescent acceptor is
indirectly proportional to the distance between fluorescent donor and acceptor (r).
to measure protein-protein interactions in vivo. In 1998, Mahajan et. al used GFP and its variant
(blue fluorescent protein) to test whether there is a protein-protein interaction between Bcl-2 and
Bax in mitochondria. Bcl-2 plays a crucial role during apopotosis (programmed cell dealth)
while Bax-mediated apoptosis happens in mitochondria. To test whether Bcl-2 and Bax work
together or independently, GFP (fluorescent acceptor) was fused with Bax and blue fluorescent
protein (BFP) (fluorescent donor) with Bcl-2. Fluorescence microscopy indicated that Bax is
localized in the mitochondria. When Bcl-2 and Bax were co-expressed within the cell, the

285

Promoter
BFP

M13

CaM

BFP

GFP

GFP

M13

CaM
Ca2+
FRET

hv

Fluorescence
BFP

GFP

Ca2+

M13
CaM

2+

Figure 4-9. Scheme for Ca

detection using GFP.

fluorescence of BFP-Bcl-2 disappeared while fluorescence enhancement of GFP-Bax was
38

observed (Figure 4-9). This result indicated that Bcl-2 and Bax due strongly interact.
addition, GFP has been used as a Ca

2+

In

sensor utilizing the FRET technique. Miyawaki and co-

workers designed a fusion protein with a combination of BFP or cyan fluorescent protein (CFP),
calmodulin, calmodulin-binding peptides (M13) and GFP or yellow fluorescent protein (YFP)
2+

for in vivo Ca

2+

sensing (Figure 4-9). Calmodulin is a Ca

29

binding protein.

Upon binding

2+

Ca , a conformation change is induced that allows M13 binding. When the fusion peptide
2+

(calmodulin-M13) binds to Ca , it changes from a dumb-bell-like extension to a compact
286

100,101

globular form.

BFP and GFP are fused to calmodulin and M13, respectively (Figure 4-9).
2+

In the presence of Ca , M13 binds to calmodulin and brings the two fluorescent proteins closer
2+

allowing for FRET to occur (Figure 4-8). Thus, at high concentration of Ca , more calmodulin
2+

is bound to Ca

increasing FRET (Figure 4-8). Moreover, FRET and GFP have been applied to

study

different

biological

processes

using

similar

strategy

such

102,103

phosphorylation,

as

Shealth Fluid

protein

104

Cell

complexation, etc.

Because of its high quantum

Hydrodynamically
Focused Stream

efficiency, GFP has been used for single
cell detection.
techniques

developed

fluorescent

activated

(FACS).

105-108

Fluorescence

One of the new
recently
cell

Detector

is

sorting

This technique has been

+

–-

applied for high throughput screening of
108-110

drugs and biocatalysis.

FACS is

a technique that combines fluorescence,
70

flow cytometry

and cell sorting. Flow

Figure 4-10. Schematic diagram of fluorescent
activated cell sorting system.

cytometry is a method to analyze different parameters for a single cell in a population. Briefly,
cells are subjected to a carrier fluid called the sheath fluid. It is then forced through an orifice

287

111

that has a size of 50 to 100 µL to generate a stream (Figure 4-10) by hydrodynamic focusing.

The hydrodynamic focusing is generated by pushing the liquid from a large diameter tube into a
capillary tube. This allows cells to flow through a single file (Figure 4-10) enabling cell sorting
based on the charge on the cell (cells are charged negatively or positively ia a changing the
nozzle potential based on the observed color) (Figure 4-10). This is a powerful technique that
7

can sort up to 10 cells per hour. In 2006, You et al. demonstrated the first quantitative analysis
112

of protein-protein interaction using both FRET and FACS.

They fused the target domain

(SH3 and PDZ) with the yellow fluorescent protein (YFP) and the bait peptide library with cyan
fluorescent protein (CFP). Both proteins were co-expressed in the cells. The YFP in the cell
would be excited and cells that showed FRET (disappearance of yellow fluorescence and
appearance of cyan fluorescence) would be selected through FACS. The sorted cells were resubmitted for several rounds of selection through FACS (Figure 4-10). Several ligands that
show good binding towards the PDZ domain were identified by FACS.
Green Fluorescent Protein is a useful protein tag that can provide different ways for
chemists, biochemists and biologists to study biological processes in vivo and to engineer protein
targets with desired functions. However, it is not without limitations. The major problem is that
molecular oxygen is required for fluorophore formation (Figure 4-5), therefore, GFP is not
76,95,113

suitable for studies under anaerobic condition.

Also, hydrogen peroxide is the by-

product of the fluorophore formation, which can be toxic to the cell (Figure 4-5). In addition,
formation of fluorophore requires at least an hour; therefore, it is not suitable for tracking fast
cellular processes. Finally, some of the mutants are misfolded at 37 ˚C, thus making the method
95

not suitable for monitoring biological processes in the human body.

288

4.2. Reengineering CRABPII into a Chromophoric and Fluorescent Protein
To provide an alternative
for GFP, we would like to take

O

R

this challenge and re-engineer
CRABPII into fluorescent and
H2N

chromophoric

protein

to

monitor biological processes in
vivo. Our engineered CRABPII

CRABPII

(R132K:R111L:L121E)
all-trans-retinal
R

N
H+

as

binds

a

PSB

through a nucleophilic lysine
residue, Lys132 (Figure 1-31) to

Red Shifted Chromophoric or
Fluorescent Protein

give

a

fully

red-shifted
114

In
Figure 4-11.
Strategy for engineered CRABPII as spectrum at 449 nm.
chromophoric or fluorescent protein. R stands for any
heterocycle.
addition, there are several alltrans conjugated aldehydic analogues that bind to bacteriorhodopsin as a PSB and show large
red-shift.

115-117

We, therefore, suggest that the binding of some of these analogs to CRABPII

mutants can turn CRABPII into a chromophoric and possibly fluorescent protein (Figure 4-11).
3

We believe CRABPII is suitable for our purpose since it has a large incassed cavity (600 Å ) that
allows the incorporation of aldehydic chromophores with different sizes. Moreover, we know
that CRABPII is highly tolerant to mutations, which would allow for the modulation of the
absorption of the bound chromophore through point mutations (see Chapter 1).

289

CRABPII mutants as fusion protein tags offer several advantages that are complimentary
to

GFP.

Changing
106,107,109

mutagenesis.

the

spectral

properties

of

GFP

require

extensive

random

CRABPII engineered tags would function as on/off switches (as opposed

to GFP), since they would turn on upon addition of chromophores. In addition, CRABPII (16
KDa) is smaller than GFP (23 KDa), therefore, the effect on the target proteins should be smaller
as compared with GFP. As mentioned, GFP requires molecular oxygen and is the right chioce
for studying processes under aerobic condition, on the other hand, a CRABPII-based fluorescent
proteins would work under both aerobic and anaerobic conditions. Moreover, formation of a
CRABPII-based fluorescent protein does not generate H2O2.

4.2.1. Binding of Merocyanine dyes to CRABPII Mutants
Cyanine and merocyanine are fluorescent dyes that are routinely used in biomedical,
118-124

biological or phar-maceutical research.
125

dyes.

They both belong to polymethine family of

The general structure for a cyanine dye contains two nitrogens that are connected by a

polymethine chain (Figure 4-12).

125,126

The difference between cyanine dye and merocyanine

dye is that one of the nitrogen in the general structure of cyanine dyes is replaced by a carbonyl
or its derivatives (Figure 4-12). They are ideal as fluorescent probes because they have high
extinction co-efficient, therefore, they are capable to capture light and being excited easily.

290

N

N

N

N

4-8

N

O
4-9

Figure 4-12. Structure of cyanine dye (4-8) and merocyanine (4-9).

Additionally, modulation of the absorption is easily done by changing the corresponding groups
attached on the nitrogen or the number of methines on the chain. In addition, cyanine dyes are
generally photostable because the protonated state of cyanine dyes generate a push-pull system
that can delocalize the positive charge efficiently (Figure 4-11). As a result of the push-pull
system, the movement (rotation, vibration and twisting) of the molecule is highly restricted. The

S2

Internal
Conversion

10-12s

Vibrational
relaxation
S1

Excitation
(10-15 s)

Intersystem
crossing
Fluorescence
(10-9s)

S0

Figure 4-13. Joblonski diagram.

291

T1
Phosphorescence
10-6s

cyanine dye’s absorption spectrum also shows that its molecular movement is restricted. Due to
the presence of vibrational and rotational movements within the molecules, each specific
conformation has its own energy level which leads to line broadening based on Franck-Condon
124,127

principle.

Therefore, restriction of freedom within the molecule by a push-pull system can

lead to a sharper absorption peak. The formation of a rigidified cyanine dye minimizes the
vibrations that decrease the chance of losing energy through non-radiative pathways from the
excited molecule, therefore, increasing the quantum yield (Figure 4-13).

128-131

In 1997, Hoischen et al. synthesized a series of all-

O

trans-retinal analogs containing the merocyanine structure,
N

such as merocyanine 4-10 and 4-11 (Figure 4-14).
4-10 (463 nm

610 nm)
O

117

These

merocyanines were used to study the relationship between the
protein-ligand interaction and wavelength regulation in
bacteriorhodopsin.

N

bacteriorhodopsin
4-11 (487 nm

These
as

a

merocyanines

PSB

and

show

bind

with

significant

755 nm)
117

bathochromic shift (Figure 4-14).
Upon forming PSB with
Figure 4-14.
Structure of
merocyanine 4-10 and 4-11
with the corresponding λmax bacteriorhodopsin, merocyanines turn into cyanine dye in
and λmax after bound with bacteriorhodopsin and form a push-pull system that
bacterio-rhodopsin.
delocalizes the positive effectively along the polyene (Figure 4-15) leading to a red-shifted
spectrum. In addition, cyanine dyes usually fluoresce allowing for the formation of a fluorescent
118,119,121,123,124

protein.

292

O

N
4-10 (!max = 463 nm)

H2N

N

N
H+

!max = 610 nm
Figure 4-15. Schematic diagram of formation cyanine dye upon binding of merocyanine
with bacteriorhodopsin as a PSB.
4.2.1.1. Synthesis of Merocyanine 4-10, 4-11 and 4-12
Merocyanines 4-10 and 4-11 were synthesized to test with the CRABPII mutants. In
addition, based on a study conducted by Nakanishi, it was found that merocyanine without the
methyl substituted on the polyene is less stable when bound with bacteriorhodopsin (5%

293

absorption decrease in 2 h for merocyanine with methyl group and in 5 hours for merocyanine
132

without methyl group).

Therefore, merocyanine 4-12 was also synthesized.

The syntheses

for all 3 merocyanines started with the formylation of 1,3,3-trimethyl-2-methylene indoline by
the Vilsmeier-Haack reaction forming methine merocyanine 4-13 in quantitative yield. The
methine merocyanine 4-13 is a common start material for all-three merocyanine syntheses. In
general, the synthesis of merocyanines 4-10, 4-11 and 4-12 is based on a modified protocol for
133

retinal synthesis established previously.

To synthesize merocyanine 4-10, merocyanine 4-13

was subjected to HWE reaction with 4-(diethylphosphono)-3-methyl-butenenitrile yielding 80%
of nitrile 4-14 with E:Z ratio 3:1.

Nitrile 4-14 was then reduced by DIBAL-H forming

merocyanine 4-10 in 40% yield with E:Z ratio about 4:1. Merocyanine 4-16 was synthesized by
HWE reaction of 4-13 with 2-(diethylphosphono)-ethanenitrile yielding nitrile 4-15. Nitrile 4-15
was then reduced by DIBAL-H yielding in 56% overall from merocyanine 4-13. To synthesize
merocyanine

4-11

and

4-12,

294

merocyanine

4-16

was

O

POCl3, DMF,

quant.

N

O
(EtO)2P

N

CN i) NaH, 0 ˚C

4-13
CN

rt/ THF;

ii)

O
N
N

4-13
4-14

overnight, 80% yield,
E:Z / ~ 3:1
i) DIBAL-H,
-78 ˚C
0 ˚C,
THF, 6 h;

O

ii) wet silica, 0 ˚C,
8 h, 40% yield,
E:Z / ~4:1

N
4-10
CN

i) NaH, 0 ˚C

rt/ THF;

ii)

O

(EtO)2P(O)CH2CN

N

N
4-15

4-13

overnight, 90% yield,
E:Z / ~3:1
i) DIBAL-H,
-78˚C
0 ˚C,
THF, 6 h;
ii) wet silica, 0˚C,
8h, 65 % yield

O

N
4-16

Scheme 4-1A. Synthesis of merocyanine 4-10, 4-11 and 4-12.
elongated by HWE reaction, yielding corresponding nitrile 4-17 and 4-19 in 3:1 E:Z

295

CN i) NaH, 0 ˚C
O
(EtO)2P

CN

rt/ THF;

ii)

O

N

N

4-16

4-17

overnight, 80 % yield
E:Z / ~3:1
i) DIBAL-H,
-78 ˚C
0 ˚C,
THF, 6 h;

O

ii) wet silica, 0 ˚C,
8h, 5% yield
E:Z / ~3:1

N
4-18

CN

i) NaH, (EtO)2P(O)CH2CN,
0 ˚C
rt/ THF;
(EtO)2P(O)CH2CN

O

ii)
N

N
4-19

4-16

overnight, 90% yield
O

i) DIBAL-H,
-78 ˚C
0 ˚C,
THF, 6 h;
ii) wet silica, 0 ˚C,
8 h, 60% yield

N
4-12

Scheme 4-1B. Synthesis of merocyanine 4-10, 4-11 and 4-12.
ratio. Both nitriles were further reduced by DIBAL-H to produce merocyanine 4-11 and alltrans-merocyanine 4-12 respectively in 4% (E:Z / 3:1) and 54%. Although merocyanines 4-10

296

and 4-11 are E/Z isomeric mixtures, upon formation of PSB with CRABPII mutants,
117

isomerization to the all-trans configuration with ensue.

Table 4-1.
UV-vis 4.2.1.2. UV-vis Spectroscopic Study with Merocyanines
absorption of meroTriple mutant R132K:R111L:L121E that binds well with allcyanines
with
CRABPII
mutant
(R132K:
R111L: trans-retinal was the first one to be tested for merocyanine binding.
L121E).
Merocyanines 4-10, 4-11 and 4-12 binds with R132K:R111L:L121E
Merocyanine

λmax
(nm)

4-10

591

4-11

577

fully shifted spectrum (Figure 4-16). Since the cyanine dyes are

4-12

612

known to be solvatochromic compounds (spectrum shift is sensitive

through the formation of a PSB (Figure 4-14) and show significant
red-shift (Table 4-1). However, all-merocyanines do not show a

134-138

to the polarity of the solvent),

we suggest that the observed

red-shift can reflect the polarity inside the
0.25

binding pocket. A more detailed study will

spectroscopic

study,

0.2

During the UV-vis
we

found

that

merocyanine 4-12, without the methyl
group along the polyene, took 24 h to bind
completely while merocyanine 4-10 took

Absorbance

be discussed later.

0.6 eq.
0.1

0.4 eq.

0.05

0.2 eq.
0.1 eq.

0

Since merocyanine 4-12 binds

slower, the following studies were focused
on merocyanines 4-10 and 4-11. Based on

0.8 eq.

0.15

about 1 h and merocyanine 4-11 took about
15 min.

1.0 eq.

300

400
500
600
Wavelength (nm)

700

Figure 4-16. UV-vis spectra of merocyanine 410
with
CRABPII
mutant
R132K:R111L:L121E.

297

the crystal structure of R132K:R111L: L121E, there are
11 solvent-exposed lysines residues (Figure 4-17).
Since formation of merocyanine PSB with lysine
generates a more stable cyanine dye, we were
wondering whether the observed UV-vis spectrum shift
(Figure 4-16) is due to non-specific PSB formations
with the solvent exposed lysines. We, therefore, tested
Figure 4-17.
Solvent exposed
the merocyanine 4-10 and 4-11 with mutant
lysines in R132K: R111L:L121E.
R132L:R111L:L121E that does not have the engineered nucleophilic Lys132. Merocyanine 4-10
forms PSB with mutant R132L:R111L:L121E, however, it takes at least 12 h
b) 0.15

a) 0.3
0.25
0.2
0.15
0.1

After
12 h

Absorbance

Absorbance

At time = 0 min

0.1

After
12 h

At time = 0 min
At time = 6 min

0.05
PSB formation

After
6h

0.05
0
0
400 450 500 550 600 650 700
Wavelength (nm)

300

400
500
600
Wavelength (nm)

700

Figure 4-18. UV-vis spectrum of mutant R132L:R111L:L121E incubated with a)
merocyanine 4-10 and b) merocyanine 4-11 in PBS buffer at room temperature.
for small-amount of PSB to be formed (Figure 4-18a). In addition, merocyanine 4-11 is unstable
and decomposes after 6 h with mutant R132L:R111L:L121E(Figure 4-18b).
indicated that non-specific binding is not an issue for merocyanine PSB formation.

298

These results

Since we have demonstrated that bound C15 aldehyde PSB with different R59 mutants
showed a different absorption shift, we incubated both merocyanine 4-10 and 4-11 with different
R59 mutants.

UV-vis spectra of both merocyanine 4-10 and 4-11 with mutant

R132K:R111L:L121E:R59E revealed a more red-shifted spectrum (Figure 4-19) as compared to
that of mutant R132K:R111L: L121E (Figure 4-16). We found that R59E showed 8 nm red-shift
for bound merocyanine 4-10 (λmax = 598 nm) as compared with mutant R132K:R111L:L121E
(λmax = 591 nm) (Table 4-1 and 4-2). To investigate whether the observed spectrum shift is due
b) 0.1

0.3

Merocyanine 49
added after 50 min

Absorbance

0.25
0.2
0.15

Merocyanine 49
added after 1 min

0.1
0.05

Absorbance

a)

Merocyanine 50
added after 1 min
0.08
Merocyanine 50
added after 12 min
0.06
0.04
0.02

0
0
400 450 500 550 600 650 700 400 450 500 550 600 650 700 750 800
Wavelength (nm)
Wavelength (nm)

Figure 4-19. UV-vis spectrum of R132K:R111L:L121E:R59E with a) merocyanine 4-10
and b) merocyaine 4-11.

to the negative charged mutation on R59, merocyanine 4-10 was tested with several R59
mutants. However, all R59 mutants showed similar λmax (Table 4-2). This result indicates that
the 8 nm blue-shift with the R59 mutants is not result of the electrostatic potential change around
the indoline ring.

299

Recently, merocyanine 4-10
with

mutant

L121E:R59W
(Figure

Table 4-2. UV-vis absorption of merocyanine 4-10 with
R132K:R111L: CRABPII mutant.
was

4-20).

crystallized

(See

Thesis
139

dissertation Dr. Xiaofei Jia)
The

observed

binding

a

Mutant

λmax (nm)

R132K:R111L:L121E:R59E

598

R132K:R111L:L121E:R59W

596

R132K:R111L:L121E:R59Y

596

Incubation at room temperature in PBS buffer (pH = 7)

conformation is the same as that of
the C15 aldehyde inside mutant R132K:R111L:L121E:R59W (Figure 4-21a).

That is, the

chromophore has rotate ~90˚ along the long-axis as compared to the bound all-trans-retinal in
R132K:R111L:L121E (Figure 4-21b). In addition, W59 is close to the end of the chromophore
(Figure 4-20) but mutations at this position do not induce any spectrum shift. That is different
from C15 aldehyde behavior (Table 4-2
and 1-5).

W59

Presumably, merocyanine is

insensitive to charged or polar mutations
because the positive charge on bound
merocyanine PSB is fully delocalized
along the polyene by the push-pull
system within the molecule, therefore,
Figure 4-20. Merocyanine 4-10 binding site in
unlike C15 aldehyde, it does not require
mutant R132K:R111L:L121E:R59W.
additional stabilization from the protein to propagate the charge delocalization as well as a
counterion for PSB stabilization (Figure 4-22).

The crystal structure of merocyanine with

R132K:R111L:L121E:R59W showed Glu121 interacts with the “PSB” through a water-mediated
300

a)

b)

Figure 4-21. a) Overlay crystal structure of R132K:R111L:L121E:R59W with C15
aldehyde (blue) and R132K:R111L:L121E:R59W with merocyanine 4-10 (purple); b)
Overlay crystal structure of R132K:R111L:L121E with all-trans-retinal (cyan) and
R132K:R111L:L121E:R59W with merocyanine 4-10 (purple).
interaction. Placement of Gln and Asp at 121 can modulate the bound C15 aldehyde PSB
absorption showing +16 and +18 nm red shift, respectively. However, as we mentioned, the
positively charged within the cyanine dye can delocalize without any assistance from the protein,

a)

b)

Figure 4-22. a) Crystal structure of C15 aldehyde with R132K:R111L:L121E:R59W; b)
Crystal structure of merocyanine 4-10 with R132K:R111L:L121E:R59W.
we, therefore, test the molecule with mutant R132K:R111L. UV-vis spectrum of R132K:R111L
with merocyanine also showed a completely red-shifted spectrum with λmax at 598 nm (Figure
4-23). This further indicates that Glu121 does not have any effect on the absorption of the bound
merocyanine PSB.
301

Several
Table 4-3. UV-vis absorption of merocyanine 4b
spectral
10 with CRABPII mutant.
a

mutants

changes

that

with

induced

bound

C15

Mutant

λmax (nm)

aldehyde PSB (Table 1-5 to 1-7) were

R132K:R111L:L121E:R59W

596

tested with merocyanine 4-10 in order to

R132K:R111L:L121E:R76W

564

investigate whether these mutants are

R132K:R111L:L121E:R76Y

496,
590 (s)

capable of modulating the absorptions of

R132K:R111L:L121E:V76R

586

bound merocyanine PSB.

R132K:R111L:L121E:V76E

584

R132K:R111L:L121E:T56W

587

R132K:R111L:L121E:T56R

588

R132K:R111L:L121E:T56I

598

Most

R132K:R111L:L121E:F15Y

596

hindered the binding of merocyanine.

R132K:R111L:L121E:F15W

591

R132K:R111L:L121E:A32E

595

C15

aldehyde,

Unlike the

merocyanine

is

insensitive to any mutations (Table 4-3).
of

the

mutations

introduced

4.2.1.2. Modulating the Absorption of
the Chromophoric Protein-CRABPII

a

(s) stands for shoulder (minor peak) of the
spectrum.
b
Incubation at room temperature in PBS buffer (pH
= 7)

Most of the CRABPII mutants
were not able to produce fully shifted

spectra with merocyanine 4-10 like bound merocyanine 4-10 with mutant R132K:R111L:L121E
(Figure 4-16). However, we found that placement of tryptophan at different positions (56, 76
and 59) along the bound the merocyanine induces spectrum shift (Figure 4-24). Based on the
model, we found that as the tryptophan is placed near the bound merocyanine PSB, a blue shifted
spectrum is obtained (Figure 4-24). We, therefore, suggested two hypotheses to explain the
effect of tryptophan on wavelength regulation.

302

As mentioned, cyanine dyes are solvatochromic compounds.

Absorbance

0.15

0.1

To understand the

mechanism for the absorption shift upon
35 mins after
merocyanine 4-10

binding merocyanine 4-10 with mutants
R132K:R111L:L121E:

1 mins after
merocyanine
4-10

R132K:R111L:

L121E:T56W

R132K:R111L:L121E:
0.05

R59W.

V76W,

and
the

merocyanine PSB UV-vis spectrum was
measured in solvents with different polarity

0
200 300 400 500 600 700 800
Wavelength (nm)

(Table 4-4).
merocyanine

The result showed that
PSB

shows

a

positive

Figure 4-23. UV-vis spectra of merocyanine
4-10 with R132K:R111L.
solvatochromism

meaning

that

the

T56W (591nm)

merocyanine PSB have bathochromic
shift as the polarity of the solvent
increases (Figure 4-25). We, therefore,
proposed that the observed spectrum

R59W
(596nm)

V76W
(564nm)

shift with CRABPII is induced by the
dielectric environment changes within
the binding pocket that are the result of

Figure 4-24.
Merocyanine binding site in
R132K:R111L:L121E:R59W. The positions with
the conformational changes induced by
tryptophan were engineered are shown (Cyan)
tryptophan mutations. To further verify with the corresponding λmax.
this hypothesis, we also measured the fluorescence of CRABPII mutants. The fluorescence of
the CRABPII protein is mainly due to the fluorescence from the 3 tryptophan residues within the

303

590!
585!

"max(nm)!

580!
575!
570!
565!
560!
555!
550!
0!

0.2!

0.4!
0.6!
0.8!
Relative Polarity!

1!

1.2!

Figure 4-25. Relationship between solvent polarity and λmax.
The

fluorescence

of

tryptophan has been used to access
different properties of the protein, such
as protein folding, polarity changes,
etc. It was found that the fluorescence
of tryptophan is sensitive to the change
of

polarity,

showing

positive

Normalized Fluorescecne

protein.

1.02
1
0.98
0.96
0.94
0.92

KLE-R59W
KLE-T56W
KLE-V76W
KLE
PSB Buffer

solvatochromism like merocyanine 4-

0.9
340 345 350 355 360 365 370
10. Based on the crystal structure of
Wavelength (nm)
Figure 4-26. Emission spectra of different mutants
CRABPII, there are three tryptophans
excited at 283 nm.
(KLE stands for
R132K:R111L:L121E).
buried inside the binding pocket.

Therefore, the fluorescence of the protein should give information about the dielectric
environment within the binding pocket.

Mutant R132K:R111L:L121E: V76W shows a

significant blue shifted fluorescent spectrum as compared with R132K:R111L:L121E,

304

R132K:R111L:L121E:R59W and
R132K:R111L:

Table 4-4. λmax of merocyanine 4-10 PSB in different
L121E:T56W solvent and corresponding polarity of the solvent.
a

(Figure 4-26). The result suggests

Solvent

Relative Polarity

λmax (nm)

that the observed hypsochromic

Hexane

0.009

556

shift

Benzene

0.099

570

Toluene

0.111

572

Chloroform

0.259

568

Dichloromethane

0.309

576

Dimethylsulfoxide

0.444

579

Acetonitrile

0.46

580

Ethanol

0.654

577

Ethylene glycol

0.79

582

Glycerol

0.812

584

Water

1.0

574

in

R132K:R111L:L121E:V76W with
merocyanine is due to the polarity
decrease

within

the

binding

pocket.
In addition, we have noticed
that the presence of tryptophan
residues

changes

the

shape

(sharpen) of the overall-spectrum.
a

Placement of a tryptophan deep

λmax of the merocyanine 4-10 PSB.

inside the binding pocket (mutant R132K:R111L:L121E: V76W)
the bound merocyanine PSB (Figure 4-27).

broadens the absorption of

Based on Franck-Condon principle, the broad

spectrum might indicate that the bound merocyanine PSB with R132K:R111L:L121E:V76W has
more freedom of movements (rotation and vibration) than that in R132K:R111L:L121E:R59W.
The line broadening also occurs as we compare the absorption spectrum of merocyanine PSB in
different solvents.

Apolar solvents broaden the absorption spectrum of merocyanine PSB

(Figure 4-28). In polar solvents, the charge delocalization can be further stabilized by the polar
solvent molecules around the chromophore (Figure 4-28).

Also, the polar solvent can further

solvate the counterion, in this case trifluoroacetate (formation of PSB in organic solvent by the

305

addition of TFA), therefore, minimizing

Normalized Absorbance

1.2

Merocyanine 4-10 KLE:T56W
KLE:V76W KLE:R59W

the interaction of counterion with PSB.

1

However, with an apolar solvent, there is

0.8

no such interaction to stabilize the positive
charge delocalization and the counterion

0.6

would be localized around the PSB,

0.4

therefore,

0.2

hindering

delocalization.
0
350 400 450 500 550 600 650 700
Wavelength (nm)

the

charge

Since the push-pull

system (charge delocalization) diminishes

the rotation and vibration of the
Figure 4-27. UV-vis spectra of merocyanine 410
and
merocyanine
4-10
with merocyanine PSB, the broadening of the
R132K:R111L:L121E (KLE) mutants having
tryptophan installed at different position (Figure absorption spectrum increases as the push4-24).
pull system is diminishing, that might also suggested that the observed line-broadening happen in
merocyanine 4-10 mutant R132K:R111L:L121E:V76W complex is due to the dielectric
environment within the protein binding site become less polar which is also the conincide with
the blue shifted observed (Figure 4-7). The binding pocket within R132K:R111L:L121E:V76W
is less polar as compared to R132K:R111L: L121E:R59W based on the fluorescence from 3
tryptophan in CRABPII, thus, it results in the broadening of the absorption spectrum of the
bound merocyanine (Figure 4-27).
In short, we have demonstrated that CRABPII mutants are able to bind mero-cyanine as a
PSB with Lys132 specifically and can create a chromophoric protein (Figure 4-27). In addition,
we are able to modulate the absorption of our chromophoric protein through mutagenesis.

306

Merocyanine 4-11 that has one

mutations on CRABPII upon binding to

Benzene
Hexane
Ethanol

1.2
1

CRABPII mutants (Table 4-5).
addition,

we

can

observe

In

similar

modulation of wavelength as a function

Absorbance

extra double bond was more sensitive to

of tryptophan mutations with the bound

0.8
0.6
0.4
0.2

merocyanine 4-11 PSB in CRABPII
0

(Figure 4-29).

Although we could not

400 440 480 520 560 600 640 680
Wavelength (nm)

observe any spectral changes when V76 Figure 4-28. UV-vis spectra of merocyanine 410 PSB with different solvents
mutant
KLE:R59W
KLE:V76W
KLE:T56W

protein

were

bound

with

merocyanine 4-10, we find merocyanine 4-

1.2

Relative Absorbance

11 is sensitive to mutations at position 76
1

(Table 4-5).
0.8

However, we found that the

merocyanine 4-11 is much less stable than

0.6

merocyanine 4-10 in buffer (Figure 4-19).

0.4

Therefore,

0.2

chromophore for CRABPII mutants.

0
400 450 500 550 600 650 700 750 800
Wavelength (nm)

4.2.1.3.

it

is

not

Fluorescent

suitable

as

Properties

a

of

Merocyanine with CRABPII mutants

Figure 4-29. UV-vis spectra of merocyanine 411 with mutants R132K:R111L:L121E:V76W
Ultimately, we would like to
(KLE:V76W),
R132K:R111L:L121E:T56W
(KLE:T56W) and R132K:R111L:L121E:R59W engineer CRABPII as a fluorescent protein
(KLE:R59W).

307

upon binding to different fluorophores

cyanine

Merocyanines and

have

fluorescence

119,121,135-137

property.

However, the

chromophore that we used can be a
problem if it fluoresces in its unbound
form

because

fluorescence

of

the

Fluorescence of
Merocyanine 4-10
Irradiated at 465nm

3 105

2 105

Fluorescence of
Merocyanine 4-10
Irradiated at 571nm

571nm

Fluorescence of
Merocyanine 4-10
PSB
Irradiated at 571nm

1 105

background

generated.

Fluorescence (cps)

(Figure 4-11).

4 105

We

0
500

hypothesized that upon forming a

600
700
Wavelength (nm)

800

Figure 4-30. Fluorescent spectra of merocyanine
the 4-10 and its PSB excited at different wavelength
and different concentration.
fluorescence of the bound merocyanine
cyanine

dye

with

CRABPII,

PSB would be more red-shifted than
Merocyanine 4-10
with KLE:R59W

3.5 10 7

therefore, we should be able to

3 107

Fluorescence (cps)

that of the unbound chromophore,

selectively

2.5 10 7
2 107

observe

the

bound

chromophore’s fluorescence from the
Merocyanine
4-10 PSB

1.5 10 7
1 107

unbound one.

Merocyanine 4-10

When merocyanine 4-10 (λmax =

5 106
0

465 nm) is irridiated at 465 nm, it
612

646 680 714 748
fluoresces at 590 nm.
When
Wavelength (nm)
Figure 4-31. Emission Spectra of merocyanine 4- merocyanine is irradiated at 571 nm in
10, merocyanine 4-10 PSB and merocyanine 4-10
with
mutant
R132K:R111L:L121E:R59W order to alleviate the effect of
(R132K:R111L: L121E:R59W) excited at 570 nm.

308

excitation on fluorescence spectral, it has

Table 4-5. UV-vis absorption of merocyanine 411
with
CRABPII
mutant
(R132K:R111L:L121E).

a small-fluorescence emission at 596 nm.
In addition, irradiation of merocyanine

Mutant

λmax (nm)

R132:R111L:L121E

612

PSB at 570 nm results in fluorescence at

R132K:R111L:L121E:R59E

682

608 nm.

R132K:R111L:L121E:R59W

676

both merocyanine and its PSB, when

R132K:R111L:L121E:R76W

564

R132K:R111L:L121E:V76R

662

R132K:R111L:L121E:V76E

680

R132K:R111L:L121E:T56W

644

R132K:R111L:L121E:T56R

684

The fluorescence emission of

excited at around 570 nm are close (Figure
4-30)

which

However,

could
the

be

a

problem.

fluorescence

of

merocyanine excited at 570 nm is much

Incubation at room temperature in PBS buffer (pH
= 7)

smaller (almost no fluorescence) than the

fluorescence of merocyanine PSB at

Fluorescence at 604 nm
(KLE:V76W)
3.5 10 7

fluorescence is enhanced more when

3 107

the

merocyanine

is

bound

with

R132K:R111L:L121E:R59W (Figure
4-32).

The

quantum

efficiency

calculation showed that there is about
10 fold increase in quantum yield

Fluorescence (cps)

same concentration (Figure 4-31). The

Fluorescence at 614 nm
(KLE:R59W)

2.5 10 7

Fluorescence at 614 nm
(KLE:T56W)

2 107
1.5 10 7
1 107
5 106
0

when the merocyanine is bound with
mutant

612

646 680 714
Wavelength (nm)

748

R132K:R111L:L121E:R59W Figure
4-32.
Fluorescence
of
mutant
R132K:R111L:
L121E:R59W
(KLE:R59W),
as compared with merocyanine PSB R132K:R111L:L121E: T56W (KLE:T56W) and
R132K:R111L:L121E:V76W (KLE:V76W) with
merocyanine 4-10 irradiated at 571 nm.

309

(Table 4-6). The quantum yield of all

Fluorescence (cps)

1 106

mutants that show completely shifted

40% Glycerol

UV-vis spectra were measured.

8 105

20% Glycerol
10% Glycerol
0% Glycerol

6 105

All

mutants with merocyanine showed at
least 5 fold increase in quantum yield as

4 105

compared with merocyanine PSB in

2 105

buffer (Table 4-6). We also observed the

0
580 600 620 640 660 680 700
Wavelength (nm)

emission spectral changes with mutant
R132K:R111L:L121E:V76W. A 10 nm

Figure 4-33. Fluorescence of merocyanine 4-10 blue shift of the fluorescence emission of
PSB in buffer with different amount of glycerol.
bound merocyanine 4-10 in mutant
R132K:R111L:L121E: V76W was observed as compared to R132K:R111L:L121E:R59W
136,137

(Figure 4-32). Since the fluorescence of cyanine dye is also solvatochromic,
shifted

emission

the blue-

for

mutant Table 4-6. Quantum yield of merocyanine 4-10
PSB and merocyanine 4-10 with CRABPII mutant
R132K:R111L:L121E:V76W may be (R132K:R111L:L121E).
due to the polarity change within the

Mutant

Quantum
Yield (%)

Merocyanine 4-10 PSB

0.93

R132K:R111L:L121E:R59W

11.4

R132K:R111L:L121E:R76W

9.0

solvent one can enhance the quantum

R132K:R111L:L121E:T56W

7.7

yield of a fluorophore, because it can

R132K:R111L:L121E:R59W:T56D

6.8

suppress the energy lost due to rotation

R132K:R111L:L121E:R59W:V76L

11.9

binding pocket as discussed before.
It has been well-documented that
by increasing the viscosity of the

and

torsional

motion

of

the

310

75,128,130,131,136

fluorophores, thus, increase the chance of fluorescence (quantum yield).

We

suggest that the quantum yield enhancement of bound merocyanine 4-10 PSB is due to the
restriction of molecular movement inflicted by the protein. To test this hypothesis, we measured
the fluorescence of merocyanine PSB in buffer with different amount of glycerol. An increase in
fluorescence intensity was observed as the viscosity (percentage of glycerol) of the buffer
increased (Figure 4-33).

As previously discussed, we believe that the enhancement of

merocyanine fluorescence inside the CRABPII mutants is due to restriction of molecular
movement.
In conclusion, we have demonstrated that we can successfully engineer CRABPII as a
fluorescent protein through incorporation of a suitable fluorophore. In addition, upon binding to
merocyanine as a PSB, the fluorescence and the absorption of the bound merocyanine is more
red-shifted which leads to the ability to
specifically excite it and observe the fluorescent
protein rather than the unbound fluorophore
Figure 4-34. Azulene and its resonance
(Figure 4-31 and 4-32). We also demonstrated form.
that upon binding the CRABPII mutants, the quantum yield of the bound fluorophore increases
(Table 4-7). This is important because greatly enhances signal to noise ratio.
4.2.2. Incorporation of Azulenic Aldehyde Chromophore with CRABPII
Azulene is a dark blue compound with a special chemical structure, a fusion of
cyclopentadiene with cyclo-heptatriene (Figure 4-34). Azulene is an aromatic compound with
10 π electrons. As a result of its unique structure, azulene can exist in two states (Figure 4-34).
From the ionic resonance structure, it is clear that the seven membered ring is more electrophilic

311

while the five membered
ring is more nucleophilic.

O

The

4-21

special

electronic

features of azulene inspired
Muthyala and co-workers
to use it in order to create a
Lys

N+
H
4-25

near

infrared

absorbing
140

bacteriorhodopsin.

They reasoned that the
azulene coupled with the
polyene PSB (4-21), would
Lys

N
H

generate
a
push-pull
4-26
Figure 4-35. Structure of azulenic aldehyde chromophore system, similar to that of
and its resonance structure after bound with
cyanine dyes that should
bacteriorhodopsin.
lead to a significant bathochromic shift (Figure 4-35). In addition, azulene already has a much
lower S1 state as compared with the cyclohexyl ring in retinal, thus, upon binding with
bacteriorhodopsin, the chromophore should be able to absorb near infrared. All the azulenic
aldehyde chromophores absorb at more than 672 nm when they form a PSB with
140

bacteriorhodopsin.

Since azulenic aldehyde compounds are such good chromophores, we

were interested in testing these analogs with CRABPII mutants.

312

4.2.2.1. Synthesis of Azulenic Aldehyde Chromophore 4-20 and 4-21

1. DMF, POCl3;
2. dil. Na2CO3,

O

quant. yield
4-22

CN
O
(EtO)2P

CN 1. NaH, 0 ˚C

rt/ THF
O

2.

4-22

4-23

rt, overnight, 92% yield
CN

4-23

i) DIBAL-H,
-78˚C
0 ˚C/
THF, 6 h;
ii) wet silica, 0 ˚C,
8 h, 55% yield,
E:Z / ~1:3

CHO

4-20

Scheme 4-2. Synthesis of azulenic aldehyde chromophore 4-20.
The azulene-1-carboxaldehyde 4-24 was synthesized by the formylation of azulene by
Vilsmeier-Haack reaction in quantitative yield. The HWE reaction of azulene-1-carboxaldehyde
with diethyl 3-cyano-2-methylallylphosphonate gave the elongated product 4-25 in 92% yield.
Reduction of nitrile 4-26 gave the corresponding aldehyde 4-20 in ~55% (Scheme 4-2).
Chromophore 4-21 was obtained by a similar sequence of reactions used of the synthesis of
aldehyde 4-20. The overall yield over two steps is 20% to give the final product aldehyde 4-21
(Scheme 4-3).

313

CN
O
(EtO)2P

CN
i) NaH, 0 ˚C

rt/ THF;

ii)
CHO
4-24
4-20
overnight, 70% yield

CN

CHO

i) DIBAL-H,
-78 ˚C
0 ˚C /
THF, 6 h;
4-24

ii) wet silica, 0 ˚C,
8 h, 30% yield,
E:Z / ~1:3

Scheme 4-3. Synthesis of azulenic aldehyde chromophore 4-21.

314
314

4-21

4.2.2.2.

UV-vis

Spectroscopic

Study with Chromophore 4-20 and
4-21

0.2 equivalent of
Azulenic 4-20 after
10mins

0.2

The UV-vis spectra for both

were measured in EtOH and buffer.
It

was

observed

that

azulenic

aldehyde 4-21 is not soluble in
buffer.

0.15
Absorbance

azulenic aldehydes 4-20 and 4-21

0.1

0.05

In order to investigate
0

whether
solubilize
aldehyde

CRABPII
the

mutants

insoluble
4-21,

can

300

400 500 600
Wavelength (nm)

700

azulene Figure 4-36. UV-vis spectra of azulenic aldehyde 420 with R132K:R111L:L121E.
mutant

R132K:R111L:L121E was added to the solution, however, a UV-vis peak consistent with a
bound chromophores not observed. Therefore, the following studies are focused on azulenic
aldehyde 4-20. Since azulene 4-20 has a similar size and structure as C15 aldehyde, we are
wondering whether azulenic aldehyde 4-20 would be also totally embedded by CRABPII
mutants, therefore, the bound azulenic aldehyde 4-20 would be more sensitive to the mutations
around. Azulenic aldehyde 4-20 shows better shifted spectra at low equivalents of chromophore
with mutant R132K:R111L:L121E as compared to C15 aldehyde (Figure 4-36). In addition, we
found that the spectra of R132K:R111L:L121E mutants with azulenic aldehyde 4-20 exist as a
mixture of PSB and unbound or SB absorptions. To investigate whether the peak around 440 nm
is unbound azulenic aldehyde 4-20 or its SB, pKa of the bound azulene PSB was measured.
The pKa of the bound azulenic aldehyde 4-20 is 9.4, which indicated that the peak around 440
315

nm at pH of 7.3 is unbound azulenic
Table 4-7. UV-vis study of azulenic aldehyde 4-20
with CRABPII mutant (R132K:R111L:L121E).
aldehyde 4-20. Moreover, during the
a

Mutant

λmax

Azulenic Aldehyde 4-20

436

Azulenic Aldehyde 4-20 PSB

496

R132K:R111L:L121E

507

R132K:R111L:L121E:R59E

551

R132K:R111L:L121E:R59Q

540

R132K:R111L:L121E:R59W

525

the

R132K:R111L:L121E:A36E

541

absorptions

of

R132K:R111L:L121E:T56D

442

chromophore

changes.

R132K:R111L:L121E:T56Q

526

R132K:R111L:L121E:T56W

507

R132K:R111L:L121E:V76W

502

acid base titration, we confirmed that
the deconvoluted peak is azulenic
aldehyde PSB peak (Figure 4-37 and
Table 4-7).

introduced at different positions along

close

is sensitive to the negative dipole.

Since

the C15 aldehyde exists in two different

0.08

Absorbance

indicated that the bound azulenic aldehyde

the

the
bound

As

the

to

the

azulenic

ring

of

0.1

A single mutation R59E

R132K:R111L: L121E (Table 4-7). This

chromophore,

chromophore based on the model,

red-shifting is induced (Figure 4-38)

induces 44 nm red-shift as compared to

bound

negative charged residues are placed

a Absorption of azulenic aldehyde 4-20 PSB based on deconvolution

(Table 4-7).

When residues are

507nm

0.06
0.04
0.02
0
300 350 400 450 500 550 600 650 700
Wavelength (nm)

Figure
4-37.
UV-vis
spectrum
of
R132K:R111L:L121E with azulenic aldehyde 420 at pH 6.8.
316

binding

modes

with

mutant

R132K:R111L

(alternate

binding

pocket)

R132K:R111L:L121E

A36E (541nm)
and

(original trajectory)

(Figure 1-68),

azulenic aldehyde 59 may also exhibit
two different binding conformations as
bound with different mutants.

As we

W59E (551nm)
proposed
previously,
mutant
Figure 4-38. Model of azulenic aldehyde 4-20 with
R132K:R111L:L121E:R59W
λmax
of
the R132K:R111L binds C aldehyde in
15
glutamate at 36 and 59 mutants is showed in
parenthesis.
the alternate binding pocket. To test
whether azulenic 4-20 would bind in the alternate binding pocket, azulenic 4-20 was tested with
a series of R132K:R111L mutants.

Introduction of Glutamate at 59, that is 6.0 Å away from the

bound chromophore (Figure 4-39),

R59
induces only 6 nm red-shift as
compared

with

R132K:R111L.

the
Moreover,

C15 aldehyde

mutant
when

neutral leucine was introduced, mutant

6.0 Å

R132K:R111L:R59L absorbs at 541
nm that is very similar to mutant
R132K:R111L. Based on our result,

Figure 4-39. Crystal structure of C15 aldehyde in
we believe 4-20 binds to the alternate alternate binding pocket with KL:T54E showing
Arg59 position.
pocket with mutant R132K:R111L
series. From the crystal structure of C15 aldehyde with mutant R132K:R111L:T54E, T54E

317

should be at the middle of the bound 4-20. Introduction of valine at position 54 induces a
bathochromic shift as a result of removal of the negative dipole at the middle of the bound
chromophore. This result further supports the hypothesis that we proposed in Section 1.2.2
(Figure 1-34) about the effect of electrostatic potential on
Table 4-8.
UV-vis study of wavelength regulation. In addition, Phe15 is close to the
azulenic aldehyde 4-20 with
CRABPII
mutant PSB of the bound chromophore (Figure 4-40).
(R132K:R111L).
Introduction of the polar tyrosine residue at position 15
Mutant
λmax
leads to a blue shifted spectrum (Table 4-8). The result
Azulenic Aldehyde 59
436
also fits with our electrostatic potential argument proposed
Azulenic Aldehyde 59
496
PSB
in Section 1.2.2.
R132K:R111L
539
4.3. Conclusion
R132K:R111L:R59E
545
R132K:R111L:R59L

541

R132K:R111L:T51V

550

R132K:R111L:T51E

533

R132K:R111L:F15Y

530

a

Absorption of azulenic aldehyde
4-20 PSB based on deconvolution.

318

In

conclusion,

demonstated
incorporation

we

successfully
of

different

have

F15

through

C15 aldehyde

aldehydic

chromophores (merocyanine 4-10 and 411 and azulenic aldehyde 4-20), the
ability of CRABPII as a potential
chromphoric

protein.

In

addition,

T54
binding of merocyanine with CRABPII,
through the formation of a PSB with the

Figure 4-40. Crystal structure of C15 aldehyde in
R132K:R111L:T54E with Thr54 and Phe15 shown.

engineered lysine, turns merocyanine
into cyanine dye that yields a significant red-shifted absorption and fluorescence properties.
These mutants can enhance the quantum yield of merocyanine PSB, thus enhancing the signal to
noise for the fluorescent measurement. Through incorporation of different chromophores and
through mutagenesis, the absorptions and fluorescence of the surrogates can be modulated. This
provides more flexibility to create different protein tags using CRABPII. Since we demonstrated
that restriction of fluorophore movement can enhance quantum efficiency, cellular retinol
binding protein II (CRBPII) may be able to enhance the quantum yield further. This is because
CRBPII has a tighter binding pocket as compared with CRABPII, therefore, it can restrict the
movement of the bound fluorophore and increase the quantum efficiency.

319

4.3. Materials and Methods
A. Biology
i. Protein Mutagenesis, Bacterial Expression and Purification
1) Protein Mutation on pET-17b vector
The PCR conditions and recipes for this chapter have been described details in Chapter 1
(Material and Methods)
2) DNA Transformation and Purification
DNA transformation, and purification of pET-17b plasmids, that contains CRABPII
mutants, using QIAGEN plasmid purification kit are mentioned in details in Chapter 1 (Materials
and Methods).
Table 4-9. Primers for Mutagenesis in Chapter 4
Mutant

Primers
name

Primers Sequence

Template

Seq.
No.

5’-GGGCCTCCCATCCAG
BBB376 AGTCTGCTCCTCAAACT
R132K:R111L
C-3’
:L121E:R59W
5’-GAGTTTGAGGAGCAG
:V76L
BBB377 ACTCTGGATGGGAGGCC
C-3’

R132K:R111L:
BB773
L121E:R59W

5’-CTACATCAAAACCTC
BBB357 CGATACCGTGCGCACCA
R132K:R111L
CAG-3’
:L121E:R59W
5’-CTACATCAAAACCTC
:T56D
BBB358 CGATACCGTGCGCACCA
CAG-3’

R132K:R111L:
BB739
L12E:R59W

320

DNA samples preparation for DNA sequencing has been detailed in Chapter 1 (Materials
and Methods)
3) Protein Expression and Purifications
Protein expression and purification of CRABPII mutants (pET-17b vector) are described in
details in Chapter 1 (Materials and Methods).
ii. Protein Characterization and Binding Assays
1) Calculation Extinction Coefficients of CRABPII mutants
The absorption extinction coefficients (ε) for CRABPII mutants are determined according
to the method described by Gill and von Hippel. The details have been mentioned in Chapter 1
(Materials and Methods)
2) Reductive Amination
The procedures for reductive amination and the electrospray mass spectrometer setting has
been fully described in Chapter 1 (Materials and Methods)
3) UV-vis and Fluorescent Spectroscopic Study
The procedures and experimental setup for UV-vis binding study have been mentioned in
details in Chapter 1 (Material and Methods)
iii. Emission Spectrum Measurement
1) Merocyanine 4-10
3 mL of 0.1 µM of merocyanine 4-10 dissolved in PBS buffer (pH 7.3) was incubated in
fluorescence quartz cuvette (1 cm path, 3.5 mL) at 24 ˚C for 10 minutes. The emission spectrum
was measured according to the following setting:
i)

Excitation wavelength: 491 nm and 571 nm respectively

321

ii) Scan range: 500 nm to 900 nm and 580 nm to 900 nm respectively
iii) Excitation slit: 1 nm
iv) Emission slit: 12 nm
v)

Integration time: 1 s

vi) Scan rate: 1 nm/s
2) Merocyanine 4-10 PSB
3 mL of 0.1 µM of merocyanine 4-10 PSB dissolved in PBS buffer (pH 7.3) was stirred
in fluorescence quartz cuvette (1 cm path, 3.5 mL) at 24 ˚C for 10 minutes. The emission
spectrum was measured according to the following setting:
i.

Excitation wavelength: 571 nm respectively

ii.

Scan range: 580 nm to 900 nm respectively

iii.

Excitation slit: 1 nm

iv.

Emission slit: 12 nm

v.

Integration time: 1 s

vi.

Scan rate: 1 nm/s

3) Merocyanine 4-10 with CRABPII mutant
3 mL of 0.5 µM of CRABPII mutants in PBS buffer

(pH 7.3) was incubated in

fluorescence quartz cuvette (1 cm path, 3.5 mL) at 24 ˚C for 10 minutes. Merocyanine 4-10 (0.1
µM in cuvette) was added into protein solution and stirred for an addition of 60 minutes at 24 ˚C.
The emission spectrum was measured according to the following setting:
i.

Excitation wavelength: 571 nm respectively

ii.

Scan range: 580 nm to 900 nm respectively

322

iii.

Excitation slit: 1 nm

iv.

Emission slit: 12 nm

v.

Integration time: 1 s

vi.

Scan rate: 1 nm/s

iv. Fluorescence Quantum Yield Measurement and Calculation
The Quantum yield measurement was measured according to the comparative method
141

developed by Willams, A. T. R. et al.

using Rhodamine-6G (Quantum Yield 95%) as a

standard.
v.

1) Generating standard curve for Rhodamine-6G
The fluorescence (excited at 480 nm, scan from 490 nm to 800 nm) and UV-vis spectra

(300 nm to 700 nm) of Rhodamine-6G at 0.01 µM, 0.02 µM, 0.03 µM and 0.04 µM were
measured. The concentration of the chromophore used must have OD480 lower than 0.1 to
prevent the self absorbance. The emission spectrum was measured according to the following
setting:
i.

Excitation wavelength: 480 nm

ii.

Scan range: 490 nm to 800 nm

iii.

Excitation slit: 1 nm

iv.

Emission slit: 12 nm

v.

Integration time: 1 s

vi.

Scan rate: 1 nm/s
The concentration of the chromophore used must have fluorescence intensity lower than

5

10 e that is the limit for the detector. The total fluorescence from 490 nm to 800 nm was

323

®

integrated by software DATAMAX . A linear plot of absorbance against total fluorescence was
obtained.
2) Merocyanine 4-10 PSB
The fluorescence and UV-vis spectra of Merocyanine 4-10 at 0.02 µM, 0.04 µM, 0.06
µM and 0.08 µM was measured. The concentration of the chromophore used must have OD570
lower than 0.1 to prevent the self absorbance. The emission spectrum was measured according
to the following setting:
i.

Excitation wavelength: 570 nm

ii.

Scan range: 580 nm to 900 nm

iii.

Excitation slit: 1 nm

iv.

Emission slit: 12 nm

v.

Integration time: 1 s

vi.

Scan rate: 1 nm/s
The concentration of the chromophore used must have fluorescence intensity lower than

5

10 e that is the limit for the detector. The total fluorescence from 580 nm to 900 nm was
integrated by software DATAMAX. A linear plot of absorbance (Abs) against total fluorescence
(Fl) was obtained.
x.

Merocyanine 4-10 with CRABPII mutants
3 mL of 0.1 µM of CRABPII mutants in PBS buffer

(pH 7.3) was incubated in

fluorescence quartz cuvette (1 cm path, 3.5 mL) at 24 ˚C for 10 minutes. 0.1 equivalent of
merocyanine 4-10 was added. The solution was incubated at 24 ˚C for 1 hour before each
fluorescence and UV-vis spectrum measurement.
324

The fluorescence and UV-vis spectra of

Merocyanine 4-10-CRABPII solution was measured after e 0.1 equivalent of merocyanine 4-10
was added. The OD571 for e measurement must be lower than 0.1 to prevent the self absorbance.
The emission spectrum was measured according to the following setting:
i.

Excitation wavelength: 570 nm

ii.

Scan range: 580 nm to 900 nm

iii.

Excitation slit: 1 nm

iv.

Emission slit: 12 nm

v.

Integration time: 1 s

vi.

Scan rate: 1 nm/s
The concentration of the chromophore used must have fluorescence intensity lower than

5

10 e that is the limit for the detector. The total fluorescence from 580 nm to 900 nm was
®

integrated by software DATAMAX . A linear plot of absorbance against total fluorescence was
obtained.
The fluorescence quantum yield of merocyanine PSB and the protein complex was
calculated based on the following equation:
Fluorescence Quantum Yield of sample (QY) = QYRhodamine-6G X slope of sample graph
plot (Abs against Fl) / slope of Rhodamine-6G plot (Abs against Fl)
B. Chemistry
General:

325

All-syntheses were carried out in a dry nitrogen atmosphere in the dark room with
minimum red lights unless otherwise were specified. All-the reaction vessels were flame-dried.
Tetrahydrofuran was distilled over sodium and benzophenone under dried N2 atmosphere.
Reactions were monitored by thin-layer chromatography (TLC, on Merck F254 silica gel 60
aluminum sheets, spots were either visible under light or UV-light (254 mm) or treated with an
oxidizing solution (KMnO4 stain). Column chromatography was performed with Silicycle silica
gel 60.
1

H-NMR spectra were recorded on a Varian Unity+ 500 spectrometer with deuterated

chloroform (CDCl3; δ = 7.24 ppm) or deuterated DMSB (DMSO-d6; δ = 2.5 ppm, br) as an
1

internal standard. H noise-decoupled

13

C spectra were recorded on a Varian Unity+ 500

spectrometer at 125MHz with deuterated chloroform (CDCl3; δ = 77 ppm) or deuterated DMSO
(DMSO-d6; δ = 39.5 ppm) as an internal standard.

Synthesis of (2E,4E)-4-(1,3,3-trimethylindolin-2-ylidene)but-2-enenitrile (4-15)

O

NaH, (EtO)2P(O)CH2CN
THF, rt.

N

CN

N
4-15

4-13

2-(Diethylphosphono)ethanenitrile (318 mg, 1.8 mmol) was added to NaH (83 mg, 60%
dispersion in oil, 1.8 mmol) in THF (25 mL) at 0 ˚C. The reaction mixture was stirred for 1 h at
room temperature. To the anion of the phosphonate was added 200 mg (1 mmol) of aldehyde (4-

326

13). The mixture was stirred for another 12 h at room temperature. The reaction was monitored
by the TLC (Rf = 0.75 (diethyl ether:40 – 60 light petroleum Ether / 20:80)) and was quenched
by ice. The aqueous layer was extracted by diethyl ether six times and the combined organic
layer was dried by anhydrous sodium sulfate (anhyd. Na2SO4) and was filtered. The product
was purified by chromatography (silica gel, diethyl ether:40 – 60 light petroleum / 20:80) to give
nitrile 4-15 (200 mg, 0.89 mmol, 41%) as a mixture of the 9-cis and all-trans compounds.
1

H-NMR (500 Mhz, CDCl3) of all-trans- (2E,4E)-4-(1,3,3-trimethylindolin-2-ylidene)but1

1

2

2-enenitrile: δ = 4.90 (d, JH-H = 16 Hz, H-15, 1H), 7.55 (d, JH-H = 16 Hz, JH-H = 13 Hz, H-14,
1

1

1

1H), 5.33 (d, JH-H = 13 Hz, H-13, 1H), 6.69 (d, JH-H = 7.9 Hz, H-3, 1H), 7.19 (td, JH-H = 7.7
2

1

1

2

Hz, JH-H = 0.7 Hz, H-2, 1H), 6.92 (t, JH-H = 7.6 Hz, H-1, 1H), 7.15 (dd, JH-H = 7.5 Hz, JH-H
= 0.7 Hz, H-6, 1H)
GC-MS = 224.3
Synthesis of (2E,4E)-4-(1,3,3-trimethylindolin-2-ylidene)but-2-enal (4-16)
CN

2) Silica-H2O

N

O

1) DIBAL-H, THF,
-78 ˚C
0 ˚C
N

4-16

4-15

A solution of 4-15 (200 mg, 0.89 mmol) in distilled THF (20 mL) was cooled to –78 ˚C.
DIBAL-H (0.59 ml of 1.5 M, 0.89 mmol) was added via syringe. The reaction mixture was
allowed to warm to 0 ˚C and was stirred for 6 h. A homogeneous mixture of water absorbed on

327

silica (1g, 0.3 g water/ gram silica) was added and was stirred for another 4 h at 0 ˚C. All-solids
were filtered off and thoroughly washed with diethyl ether. The organic layer was dried by
anhyd. Na2SO4 and solvents was evaporated in vacuo. The product was purified by
chromatography (silica gel, diethyl ether:40 – 60 light petroleum ether/ 40:60) to give all-trans
4-16 (120 mg, 0.53 mmol)
1

H-NMR (500 Mhz, CDCl3) of (2E,4E)-4-(1,3,3-trimethylindolin-2-ylidene)but-2-enal:
1

1

1

δ = 9.45 (d, JH-H = 8 Hz, H-16, 1H), 7.70 (t, JH-H = 13 Hz, H-14, 1H), 7.22 (td, JH-H = 7.8
2

1

1

2

Hz, JH-H = 1.1 Hz, H-2, 1H), 7.18 (d, JH-H = 7.2 Hz, H-6, 1H), 6.96 (td, JH-H = 7.5 Hz, JH-H
1

1

2

= 0.8 Hz, H-1, 1H), 6.75 (d, JH-H = 7.8 Hz, H-3, 1H), 5.94 (dd, JH-H = 14.2 Hz, JH-H = 8.1
1

Hz, H-15, 1H), 5.53 (d, JH-H = 12.5 Hz, H-13, 1H), 3.22 (s, H-12, 3H), 1.61 (s, H-10, 6H)
GC-MS = 227.2
Synthesis of (2E,4E,6E)-6-(1,3,3-trimethylindolin-2-ylidene)hexa-2,4-dienenitrile (4CN

NaH, (EtO)2P(O)CH2CN
O THF, rt
N

N

4-16

4-19

2-(Diethylphosphono)ethanenitrile (157 mg, 0.88 mmol) was added to 36.7 mg of 60%
NaH (36.7 mg, 0.88 mmol, 60% dispersion in oil) dispersion in oil suspended in THF (25 mL) at
0 ˚C. The reaction mixture was stirred for 1 h at room temperature. To the anion of the
phosphonate was added of aldehyde (4-16) (100 mg, 0.4 mmol). The mixture was stirred for

328

another 12 h at room temperature. The reaction was monitored by the TLC (Rf = 0.75 (diethyl
ether:40 – 60 light petroleum Ether / 20:80) and was quenched by ice. The aqueous layer was
extracted by diethyl ether three times and the combined organic layer was dried by anhydrous
sodium sulfate (anhyd. Na2SO4) and filtered. The product was purified by chromatography
(silica gel, diethyl ether:40 – 60 light petroleum / 20:80) to give 4-19 (100 mg, 0.36 mmol) of as
a mixture of the 16-cis and all-trans compounds.
GC-MS = 250.0
Synthesis of (2E,4E,6E)-6-(1,3,3-trimethylindolin-2-ylidene)hexa-2,4-dienal (4-12)
CN 1) DIBAL-H, THF,
-78 ˚C
0 ˚C
2) Silica-H2O
N

O

N

4-19

4-12

A solution of 4-19 (100 mg, 0.36 mmol) in distilled THF (10 mL) was cooled to –78 ˚C.
DIBAL-H (0.4 ml, 1.5 M, 0.23 mmol) was added via syringe. The reaction mixture was allowed
to warm to 0 ˚C and was stirred for 6 h. A homogeneous mixture of water absorbed on silica
(1g, 0.3 g water/ gram silica) was added and stirring continued for another 4 h at 0 ˚C. All-solids
were filtered off and thoroughly washed with diethyl ether. The organic layer was dried by
anhyd. Na2SO4 and solvents was evaporated

in vacuo.

The product was purified by

chromatography (silica gel, diethyl ether: 40 – 60 light petroleum ether/ 40:60) to give 4-12 (50
mg, 0.19 mmol) of as all-trans compounds.

329

1

H-NMR

(500

Mhz,

CDCl3)

of

all-trans-(2E,4E,6E)-6-(1,3,3-trimethylindolin-2-

1

1

2

ylidene)hexa-2,4-dienal: δ = 9.47 (d, JH-H = 8 Hz, H-18, 1H), 6.00 (dd, JH-H = 15 Hz, JH-H =
1

2

1

8 Hz, H-17, 1H), 7.26 (dd, JH-H = 14 Hz, JH-H = 13 Hz, H-16, 1H), 6.20 (dd, JH-H = 14 Hz,
2

1

1

2

JH-H = 12 Hz, H-15, 1H), 5.22 (d, JH-H = 13 Hz, H-13, 1H), 7.19 (td, JH-H = 7.6 Hz, JH-H =
1

2

1

1.1 Hz, H-2, 1H), 7.15 (dd, JH-H = 7.8 Hz, JH-H = 1.1 Hz, H-6, 1H), 6.90 (td, JH-H = 7.6 Hz,
2

1

JH-H = 0.9 Hz, H-1, 1H), 6.68 (d, JH-H = 7.8 Hz, H-3, 1H), 3.16 (s, H-12, 3H), 1.58 (s, H-10,

6H)
GC-MS=253.2
Synthesis of (2E,4E,6E)-3-methyl-6-(1,3,3-trimethylindolin-2-ylidene)hexa-2,4-dienenitrile
(4-14)

CN
O
(EtO)2P

i) NaH, 0 ˚C

ii)

CN
rt, THF;
O
N

N

4-13
4-14

overnight

4-(Diethylphosphono)-3-methylbut-2-enenitrile (174 mg, 1 mmol) was added to NaH (42
mg, 60% dispersion in oil) in THF (25 mL) at 0 ˚C. The reaction mixture was stirred for 1 h at
room temperature. To the anion of the phosphonate was added aldehyde (4-13) (100 mg, 1
mmol). The mixture was stirred for another 12 h at room temperature. The reaction was
monitored by the TLC (Rf = 0.72, 0.70 (diethyl ether:40 – 60 light petroleum Ether / 20:80)) and
was quenched by ice. The aqueous layer was extracted by diethyl ether six times and the

330

combined organic layer was dried by anhydrous sodium sulfate (anhyd. Na2SO4) and was
filtered. The product was purified by chromatography (deactivated silica gel, diethyl ether:40 –
60 light petroleum / 20:80) to give 4-14 (200 mg, 0.9 mmol) of as a mixture of 16-cis and alltrans compounds.
1

H-NMR

(500

Mhz,

CDCl3)

of

(2E,4E,6E)-3-methyl-6-(1,3,3-trimethylindolin-21

ylidene)hexa-2,4-dienenitrile: δ = 4.97 (s, H-18, 1H), 5.32 (d, JH-H = 13 Hz, H-13, 1H), 6.04 (d,
1

1

2

1

JH-H = 14 Hz, H-15, 1H), 7.15 (dd, JH-H = 14.5 Hz, JH-H = 12.0 Hz, H-14, 1H), 7.17 (td, JH2

H

1

2

= 7.7 Hz, JH-H = 1.2 Hz, H-2, 1H), 7.13 (dd, JH-H = 7.3 Hz, JH-H = 1.0 Hz, H-6, 1H), 6.87
1

2

1

(td, JH-H = 7.5 Hz, JH-H = 0.9 Hz, H-1, 1H), 6.64 (d, JH-H = 7.96 Hz, H-3, 1-H), 3.12 (s, H-12,
3H), 2.20 (s, H-17, 3H), 1.56 (s, H-10, 6H).
13

C-NMR (126 Mhz, CDCl3) of (2E,4E,6E)-3-methyl-6-(1,3,3-trimethylindolin-2-

ylidene)hexa-2,4-dienenitrile: δ = 119.6 (s, C-19), 91.3 (s, C-18), 157.5 (s, C-16), 133.1 (s, C15), 127.9 (s, C-14), 95.5 (s, C-13), 160.8 (s, C-8), 45.9 (s, C-9), 138.9 (s, C-5), 123.2 (s, C-6),
120.2 (s, C-1), 121.6 (s, C-2), 106.3 (s, C-3), 144.7 (s, C-4), 16.7 (s, C-17), 28.5 (s, C-10), 29.1
(s, C-12)

331

Synthesis of (2E,4E,6E)-3-methyl-6-(1,3,3-trimethylindolin-2-ylidene)hexa-2,4-dienal (4-10)

CN

O
i) DIBAL-H,
-78 ˚C
0 ˚C,
THF, 6 h;
ii) wet silica, 0 ˚C, 8 h

N

N

4-14

4-10

A solution of 4-14 (110 mg, 0.41 mmol) in THF (10 mL) was cooled to –78 ˚C. 0.248 ml
of 2.5 M (0.61 mmol) DIBAL-H was added via syringe. The reaction mixture was allowed to
warm to 0 ˚C and was stirred for 6 h. A homogeneous mixture of water absorbed on silica (0.7g,
0.3 g water/ gram silica) was added and stirring continued for another 4 h at 0 ˚C. All-solids
were filtered off and thoroughly washed with diethyl ether. The organic layer was dried by
anhyd. Na2SO4 and solvents was evaporated

in vacuo. The product was purified by

chromatography (silica gel, diethyl ether:40 – 60 light petroleum ether/ 30:70) to give 4-10 (70
mg, 0.26 mmol) as cis-trans mixture.
1

H-NMR (500 MHz, CDCl3) of (2E,4E,6E)-3-methyl-6-(1,3,3-trimethylindolin-21

1

ylidene)hexa-2,4-dienal: δ = 10.0 (d, JH-H = 8.3 Hz, H-19, 1H), 5.89 (d, JH-H = 8.3 Hz, H-17,
1

1

1

1H), 6.11 (d, JH-H = 14.7 Hz, H-15, 1H), 7.38 (dd, JH-H = 14.7 Hz, H-14, 1H), 5.43 (d, JH-H =
1

2

1

12 Hz, H-13, 1H), 7.18 (td, JH-H = 7.6 Hz, JH-H = 1.3 Hz, H-2, 1H), 7.14 (dd, JH-H = 7.29 Hz,

332

2

1

2

1

JH-H = 0.4 Hz, H-6, 1H), 6.88 (td, JH-H = 7.4 Hz, JH-H = 0.9 Hz, H-1, 1H), 6.66 (d, JH-H =

7.9 Hz, H-3, 1H), 3.15 (s, H-12, 3H), 2.31 (s, H-18, 3H), 1.59 (s, H-10, 6H)
13

C-NMR (124 MHz, CDCl3) of (2E,4E,6E)-3-methyl-6-(1,3,3-trimethylindolin-2-

ylidene)hexa-2,4-dienal: δ = 13.2 (s, C-18), 28.5 (s, C-10), 29.2 (s, C-12), 46.0 (s, C-9), 96.4 (s,
C-13), 106.5 (s, C-3), 120.4 (s, C-1), 121.6 (s, C-2), 125.4 (s, C-17), 126.2 (s, C-15), 127.9 (s, C6), 133.7 (s, C-14), 138.8 (s, C-5), 144.6 (s, C-4), 156.3 (s, C-16), 161.2 (s, C-8), 190.5 (s, C-19)
Synthesis

of

(2E,4E,6E,8E)-3-methyl-8-(1,3,3-trimethylindolin-2-ylidene)octa-2,4,6-

trienenitrile (4-19)

i) NaH, 0 ˚C

CN
rt, THF;

(EtO)2P(O)CH2CN

O

ii)

N
N

4-16

4-19

overnight

4-(Diethylphosphono)-3-methylbut-2-enitrile (220 mg, 1 mmol) was added to NaH (44
mg, 60% dispersion in oil) THF (20 mL) at 0 ˚C. The reaction mixture was stirred for 1 h at
room temperature. To the anion of the phosphonate was added aldehyde (4-16). The mixture
was stirred for another 12 h at room temperature. The reaction was monitored by the TLC (Rf =
0.72, 0.70 (diethyl ether:40 – 60 light petroleum Ether / 20:80)) and was quenched by ice. The
aqueous layer was extracted by diethyl ether six times and the combined organic layer was dried
by anhydrous sodium sulfate (anhyd. Na2SO4) and was filtered. The product was purified by

333

chromatography (deactivated silica gel, diethyl ether:40 – 60 light petroleum / 20:80) to give 419 (120 mg, 0.45 mmol) of as a mixture of 16-cis and all-trans compounds.
Synthesis of (2E,4E,6E,8E)-3-methyl-8-(1,3,3-trimethylindolin-2-ylidene)octa-2,4,6-trienal
(4-12)

O

CN i) DIBAL-H,
-78 ˚C
THF, 6 h;

0 ˚C,
N

ii) wet silica, 0 ˚C,
8h

N

4-12

4-19

A solution of 4-19 (120 mg, 0.41 mmol) in THF (10 mL) was cooled to –78 ˚C. DIBAL-H
(0.199 ml, 2.5 M, 0.61 mmol) was added via syringe. The reaction mixture was allowed to warm
to 0 ˚C and was stirred for 6 h. A homogeneous mixture of water absorbed on silica (0.7g, 0.3 g
water/gram silica) was added and stirring continued for another 4 h at 0 ˚C. All-solids were
filtered off and thoroughly washed with diethyl ether. The organic layer was dried by anhyd.
Na2SO4 and solvents was evaporated in vacuo. The product was purified by chromatography
(silica gel, diethyl ether:40 – 60 light petroleum ether/ 30:70) to give 4-12 (70 mg, 0.26 mmol) as
cis-trans mixture.
Synthesis of Azulene-1-carbaldehyde (4-22)
1. DMF, POCl3, 0 ˚C
2. dil. Na2CO3, 0 ˚C

O

rt

4-22

334

˚

Azulene (500 mg, 3.9 mmol) dissolved in DMF (5 mL) was stirred at 0 C. POCl3 (2.5 g,
˚

16.5 mmol) was added dropwise. The mixture was then stirred at 0 C for 30 min and was
followed by 2 h at rt. The reaction was quenched by ice-cold Na2CO3. An equal volume of
ethyl acetate was added and the resulting mixture was stirred at 0 ˚C for 12 h. Organic layer was
then separated.

The aqueous layer was extracted with ethyl acetate for three times.

The

combined organic layer was washed with brine and was dried over anhydrous MgSO4. The
solvent was evaporated under vacuo. The final product 4-22 (560 mg, 0.356 mmol) was purified
by column chromatography (30% EtOAc: Hex: 0.05% Et3N) yielding 92%.
1

H NMR (CDCl3, 500 Mhz): δ 10.36 (s, 1H), 9.59 (d, J = , 1H), 8.51 (d, J = , 1H), 8.27 (d,

J = , 1H), 7.85 (t, J = , 1H), 7.64 (t, J = , 1H), 7.54 (t, J = , 1H), 7.34 (d, J = , 1H)
13

C NMR (CDCl3, 125 Mhz): δ 191, 138.8, 137.4, 134.1, 133.5, 128.7, 128.2, 127.5,

125.3, 124.2, 120.1
Synthesis of (2E,4E)-5-(azulen-1-yl)-3-methylpenta-2,4-dienenitrile (4-23)

O
(EtO)2P

CN 1. NaH, 0 ˚C

CN

rt, THF
O

2.

4-22

4-23

rt, overnight

To a stirred suspension of NaH (213 mg, 5.1 mmol, 60% in mineral oil) in THF (50 mL) at
˚

0 C, ylide (1.11 g, 5.1 mmol) was added dropwise. The mixture was stirred at rt for 1 h.
Azulene-1-carbaldehyde (400 mg, 2.6 mmol) was dissolved in 10 mL THF and was added to the
335

mixture dropwise. The mixture was stirred at rt for 12 h. Sat. NH4Cl solution was added and the
organic layer was separated. The aqueous layer was extracted with ethyl acetate for three times.
The combined organic layer was washed with brine and was dried over anhydrous MgSO4. The
crude was concentrated under vacuo and was purified by column chromatography (20% EtOAc:
Hex: 0.05% Et3N) yielding 88% of final product (500 mg, 2.3 mmol).
1

H NMR (CDCl3, 500 Mhz): δ 8.43 (d, J = 10 Hz, 1H), 8.24 (d, J = 9 Hz, 1H), 8.15 (d, J =

4.5 Hz), 1H), 7.60 (t, J = 10 Hz, 1H), 7.48 (d, J = 16 Hz, 1H), 7.39 (d, J = 2.5 Hz, 1H), 7.22 (t, Ji
= 7Hz, 1H), 7.18 (t, J = 7 Hz, 1H), 6.91 (d, J = 16 Hz, 1H), 5.27 (s, 1H), 2.36 (s, 3H)
13

C NMR (CDCl3, 125 Mhz): δ 138.8, 137.4, 133.8, 133.5, 127.5, 125.7, 125.2, 124.1,

119.9, 95.2, 16.7
Synthesis of (2E,4E)-5-(azulen-1-yl)-3-methylpenta-2,4-dienal (4-20)

CN

4-23

i) DIBAL-H,
-78 ˚C
0 ˚C,
THF, 6h;
ii) wet silica, 0 ˚C,
8h

CHO

4-20

To a stirred solution of nitrile 4-23 (400 mg, 1.8 mmol) in 30 mL THF at –78 ˚C, DIBALH (1.46 mL, 3.6 mmol, 2.5 M in toluene) was added dropwise. The reaction mixture was then
stirred at rt for 4 h. The mixture was brought down to 0 ˚C followed by addition of wet silica.
The reaction was then further stirred at 0 ˚C for 2 h. The mixture was filtered and the filtrate

336

was concentrated under vacuo and further purified by column chromatography yielding 55% of
aldehyde 4-20 (220 mg, 1.0 mmol) in a mixture of isomer.
1

H NMR (CDCl3, 500 Mhz): δ 10.34 (d, J = 8 Hz, 1H), 10.14 (d, J = 8 Hz, 1H), 8.58 (d, J

= 10 Hz, 1H), 8.46 (d, J = 10 Hz, 1H), 8.45 (d, J = 10 Hz, 1H), 8.30 (d, J = 9 Hz, 1H), 8.25 (d, J
= 9 Hz, 1H), 8.24 (d, J = 9 Hz, 1H), 8.21 (d, J = 5 Hz, 1H), 8.16(d, J = 16 Hz, 1H), 7.92 (d, J =
16 Hz, 1H), 7.21 (d, J = 9 Hz, 1H), 7.18 (d, J = 9 Hz, 1H), 6.99 (d, J = 16 Hz, 1H), 6.81 (d, J =
16 Hz, 1H), 6.10 (d, J = 8 Hz, 1H), 5.89 (d, J = 8 Hz, 1H), 5.33 (s, 1H), 5.09 (s, 1H), 2.47 (s,
3H), 2.41 (s, 3H)
Synthesis of (2E,4E,6E,8E)-9-(azulen-1-yl)-3,7-dimethylnona-2,4,6,8-tetraenenitrile (4-24)
CN
CN
O
(EtO)2P

i) NaH, 0 ˚C

rt, THF;

ii)
CHO

4-24

4-20
overnight
To a stirred suspension of NaH (56 mg, 1.35 mmol, 60% in mineral oil) in THF (30 mL) at
0 ˚C, ylide (293 mg, 1.35 mmol) was added dropwise. The mixture was stirred at rt for 1h.
Aldehyde 4-20 (150 mg, 0.68 mmol) was dissolved in 5 mL THF and was added to the mixture
dropwise. The mixture was stirred at rt for 12 h. Sat. NH4Cl solution was added and the organic
layer was separated. The aqueous layer was extracted with ethyl acetate for three times. The
combined organic layer was washed with brine and was dried over anhydrous MgSO4. The

337

crude was concentrated under vacuo and was purified by column chromatography (20% EtOAc:
Hex: 0.05% Et3N) yielding 67% of final product 4-24 (130 mg, 0.46 mmol).
1

H NMR (CDCl3, 500 Mhz): δ 8.44 (d, J = 11 Hz, 1H), 8.28 (d, J = 4.5 Hz, 1H), 8.26 (d, J

= 9 Hz, 1H), 8.25 (d, J = 9 Hz, 1H), 8.15 (d, J = 4 Hz, 1H), 7.60 (t, J = 9.3 Hz, 1H), 7.59 (t, J =
9.3 Hz, 1H), 7.51 (d, J = 5 Hz, 1H), 7.47 (d, J = 15 Hz, 1H), 7.42 (d, J = 4.3 Hz, 1H), 7.40 (d, J =
4.3 Hz, 1H), 7.23 (d, J = 10 Hz, 1H), 7.23 (d, J = 10 Hz, 1H), 7.20 (t, J = 10 Hz, 1H), 7.18 (t, J =
10 Hz, 1H), 6.91 (d, J = 15.8 Hz, 1H), 5.26 (s, 1H), 5.08 (s, 1H), 2.35 (s, 3H), 2.21 (s, 3H)
Synthesis of (4E,6E,8E)-9-(azulen-1-yl)-3,7-dimethylnona-2,4,6,8-tetraenal (4-21)
CN

CHO
i) DIBAL-H,
-78 ˚C
0 ˚C,
THF, 6h;

4-24

ii) wet silica, 0 ˚C,
8h

4-21

To a stirred solution of nitrile 4-24 (100 mg, 0.35 mmol) in THF (20 mL) at –78 ˚C,
DIBAL-H (0.281 mL, 0.7 mmol, 2.5M in toluene) was added dropwise. The reaction mixture
was then stirred at rt for 4 h. The mixture was brought down to 0 ˚C followed by addition of wet
silica. The reaction was then further stirred at 0 ˚C for 2 h. The mixture was filtered and the
filtrate was concentrated under vacuo and further purified by column chromatography yielding
30% of aldehyde 4-21 (30 mg, 0.1 mmol) in a mixture of isomer.

338

1

H NMR (CDCl3, 500 Mhz): δ 10.21 (d, J = 8 Hz, 1H), 10.10 (d, J = 8 Hz, 1H), 8.38 (d, J

= 10 Hz, 1H), 8.17 (s, 1H), 8.16 (d, J = 6 Hz, 1H), 7.51 (t, J = 10 Hz, 1H), 7.37 (d, J = 5 Hz,
1H), 7.31 (d, J = 16 Hz, 1H), 7.18 (dd, J = 15 Hz, 12.5 Hz, 1H), 7.10 (t, J = 10 Hz, 1H), 7.07 (t, J
= 10 Hz, 1H), 7.00 (d, J = 16 Hz, 1H), 6.38 (d, J = 15 Hz, 1H), 6.36 (d, J = 12 Hz, 1H), 5.98 (d, J
= 8 Hz, 1H), 5.83 (d, J = 8 Hz, 1H), 2.36 (d, J = 1 Hz, 3H), 2.32 (d, J = 1 Hz, 3H), 2.19 (s, 3H),
2.14 (d, J = 1 Hz, 1H)
13

C NMR (CDCl3, 125 Mhz): δ 191.1, 154.9, 143.5, 141.8, 138.5, 136.9, 135.9, 134.2,

133.6, 133.4, 132.7, 130.8, 129.9, 128.8, 127.5, 124.3, 123.1, 122.2, 119.6, 13.2,

339

BIBLIOGRAPHY

340

Bibliography
(1)
Jackson, D. A.; Berg, P.; Symons, R. H. Proceedings of the National Academy of
Sciences of the United States of America 1972, 69, 2904-&.
(2)
Morrow, J. F.; Berg, P. Proceedings of the National Academy of Sciences of the
United States of America 1972, 69, 3365-&.
(3)
Guzman, L. M.; Belin, D.; Carson, M. J.; Beckwith, J. Journal of Bacteriology
1995, 177, 4121-4130.
(4)
Gossen, M.; Bujard, H. Proceedings of the National Academy of Sciences of the
United States of America 1992, 89, 5547-5551.
(5)
Gorman, C. M.; Moffat, L. F.; Howard, B. H. Molecular and Cellular Biology 1982,
2, 1044-1051.
(6)

Luckow, V. A.; Summers, M. D. Bio-Technology 1988, 6, 47-55.

(7)

Choi, P. J.; Cai, L.; Frieda, K.; Xie, S. Science 2008, 322, 442-446.

(8)
Rosenfeld, N.; Young, J. W.; Alon, U.; Swain, P. S.; Elowitz, M. B. Science 2005,
307, 1962-1965.
(9)
Hall, D. A.; Zhu, H.; Zhu, X. W.; Royce, T.; Gerstein, M.; Snyder, M. Science
2004, 306, 482-484.
(10)

Valdivia, R. H.; Falkow, S. Science 1997, 277, 2007-2011.

(11) He, T. C.; Zhou, S. B.; da Costa, L. T.; Yu, J.; Kinzler, K. W.; Vogelstein, B.
Proceedings of the National Academy of Sciences of the United States of America
1998, 95, 2509-2514.
(12)

Smith, D. B.; Johnson, K. S. Gene 1988, 67, 31-40.

(13)

Guan, K. L.; Dixon, J. E. Analytical Biochemistry 1991, 192, 262-267.

(14) Hochuli, E.; Bannwarth, W.; Dobeli, H.; Gentz, R.; Stuber, D. Bio-Technology
1988, 6, 1321-1325.
(15) Maina, C. V.; Riggs, P. D.; Grandea, A. G.; Slatko, B. E.; Moran, L. S.;
Tagliamonte, J. A.; McReynolds, L. A.; Diguan, C. Gene 1988, 74, 365-373.
(16) Uhlen, M.; Forsberg, G.; Moks, T.; Hartmanis, M.; Nilsson, B. Current Opinion in
Biotechnology 1992, 3, 363-369.

341

(17)

Hochuli, E.; Dobeli, H. Biological Chemistry Hoppe-Seyler 1987, 368, 748-748.

(18)
184.

Hochuli, E.; Dobeli, H.; Schacher, A. Journal of Chromatography 1987, 411, 177-

(19) Hopp, T. P.; Prickett, K. S.; Price, V. L.; Libby, R. T.; March, C. J.; Cerretti, D. P.;
Urdal, D. L.; Conlon, P. J. Bio-Technology 1988, 6, 1204-1210.
(20) Huh, W. K.; Falvo, J. V.; Gerke, L. C.; Carroll, A. S.; Howson, R. W.; Weissman,
J. S.; O'Shea, E. K. Nature 2003, 425, 686-691.
(21) Stokoe, D.; Macdonald, S. G.; Cadwallader, K.; Symons, M.; Hancock, J. F.
Science 1994, 264, 1463-1467.
(22)

Patterson, G. H.; Lippincott-Schwartz, J. Science 2002, 297, 1873-1877.

(23) Giepmans, B. N. G.; Adams, S. R.; Ellisman, M. H.; Tsien, R. Y. Science 2006,
312, 217-224.
(24)

Makrides, S. C. Microbiol. Rev. 1996, 60, 512-&.

(25) Karpeisky, M. Y.; Senchenko, V. N.; Dianova, M. V.; Kanevsky, V. Y. Febs
Letters 1994, 339, 209-212.
(26)

Kim, J. S.; Raines, R. T. European Journal of Biochemistry 1994, 224, 109-114.

(27)

Kim, J. S.; Raines, R. T. Analytical Biochemistry 1994, 219, 165-166.

(28)

Stofkohahn, R. E.; Carr, D. W.; Scott, J. D. Febs Letters 1992, 302, 274-278.

(29) Blumenthal, D. K.; Takio, K.; Edelman, A. M.; Charbonneau, H.; Walsh, K.; Titani,
K.; Krebs, E. G. Biophysical Journal 1985, 47, A76-A76.
(30) Enright, A. J.; Iliopoulos, I.; Kyrpides, N. C.; Ouzounis, C. A. Nature 1999, 402,
86-90.
(31) Yangyen, H. F.; Chambard, J. C.; Sun, Y. L.; Smeal, T.; Schmidt, T. J.; Drouin, J.;
Karin, M. Cell 1990, 62, 1205-1215.
(32) Chien, C. T.; Bartel, P. L.; Sternglanz, R.; Fields, S. Proceedings of the National
Academy of Sciences of the United States of America 1991, 88, 9578-9582.
(33) Karimova, G.; Pidoux, J.; Ullmann, A.; Ladant, D. Proceedings of the National
Academy of Sciences of the United States of America 1998, 95, 5752-5756.
(34) Karimova, G.; Dautin, N.; Ladant, D. Journal of Bacteriology 2005, 187, 22332243.

342

(35)

Kruse, T.; Bork-Jensen, J.; Gerdes, K. Molecular Microbiology 2005, 55, 78-89.

(36) Joung, J. K.; Ramm, E. I.; Pabo, C. O. Proceedings of the National Academy of
Sciences of the United States of America 2000, 97, 7382-7387.
(37) Ito, T.; Chiba, T.; Ozawa, R.; Yoshida, M.; Hattori, M.; Sakaki, Y. Proceedings of
the National Academy of Sciences of the United States of America 2001, 98, 45694574.
(38) Mahajan, N. P.; Linder, K.; Berry, G.; Gordon, G. W.; Heim, R.; Herman, B.
Nature Biotechnology 1998, 16, 547-552.
(39)

Xia, Z. P.; Liu, Y. H. Biophysical Journal 2001, 81, 2395-2402.

(40) Zhang, Y. B.; Howitt, J.; McCorkle, S.; Lawrence, P.; Springer, K.; Freimuth, P.
Protein Expression and Purification 2004, 36, 207-216.
(41) Davis, G. D.; Elisee, C.; Newham, D. M.; Harrison, R. G. Biotechnology and
Bioengineering 1999, 65, 382-388.
(42)

Kapust, R. B.; Waugh, D. S. Protein Science 1999, 8, 1668-1674.

(43)

Baneyx, F. Current Opinion in Biotechnology 1999, 10, 411-421.

(44) Jaakola, V. P.; Griffith, M. T.; Hanson, M. A.; Cherezov, V.; Chien, E. Y. T.; Lane,
J. R.; Ijzerman, A. P.; Stevens, R. C. Science 2008, 322, 1211-1217.
(45) Champe, P. C., Harvey, R. A., Ferrier, D. R. Biochemistry; 4
Williams & Williams: Philadelphia, 2008.

th

ed.; Lippincott

(46) Nelson, D. L., Cox M. M. Lehninger Principles of Biochemistry; 5
Freeman: New York, 2008.
(47)

ed.; W. H.

Pandey, A.; Mann, M. Nature 2000, 405, 837-846.

(48)

th

Janssens, V.; Goris, J. Biochemical Journal 2001, 353, 417-439.

(49) Wodarz, A.; Nusse, R. Annual Review of Cell and Developmental Biology 1998,
14, 59-88.
(50)

Buck, C. A.; Horwitz, A. F. Annual Review of Cell Biology 1987, 3, 179-205.

(51)

Bockaert, J.; Pin, J. P. Embo Journal 1999, 18, 1723-1729.

(52)

Strahl, B. D.; Allis, C. D. Nature 2000, 403, 41-45.

(53)

Helenius, A.; Aebi, M. Science 2001, 291, 2364-2369.
343

(54) Valko, M.; Leibfritz, D.; Moncol, J.; Cronin, M. T. D.; Mazur, M.; Telser, J.
International Journal of Biochemistry & Cell Biology 2007, 39, 44-84.
(55)

Karin, M. Nature 2006, 441, 431-436.

(56)

Withey, J. H.; Friedman, D. I. Current Opinion in Microbiology 2002, 5, 154-159.

(57) Forman, H. J.; Fukuto, J. M.; Torres, M. American Journal of Physiology-Cell
Physiology 2004, 287, C246-C256.
(58) Gabe, Y.; Urano, Y.; Kikuchi, K.; Kojima, H.; Nagano, T. Journal of the American
Chemical Society 2004, 126, 3357-3367.
(59) Sharma, V.; Piwnica-Worms, D. Journal of Magnetic Resonance Imaging 2002,
16, 336-351.
(60)

Blasberg, R. European Journal of Cancer 2002, 38, 2137-2146.

(61) Tanaka, K.; Miura, T.; Umezawa, N.; Urano, Y.; Kikuchi, K.; Higuchi, T.; Nagano,
T. Journal of the American Chemical Society 2001, 123, 2530-2536.
(62) Zhang, J.; Campbell, R. E.; Ting, A. Y.; Tsien, R. Y. Nature Reviews Molecular
Cell Biology 2002, 3, 906-918.
(63)

Massoud, T. F.; Gambhir, S. S. Genes & Development 2003, 17, 545-580.

(64)

Baker, N. R. Annual Review of Plant Biology 2008, 59, 89-113.

(65) Schwarzlander, M.; Fricker, M. D.; Muller, C.; Marty, L.; Brach, T.; Novak, J.;
Sweetlove, L. J.; Hell, R.; Meyer, A. J. Journal of Microscopy-Oxford 2008, 231, 299316.
(66) Rauch, S. L.; Whalen, P. J.; Shin, L. M.; McInerney, S. C.; Macklin, M. L.; Lasko,
N. B.; Orr, S. P.; Pitman, R. K. Biological Psychiatry 2000, 47, 769-776.
(67) Li, J. N.; Zhang, Z.; Rosenzweig, J.; Wang, Y. Y.; Chan, D. W. Clinical Chemistry
2002, 48, 1296-1304.
(68) Weissleder, R.; Tung, C. H.; Mahmood, U.; Bogdanov, A. Nature Biotechnology
1999, 17, 375-378.
(69) Saiki, R. K.; Scharf, S.; Faloona, F.; Mullis, K. B.; Horn, G. T.; Erlich, H. A.;
Arnheim, N. Science 1985, 230, 1350-1354.
(70) Edwards, B. S.; Oprea, T.; Prossnitz, E. R.; Sklar, L. A. Current Opinion in
Chemical Biology 2004, 8, 392-398.

344

(71) Subramaniam, R.; Desveaux, D.; Spickler, C.; Michnick, S. W.; Brisson, N.
Nature Biotechnology 2001, 19, 769-772.
(72) Bachovchin, D. A.; Brown, S. J.; Rosen, H.; Cravatt, B. F. Nature Biotechnology
2009, 27, 485-485.
(73) Cervantes, S.; Prudhomme, J.; Carter, D.; Gopi, K. G.; Li, Q.; Chang, Y.-T.; Le
Roch, K. G. BMC Cell Biology 2009, 10, Article No.: 45.
(74) Shimomura, O.; Johnson, F. H.; Saiga, Y. Journal of Cellular and Comparative
Physiology 1962, 59, 223-&.
(75) Johnson, F. H.; Gershman, L. C.; Waters, J. R.; Reynolds, G. T.; Saiga, Y.;
Shimomura, O. Journal of Cellular and Comparative Physiology 1962, 60, 85-&.
(76) Heim, R.; Prasher, D. C.; Tsien, R. Y. Proceedings of the National Academy of
Sciences of the United States of America 1994, 91, 12501-12504.
(77) Morise, H.; Shimomur.O; Johnson, F. H.; Winant, J. Biochemistry 1974, 13,
2656-2662.
(78) Ward, W. W.; Prentice, H. J.; Roth, A. F.; Cody, C. W.; Reeves, S. C.
Photochem. Photobiol. 1982, 35, 803-808.
(79) Cubitt, A. B.; Heim, R.; Adams, S. R.; Boyd, A. E.; Gross, L. A.; Tsien, R. Y.
Trends Biochem.Sci. 1995, 20, 448-455.
(80)

Shimomura, O. Febs Letters 1979, 104, 220-222.

(81) Ormo, M.; Cubitt, A. B.; Kallio, K.; Gross, L. A.; Tsien, R. Y.; Remington, S. J.
Science 1996, 273, 1392-1395.
(82) Cody, C. W.; Prasher, D. C.; Westler, W. M.; Prendergast, F. G.; Ward, W. W.
Biochemistry 1993, 32, 1212-1218.
(83) Prasher, D. C.; Eckenrode, V. K.; Ward, W. W.; Prendergast, F. G.; Cormier, M.
J. Gene 1992, 111, 229-233.
(84)

Reid, B. G.; Flynn, G. C. Biochemistry 1997, 36, 6786-6791.

(85) Cormack, B. P.; Bertram, G.; Egerton, M.; Gow, N. A. R.; Falkow, S.; Brown, A. J.
P. Microbiology-(UK) 1997, 143, 303-311.
(86)

Heim, R.; Cubitt, A. B.; Tsien, R. Y. Nature 1995, 373, 663-664.

(87) Elsliger, M. A.; Wachter, R. M.; Hanson, G. T.; Kallio, K.; Remington, S. J.
Biochemistry 1999, 38, 5296-5301.

345

(88)

Thastrup, O., Tullin, S., Kongsbak Poulsen, L., Bjørn S.; Patent., U., Ed. 1995.

(89) Chalfie, M.; Tu, Y.; Euskirchen, G.; Ward, W. W.; Prasher, D. C. Science 1994,
263, 802-805.
(90)

Pollok, B. A.; Heim, R. Trends in Cell Biology 1999, 9, 57-60.

(91)

Doyle, T. C.; Burns, S. M.; Contag, C. H. Cellular Microbiology 2004, 6, 303-317.

(92)

Davis, T. N. Current Opinion in Chemical Biology 2004, 8, 49-53.

(93) Shcherbo, D.; Merzlyak, E. M.; Chepurnykh, T. V.; Fradkov, A. F.; Ermakova, G.
V.; Solovieva, E. A.; Lukyanov, K. A.; Bogdanova, E. A.; Zaraisky, A. G.; Lukyanov, S.;
Chudakov, D. M. Nature Methods 2007, 4, 741-746.
(94) Chudakov, D. M.; Lukyanov, S.; Lukyanov, K. A. Trends in Biotechnology 2005,
23, 605-613.
(95)

Tsien, R. Y. Annual Review of Biochemistry 1998, 67, 509-544.

(96) Harpur, A. G.; Wouters, F. S.; Bastiaens, P. I. H. Nature Biotechnology 2001, 19,
167-169.
(97) Kneen, M.; Farinas, J.; Li, Y. X.; Verkman, A. S. Biophysical Journal 1998, 74,
1591-1599.
(98)

Forster, T. Annalen Der Physik 1948, 2, 55-75.

(99)

Wu, P. G.; Brand, L. Analytical Biochemistry 1994, 218, 1-13.

(100) Porumb, T.; Crivici, A.; Blackshear, P. J.; Ikura, M. European Biophysics Journal
with Biophysics Letters 1997, 25, 239-247.
(101) Ikura, M.; Clore, G. M.; Gronenborn, A. M.; Zhu, G.; Klee, C. B.; Bax, A. Science
1992, 256, 632-638.
(102) Ting, A. Y.; Kain, K. H.; Klemke, R. L.; Tsien, R. Y. Proceedings of the National
Academy of Sciences of the United States of America 2001, 98, 15003-15008.
(103) Zhang, J.; Ma, Y. L.; Taylor, S. S.; Tsien, R. Y. Proceedings of the National
Academy of Sciences of the United States of America 2001, 98, 14997-15002.
(104) Janetopoulos, C.; Jin, T.; Devreotes, P. Science 2001, 291, 2408-2411.
(105) ; Dickinson, B., Ed. 2007.
(106) Cormack, B. P.; Valdivia, R. H.; Falkow, S. Gene 1996, 173, 33-38.

346

(107) Galbraith, D. W.; Anderson, M. T.; Herzenberg, L. A. Methods in Cell Biology, Vol
58 1999, 58, 315-341.
(108) Yang, G. Y.; Withers, S. G. Chembiochem 2009, 10, 2704-2715.
(109) Crameri, A.; Whitehorn, E. A.; Tate, E.; Stemmer, W. P. C. Nature Biotechnology
1996, 14, 315-319.
(110) Herzenberg, L. A.; De Rosa, S. C. Immunology Today 2000, 21, 383-390.
(111) Hung, C. I.; Ke, B. J.; Huang, G. R.; Hwei, B. H.; Lai, H. F.; Lee, G. B. Journal of
Fluids Engineering-Transactions of the Asme 2001, 123, 672-679.
(112) You, X.; Nguyen, A. W.; Jabaiah, A.; Sheff, M. A.; Thorn, K. S.; Daugherty, P. S.
Proceedings of the National Academy of Sciences of the United States of America
2006, 103, 18458-18463.
(113) Inouye, S.; Tsuji, F. I. Febs Letters 1994, 351, 211-214.
(114) Vasileiou, C.; Vaezeslami, S.; Crist, R. M.; Rabago-Smith, M.; Geiger, J. H.;
Borhan, B. Journal of the American Chemical Society 2007, 129, 6140-6148.
(115) Hota, P. K.; Singh, A. K. Letters in Organic Chemistry 2007, 4, 300-305.
(116) Singh, A. K.; Hota, P. K. Photochemistry and Photobiology 2007, 83, 50-62.
(117) Hoischen, D.; Steinmuller, S.; Gartner, W.; Buss, V.; Martin, H. D. Angew.
Chem.-Int. Edit. Engl. 1997, 36, 1630-1633.
(118) Cossarizza, A.; Baccaranicontri, M.; Kalashnikova, G.; Franceschi,
Biochemical and Biophysical Research Communications 1993, 197, 40-45.

C.

(119) Wang, S. P.; Deng, W. J.; Sun, D.; Yan, M.; Zheng, H.; Xu, J. G. Organic &
Biomolecular Chemistry 2009, 7, 4017-4020.
(120) Dragsten, P. R.; Webb, W. W. Biochemistry 1978, 17, 5228-5240.
(121) Achilefu, S.; Dorshow, R. B.; Bugaj, J. E.; Rajagopalan, R. Investigative
Radiology 2000, 35, 479-485.
(122) Sieber, F.; Spivak, J. L.; Sutcliffe, A. M. Proceedings of the National Academy of
Sciences of the United States of America-Biological Sciences 1984, 81, 7584-7587.
(123) Rye, H. S.; Yue, S.; Wemmer, D. E.; Quesada, M. A.; Haugland, R. P.; Mathies,
R. A.; Glazer, A. N. Nucleic Acids Research 1992, 20, 2803-2812.

347

(124) Toutchkine, A.; Kraynov, V.; Hahn, K. Journal of the American Chemical Society
2003, 125, 4132-4145.
(125) Jahannes, H. H., Braunschweig University of Technology, 2000.
(126) Toutchkine, A.; Nguyen, D. V.; Hahn, K. M. Bioconjugate Chemistry 2007, 18,
1344-1348.
(127) Honig, B.; Dinur, U.; Nakanishi, K.; Baloghnair, V.; Gawinowicz, M. A.; Arnaboldi,
M.; Motto, M. G. Journal of the American Chemical Society 1979, 101, 7084-7086.
(128) Ismail, L. F. M.; Antonious, M. S.; Mohamed, H. A.; Ahmed, H. A. H. Proceedings
of the Indian Academy of Sciences-Chemical Sciences 1992, 104, 331-338.
(129) Oster, G.; Nishijima, Y. Journal of the American Chemical Society 1956, 78,
1581-1584.
(130) Chang, C. C.; Chu, J. F.; Kuo, H. H.; Kang, C. C.; Lin, S. H.; Chang, T. C. Journal
of Luminescence 2006, 119, 84-90.
(131) Law, K. Y. Chemical Physics Letters 1980, 75, 545-549.
(132) Derguini, F.; Caldwell, C. G.; Motto, M. G.; Baloghnair, V.; Nakanishi, K. Journal
of the American Chemical Society 1983, 105, 646-648.
(133) Creemers, A. F. L.; Lugtenburg, J. Journal of the American Chemical Society
2002, 124, 6324-6334.
(134) Buncel, E.; Rajagopal, S. Accounts of Chemical Research 1990, 23, 226-231.
(135) Wurthner, F.; Archetti, G.; Schmidt, R.; Kuball, H. G. Angewandte ChemieInternational Edition 2008, 47, 4529-4532.
(136) Yu, A. C.; Tolbert, C. A.; Farrow, D. A.; Jonas, D. M. Journal of Physical
Chemistry A 2002, 106, 9407-9419.
(137) Ishchenko, A. A.; Derevyanko, N. A.; Svidro, V. A. Dyes and Pigments 1992, 19,
169-177.
(138) Arai, S.; Yamazaki, M.; Nagakura, K.; Ishikawa, M.; Hida, M. Journal of the
Chemical Society-Chemical Communications 1983, 1037-1038.
(139) Jia, X. F., Michigan State University, 2008.
(140) Liu, R. S. H.; Muthyala, R. S.; Wang, X. S.; Asato, A. E.; Wang, P.; Ye, C. Org.
Lett. 2000, 2, 269-271.

348

(141) Williams, A. T. R., Winfield, S. A., Miller, J. N. Analyst 1983, 108.

349