PLACE IN RETURN BOX to remove this checkout from your record. TO AVOID FINES return on or before date due. MAY BE RECALLED with earlier due date if requested. DATE DUE DATE DUE DATE DUE 5/08 KzlProleocsPresIClRC/Dateoue.indd SYNTHESIS, CHARACTERIZATION, AND APPLICATIONS OF FUNCTIONALIZED POLYGLYCOLIDES By Erin B. Vogel A DISSERTATION Submitted to Michigan State University in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY Chemistry 2008 ABSTRACT SYNTHESIS, CHARACTERIZATION, AND APPLICATIONS OF FUNCTIONALIZED POLYGLYCOLIDES By Erin B. Vogel Poly(L-lactide) has been widely used as a degradable material in environmental and biomedical fields; however, for many applications polylactide is not ideal due to its high crystallinity, hydrophobicity, and lack of functionalization. Strategies that provide routes to functionalized polyesters are necessary to produce materials with desired physical and chemical properties. We exploited two methods for polyester functionalization: the functional monomer approach and the post-polymerization modification approach. Copolymers of Iactide and functionalized glycolides can provide a range of polymer properties that can be tailored based on copolymer composition. However, if the two comonomers have unequal reactivity towards polymerization, “blocky” polymers can form with unpredictable properties. We copolymerized rec-Iactide with five disubstituted Iactide monomers and determined their relative reactivity ratios, which can be used to predict copolymer architecture. Our results indicated that the size of the substituent on the Iactide framework directly influences monomer reactivity ratios and homopolymerization rates. Adding functional groups, such as alcohols, increases the versatility of conventional polylactide, and permits further elaboration to enhance polymer hydrophilicity, provide attachment sites for biologically relevant molecules, and serve as branching sites in the synthesis of complex polymer architectures. We synthesized 3-benzyloxyoctyl glycolide, an AB glycolide monomer derived from oleic acid that places a benzyl-protected primary alcohol eight methylene units from the polymer backbone. Copolymerization of 3-benzyloxyoctyl glycolide with L-lactide yielded high molecular weight copolymers. Quantitative hydrogenolysis of the benzyl groups provided functional polylactides with pendent primary alcohols distributed along the polymer backbone. Initiation of L-lactide polymerization from a copolymer with 5 mol% alcohol sites generated amorphous PLA comb polymers that crystallized upon capping the hydroxy groups at the termini of the teeth. ‘- Post-polymerization modification via “click” chemistry provides a convenient route to functionalized polyesters. Using poly(propargylglycolide), an acetylene-functionalized polyglycolide, we prepared and characterized a family of degradable comb polymers that individually self-aggregate in aqueous solutions. The polymers encapsulate and release hydrophobic and hydrophilic molecules and display thermoresponsive behavior from room temperature to > 60 °C. Furthermore, we described the synthesis, characterization, and applications of chemically cross-linked comb polymers that retain chemical functionality. The organic nanoparticles exhibit prolonged controlled release and are biocompatible. Be exalted, O God, above the heavens; Let your glory be over all the earth. ACKNOWLEDGEMENTS I would like to thank my advisor, Dr. Gregory Baker, who not only taught me the basics of research and polymer chemistry, but also how to “be a scientist.” Thanks to Drs. Christina Chan and S. Patrick Walton for meaningful conversations and an intriguing collaboration. Thanks to my guidance committee, Dr. Babak Borhan, Dr. Kevin Walker, Dr. Merlin Bruening, and Dr. Mitch Smith, for helpful discussions and suggestions. Thanks to Dan Holmes, Kermit Johnson, Riu Huang, Kathy Severin, Dan Jones, Alicia Pastor, and Ewa Danielewicz for teaching me various characterization techniques. I would also like to thank past and current graduate students and group members: quei, Feng, Leslie, DJ, Bao, Ying, Ping, Jon, Qin, Sampa, Tom, Tiffany, Joe, Amanda, Aman and Nicki for their assistance and friendship. They have taught me many things and made life at MSU enjoyable. Finally I would like to thank my family for their endless encouragement and prayers and my husband, Mark, for his patience, support, and love. This would have not been possible without you. TABLE OF CONTENTS LIST OF TABLES ................................................................................................. ix LIST OF FIGURES ................................................................................................ x LIST OF SCHEMES ........................................................................................... xix LIST OF ABBREVIATIONS ............................................................................... xxii Chapter ‘1 Introduction ........................................................................................ 1 Structure and Properties of Functionalized Polyglycolides ................................ 1 Functional Monomer Approach .......................................................................... 2 Synthesis and Polymerization of Functionalized Glycolide Monomers ........... 2 Alkyl-Substituted Polyglycolides ..................................................................... 4 Aryl-Substituted Polyglycolides ...................................................................... 9 Hydroxy-Functionalized Glycolides .............................................................. 11 Carboxylic Acid-Functionalized Polyglycolides ............................................. 16 Amine-Substituted Glycolides ...................................................................... 19 Peg-Grafted Glycolides ................................................................................ 20 Alkenyl- and AlkynyI-Substituted Polyglycolides .......................................... 22 Post-Polymerization Modification ..................................................................... 26 Direct Grafting of Functional Groups ............................................................ 26 Lactide Initiation from Pendant Alcohol Groups ........................................... 28 000 Coupling Reactions ............................................................................. 30 Ketoximine Coupling Reactions ................................................................... 32 Michael Addition Reactions .......................................................................... 34 Atom Transfer Radical Addition Reactions ................................................... 35 Olefin Cross Metathesis ............................................................................... 36 “Click” Chemistry .......................................................................................... 38 Conclusion ................................................................................................... 41 vi Chapter 2 Reactivity Ratios of Substituted Lactide Monomers ......................... 42 Introduction ...................................................................................................... 42 Results and Discussion ................................................................................... 46 Monomer Syntheses .................................................................................... 46 Capolymerization and Copolymer Isolation .................................................. 47 Monomer Reactivity Ratios and Copolymer Architectures ........................... 48 Reactivity Ratio Determination ..................................................................... 50 Conclusion ....................................................................................................... 76 Experimental Section ....................................................................................... 77 Chapter 3 Branched Poly(lactide)s ................................................................... 88 Introduction ...................................................................................................... 88 Results and Discussion ................................................................................... 89 Monomer Synthesis ..................................................................................... 89 Bulk Polymerization ...................................................................................... 92 Solution Polymerization ................................................................................ 94 Copolymerization ......................................................................................... 99 Physical Properties of poly(BOG-co-LLA) .................................................. 101 Be nzyl Deprotection ................................................................................... 101 Degradation of poly(HOG-co-LLA) ............................................................. 103 Graft Polymerizations ................................................................................. 105 Side Chain Crystallinity of poly((HOG-co-LLA)-g-LLA) ............................... 110 Conclusion ................................................................................................. 1 13 ExDerimental Section ..................................................................................... 1 14 Chapter 4 Post-Polymerization Modification of Polyglycolides via “Click” Chemistry 126 Introduction .................................................................................................... 1 26 Results and Discussion ................................................................................. 130 Post-Polymerization Modification ............................................................... 130 FOrmation and Characterization of Unimolecular Micelles ......................... 135 HYCI rophobic Guest Encapsulation ............................................................. 141 vii Prodrug Conjugation .................................................................................. 145 Lower Critical Solution Temperature Materials ........................................... 149 Degradation of Poly(propargylgcholide)-graph-poly(ethylene glycol) ........ 154 Conclusion ................................................................................................. 156 Experimental Section ..................................................................................... 157 Chapter 5 Biodegradable Cross-Linked Nanoparticles ...................................... 174 Introduction .................................................................................................... 1 74 Results and Discussion ................................................................................. 178 Synthesis of Cross-Linked Nanoparticles ................................................... 178 Characterization of Cross-Linked Nanoparticles ........................................ 181 Estimation of Cross-linked Nanoparticle Size ............................................. 183 Conjugation to Cross-linked Nanoparticles ................................................ 184 Controlled Release from Cross-linked Nanoparticles ................................. 187 Biocompatibility .......................................................................................... 1 88 Conclusion ................................................................................................. 190 EXperimental Section ..................................................................................... 191 Appendix ........................................................................................................... 198 References ........................................................................................................ 257 viii LIST OF TABLES Table 1. Mole fractions of Iactide in the feed (f1) and copolymer (F1) for the copolymerization of Iactide and ethylglycolide .............................................. 51 Table 2. Reactivity ratios calculated for rec-Iactide (r1) and disubstituted glycolides (r2). .............................................................................................. 56 Table 3. T93 of poly(Iactide-co-ethylglycolide) copolymers. ................................ 57 Table 4. Mole fraction of Iactide in the feed (f1) and copolymer (F1) for the copolymerization of Iactide and n-hexylglycolide .......................................... 59 Table 5. Mole fractions of Iactide in the feed (f1) and copolymer (F1) for the copolymerization of Iactide and isopropylglycolide. ...................................... 62 Table 6. T95 of poly(Iactide-co-isopropylglycolide) copolymers. ......................... 65 Table 7. Mole fraction of Iactide in the feed (f1) and copolymer (F1) for the copolymerization of Iactide and cyclohexylglycolide ..................................... 66 Table 8. Solution polymerization rates for the homopolymerization of substituted glycolides.a ................................................................................................... 70 ‘I'able 9. Mole fractions of Iactide in the feed (f1) and copolymer (F1) for the co polymerization of Iactide and phenyllactide. ............................................. 71 Table 1 0. T93 of poly(Iactide-co-mandelide) copolymers. .................................... 75 Table 1 1. GPC and DSC characterization of BOG homo— and co-polymers before and after benzyl ether deprotection.‘al ........................................................... 94 Table 1 2. Macroinitiation of L-Iactide from poly(HOG-co-LLA) containing 5 mol °/o HOG.a ........................................................................................................ 106 Table 1 3. Properties and size of PPGL-g-mPEG unimolecular micelles. ......... 137 Table 1 4. Characterization of cross-linked nanoparticles by DLS and AFM ..... 182 LIST OF FIGURES Figure 1 - Linear alkyI-substituted polyglycolides and their glass transition temperatures. ................................................................................................. 5 Figure 2. Branched alkyl-substituted polyglycolides and their glass transition temperatures. ................................................................................................. 7 Figure 3. AB-substituted polyglycolides and their glass transition temperatures...8 Figure 4- Aryl-substituted polyglycolides and their glass transition temperatures. ..................................................................................................................... 10 Figure 5 . 75 MHz 13C NMR carbonyl regions of propargylglycolide homopolymer (PPGL), copolymers. and polylactide (run in CDCI3). TUnique Iactide carbonyl resonances in poly(lactide-co-propargylglycolide) (PLA-co-PPGL).33 .......... 25 Figure 6. 500 MHz 1H NMR of poly(LA-co-nHG) at 8 % conversion. The molar fraction of Iactide in the copolymer (F1) was determined using the relative integrations of the two comonomers (see text, f1 = 0.81, F1 = 0.84) ............. 50 Figure 7. Nonlinear least squares fit of the copolymerization data of rac-Iactide and rec-ethylglycolide to equation 7. The reactivity ratios were determined to be rLA=1.29 and rEG = 0.71 ........................................................................... 52 Figure 8. Fineman-Ross plot of the rec-Iactide and racethylglycolide co polymerization data. A least squares linear fit to the data gave y = 1.4x + 0.82 (ILA = 1.4 and res. = 0.82). ..................................................................... 53 Fi9ure 9. Kelen-Tudos plot for the copolymerization of rec-Iactide and rac- ethylglycolide. Extrapolation to E = 0 and E = 1 gave reactivity ratio values of rLA = 1.3 and res = 0.72. ............................................................................... 55 Fi9ure 1 0. Glass transition temperatures of ethylglycolide (EG) and Iactide (LA) Copolymers as a function of polymer composition. DSC samples were run UDder a nitrogen atmosphere at a heating rate of 10 °C/min ........................ 58 Fi9ure 1 1. Nonlinear least squares fit of the copolymerization data of racolactide and rac-n-hexylglycolide to equation 7. The reactivity ratios were determined to be rLA=1.30 and rnHG = 0.15. .................................................................... 59 Figure 1 2. Fineman-Ross plot of the rac-lactide and rac-n-hexylglycolide copolymerization data. A least squares linear fit to the data gave y = 1.4x + 0.24 (r._A = 1.4 and rnHG = 0.24). ................................................................... 60 Figure 1 3. Kelen-Tudt’is plot for the copolymerization of rec-Iactide and rac-n- hexylglycolide. Extrapolation to E = 0 and E = 1 gave reactivity ratio values of ILA = 1.3 and anG = 0.16. ............................................................................. 61 Figure 1 4. Nonlinear least squares fit to the copolymerization data of rac-lactide and rac-isopropylglycolide to equation 7. The reactivity ratios were determined to be rm: 2.5 and riPG = 0.080. ................................................. 62 Figure 15. Fineman-Ross plot of the rec-Iactide and rac-isopropylglycolide copolymerization data. A least squares linear fit to the data gave y = 2.7x + 0.081 (rLA = 2.7 and ripe = 0.081 ). ................................................................ 63 Figure 1 6. Kelen-Tudés plot for the copolymerization of rec-Iactide and rac- isopropylglycolide. Extrapolation to g = 0 and E = 1 gave reactivity ratio values Of ILA = 2.8 and ripe = 0.10. ............................................................... 64 Figure 1 7. Nonlinear least squares fit of the copolymerization data for rec-Iactide and raccyclohexylglycolide to equation 7. The reactivity ratios were determined to be of rLA= 3.3 and rcHG = 0.015 .............................................. 66 Figure 1 8. Fineman-Ross plot of the rec—Iactide and rec-cyclohexylglycolide COpolymerization data. A least squares linear fit to the data gave y = 3.3x + 0.015 (fLA= 3.1 and chG = 0.015). ................................................................ 67 Figure 19. Kelen-Tudbs plot for the copolymerization of rec-Iactide and rac- Cyclohexylglycolide. Extrapolation to E = 0 and E = 1 gave reactivity ratio values of rLA= 3.0 and ram; = 0.011. ............................................................. 68 l:i9Ure 20. Nonlinear least squares fit of the copolymerization data of rec-Iactide and rac-phenyllactide to equation 7. The reactivity ratios were determined to be ILA: 1.05 and I'm = 1.22. ......................................................................... 71 Figure 21. Fineman-Ross plot of the rac-Iactide and rac-phenyllactide Copolymerization data. A least squares linear fit to the data gave y = 1.0x + 0.12(rLA= 1.0 and rpL= 1.2) ......................................................................... 72 Fi9ure 22. Kelen-Tudos plot for the c0p0lymerizati0n of rec-Iactide and rac- Dhenyllactide. Extrapolation to E = 0 and E = 1 gave reactivity ratio values rLA = 1 .0 and rpL = 1.2. ....................................................................................... 73 Fi9ure 23. Glass transition temperatures of mandelide (MD) and Iactide (LA) COpolymers as a function of polymer composition. DSC samples were run Under a nitrogen atmosphere at a heating rate of 10 °C/min. ...................... 74 FiSIure 24. Instantaneous copolymer composition (F1) as a function of monomer feed ratio (11) for the copolymerization of rac-ethylglycolide and rac-lactide xi (res = 0.72, rLA = 1.29, M2 = 0.92) (O) and copolymerization of rac- and mesa-ethylglycolide with rac-lactide (res = 0.65, rLA = 1.48, Me = 0.96) (0). ..................................................................................................................... 76 Figure 25. 500 MHz 1H NMR of BOG (A) and poly(BOG-co-LLA) (B). ............... 92 Figure 26. Molecular weight versus conversion for the polymerization of BOG. Solution polymerization conditions: 90 °C, [SnOct2]1[BBA]:[m0nomer]= 1:1 :1 00. Mn determined via GPO using poly(styrene) standards. Each point corresponds to an average of three samples and the error bars represent the standard deviation. ....................................................................................... 96 Figure 27. Solution polymerization of BOG with SnOct2 at 90 °C. All reactions were carried out in a 0.2 M BOG solution in toluene with a [SnOctz]:[BBA]: [monomer] ratio of 1:1:100. Each point corresponds to an average of three samples and the error bars represent the standard deviation. ..................... 97 Figure 28. Solution polymerization kinetics of BOG with SnOct2 at 90 °C. All reactions were carried out at 0.2 M BOG solution in toluene with a [SnOctg]: [BBA]:[monomer] ratio of 1:1:100. The equilibrium monomer concentration, [M]eq, is assumed to be 0.020 M. The line is a least squares fit to the data. ..................................................................................................................... 99 Figure 29. ‘H NMR spectrum for poly(BOG-co-LLA) (5 mol% BOG) (top) and poly(HOG-co-LLA) (bottom) taken in CDCIs at 500 MHz. Absence of peaks at 7.2-7.4 ppm in the spectra indicates complete removal of the benzyl ether Protecting groups. ...................................................................................... 103 Fi9ure 30. Normalized molecular weight decrease for poly(HOG-co-LLA) (50 mol % HOG) (A), poly(rac-LA) (El) and PLLA (0) during hydrolytic degradation at 55 °C in pH = 7.4 phosphate buffer saline solution. Each point corresponds to the average of 3 samples and the line is a least squares fit to the poly(HOG-co-LLA) data ........................................................................ 105 Fi9ure 31. 500 MHz 1H NMR spectrum of poly(HOG-co-LLA) (5 mol% HOG) (top) and poly((HOG-co-LLA)-g-LLA) (bottom) taken in CDCla. .......................... 107 Figure 32. Double logarithmic plot of rms radius of gyration versus molecular mass of linear PLA (broken red line) and poly((HOG-co-LLA)-g-LLA) (solid blue line) in THF at 35 °C. The smaller slope for poly((HOG-co-LLA)-g-LLA), a = 0.24, compared to linear PLA, 3 = 0.81, is consistent with the branched arChitecture of poly((HOG-co-LLA)-g-LLA). The HOG content in poly((HOG- CO-LLA)-g-LLA) sample was 5 mol °/o, and the degree of polymerization of the LLA branches was 60 ........................................................................... 108 xii Figure 33. Mark-Houwink plots for linear (broken red line) and branched (solid blue line) PLA in THF at 35 °C using light scattering and dilute-solution viscometry data, extracted from GPC—MALLS. The smaller sl0pe for poly((HOG-co-LLA)-g-LLA) is consistent with the branched architecture of p0ly((HOG-co-LLA)-g-LLA). The HOG content in poly((HOG-co-LLA)-g-LLA) was 5 mol%, and the degree of polymerization of the LLA branches was 60. Linear PLA, a = 0.74; branched PLA, a = 0.34. .......................................... 110 Figure 34. Acylation of poly((HOG-co-LLA)-g-LLA) using DCC coupling .......... 111 Figure 35. DSC second heating scans of poly((HOG-co-LLA)-g-LLA) at a heating rate of 10 °C/min. The degree of polymerization of the PLLA branches is 40; (OH terminated (.2 bottom), acetate terminated (- middle), and stearic acid terminated (— top)). .................................................................................. 112 Figure 36. Relationship between AH,us and the branch length for poly((HOG-co- LLA)-g-LLA). The terminal —CH20H groups of the branches were acetylated. ................................................................................................................... 113 Figure 37. The comb architecture of poly(propargylglycolide)-graft—poly(ethylene glycol monomethyl ether) and formation of a unimolecular micelle in aqueous solution. The hydrophobic backbone and hydrophilic side chains are illustrated in red and blue, respectively. ..................................................... 129 Figure 38. 500 MHz 1H NMR spectra of poly(propargylglycolide) and its click fu nctionalization with mPEG-azide. Integration of a = c = a’ = c’ = 1 ......... 132 Figure 39. IR spectra of unfunctionalized PPGL (bottom) and PPGL-g-mPEG (tOp). Notably, the alkyne stretch at 3200 crn'1 completely disappears upon click functionalization and a new stretch at 3300-2500 cm'1 from the NH triazole stretch appears. ............................................................................. 133 l:igure 40. GPC traces of PPGL-g-mPEG polymers before and after exposure to ion exchange resin beads. The Cu(|l) concentration was reduced to < 10 pr. Original polymer (Mn = 56,200; PDI = 1.07; solid blue line); dialysis solution without ion exchange resin (Mn = 56,600; PDI = 1.11; dashed green line); polymer after 24 hours exposure to ion exchange resin in DMF (Mn = 55,900; PDI = 1.12; dotted red line). .......................................................... 134 Fi9ure 41. Hydrodynamic radius of PPGL-g-mPEG micelles as determined by dynamic light scattering (Fin = 31 nm; 2 mg/mL solution in Milli-Q water)...138 Fl9ure 42. Organization of traditional micelles and unimolecular micelles from block copolymers and PPGL-g-mPEG, respectively. The red and blue Spheres indicate hydrophobic and hydrophilic polymer segments, respectively. ............................................................................................... 1 39 xiii Figure 43. AFM height image of PPGL-g—mPEG on a silicon wafer substrate; height = 3.0 :I: 0.1 nm. ................................................................................ 140 Figure 44. TEM image of PPGL-g-mPEG stained with uranyl acetate; diameter = 10 :9: 2 nm. Image taken at 200,000x; scale bar represents 100 nm .......... 141 Figure 45. UV-vis spectra of azobenzene loaded polymeric micelles (top, blue), polymeric micelles (middle, green), and azobenzene (bottom, red) in Milli-Q water at room temperature (polymer concentration = 1 mg/mL) ................. 142 Figure 46. Continual release of azobenzene from loaded polymeric micelles suspended in Milli—Q water at room temperature over 10 hours. At each time point, the azobenzene remaining in the micelles was determined (at 311 nm) by UV-vis spectroscopy .............................................................................. 143 Figure 47. UV-Vis spectra of azobenzene-PMMA (Mn = 10 kD) loaded polymeric micelles (top, blue), polymeric micelles (middle, green), and azobenzene- PM NIA (bottom, red) in Milli-Q water at room temperature. ........................ 145 Figure 48. Release of L-homoserine Iactone from PPGL-g-mPEG polymer in Milli-Q water (pH = 6.0; A) and PBS solution (pH = 7.4; x). ...................... 149 Figure 49. Visualization of thermoresponsive poly(glycolide) (59 % mDEG; 41 % decyl) below (left) and above (right) its LCST (26 °C). ............................... 151 Fi9ure 50. Cloud point determination at 450 nm for PPGL “clicked” with a mixture Of hydrophobic alkyl and hydrophilic PEG chains. The composition of the “clicked" polymer, (59 % mDEG chains, 41 % decyl chains) was determined by ‘H NMR. ................................................................................................ 152 Figure 51. Relationship between the cloud point observed at 450 nm and the mole percentage of PEG chains “clicked" onto the PPGL homopolymer. The broken line is a least squares fit to the data. .............................................. 153 I:i9ure 52. TEM images of thermoresponsive poly(glycolide) (59 % mDEG; 41 °/o decyl) below (top) and above (bottom) its LCST (26 °C) ............................ 154 Figure 53. Molecular weight change of PPGL-g—mPEG during hydrolytic degradation in Milli-Q water at 36 °C. Each point represents an average of three samples. ........................................................................................... 155 Figure 54. Molecular weight change of PPGL-g-mPEG (El) and rac-PLA (A) during hydrolytic degradation. Degradation experiments were run at 36 °C for PPGL-g-mPEG and 50 °C for rac-PLA. The lines are least squares fits to a random chain scission model. The data points represent an average of 3 sathples ...................................................................................................... 156 xiv Figure 55. IR spectra of poly(PGL (bottom), poly(PGL[EG75X25] (middle), and cross-linked poly(PGLlEG75CL25] (t0p). Upon PEG functionalization and cross-linking, the alkyne stretching band (3340-3270 nm) was replaced by the triazole stretching band (N-H) (3300-2500 nm). In addition, no residual azide stetching bands were observed at 2100 nm. .................................... 180 Figure 56. 500 MHz 1H NMR of p0ly(PGL[EG75X25]) before (bottom) and after (top) cross-linking with 1,5-diazid0pentane (integration of signal czc’ = 1:0.75 (bottom) and 1:1 (top). ............................................................................... 181 Figure 57. Representative AFM image of cross-linked nanoparticles (polymer 2a, Table 14, 200 ug/mL solution spin-coated on a Si wafer substrate). ......... 182 Figu re 58. Representative TEM image of cross-linked nanoparticles (polymer 2a, Table 14, 100 pg/mL solution; 370,000x magnification; scale bar = 50 nm). ................................................................................................................... 183 Figu re 59. Reaction of 3-azido-7-hydr0xycoumarin with polymer micelles. Vial 1: polymeric micelles and dye in the absence of “click” catalyst; Vial 2: polymer micelles and dye after the “click” reaction; Vial 3: dye and components for “click” reaction without polymer micelles. ................................................... 185 Fig u re 60. Fluorescence intensity of 1-step and 2-step click reactions on p0ly(PGL[EG75X2s]) substrates. (From top to bottom): a single click step using an equimolar mixture of the dye precursor and 1,5-diazid0pentane (blue), a two step process where 5 % of the remaining alkynes were cross- Iinked, followed by a second click reaction using an excess of the dye precursor (red); the same protocol, with 10 °/o cross-linking (orange), and the same protocol, with 100 % cross-linking (green) ........................................ 186 F'Q u re 61. Release of azobenzene from nanoparticles suspended in Milli-Q water at room temperature (1 mg/mL). Cross-linked nanoparticles (I) show 2x slower release than noncross-Iinked nanoparticles (D). The loss of dye from the nanoparticles was monitored at 311 nm by UV spectroscopy ...... 188 R — ' Q u re 62. Cortical astrocytes cultured in media containing 10 ug/mL of cross- Iinked nanoparticles for 24 hours. The red color emanates from rhodamine dye covalently linked to the nanoparticles that have entered cells. ............ 189 I: ' Q L: re 63. Evaluation of nanoparticle toxicity on cell viability reported in relative fluorescence units (RFU). Noncross-linked nanoparticles (poly(PGL[E675X2s]). NonCL NP) and cross-linked nanoparticles (poly(PGL[EGyscL25]), CL NP) correspond to polymers mentioned in the text. LipoFectamine (LF2k), linear poly(ethylene imine) (Mn = 25,000) (LPEI), and poly(L-Iysine) (M" = 150,000) (PLL) were used as controls. LF2k is reported as dilutions of the purchased solution. ....................................................... 190 Appendix 1. 1H NMR spectra for the determination of F1 for the capolymerization of rec-Iactide and rec-ethylglycolide. .......................................................... 199 Appendix 2. 1H NMR spectra for the determination of F1 for the copolymerization of rec-Iactide and rec-isopropylglycolide. ................................................... 200 Appendix 3. 1H NMR spectra for the determination of F1 for the copolymerization of rec-Iactide and rac-n-hexylglycolide. ...................................................... 201 Appendix 4. 1H NMR spectra for the determination of F1 for the copolymerization of rec-Iactide and rec-cyclohexylglycolide. ................................................. 202 Appendix 5. 1H NMR spectra for the determination of F1 for the copolymerization of rac-lactide and rac-phenyllactide ............................................................ 203 Appendix 6. 1H NMR spectra of 9,9-dimethcxynonanoate. .............................. 204 Appendix 7. ”C NMR spectra of 9,9-dimeth0xyn0nan0ate. ............................. 205 Appendix 8. 1H NMR spectra of 9,9-dimethoxynonanol. .................................. 206 Appendix 9. 130 NMR spectra of 9,9-dimethoxynonanol. ................................. 207 Appendix 10. 1H NMR spectra of ((9,9-dimeth0xyn0nyloxy)methyl)benzene. ..208 AD pendix 11. 1"C NMR spectra of ((9,9-dimethoxynonyloxy)methyl)benzene..209 AD Dendix 12. 1H NMR spectra of (9-benzyl0xy)nonanal. ................................. 210 AD Dendix 13. 1H NMR spectra of 10-(benzyloxy)-2-hydroxydecanenitrile. ....... 211 AD Dendix 14. 1H NMR spectra of 10-(benzyloxy)-2-hydroxydecanamide. ........ 212 AD Dendix 15. 1H NMR spectra of 10-(benzyl0xy)-2-hydroxydecanoic acid. ..... 213 AD Dendix 16. 13C NMR spectra of 10-(benzyloxy)—2-hydroxydecanoic acid. ..... 214 ADIZ>endix 17. 1H NMR spectra 3-(8-(benzyloxy)octyI-1,4-di0xane-2,5-dione (BOG) ......................................................................................................... 215 A'DIZ-endix 18. 13C NMR spectra 3-(8-(benzyloxy)octyl-1,4-dioxane-2,5-dione (BOG) ......................................................................................................... 216 A IDFJendix 19. 1H NMR spectra of poly(BOG) .................................................... 217 Appendix 20. “’0 NMR spectra of poly(BOG). .................................................. 218 xvi Appendix 21. 1H NMR spectra of poly(BOG-co-LLA). ...................................... 219 Appendix 22. 13C NMR spectra of poly(BOG-co-LLA). ...................................... 220 Appendix 23. 1H NMR spectra of poly(HOG) ..................................................... 221 Appendix 24. 13C NMR spectra of poly(HOG). .................................................. 222 Appendix 25. 1H NMR spectra of poly(HOG-co-LLA). ...................................... 223 Appendix 26. 13C NMR spectra of poly(HOG-co-LLA) ....................................... 224 Appendix 27. 1H NMR of poly((HOG-co-LLA)-g—LLA) ....................................... 225 Appendix 28. 1H NMR of poly((HOG-coLLA)-g—LLA), stearic acid terminated..226 Appendix 29. 1H NMR spectra of propargylglycolide ........................................ 227 Appendix 30. 13C NMR spectra of propargylglycolide. ...................................... 228 Appendix 31. 1H NMR spectra of poly(propargylglycolide). .............................. 229 Appendix 32. 13C NMR spectra of poly(propargylglycolide) ............................... 230 Appendix 33. 1H NMR spectra of 16-azid0-2,5,8,11,14-pentaoxahexadecane (mPEG). ..................................................................................................... 231 AD pendix 34. 13C NMR spectra of 16-azido-2,5,8,11,14-pentaoxahexadecane (mPEG). ..................................................................................................... 232 AD pendix 35. 1H NMR spectra of PPGL-g-mPEG. ........................................... 233 AD pendix 36. ”C NMR spectra of PPGL-g—mPEG. ........................................... 234 AD pendix 37. 1H NMR spectra of 2—br0mopropionic acid-2-(4- phenylazophenyl)ethyl ester. ..................................................................... 235 AD pendix 38. ”C NMR spectra of 2-bromopropionic acid-2-(4- phenylazophenyl)ethyl ester. ..................................................................... 236 Appendix 39. 1H NMR spectra of 3-azidopr0panol. .......................................... 237 Appendix 40. 13C NMR spectra of 3-azidopropan0l ........................................... 238 A‘3F3endix 41. 1H NMR spectra of 3-azidopropyl-3-oxobutanoate. .................... 239 A Dpendix 42. ”C NMR spectra of 3-azidopr0pyl 3-oxobutanoate. .................... 240 xvii Appendix 43. 1H NMR spectra of 3-azidopropyl 3-(2-oxo-tetrahydr0furan)-3- ylimino)butanoate ....................................................................................... 241 Appendix 44. 1H NMR spectra of L-homoserine Iactone hydrobromide. .......... 242 Appendix 45. 13C NMR spectra of L-homoserine Iactone hydrobromide. .......... 243 Appendix 46. 1H NMR spectra of 3-azidopropyI-3-(3-0xoisoxazolidin-4- ylimino)butanoate ....................................................................................... 244 Appendix 47. 13C NMR spectra of 3-azidopropyl-3-(3-oxois0xazolidin-4- ylimino)butanoate ....................................................................................... 245 Appendix 48. 1H NMR spectra of 1-azid0-2-(2-(2-methoxyethoxy)ethoxy)ethane (mDEG). ..................................................................................................... 246 Appendix 49. 13C NMR spectra of 1-azid0-2-(2-(2-methoxyethoxy)eth0xy)ethane (mDEG). ..................................................................................................... 247 Appendix 50. 1H NMR spectra of 1-azidodecane. ............................................ 248 Appendix 51. 13C NMR spectra of 1-azidodecane. ............................................ 249 Appendix 52. 1H NMR spectra of PPGL-g-n-decyl-co—mDEG. ......................... 250 Appendix 53. ”C NMR spectra of PPGL-g-n-decyI-co-mDEG. ......................... 251 AD pendix 54. 1H NMR spectra of 1,5-diazidopentane. ..................................... 252 AD pendix 55. “’0 NMR spectra of 1,5-diaziopentane. ...................................... 253 A D pendix 56. 1H NMR spectra of cross-linked PPGL nanoparticle by 1,5- diazidopentane ........................................................................................... 254 AD pendix 57. 1H NMR spectra of 3-azido-7-hydroxycoumarin. ........................ 255 Appendix 58. 13C NMR spectra of 3-azido-7-hydroxycoumarin. ....................... 256 Images in this dissertation are presented in color. xviii LIST OF SCHEMES Scheme 1. Routes to functionalized Iactide-based polymers. Lactide monomers are polymerized via catalytic ring opening polymerization. FG1 and FG2 represent functional groups (alcohols, amines, alkyl chains, acids, etc.) that can be appended to the polymer .................................................................... 2 Scheme 2. NatureWork’sO route to high molecular weight poly(Iactide) ............... 3 Scheme 3. Synthesis of functionalized glycolide monomer. FG1 represents functional groups (alcohols, amines, alkyl chains, acids, etc.) that can be appended to the functional monomer. ............................................................ 4 Scheme 4. Synthetic routes to alkyl substituted polyglycolides. ............................ 5 Scheme 5. Comparison between copolymerization of two symmetrical monomers and homopolymerization of AB monomer. ................................... 8 Sch eme 6. Synthesis, polymerization, and deprotection of 3-benzyl0xymethyl- 1,4-di0xane-2,5-dione (3-BMG) .................................................................... 12 30h eme 7. Synthesis, polymerization, and deprotection of 3(8)- benzyloxylmethyl-1,4-dioxane-2,5-dione (OBn-L—Ser-aIt-Gly) and 3(8)- (benzyloxymethyI)-6(S)—methyl-1 ,4-dioxane-2,5-di0ne (OBn-L-Ser-aIt-Lac).13 Sc h eme 8. Synthesis, polymerization, and partial deprotection of 3-(1,2,3,4- tetraoxobutyI-diisopropylidene)-di0xane-2,5-dione (DiPAGYL) and 3-methyl- 6-(1 ,2,3,4-tetraoxobutyI-diisopropylidene)-di0xane-2,5-dione (DiPALAC)....16 S ch eme 9. Synthesis, polymerization, and deprotection of 3(3)- [(benzyloxycarbonyl)methyl]-1 ,4-di0xane—2,5Oione. .................................... 1 7 Sch eme 10. Synthesis, polymerization, and deprotection of malide dibenzyl ester prepared from either L-malic acid (L-MA) 0r L-aspartic acid (L-Asp) ............ 18 SCI'“Ieme 11. Synthesis, polymerization, and deprotection of OBn-L-Glu-aIt-Lac. ..................................................................................................................... 19 Scheme 12. Synthesis, polymerization, and deprotection of Cbz-L-Lys-aIt-Lac. 20 i Qheme 13. Synthesis and polymerization of oligo(ethylene glycol) monomethyl ether-modified polyglycolides. ...................................................................... 22 S<2I'Weme 14. Synthesis of and polymerization of diallylglycolide. ........................ 23 xix Scheme 15. Synthesis and polymerization of propargylglycolide. ...................... 24 Scheme 16. Functionalized poly(e-caprolactone) via anionic derivatization ....... 27 Scheme 17. Synthesis of functionalized polylactide via free radical addition of maleic anhydride and functionalization of maleic anhydride with hexanediamine ............................................................................................. 28 Scheme 18. Synthesis of comb PLA via macroinitiation of Iactide from pendant alcohol groups. ............................................................................................. 29 Scheme 19. Synthesis of poly(vinyl alcohol-graft-poly(lactide-co—glycolide)) via macroinitiation of Iactide and glycolide from poly(vinyl alcohol). .................. 30 Scheme 20. Synthesis of p0lyester-g-poly(ethylene glycol) copolymers via DCC coupling of terminal alcohol groups and succinic anhydride derivatives. ..... 31 Sch eme 21. Hydroxylation and functionalization of poly(L-lactide-co- diallylglycolide). ............................................................................................ 32 Sch eme 22. Synthesis of poly(e-capr0lactone-graft—poly(ethylene oxide)) via ketoximine ether functionalization. ............................................................... 33 30h eme 23. Functionalization and cross-linking of poly(e-caprclactone-co-2- oxepane-1,5-dione) with 1-amin00xydecane or 1,6-bis(aminooxy)hexane. .34 Sch eme 24. Functionalization and grafting of poly(e-caprolactone) by Michael- type addition ................................................................................................. 35 SC h eme 25. Functionalization of poly(e-caprolactone) by atom transfer radical addition. ....................................................................................................... 36 8 ch eme 26. Functionalization of poly(L-lactide-co—diallylglycolide) using olefin cross metathesis. ......................................................................................... 38 S Q h eme 27. Emrick’s synthesis of PEG-functionalized polyesters by click chemistry ...................................................................................................... 39 3 Q"Weme 28. JérOme’s synthesis of functionalized polyesters by Cu(l) catalyzed click chemistry .............................................................................................. 40 S'S'P'Weme 29. Synthesis of alkyl- and poly(ethylene glycoI)-functi0nalized poly(propargylglycolide). .............................................................................. 41 S Cheme 30. Synthesis of disubstituted Iactide monomers, ethylglycolide (EG), isopropylglycolide (iPG), n—hexylglycolide (nHG), cyclohexylglycolide (cHG) and phenyllactide (PL). ................................................................................ 47 Scheme 31. Separation of meso- and rac- monomers, and copolymerization with rac-lactide. R denotes the substituent on the disubstituted glycolide. .......... 48 Scheme 32. Synthesis of the AB glycolide 3-(8-(benzyloxy)octyl)-1,4-di0xane-2,5- dione. ........................................................................................................... 90 Scheme 33. Synthesis of BOG-co-LLA copolymers. Polymerizations were run at 130 “C (bulk) or 90 °C (toluene solution) using SnOctz and BBA as catalyst and initiator, respectively ............................................................................ 101 Scheme 34. Mechanism for the c0pper(l)-mediated 1,3-dip0lar cycloaddition between alkynes and azides, also known as “click” chemistry ................... 128 Scheme 35. Synthesis, polymerization, and functionalization of poly(propargylglycolide). ............................................................................ 131 Scheme 36. Synthesis of azobenzene PMMA initiator and ATRP polymerization of poly(methyl methacrylate) using CuBr/2,2’-bipyridine in anisole at room temperature ................................................................................................ 1 44 Sch eme 37. Synthesis of L-cycloserine hydrolyzable linker and conjugation to PPGL-g—mPEG ........................................................................................... 147 Sch eme 38. Synthesis of L-homoserine Iactone hydrobromide and conjugation to PPGL-g—mPEG ........................................................................................... 148 SC h eme 39. Synthesis of thermoresponsive polyglycolides. ............................ 151 Sc h eme 40. Cross-linking of poly(PGL[EG75X25]) unimolecular micelles by 1,5- diazidopentane ........................................................................................... 1 79 Sc'"Ieme 41. Synthesis of 3-azido7-hydroxycoumarin and click attachment to poly(PG L[EG75X25). .................................................................................... 1 84 xxi 3-BMG 3-BMMG 3-MG AF M BBA BOG cHG Cl— cmc LIST OF ABBREVIATIONS 3-benzyloxymethyl glycolide 3-benzyloxymethyl methyl glycolide 3-methylglycolide atomic force microscopy 4-tert-butylbenzyl alcohol 3-benzyl0xy0ctyl glycolide broad cyclohexylglycolide cross-linked critical micelle concentration doublet chemical shift Dalton N,N'-dicyclohexylcarbodiimide doublet of doublets dynamic light scattering 4-dimethylaminopyridine N,N-dimethyl formamide deuterated dimethyl sulfoxide differential scanning calorimetry xxii dt EG e-CL FG F M F-R GC GPC HOG iPG IR doublet of triplets ethylglycolide epsilon-caprolactone functional group functional monomer Fineman-Ross graph gas chromatography gel permeation chromatography 3-hydroxyoxyoctyl glycolide Henz isopropylglycolide infrared spectroscopy coupling constant kiloDalton Kelen-Tudbs liquid chromatography L-cycloserine lower critical solution temperature L-homoserine Iactone L4acfide light scattering xxiii M1 or M2 m/z MALLS MD mg MHz mL mp mPEG MS MWCO nDG nHG NLS NMR NOE multiplet moles per liter monomer 1 or monomer 2 mass of ion (atomic units)fits charge number multi angle laser light scattering mandelide milligram megaHertz milliliter number average molecular weight melting point poly(ethylene glycol) monomethyl ether mass spectrometry weight average molecular weight molecular weight cutoff wavenumber n-decylglycolide n-hexylglycolide nonlinear least squares nuclear magnetic resonance nuclear Overhauser effect xxiv NP PBS PDI PEG PEI PEO PGL PL PLA PLL PLLA PMMA POM ppb PPGL pm“ PPNI psi PTFE nanoparticle pentet para phosphate buffered saline polydispersity index poly(ethylene glycol) poly(ethylene imine) poly(ethylene oxide) propargylglycolide phenyllactide polylactide poly(L-lysine) poly(L-lactide) poly(methyl methacrylate) polarized optical microscopy part per billion poly(propargylglycolide) parts per million post-polymerization modification pound per square inch poly(tetrafluoroethylene) quanet XXV rac ROP RT 86C SnOctz TEM tert TsOH UV-vis reactivity ratio racemic ring opening polymerization room temperature singlet secondary tin(ll)-2-ethylhexanoate triplet transmission electron microscopy tertiary glass transition temperature tetrahydrofuran melting temperature p-toluenesulfonic acid ultraviolet-visible xxvi Chapter 1 Introduction Structure and Properties of Functionalized Polyglycolides Functionalized polylactides address two critical issues in polymer science: degradability and utility. An important physical characteristic of polymers is the glass transition temperature ( Tg) since it defines the use temperature range for many applications."2 Heating polymers through their Tg affects many physical properties and causes increased permeability, loss of dimensional stability, and increased resilience. Since the T9 of polylactide (PLA) is fairly low (50 — 60 °C), crystalline poly(L-lactide) (PLLA) is used in structural applications such as fibers and packaging? orthopedic fixations,“'5 sutures,6 and scaffolds for tissue engineering.7 Low T9 polylactides are rubbery at room temperature, degrade faster, and are preferred for applications such as controlled drug delivery. Although racemic polylactide (rac-PLA) has a fairly low Tg, its hydrophobicity and lack of functional group diversity prevents its widespread use as a degradable drug delivery vehicle. Two strategies have been developed to introduce functionality to the polylactide scaffold: the functional monomer (FM) approach and the post- polymerization modification (PPM) approach (Scheme 1). Both routes incorporate functional groups onto the polymer backbone to enhance properties or act as access points for further modification. Tradltlonal Polymerization Approach (sole modification sites 1010 ROH,Sn" (who E/LOIIV ———> O O ROP Iactide hydrophobic PLA Functional Monomer Approach F6 0 O (I O 1 FG ————> R0 0 . F601 functional monomer Post-Polymerization Modification Approach II 9’ ROH, PSn ‘ o H no 0 o .i n monomer with sites for subsequent modification functional polymer synthon Scheme 1. Routes to functionalized Iactide-based polymers. Lactide monomers are polymerized via catalytic ring opening polymerization. FG1 and FG2 represent functional groups (alcohols, amines, alkyl chains, acids, etc.) that can be appended to the polymer. Functional Monomer Approach Synthesis and Polymerization of Functionalized Glycolide Monomers Prior to the commercialization of poly(Iactide) by NatureWorks®, large- scale production of polylactide was too costly for all but medical applications. However, development of the low-cost “from corn to plastic” process has allowed for continuous processing of PLA and significant cost reduction. The process begins when field corn is harvested and the corn sugar, or dextrose, is isolated. The dextrose is fermented and distilled into lactic acid, which can undergo acid- catalyzed oligomerization forming low molecular weight polylactide. The oligomers are then thermally cracked to the dimer, L-Iactide, which is polymerized via ring opening polymerization (ROP) (Scheme 2). Since only a few other a-I’IYOTOXY acids such as 2-hydroxyisocaproic acid8 and B-phenyllactic acid9 can be isolated via fermentation most functionalized a-hYCIi'OXy acids rely on the synthesis of unique materials. As shown in Scheme 3, functionalized glycolide monomers are prepared by synthesizing functionalized a-hydroxy acids, which are dimerized in dilute organic solvents. o O H+ H.0JKrOH > 13.0 (DWI/k0 H O n lactic acid fermentation oli omer \ OH 9 HO 0 l HO OH Z20 OH dextrose O YL A M 10f R. O H ‘ O O O O O n Iactide poly(Iactic acid) (PLA) Scheme 2. NatureWork’s® route to high molecular weight poly(Iactide). + O o dilute solvent 0 0 FG‘ functional a-hydroxy acid functional monomer Scheme 3. Synthesis of functionalized glycolide monomer. FG1 represents functional groups (alcohols, amines, alkyl chains, acids, etc.) that can be appended to the functional monomer. Alkyl-Substituted Polyglycolides Alkyl-substituted polyglycolides have been synthesized by polymerizing cyclic dimers derived from the corresponding ct-hydroxy acids, which are obtained by enzymatic a-hydroxylation of fatty acids or by standard synthetic methods. Most standard syntheses follow one of two routes: reaction of Grignard reagents with diethyl oxalate, followed by reduction and hydrolysis of the a-ketoester, or hydrolysis of the cyanohydrin formed upon reaction of potassium cyanide with the corresponding aldehyde (Scheme 4). Using these methods, polyglycolides with n- alkyl chains up to 16 carbon atoms have been prepared. Usually, increasing the steric bulk of the substituents attached to the polymer backbone increases T9, as seen in p0lyolefins, but the T93 measured by differential scanning calorimetry (DSC) for polyglycolides show the opposite trend, longer alkyl chains decrease the T9 (Figure 1).10 Apparently, the linear alkyl chains decrease the dipole-dipole interactions between the ester groups in the polymer backbone. Pd/C o -OH OH OH ' + R H R CN 0 H+ O 0 R H A catalytic ROP RIGID no \IIAO o o n R o n Scheme 4. Synthetic routes to alkyl substituted polyglycolides. R0 0 O H O n poly(Iactide) poly(ethylglycolide) poly(hexylglycollde) poly(octylglycolide) O __ O _ - O _ 0 Figure 1. Linear alkyl-substituted polyglycolides and their glass transition temperatures. lf linear chains decrease the Tg, then branched alkyl chains would more effectively increase the T9. Although several polyglycolides with branched chains have been synthesized, the structural diversity of polyglycolides is still modest. For example, poly(isopropylglycolide), a Iactide analogue of polyvaline, and poly(isobutylglycolide) have been prepared. In poly(isopropylglycolide) the isopropyl group is directly attached to the polymer backbone, whereas in poly(isobutylglycolide) the iSOpropyl group is placed one carbon from the polymer backbone (Figure 2).10 The T9 of poly(isopropylglycolide) and poly(isobutyl glycolide) are 50 and 23 °C, respectively, illustrating the importance of placing the bulky group adjacent to the polymer backbone. Clearly, the isopropyl group is more sterically demanding than the methyl group of Iactide, but its failure to significantly increase T9 again points to the importance of dipole-dipole screening in determining Tg. A significant increase in the T9 was seen in poly(cyclohexylglycolide),11 with a T9 of 98 °C. Apparently, the cyclohexyl group was sufficiently bulky to hinder rotation of the polymer backbone and increase the T9. o o 0 no OjI/Lo H no 0 O H no 0 o H O n O n O n poly(Iactide) poly(isopropylglycolide) poly(isobutylglycolide) poly(cyclohexylglycolide) _ O __ O __ O _ Tg_55 C Tg-50 C Tg-15 C Tg_98 °C Figure 2. Branched alkyl-substituted polyglycolides and their glass transition temperatures. Non-symmetrical monomers provide another route to structural diversity in polylactide-related polymers. Polymerization of AB monomers, where two different glycolic acid derivatives are incorporated into one glycolide ring, is analogous to copolymerization. However, the growing chain end in an AB system discriminates between two different sites on the same glycolide monomer rather than between two comonomers (Scheme 5). For glycolides with simple alkyl and aryl substituents, steric effects largely define the rate of ring opening polymerization and increasing the difference in the relative size of the two ring substituents favors alternation. Shown in Figure 3 are six examples of simple AB monomers and their polymers prepared by Baker and coworkers“"12 The n-hexyl derivative was also reported by Trimaille etatl.13 A O o B O O 0 A 0 B I I + I I —> so 0184690180 ” O O A O O B A O B O n Random A O O O B O B 010:. —“’ R0 °Ykokr°rko ” A O A O ” Alternating Scheme 5. Comparison between copolymerization of two symmetrical monomers and homopolymerization of AB monomer. o o 0 O O 0 IL—rno . O. I —».. . or o O o 0 0 O n O rg=77c O 7 o O O 0 O O H I —+Ro 01X)” 1 “TR0 0 o o o 0 0 o n o _ o Tg=SOC 79-730 0 o 0 o o 0 I THO O 0” o JVVV—Wo ° 0” o o o n o n o _ o Tg=85C Tg—13C Figure 3. AB—substituted polyglycolides and their glass transition temperatures. 8 AryI-Substituted Polyglycolides By analogy to simple p0lyolefins, aryl-substituted polyglycolides are promising candidates for high Tg polymers. For example, while atactic polypropylene has a T9 of ~ -11 °C, polystyrene's T9 is 109 °C. Inspired by the high T9 of polystyrene and the availability of phenyllactic acid via various biosynthetic pathways,14 polymers based on phenyllactide and its derivatives were synthesized and studied (Figure 4).‘5'16 Despite the addition of a bulky aromatic ring and the potential for Jt-J'c interactions in the polymer, the T9 for high molecular weight polymer was only 50 °C. Although the T9 was comparable to that of poly(isopropylglycolide), it is significantly higher thanpoly(isobutylglycolide), which also contains a methylene group between the bulky group and the polymer backbone. As expected, the steric effects of alkyl groups or aromatic rings decrease as they are moved farther from the polymer backbone. To further increase the steric bulk, methyl groups were added to the aromatic rings. 4» Methylphenyllactide was synthesized and polymerized to high molecular weight, increasing the T9 to 59 °C, an improvement over poly(phenyllactide), but still far too low to be considered a high T9 material. The meta and ortho derivatives were also investigated and are suspected to have greater effects on increasing the T9 due to higher steric demands, however, their syntheses were problematic, and no reliable T9 data was reported. O H no 0 o H no 0 O H no 0 o H 60 O O O n O n O n n poly(phenyllactide) poly(o-methylphenyl)lactide poly(m-methylphenyl)lactide poly(p-methyiphenyl)lactide T9 = 50 °c Tg = 51 (7) °C T9 = 42 (7) °c T9 = 59 °C 0 O O H R0 0 O H R0 0 O H R0 0 O O n O J n O n poly(methylbenzylglycolide) polymandelide) po|y(cyc|ohexy|g|yco|jde) o _ 0 __ o Tg=?? C Tg_100 C Tg_98 C Figure 4. Aryl-substituted polyglycolides and their glass transition temperatures. Comparing the T93 of polystyrene and poly(allylbenzene), a structural analogue of poly(phenyllactide), suggests that using methylene groups to link the aromatic ring to the polymer backbone causes an ~30° drop in T9. Using this logic, direct attachment of the aromatic ring to the polyglycolide backbone should significantly increase T9. Thus, polymandelide, the cyclic dimer of mandelic acid should provide the appropriate structure. Like lactic acid, mandelic acid is a naturally occurring metabolite and in principle, poly(mandelide) synthesis could be a bio-based process. Early attempts to synthesize poly(mandelide) were hindered by impurities in the monomer synthesis that served as initiators in its polymerization and limited its molecular weight. The synthesis of poly(mandelide) 10 presented several challenges, most notably racemization of stereochemistry and poor monomer solubility, however, Baker et al. was able to generate high molecular weight polymers with Tgs of ~100°.12 The materials were colorless, glassy amorphous solids, similar to polystyrene. However, poly(mandelide) may not be a perfect substitute for polystyrene due to yellowing that develops during melt processing of the polymer most likely a result of the highly reactive methine protons which are a to both a carbonyl and a benzene ring. One way to alleviate the problems associated with poly(mandelide) would be to use the analogous structure without the reactive aromatic ring, which would minimize the reactivity of the methine protons. Jing et al. showed that both mandelic acid and mandelide could be hydrogenated directly to the corresponding cyclohexyl derivatives. Polymerization of cyclohexylglycolide yielded high molecular weight polymer with a T9 of 98 °C, comparable to both polystyrene and poly(mandelide), but with the decreased reactivity of the methine protons, no degradation was observed.11 Hydroxy—Functionalized Glycolides One of the synthetic challenges of hydroxy-functionalized glycolides is the presence of the alcohol group itself. Water and alcohols are known initiators in the polymerization of Iactide, and therefore the alcohol groups must either be introduced post-polymerization or via a protection/deprotection scheme. Feng and co-workers introduced the first hydroxy-functionalized glycolide, 3- 11 benzyloxymethyl-1,4-dioxane-2,5-dione (3-BMG), starting from 3-chloro- propanediol (Scheme 6).17 Ring-opening polymerization catalyzed by tin(|l)-2- ethylhexanoate (SnOctz) generated an amorphous polymer with a T9 of 34 °C. Deprotection via catalytic hydrogenation removed the pendant benzyl groups, resulting in a water-soluble polymer, poly(3-hydroxymethyl-1,4-dioxane-2,5-dione) (poly(3-HMG)). The contact angle and the T9 of the polymer both decreased upon deprotection. o o o HOrOH HNO3 HO HO B’\/U\BI i0 O _——> OH ——> OH —-———-> e 0 CI Na 0 O 0 Cl 69 O ’ 3-BMG | \= E) / ROP HO I O i H O O 2 no 0 o H <——— R0 0 o H O " Pd/C o n poly(HMG) poly(BMG) O O Tg=20 c 79:34 0 Scheme 6. Synthesis, polymerization, and deprotection of 3-benzyloxymethyl-1,4- dioxane-2,5-dione (3-BMG). Hennink et al. and Weck et al. concurrently reported a more versatile route to substituted Iactide monomers bearing hydroxy groups (Scheme 7).“3'19 Starting from optically pure O-Bn-L-serine, they obtained optically pure glycolides. In the synthesis of Iactide derivatives however, rac-2-bromopropionyl bromide was used 12 in the intramolecular ring closing reaction and two diastereomers were formed. Hennink and co-workers separated the two diastereomers via flash column chromatography and used the (SS) isomer for polymerizations. OBn-L-Ser-aIf-Gly o o o O: 0 no: —> O : H —> O : H i T o\/© no 0 no 0 O O ’/ HOI/\ n I)? n O 1 . poly(OBn-L-Ser-aIt-Gly) poly(L~Ser-aIi-Gly) Br\)LBr 2. NaHCO3 O O NaNOg HO/H\(NH2 _ HOJKEOH 0 H2304 O I \ / O 1. / I Br\./ILCI . I 5 \ 2. NaHCO3 o 0\_ o “Ox o O i , = . / . —*R0 (3on RO OrOH -. O \ I O n o n O O ’I/ OBn-L'sef'a”‘La° poly(OBn-L-Ser-aIf-Lac) poly(L-Ser-aIf-Lac) Scheme 7. Synthesis, polymerization, and deprotection of 3(S)-benzyloxylmethyl- 1,4-dioxane-2,5~dione (OBn-L-Ser-alt-Gly) and 3(S)-(benzyloxymethyl)-6(S)- methyl-1,4-dioxane-2,5-dione (OBn-L-Ser-alt-Lac). Hennink also compared the polymerization rates of the Iactide- and glycolide-substituted benzyloxymethyl monomers, using ethylzinc phenolate and 2- propanol as the catalyst and initiator, respectively. Crystal structure analysis revealed minor differences in the placement of the OBn and methyl substituent in OBn-L-Ser-aIt-Lac monomer, with both substituents in equatorial positions of the 13 dilactone ring. Since the groups presented almost identical steric demands, both carbonyls were nearly equally accessible and polymerization yielded a random polymer. However in OBn-L-Ser—aIt-Gly, replacement of the methyl group with hydrogen results in a clear structural difference and preferential opening at the less hindered carbonyl, creating an alternating polymer. DSC analysis showed that both polymers were completely amorphous with T93 of 30 °C and 25 °C for poly(OBn-L-Ser-aIt-Lac) and poly(OBn-L-Ser-aIt-Gly), respectively. The protecting groups were removed from the two polymers via catalytic hydrogenation using Pd/C (10 °/o) and THF as a solvent. The homopolymers were completely deprotected after 24 hours and GPC showed no chain scission. Deprotected poly(L-Ser-aIt-Lac) was completely amorphous with a T9 of 30 °C; whereas poly(L-Ser-aIt-Gly) was semi-crystalline with a T9 of —4 °C and a Tm of 135 °C. Neither polymer was soluble in aqueous solution.19 Weck and co-workers obtained 21 kD poly(OBn-L-Ser-aIt-Gly) via ring- opening polymerization using SnOctz as the catalyst. The amorphous polymer had a T9 of 18 °C and was deprotected by hydrogenolysis using Pd(OH)2; however no molecular weight or T9 data was presented for the deprotected polymers.18 A second approach to hydroxy-modified glycolides was described by Vert and co—workers in their syntheses of 3-(1,2,3,4-tetraoxobutyl-diisopropylidene)- dioxane—2,5—dione (DiPAGYL) and 3-methyl-6-(1,2,3,4-tetraoxobutyl- 14 diisopropylidene)-dioxane-2,5-dione (DiPALAC), a monomer composed of a glycolyl and a D-gluconyl unit derived from gluconic acid (Scheme 8).20 After selective protection of the four pendant alcohol groups with 2,2-dimethoxy propane and intramolecular ring closure with bromoacetyl chloride or 2-bromopropionyl chloride, the monomers were isolated as white crystals. Ring opening polymerization of DiPAGYL led to a 20 kD brittle amorphous polymer with a T9 of 95 °C. All attempts to polymerize DiPALAC led to oligomers with molecular weights of ~2 kD. Several methods to remove the isopropylidine groups were investigated, all under acidic conditions, but most degraded the polyester backbone.21 The best results were accomplished using iodine in methanol or aqueous acetic acid; however even these methods showed modest degradation. The maximum deprotection obtained was 60 %, as determined by NMR. At 60 % percent deprotection, a fully protected 32 kD polymer degraded to 5 kD (versus 27 kD theoretical value). The partially deprotected polymers were soluble in hydrophilic solvents such as methanol and ethanol and were partially soluble in water. 15 OH OMe OH 0 OH R = H (poly(DiPAGLY)) R = H (DiPAGLY) n = Me (poly(DiPALAC)) n = Me (DiPALAC) Scheme 8. Synthesis, polymerization, and partial deprotection of 3-(1,2,3,4» tetraoxobutyl-diisopropylidene)-dioxane-2,5-dione (DiPAGYL) and 3-methyl-6- (1 ,2,3,4—tetraoxobutyl-diisopropylidene)-dioxane-2,5—dione (DiPALAC). Carboxylic Acid-Functionalized Polyglycolides Biomedical and pharmaceutical applications often require water-soluble biodegradable polymers. Since the hydrolysis rates of widely used poly(a-hydroxy acids) are often slow and limited by crystallinity, methods to decrease crystallinity and hydrophobicity are important for in vivo applications. One method for imparting hydrophilic character to polymers is to functionalize them with hydrophilic groups, such as carboxylic acids. Taguchi and co-workers described the synthesis of 3(8)- [(benzyloxycarbonyl)methyl]-1,4-dioxane-2,5-dione which was polymerized to an alternating copolymer consisting of glycolic acid and benzyl a-(S)-malate units 16 (Scheme 9).22 Ring-opening polymerizations were run in solution and bulk utilizing SnOctz or EtQZn (diethyl zinc) as the catalyst. Molecular weights ranged from 2 kD to 21 kD. The polymers were deprotected via catalytic hydrogenolysis using platinum oxide or palladium carbon yielding crystalline, water-soluble polymers. No molecular weight data were reported for the deprotected polymers. 0 + ”0W1 1.PhCH20H,H _ ©\/OY\I/CHL 1.Br\)LCI Lord 0 OH OH o o ”/U‘O O NH2 2. NaNOQ, H+ o OH 2. NaHC03 /\© ROP E? on o lo oék: 1 _ H2 0 04KE RO ‘ O\n/'\O'H = HO O\[]/’\0 H O In Pd/C o " Scheme 9. Synthesis, polymerization, and deprotection of 3(S)- [(benzyloxycarbonyl)methyl]-1 ,4-dioxane-2,5-dione. Malide dibenzyl ester, the cyclic dimer of B-benzyl rnalate, was prepared by Ouchi and co-workers23 using either L-malic acid (L-MA) or L-aspartic acid (L-AA); however higher yields were obtained using L—MA (Scheme 10). Ring-opening polymerization was carried out at 220 °C using organotin compounds as catalysts. The protecting groups were removed by hydrogenation using palladium-carbon in ethyl acetate. 17 o CCI3CHO'H20 O PhCHZOH 0 HO 2 2 NOH r O o r O o o OH A occ A0 L-MA I HCI O 1. PhCHQOH, H+ @ O ”ONOH ' 0N0“ o NH2 2. NaNOZ, H“ O OH L-Asp g ZnO OH o v 0 0 H 0 0 I @o o o o H 2 o H o O " Pd/C O n 0 0 OAK) o 0 Scheme 10. Synthesis, polymerization, and deprotection of malide dibenzyl ester prepared from either L-malic acid (L-MA) or L-aspartic acid (L-Asp). More recently Weck and co-workers prepared a glutamic acid polylactide derivative, OBn-L—Glu-aIt-Lac. Diazotization of glutamic acid to its corresponding a-I‘IYOI'OXY acid and reaction with (S)-2-bromopropionic acid in the presence of N- N’-dicyclohexylcarbodiimide (DCC) and 1-hydroxybenzotriazole followed by intramolecular ring closure led to the functionalized monomer. Ring-opening polymerizations were carried out in the presence of SnOctg at 140 °C (Scheme 11).18 The highest molecular weight obtained was 3 kD, reportedly due to 18 interaction between the monomer’s side chain and the catalyst. A T9 of 10 °C was reported; however this value may change significantly with higher molecular weight polymers. Deprotection of the polymers via hydrogenolysis using Pd(OH)2 resulted in quantitative removal of the protecting groups as seen from the 1H NMR. O O 0 far , OH o 0 NH NaN02 OH 5 - T O HO 2 > HO > )L O O ' O O O OBn-L-Glu-aIt-Lac 2. Kl, acetone / \ I 3. DIEA, acetone ROP O 75 0,10 hr OBn-L-Glu OH O 0 05K: O A ‘ O ‘ H no CHI/\o , H <——— nO \n/\O 0 H Pd(OH)2/C 0 " poly(L-Glu-aIt-Lac) poly(OBn-L-Glu-aIt-Lac) Scheme 11. Synthesis, polymerization, and deprotection of OBn-L—Glu-alt-Lac. Amine-Substituted Glycolides Although many morpholine—2,5-diones have been described,2427 very few 1.18 amine functionalized Iactide derivatives exis Gerhard et al. synthesized a Iactide analog of polylysine, Cbz-L-Lys-aIt-Lac (Scheme 12). Ring-opening polymerization provided an 8 k0 amine protected polymer with a T9 of 20 °C, and 19 complete deprotection of the benzyl carbamate using HBr (33 %)/AcOH caused little backbone degradation. Molecular weight data for the polymer were not obtained due to poor solubility in CHzClz; however, the GPC trace of a 30 kD polylactide had the same Mn and peak shape before and after exposure to the deprotection conditions. 0 N NO NH2 8 2 ":TO HO MNO H2804 on Cbz-L-Lys-a/f-Lac 2. KI acetone H NYC HN Y0 O 3. DIEA acetone ROP I CI 75 °,C 10 hr Cbz-L-Lys Q 0 2k NH2 0 o x, o : .pr ._ “heir poly(L-Lys-aIf-Lac) poly(Cbz-L-Lys-aIi-Lac) NH Scheme 12. Synthesis, polymerization, and deprotection of Cbz—L-Lys-aIf-Lac. Peg-Grafted Glycolides Recent reports have shown that introducing oligo(ethylene glycol) groups to hydrophobic polymers render them hydrophilic.”32 In addition, these polymers often exhibit lower critical solution temperature (LCST) behavior, making them ideal targets in applications such as drug delivery systems, smart surfaces, and 20 bioseparations. At the LCST, these materials undergo a solution-gel transition that corresponds to the entropically driven expulsion of solvating water molecules from the polymer. Recent research on responsive LCST materials, especially those that contain mPEG segments, suggests that tuning the hydrophobic/hydrophilic balance in materials can shift the LCST over a broad temperature range. Lactide derivatives with 1-4 oligo(ethylene glycol) segments have been synthesized and polymerized using SnOctz (Scheme 13).33 The synthetic approach to the monomers entails reaction of an oligo(ethylene glycol) monomethyl ether with 1,6- dibromohexane to generate the corresponding bromine capped oligo(ethylene glycol) monomethyl ether. Reaction with magnesium followed by the addition of diethyl glyoxalate provides the corresponding ct-keto ester. Catalytic hydrogenation of the ct-keto-ester, followed by hydrolysis, gave the corresponding ct-hydroxy acid, which was dimerized to the PEG-substituted glycolide. Since high molecular weight polymers require complete removal of trace water and alcohol impurities prior to polymerization, the PEG-functionalized monomers were dried under vacuum at 100 °C. High molecular weight polymers catalyzed by SnOct2 were obtained: Mn = 149,500, 56,400, 59,800 and 10,600 g/mol for polymers containing 1, 2, 3, or 4 ethylene glycol units per repeat unit, respectively. 21 Br Br 1. Mg NaH n _ 1 2 3 4 2. diethyl oxalate n = 1, 2, 3 4 — 9 I s O O HOfiO/WQKU/ \/ O n = 1, 2, 3, 4 1. 5 °/o Pt/C/ACOH 2. KOH 0L 0 OH n O AHOVHZTO/VWSI/ O n=1,2,3,4 I H. TOP— IOH"OVV\?1:7:Wo/rvoi n n=1,2,3,4 Scheme 13. Synthesis and polymerization of oligo(ethylene glycol) monomethyl ether-modified polyglycolides. Alkenyl- and Alkynyl-Substituted Polyglycolides The versatility of post-polymerization reactions has driven the synthesis of monomers that incorporate functional groups tolerant of polymerization conditions, but available for future modification. For example, pendant alkene groups could be used in cross metathesis reactions, which inspired the synthesis of diallylglycolide (DAG), shown in Scheme 14.34 Diallylglycolide was polymerized in toluene at 80 °C for 3 days using SnOctz to give a tacky solid with a T9 of —10 °C. Melt 22 polymerization was not as effective as it lead to poor control over molecular weight and polydispersity index (PDI) due to the higher reaction temperature. Copolymerization of Iactide with DAG spaced the reactive sites along the polymer backbone. O O H WBr \ H+ \ O O OHHZO B'Cl z = OH _——> \/j/: IA 0 ' 3' ” OH \ THF 0 O diallylglycolide SnOct2 BBA poly( alllylglycolide) T9 = '10 00 Scheme 14. Synthesis of and polymerization of diallylglycolide. A second example is the synthesis of alkyne functionalized monomers that can be used in “click” chemistry reactions, specifically 1,3-dipolar cycloaddition reactions between an alkyne and azide. Propargylglycolide (PGL), an alkyne functionalized Iactide derivative, was synthesized in three steps beginning with a Reformansky-type reaction between propargyl bromide and ethyl glyoxalate in the presence of activated zinc (Scheme 15).33 Bulk polymerization of the monomer in the presence of SnOctz shows a linear relationship between the number average molecular weight (Mn) and conversion, indicating a reasonably controlled 23 homopolymerization; the polymer was amorphous with a T9 of 30 °C. Incorporation of propargylglycolide in a copolymer with Iactide was consistent with the feed ratios and 130 NMR is consistent with propargylglycolide being evenly distributed throughout the copolymer (Figure 5). 1. Zn TH F/ether H+ / O 0 _—fi + O\/ / MgBr /\O \/,2 0H20, WOH reflux /1 I/ O O O reflux % O O H O \ poly(propargyl glycolide) _ 0 T9 — 30 C propargyl glycolide 3110th Scheme 15. Synthesis and polymerization of propargylglycolide. 24 x’°~;rLo" PLA K} T PPGL-co—PLA U PPGL-block—PLA I I I I I I I I l l I I , I I I I I I I I I T ppm 169.0 168.0 167.0 166.0 Figure 5. 75 MHz ”C NMR carbonyl regions of propargylglycolide homopolymer (PPGL), copolymers, and polylactide (run in CDCI3). 1‘Unique Iactide carbonyl resonances in poly(Iactide-co-propargylglycolide) (PLA-co-PPGL).33 25 Post-Polymerization Modification As shown in Scheme 1, the functional monomer approach involves multi- step synthetic procedures each time modification is desired. Moreover, the functionality that is initially introduced must be compatible with polymerization conditions. Similarly, post-polymerization modification requires careful control of reaction conditions to avoid backbone degradation, but in contrast to the functional monomer approach, it offers the distinct advantage that an array of polymers can in principle be accessed from a single monomer. Having a single, simple procedure for placing a broad spectrum of pendant functional groups onto polyester substrates is highly desirable. The ideal post-polymerization modification approach should utilize a chemical process that does not degrade the polyester backbone and be compatible with a wide range of functional groups, solvents, and conditions. Although this document primarily focuses on glycolide and polyglycolide derivatives, most examples of post-polymerization modification on polyesters use various poly(s—caprolactone) derivatives and thus these examples will be used as background in this area. Direct Grafting of Functional Groups Directly grafting functional groups to a polymer backbone is appealing because of the vast number of groups that can potentially be attached to a single polymer scaffold. For example, Vert reported the derivatization of e-caprolactone (s-CL) using lithium diisopropylamide (LDA) and electrophiles such as methyl 26 iodide, benzylaldehyde, carbon dioxide, naphthoyl chloride, bromoacetylated ct- hydroxy-w-methoxy-poly(ethylene glycol) and polylysine (Scheme 16).”38 Successful implementation of this method however, is limited due to polymer degradation by main chain cleavage and low grafting densities, typically 10 %. MO} LDA MO] no H > no n THF G Li H n Mel 6) Br /\Cn)/O‘(/\O)Ip O (:I/HH 002 0 o no H n O V O O O o 0 no H no H KL n n O OH o OH 5’ Scheme 16. Functionalized poly(e-caprolactone) via anionic derivatization. :I Luo and co-workers described a similar approach, using free radical reactions.39 Radical reactions initiated by benzoyl peroxide, a radical initiator, covalently grafted maleic anhydride to PLA (Scheme 17). Amidation of maleic anhydride with hexanediamine resulted in the formation of carboxylic acid and amine side chains. The use of free radicals is also limited by significant backbone degradation, losing up to 20 °/o of the original Mn, and very low grafting densities, a maximum of 3 °/o. 27 stoop.) GAVE) rotWirLotittcoii: DAD 0 ll 0 o O O o O ij/flfNW o o H no 0 O ‘ no 0 o O O n O O 11 OH HN £21 H2N Scheme 17. Synthesis of functionalized polylactide via free radical addition of maleic anhydride and functionalization of maleic anhydride with hexanediamine. Lactide Initiation from Pendant Alcohol Groups One way to control PLA’s poor compatibility with soft tissue is to modify its crystallinity and degradation rate by introducing branching into the polymer structure. Branching affects physiochemical properties such as crystallinity, T9, and Tm. Recently a number of examples of branched polymer structures have been reported including hyperbranchedfwz star-shaped,”45 and dendritic“‘5'48 polymers, all showing interesting rheological and mechanical properties. For example, Ouchi ef al. synthesized poly(cyclo[Gly-Ser(Ole)]-co-LA), a copolymer of a protected cyclodepsipeptide consisting of glycolic acid (Gly) and O-benzyl-L- serine (Ole) and Iactide (Scheme 18).“50 The functionalized monomer was copolymerized with Iactide, spacing the alcohol groups along the polymer 28 backbone. Comb polymers were synthesized by polymerizing Iactide “teeth” from pendant alcohol groups. In vitro degradation suggests that comb-type PLA degrades significantly faster than linear PLA due to increased number of hydroxy end groups. wiltgrowwrlolk~ Nowell: ”3 p.121" ll 0 H o no N\ll/\O>8 p Scheme 24. Functionalization and grafting of poly(s-caprolactone) by Michael- type addition. Atom Transfer Radical Addition Reactions Atom transfer radical addition (ATRA) is the radical substitution of a radical initiator with a “grafting group,” frequently containing a double bond. One of the most commonly used monomers is a-chloro-e-caprolactone because its activated chloride is well suited for ATRA. a—Chloro-e-caprolactone can be easily 35 synthesized by Baeyer—Villiger oxidation of 2-chlorohexanone.67 Various olefins have been grafted to polymers via ATRA such as 3-buten-1-ol, 1,2-epoxyhex-5- ene, and a-methoxy-w-acrylate-PEO using CuBr/MeeTREN (copper(|) bromide/ tris[2-(dimethylamino)ethyl]amine) in DMF (Scheme 25). Unfortunately, many 68,69 chlorinated units were lost by reduction during ATRA, which decreased the grafting efficiency. Aiolwlp mela Cu Br Cu Br Mes-TR EN M66- -TRE Mes-THEN mltgolwi O. low): I 0 Che); Ollg/vwl: Scheme 25. Functionalization of poly(e-caprolactone) by atom transfer radical addition. Olefin Cross Metathesis Olefin cross-metathesis and ring-opening olefin metathesis have become standard procedures in polymer chemistry; however they have only been applied to PLA in one example. Radano et al. performed olefin cross-metathesis using 36 Grubbs’ 1St generation ruthenium carbene catalyst on an allyl-functionalized copolymer, poly(Iactide-co-diallylglycolide), using 9-decenyl-1-oxy-(4- nitrobenzene), allylbenzene, 2-(2-propenyloxy)benzaldehyde, 9—decane-1-oxy-(terf- butyldimethylsilane), or 9-decen-1-al. (Scheme 26)."4 Although the polymer showed no signs of degradation, the polymer was discolored indicating possible ruthenium contamination. Using a nitrobenzene UV tag, the conversion of olefins undergoing metathesis could be calculated, typically 26-29 %. In addition, a viscous gel formed in the absence of an olefin substrate, presumably the result of cross-linking the pendant allyl groups. To prevent cross-linking, five equivalents of the olefin substrate were required. 37 | 0 o no 0 OWOYLO H O o n OW I O H [j a 075 02N W CI2(PPh3)2RU=CHPh — d Cl2(PPh3)2Ru=CHPh OQNO Cl2(PPh3)2Ru=CH Ph 0 O:MHOIH H Scheme 26. Functionalization of poly(L-Iactide-co—diallyiglycolide) using olefin cross metathesis. “Click” Chemistry Because of its high selectivity, reliability, and tolerance to a broad range of functional groups and reaction conditions, “click” chemistry, specifically the copper(l)-mediated 1,3—dipolar cycloaddition of azides and alkynes, is a powerful strategy for elaborating polymer architectures. The Emrick group first described the 38 use of aqueous “click” chemistry to graft azide-terminated PEO and peptides onto polyesters containing pendant acetylene groups (Scheme 27).70 Later, Jéréme and co-workers found Emrick’s conditions caused significant backbone degradation during functionalization?"72 Using less severe conditions (THF as the solvent), they were able to introduce PEO, tertiary amines and ammonium salts onto caprolactone—based polyesters having pendant azides (Scheme 28). O >4fi\//\v/\\/ /\O O O H % n NBA/0V0); CUSO4'5H20 sodium ascorbate O M O o A O H n CUSO4'5H20 sodium ascorbate N3\/\/\)J\GRGDS / NNO‘GAO); N=N' WIGRGDS 0 Scheme 27. Emrick’s synthesis of PEG-functionalized polyesters by click chemistry. 39 Sé’b Now/Q 69% Br 0 /\N® G \fif/VOVO)’ // Q Cul 0 j DBU O H Cul no W0 n DBU N’N O N{ , on O , H 0 R0 0W0 n ,N 0 Scheme 28. Jérbme’s synthesis of functionalized polyesters by Cu(l) catalyzed click chemistry. Recently we introduced an acetylene modified Iactide derivative capable of post-polymerization modification by click chemistry (Scheme 29). Using milder conditions, DMF and room temperature, we were able to functionalize the polyester with alkyl and poly(ethylene glycol) branches. We showed good control over the degree of functionalization and no backbone degradation under the new click conditions.” 40 N3 CuSO4-5H20 N sodium ascorbate N’N \/\/\/\/\/ DMF, RT H O O O J” N 3/\/ \/‘(~0/\arp \ CUSO4'5H20 sodium ascorbate DMF, RT HO O //O:§—// \ \N,N\/\0/\éo\/io/ Scheme 29. Synthesis of alkyl- and poly(ethylene glycol)-functionalized poly(propargylglycolide). Conclusion Functionalized polyglycolides greatly enhance PLA’s role in biomedical and structural applications. Modification via either the functional monomer approach or the post-polymerization modification approach provides opportunities to modify the polymers’ Tg, Tm, and mechanical properties. Large bulky substituents increase the T9 relative to PLA and long flexible chains degrease the T9 of the homopolymer. The addition of functional groups provides access to many polymer architectures via post-polymerization. 41 Chapter 2 Reactivity Ratios of Substituted Lactide Monomers Introduction Biodegradable polyesters are used widely in pharmaceutical, biomedical, and environmental applications.‘7'2“H""‘H7"‘76 Many of these applications require specific material properties such as a high T9, controlled crystallinity, a specific degradation rate, and high mechanical strength that cannot be obtained from the homopolymers alone. Typical approaches used to control these properties include ”'79 preparation of block copolymers,8"'83 and the manipulation of stereochemistry, synthesis of specialized monomers.‘°"3"6 Perhaps the simplest method, however, is the copolymerization of two or more monomers which can provide multiple materials from a few starting monomers.”86 For example, copolymers of Iactide and glycolide or e-caprolactone are now commercially available and have been used use as sutures, implants, drug carriers, and cell scaffolds for tissue engineering?”88 Although copolymerizations are often assumed to be completely “random”, few are actually random from a statistical point of view due to the higher reactivity of one monomer towards polymerization. The products of such copolymerizations are often “blocky” and exhibit multi-phasic properties. Access to homogeneous polymers is particularly important in drug delivery and other applications because the polymer degradation rate and drug release profile are affected by the polymer crystallinity. Drugs diffuse more rapidly through amorphous domains than through 42 the crystalline regions,89 and in poly(lactide-co-glycolide) and poly(Iactide-co-e- caprolactone) crystalline regions degrade far slower than amorphous regions.90 In both systems, crystallization is the outcome of different monomer reactivities and the nonrandom placement of the monomers in the copolymers. In the copolymerization of Iactide and :s-caprolactone86 Iactide is approximately 30 times more reactive, whereas in the copolymerization of Iactide and glycolide,91 glycolide is 11 times more reactive than Iactide. In a copolymerization of two monomers, M1 and M2, four rate equations describe the rates of homo- and cross-propagation when the addition of monomer to a growing chain is only dependant on the chain end. In each expression, the rate of monomer addition is described by the rate constant, k. For example, when M1 adds to a propagating chain ending in M2, the rate constant is defined as k21. M1 k § WM1M1* (1) 11 WM1* M 2 > ""WW‘M1M2"r (2) k12 M1 .. k > WM2M1 (3) 21 M2 I» MM2M2* (4) k22 k11 k22 r1=-— and r2=— (5) k12 k21 43 Unfortunately, absolute rate constants are often inaccessible. Therefore the reactivity ratios, r1 and r2 (equation 5), are extracted from competitive reactions in lieu of the rate constants to describe the reactivity of the monomers in the copolymerization. The instantaneous polymer composition of the copolymer is described by the relative incorporation of M1 and M2 expressed in terms of the monomer reactivity ratios, r1 and r2 (equation 6). Reactivity ratios are usually extracted from polymerization data using the Fineman-Ross (F--R)92 or Kelen- Tudos (K-T)93 methods however more statistically sound methods such as non- linear least squares94 are generally preferred. dIM1] _ [M1l(f1[M1l+[M2]) 6 WM [M21([M1l+r2[M2]) ( ) Monomer reactivity ratios provide a straightforward method for predicting copolymer architectures and ultimately copolymer properties such as T9 and degradation profiles. Once determined, monomer reactivity ratios can be used to predict copolymer architectures. Random copolymers are formed when r1 = r2, and “blocky” copolymers containing long homopolymer segments form when r1 > r2 or r2 > r1. Although the reactivity ratios for many nondegradable polymers have been determined,”98 the reactivity ratios of most substituted Iactide derivatives are unknown.”100 The few examples include the copolymerization of Iactide with glycolide (GA) (ILA = 0.22 and rep, = 3.4),“ 3-methylglycolide (3-MG) (M = 0.92 44 and r3-MG = 1.71 ),95 e-caprolactone (ILA = 26.5 and QCL = 0.92),86 3-(1,2-3,4- tetraoxobutyl-diisopropylidene)-1,4-dioxane-2,5-dione (DiPAGYL) (ILA = 1.24 and rmpAGYL = 0.4)96 and 3-benzylmethylglycolide (3-BMG) (ILA = 0.37 and ’3-BMG = 1.96).97 Since these monomers are based on the same glycolide structure, the differences in reactivity are primarily based on steric hindrance around the ring carbonyls. Functionalized Iactide derivatives have been shown to have wide success in medical applications‘7'25'26'73'75 and as effective initiators in the synthesis of comb or brush polymers.“95°33"102 Understanding the placement of the functionalized monomer in the polymer backbone is especially important when it contains a functional group that can undergo post-polymerization modification or serve as an initiator for the synthesis of comb polymers. Leemhuis and co-workers19 evaluated the thermal properties of L-lactide copolymers comprised of 2 75 % L-lactide and 2 functional monomers with benzyl-protected hydroxy groups, 3-benzyloxymethyl glycolide (3-BMG) or 3-benzyloxymethyl methyl glycolide (3-BMMG); copolymers of L-lactide and 3-BMG contained highly crystalline regions, but copolymers with 3- BMMG were amorphous, indicating that the reactivity of 3-BMMG was similar to Iactide. The difference between the two monomers is that 3-BMG lacks a methyl group at the 6-position, which increases its reactivity and leads to blocky polymers. As part of our interest in designing, synthesizing, and characterizing 10,12,16.73,98-100 functionalized Iactide monomers, we copolymerized rec-Iactide with 45 five disubstituted Iactide monomers and determined their reactivity ratios. We compared the reactivity ratio data with the thermal properties of rac-lactide copolymers and homopolymerization rate data to develop simple structure- reactivity relationships that will guide the design of future functionalized Iactide monomers. Results and Discussion Monomer Syntheses The syntheses of the substituted Iactide derivatives—ethylglycolide (EG), isopropylglycolide (iPG), n-hexylglycolide (nHG), cyclohexylglycolide (cHG) and phenyllactide (PL) are shown in Scheme 30. The a-hydroxy acids used to prepare the disubstituted monomers were commercial products, as in the case of DL- phenyllactic acid, or synthesized. 2-Hydroxybutanoic acid (2) was prepared from reaction of propanal (1) with potassium cyanide followed by acidic hydrolysis of the cyanohydrin. Diazotization of DL-valine followed by alcohol displacement of the diazonium salt provided 2- hydroxy-3-methyl butyric acid (5).101 2-Hydroxyoctanoic acid (8) was prepared from reaction of n-hexyl Grignard with diethyl glyoxalate, followed by hydrogenation over Pd/C to give the a-hydroxy ester, which was hydrolyzed to 8. Catalytic hydrogenation of mandelic acid provided 2-cyclohexyl-2-hydroxyacetic acid (11).100 The a-hydroxy acids were dimerized to the symmetrical monomers using a '10 previously reported protoco The mesa-isomer was separated from the racemic- 46 isomers via recrystallization or column chromatography to eliminate potential isomer effects in determining the reactivity ratios (Scheme 31). die, p-TsOH, O O toluene. A t L reflux o o 56% 3 ethylglycolide (EG) p-TsOH , Jot... 1. KCN H20, /\'rH NaHSO3 0 °C 0 2 HCI reflux 63 °/o YIGId 1 )fi/IOK NaNOz, OH H2804, H20 NH2 . 75 % yield 4 1. Mg THF 2 diethyl oxalate then HCI \M/Br 3.1500 psi H2 =\NVOKH[JW 5% Pt/Al203, 5 days 7 4. KOH, EtOH OH 1500 psi H2, OH 5 % Rh/C, AcOH, O RT, 8 h 81 % yield 10 11 O OH OH OH OH O to o toluene, A reflux r o OI’/ 16 % 6 isopropylglycolide (iPG) .. O 0 ° NVI hA/ 38 /o O O 9 n-hexylglycolide (nHG) p-TsOH, toluene reflux 21 % 12 cyclohexylglycolide (cHG) p-TsOH, reflux 49 % 14 phenyllactide (PL) Scheme 30. Synthesis of disubstituted Iactide monomers, ethylglycolide (EG), isopropylglycolide (iPG), n-hexylglycolide (nHG), cyclohexylglycolide (cHG) and phenyllactide (PL). Copolymerization and Copolymer Isolation The racemic glycolides were polymerized with rac-lactide at various mole fractions under solvent free conditions using SnOct2 and t-butylbenzyl alcohol 47 (BBA) as the catalyst and initiator, respectively ([monomer]:[catalyst]:[initiator] = [400:1:1]) (Scheme 31). Since the monomer pool changes with conversion, the monomers were immersed in a preheated oil bath for 45 seconds and then quenched in ice to ensure low conversion (< 10 %). The monomer/polymer mixture was dissolved in CH2CI2 and passed through a short silica column to separate the polymer from the monomers. After the separation was complete, the copolymer adsorbed onto the silica was transferred into a 25 mL round bottom flask containing MeOH. The mixture was stirred for 4 hours and then centrifuged. The colorless polymer solution was decanted from the silica gel, evaporated to dryness, and analyzed by 1H NMR. mesa n,“ o o ——» I I O O H —:;I I; :I °I"O—+ ”,«IW, «I (9.3) 1-x 0 (8,8) (RR) racemic Scheme 31 . Separation of meso- and rac- monomers, and copolymerization with rec-Iactide. R denotes the substituent on the disubstituted glycolide. Monomer Reactivity Ratios and Copolymer Architectures We determined the relative incorporation of each monomer in the copolymers by comparing the integration of the methine peaks of each monomer in the copolymer 1H NMR spectra. The molar fraction of Iactide in the copolymer was 48 determined by equation 7, where F1 is the molar ratio of Iactide in the 00polymer and ILA and IM2 are the total integration of the methine protons of rac-lactide and the comonomer, respectively. I F1 -' LA (7) LA+'M2 For example, in the copolymerization of Iactide and n-hexylglycolide (molar Iactide feed ratio (ILA) = 0.81 ) (Figure 6), the signals from the methine protons of nHG and LA overlap, and the composition of the copolymer was calculated using the 3:221 ratio between the methyl protons (CH3, 1.9-1.8 ppm), methylene protons (CH2, 201.7 ppm), and methine protons of nHG (CH 5.2-2.0 ppm). We arbitrarily set the integration of the methylene protons (lnHG CH2) to be 2.00, which is consistent with the integration of the methyl protons (InHG CH3 = 3.05). Subtracting the expected integration due to the nHG methine protons (InHG OH = 1.00) from the total methine integration (InHG and LA CH = 6.32) gave the relative integration of the Iactide methine protons (ILA CH = 5.32). These values were used in equation 7 to calculate F1 (f1 = 0.81, F1 = 0.84). 49 o o C a O b 0 a and b l 1 fi l .l ‘ d c l Ix /\ " _ -, _ .2 _.____,_ A_ _ _ ___ _‘ __ J _J.____L_ J, . _’ ____. __ _,L-/ - A! ‘ ‘-. K..- 4 2 Y T I Y I T Y I Y I I f Y W T T ‘7— 'Y Y I— T Y Y Y T T Y Y Y T Y Y Y T T f ‘l T I I Y Y Y I T Y Y Y Y 5.0 4.5 4.0 3.5 3.0 2.5 2.0 1.5 1.0 ppm Figure 6. 500 MHz 1H NMR of poly(LA-co-nHG) at 8 % conversion. The molar fraction of Iactide in the copolymer (F1) was determined using the relative integrations of the two comonomers (see text, f1 = 0.81, F1 = 0.84). Reactivity Ratio Determination We copolymerized raclactide and rec-ethylglycolide at eight different mole fractions of Iactide (f1). The final compositions of the copolymers at low conversion in terms of the mole fraction of Iactide (F1) are given in Table 1. Using Origin (OriginLab, Massachusetts, USA), we plotted the instantaneous copolymer composition (F1) versus the comonomer feed composition (f1) for the various Iactide feeds. The reactivity ratios were extracted from the data via a nonlinear least squares (NLS) fit of the data to equation 8 using an iterative process for the selection of r1 and r2 values,102 50 rf2 + ff F'f.‘ + 2f1f2 + {2,2 where F1 and f1 are the mole fractions of lactide in the copolymer and in the feed, respectively, and f2 is the mole fraction of the disubstituted lactide derivative in the feed. The reactivity ratios calculated for the copolymerization of rac-LA and rac-EG were rLA =1 .29 and use = 0.71 (Figure 7). Table 1. Mole fractions of lactide in the feed (f1) and copolymer (F1) for the copolymerization of lactide and ethylglycolide. f1 F1 0.95 0.97 0.89 0.92 0.86 0.89 0.75 0.80 0.71 0.77 0.55 0.60 0.36 0.45 0.16 0.20 51 0.8 - 0.6 ~ 0.4 ~ 0.2 - O 1 l I I 0 0.2 0.4 0.6 0.8 1 ’1 Figure 7. Nonlinear least squares fit of the copolymerization data of rec-lactide and rec-ethylglycolide to equation 7. The reactivity ratios were determined to be rLA =1.29 and E6 = 0.71. The traditional methods for determining reactivity ratios were procedures that linearized the copolymerization data. For example, the Fineman-Ross method for analyzing copolymerization data expresses the copolymerization equation in terms of the mole fractions of the monomers, M1 and M2 in the feed (f) and the mole fraction of monomers, M1 and M2, in the copolymer (F1): 2 F F f—11(f1-1)-r1%-’2 (9) 52 "1(1 - 2F1) where r1 and r2 are the reactivity ratios of the two monomers.92 A plot of F (1 f) 1 - 1 #F—1 versus}—((1—1—f—)22- should yield a straight line with a slope of r1 and a y-intercept of 1 - 1 r2. An analysis of the copolymerization using the Fineman-Ross method is shown in Figure 8. A linear least squares fit to the data provides y = 1.4x + 0.82 (ILA = 1.4 and use = 0.82). f1(1-2F1) / ran-1'1) a -8 1 . -6 4 -2 0 qRHJHHUfiF Figure 8. Fineman-Ross plot of the rec-lactide and rac-ethylglycolide copolymerization data. A least squares linear fit to the data gave y = 1.4x + 0.82 (n_A = 1.4 and age. = 0.82). 53 Like the Fineman-Ross method, the Kelen-Tudos method also determines the reactivity ratios by linearization of the experimental data; however, it introduces an arbitrary constant 01 that distributes the experimental data symmetrically on the plot. The results are expressed in the following form: r r a a where n and E are functions of the parameters G and H: and g- H 11 a+H a+H ( ) n: _ F2 F——1(’; 1) ,=H ,and a is a constant that IS equal to (Hm max mem)}§ where G- 1 f11 with Hmax and Hmin being the highest and lowest H values, respectively.93 Plotting 11 versus l; generates a straight line that gives —r2/a and r1 on extrapolation to lg = 0 and E = 1, respectively. A Kelen-Tudos analysis for the copolymerization of rac- lactide and rec-ethylglycolide is shown in Figure 9. A linear least squares fit of the data and extrapolation to g = 0 and i; = 1 gave reactivity ratio values of rLA = 1.3 and r53 = 0.72. Notably, the results from the three analyses gave similar values and are summarized in Table 2. Although the F-R and K-T methods give a good first approximation of the reactivity ratios, they have two disadvantages. First, both methods use the initial feed composition instead of the instantaneous feed composition. If the reactivity’s of the two monomers are not identical, the more reactive monomer enters the copolymerization more rapidly than the less reactive 54 monomer, and thus the residual monomer mixture progressively changes its composition. The second disadvantage is the arbitrariness contained in the visual best-fit line that is drawn upon linearizing the experimental data. 1.5 0.2 0.4 0.6 0.8 1 Figure 9. Kelen-Tudos plot for the copolymerization of rec-lactide and rac- ethylglycolide. Extrapolation to E = 0 and E = 1 gave reactivity ratio values of rLA = 1.3 and E6 = 0.72. 55 Table 2. Reactivity ratios calculated for rec-lactide (r1) and disubstituted glycolides (r2)- Method a b b entry Monomers r1 r2 r1 r2 F-R 1.4 0.82 1.1 1 EG/LA K-T 1 .3 0.72 0.93 NLS 1.29 (x 0.07) 0.71 (:1: 0.05) 0.92 F-R 1.4 0.24 0.33 2 nHG/LA K-T 1.3 0.16 0.22 NLS 1.30 (:1: 0.03) 0.15 (:1: 0.01) 0.20 F-R 2.7 0.081 0.22 3 iPG/LA K-T 2.8 0.10 0.29 NLS 2.5 (1 0.2) 0.08 (:1: 0.01) 0.21 F-R 3.1 0.015 0.046 4 cHG/LA K-T 3.0 0.011 0.034 NLS 3.3 (:1: 0.1) 0.015 (:1: 0.006) 0.049 F-R 1.0 1.2 1.2 5 PL/LA K-T 1.0 1.2 1.2 NLS 1.05 (:1: 0.02) 1.22 (x 0.03) 1.2 (a) Fineman-Ross (F-R), Kelen-Tudés (K-T) and nonlinear least squares (NLS) methods. (b) Errors reported are standard deviations. The product of the two reactivity ratios gives information regarding the copolymer microstructure. For example, when the product of r1 r2 is close to unity, the copolymerization is termed “ideal” or “random.” Using the NLS method of analysis, the r1 r2 value for the copolymerization of lactide and ethylglycolide was 0.92, suggesting that both Iactide and ethylglycolide should be randomly distributed along the polymer backbone. If this is correct, DSC analysis of these copolymers should show only one glass transition temperature that increases with increasing lactide content (T9 (LA) = 66 °C, T9 (EG) = 15 °C). We synthesized a 56 series of copolymers (monomer ratios LA/EG: 5/1, 3/1, 1/3, 1/5) via ring opening polymerization catalyzed SnOct2 (Table 3).103 Thermal analysis of these copolymers revealed homogeneous materials with a single, well-defined T9 that systematically increased as lactide composition increased (Figure 10). Table 3. T93 of poly(Iactide-co-ethylglycolide) copolymers. Polymer Mn (kD)a PDI _Tg (°C)b poly( LA) 35.2 1 .89 66 poly(LA/EG) (5/1) 41.5 1.85 42 poly(LA/EG) (3/1) 47.8 1 .76 38 poly(LA/EG) (1/1) 41.7 1.88 30 poly(LA/EG) (1 13) 50.8 1 .92 23 poly(LA/EG) (1 /5) 43.6 1.85 19 poly(EG) 45.6 1 .78 15 (a) determined by GPC using polystyrene standards. (b) determined by DSC, heating rate 10 °C/min. 57 Polylactide ___/__ I EG/LA=1/5 .9 f— E (T) EG/LA=1/3 '5 8 C EG/LA=1 f."— 3 E EG/LA=3 [‘— 76 EG/LA=5 J‘— O) .C Poly(ethylglycolide) -50 0 50 100 Temperature ( °C) Figure 10. Glass transition temperatures of ethylglycolide (EG) and lactide (LA) copolymers as a function of polymer composition. DSC samples were run under a nitrogen atmosphere at a heating rate of 10 °C/min. Lactide and ethylglycolide are structurally similar, favoring the formation of random copolymers, however, the reactivity ratio for lactide was higher than ethylglycolide, presumably due to the steric difference between the two monomers. This trend should be more prominent as the alkyl chain is lengthened. We copolymerized lactide and n-hexylglycolide at 6 feed ratios (Table 4) and analyzed the data using to nonlinear least squares (Figure 11), Fineman-Ross (Figure 12), and Kelen-Tudos (Figure 13) methods. A summary of the kinetic data for the copolymerization of lactide and n-hexylglycolide is listed in Table 2. Again, the three methods are in good agreement, and as expected, the ILA and rnHG values differ considerably, ILA = 1.30 and rnHG = 0.15. 58 Table 4. Mole fraction of lactide in the feed (f1) and copolymer (F1) for the copolymerization of lactide and n-hexylglycolide. f1 F1 0.90 0.93 0.81 0.86 0.70 0.79 0.61 0.74 0.32 0.54 0.15 0.49 0.6 - 0.4 H 0.2 . O 1 0 0.2 0.4 0.6 0.8 1 f1 Figure 11. Nonlinear least squares fit of the copolymerization data of rec-lactide and rac-n-hexylglycolide to equation 7. The reactivity ratios were determined to be ILA =1 .30 and rnHG = 0.15. 59 f1(1-2F1) / F1(1-f1) .L -10 L 1 L -7 -5 -3 -1 1 f12(Fm-1)/ F1(1-f1)2 Figure 12. Fineman-Ross plot of the rec—lactide and rac-n-hexylglycolide copolymerization data. A least squares linear fit to the data gave y = 1.4x + 0.24 (ILA = 1.4 and rnHG = 0.24). 60 1.5 1.25 O 0.75 - 0.5 P 0.25 - -0.25 . J Figure 13. Kelen-Tudos plot for the copolymerization of rec-lactide and rac-n- hexylglycolide. Extrapolation to E = 0 and E = 1 gave reactivity ratio values of rLA = 1.3 and rnHG = 0.16. If n-alkyl chains decrease the reactivity of the disubstituted lactide monomers towards copolymerization, then branched chains should further decrease the reactivity, especially when the branch is close to the ring carbonyl. We copolymerized lactide and isopropylglycolide at seven feed ratios (Table 5) and analyzed the data according to nonlinear least squares (Figure 14), Fineman- Ross (Figure 15), and Kelen-Tudos (Figure 16) analyses. A summary of the copolymerization data for lactide and isopropylglycolide is listed in Table 2. Again, the difference between the reactivity ratios of LA and iPG increased (ILA = 2.5, mag = 0.08), suggesting a blocky or tapered architecture (r1 r2 =0.21). 61 Table 5. Mole fractions of lactide in the feed (f1) and copolymer (F1) for the copolymerization of lactide and isopropylglycolide. 1.2 f1 F1 0.78 0.91 0.67 0.87 0.59 0.83 0.35 0.65 0.31 0.63 0.19 0.55 0.10 0.43 Figure 14. Nonlinear least squares fit to the copolymerization data of rec-lactide and rac-isopr0pylglycolide to equation 7. The reactivity ratios were determined to be ILA = 2.5 and ripe. = 0.080. 62 0 - I g: -0.5 ‘7 '5:— -1 « ll. \ q: -1 5 ~ ll. N :‘L -2 ~ ..r -2.5 - -3 _ '3.5 L L '1-5 -1 -0.5 0 r1 2(1:1-1) / F1 (1-1'1)2 Flgure15. Fineman-Ross plot of the rec-lactide and rec-isopropylglycolide copolymerization data. A least squares linear fit to the data gave y = 2.7x + 0.081 (mg = 2.7 and an = 0.081). 63 1.6 ~ 1.2 ~ 0.8 I 0.4 ~ 0 0.2 0.4 0.6 0.8 E Figure 16. Kelen-Tudos plot for the copolymerization of rec-lactide and rac- i30propylglycolide. Extrapolation to E = 0 and E = 1 gave reactivity ratio values of ILA = 2.8 and an = 0.10. To determine whether this difference would manifest itself in the thermal properties of the copolymer, we prepared a series of lactide-co-isopropylglycolide copolymers (monomer ratios LA/IPG: 3/1, 1/1, 1/3).104 Again the copolymers only showed a single T9 and behaved as single-phase materials (Table 6). Despite the presence of only one T9, the value of r1 r2 (0.21) suggests that the polymer should be slightly blocky, although the blocks may not be sufficiently long to cause phase separation. 64 Table 6. T93 of poly(Iactide-co-isopropylglycolide) capolymers. Polymer Mn (kma PDI j! (0(3)b poly(LA) 35.4 1 .78 56 poly(LA/IPG) (3/1) 27.8 1 .82 48 poly(LA/IPG) (1/1) 25.7 1.76 48 poly(LAfiPG) (3/3) 28.2 1.90 43 poly(iPG) 32.6 1 .74 42 (a) determined by GPC using polystyrene standards. (b) determined by DSC, heating rate 10 °C/min. A bulky substituent situated close to the ring carbonyls should have the greatest effect on relative reactivity data. We copolymerized lactide and cyclohexylglycolide at six feed ratios (Table 7) and plotted the data according to nonlinear least squares (Figure 17), Fineman-Ross (Figure 18), and Kelen-Tudos (Figure 19) analyses. A summary of the copolymerization data for lactide and cyclohexylglycolide is listed in Table 2. All three methods confirmed the sluggish reactivity of cHG in the copolymerization with lactide (ILA = 3.3 and rcHG = 0.015, r1 r2 =0.049). 65 Table 7. Mole fraction of lactide in the feed (f1) and copolymer (F1) for the copolymerization of lactide and cyclohexylglycolide. f1 F1 0.91 0.97 0.86 0.95 0.70 0.89 0.55 0.83 0.32 0.72 0.15 0.59 1 0.8 - 0.6 - If: 0.4 l 0.2 - O l L n 1 0 0.2 0.4 0.6 0.8 1 ’1 Figure 17. Nonlinear least squares fit of the copolymerization data for rec-lactide and rec-cyclohexylglycolide to equation 7. The reactivity ratios were determined to be of rLA = 3.3 and rcHG = 0.015. 66 -1 _ 3:3“ =2 -3 - U. \ 1r -5 r ‘1‘ Z;- -7 - -9 _ '11 ' L 1 1 '4 -3 -2 -1 0 1 f12(F1-1) / F1 (143)2 Figure 18. Fineman-Ross plot of the rac-lactide and rec-cyclohexylglycolide copolymerization data. A least squares linear fit to the data gave y = 3.3x + 0.015 (ILA = 3.1 and rcHG = 0.015). 67 2.6 2.2 ~ 1.8 L 1.41 0.6 - 0.2 - Figure 19. Kelen-Tudos plot for the copolymerization of rec-lactide and rac- cyclohexylglycolide. Extrapolation to E = 0 and E = 1 gave reactivity ratio values of ILA = 3.0 and rcHG = 0.011. Since the electronic structures of alkyl glycolides are similar, we expect that reactivity should primarily depend on the steric constraints of the monomer. If true, there should be a relationship between the reactivity ratio and rate of homopolymerization. We compared the h0m0polymerization rates for a series of structurally similar monomers and found that the reactivity ratios reflected their polymerization rates. For example, the homopolymerization rates for lactide, ethylglycolide, n-hexylglycolide, and n-decylglycolide were 0.051, 0.048, 0.015, and 0.0062 L-s'1-mol", respectively (Table 8). Notably, as the length of the n—alkyl chain increased, both the homopolymerization rate and reactivity ratio for the copolymerization with lactide decreased. The trend suggests that comparison of 68 homopolymerization rates can be used as a rough estimate of reactivity ratios for electronically similar monomers. Furthermore, a comparison between isopropylglycolide, cyclohexylglycolide, and methylcyclohexylglycolide, a non- symmetrical AB monomer where a methyl group replaces one cyclohexyl groups, suggests that the rates of homopolymerization scale with the size of the ring substituents. The homopolymerization rate of Iactide was approximately 10x faster than isopropylglycolide and 360x faster than cyclohexylglycolide. Yet when compared with cHG, the polymerization rate of methylcyclohexylglycolide was ~100x faster, implying preferential ring opening at the carbonyl adjacent to the methyl group. Such differences in reactivity leads an alternating sequence of methyl and cyclohexyl groups situated along the polymer backbone.105 Although the polymerization rates scale with the reactivity ratios, the homopolymer Tgs do not (Table 8). The lack of correlation indicates that the interaction between polymer chains is significantly different than the ring-opening step during polymerization. 69 Table 8. Solution polymerization rates for the homopolymerization of substituted glycolides.a . Monomer kp_(L°S'1'm0|'1) x 103 [LAD r1012" 7;L(°C)° rec-lactide 0.051 -- -- 55 rec-ethylglycolide 0.048 1 .3 0.71 15 rac-n-hexylglycolide 0.015 1 .3 0.20 -37 rac-n-decylglycolide 0.0062 -- -- nd rec-isopropylglycolide 0.0053 2.5 0.08 56 rec—cyclohexylglycolide 0.00014 3.3 0.02 98 racmethylgclohexylglycolide 0.01 4 -- -- 73 (a) solution polymerization catalyzed by SnOct2 at 90 °C. (b) reactivity ratio for the copolymerization of rec-lactide (rm) and comonomer (rM2) as determined by nonlinear least squares analysis. (c) determined by DSC, heating rate 10 °C/min. nd indicates that the T9 was not detected (Tm of nDG = -18 °C). The most surprising monomer in this study was rec—phenyllactide, a benzyl substituted glycolide. Although the benzyl groups should increase the steric bulk around the ring carbonyls, they had little effect on reactivity. We copolymerized lactide and phenyllactide at seven feed ratios (Table 9) and plotted the data according to nonlinear least squares (Figure 20), Fineman-Ross (Figure 21), and Kelen-Tudos (Figure 22) analyses. A summary of the kinetic data for the copolymerization of lactide and phenyllactide is listed in Table 2. Nonlinear least squares analysis gave reactivity ratios of ILA = 1.05 and rpL = 1.22 and an r1 r2 value of 1.2. The results suggest that lactide and phenyllactide form random copolymers and phenyllactide may be slightly more reactive than lactide. 70 Table 9. Mole fractions of lactide in the feed (f1) and copolymer (F1) for the copolymerization of lactide and phenyllactide. f1 F1 0.90 0.87 0.85 0.82 0.70 0.67 0.56 0.51 0.33 0.30 0.16 0.14 0.8 ~ 0.6 - 0.4 l 0.2 H O 1 l l 0 0.2 0.4 0.6 0.8 1 f1 Figure 20. Nonlinear least squares fit of the copolymerization data of rec-lactide and rac-phenyllactide to equation 7. The reactivity ratios were determined to be rLA = 1.05 and rpL = 1.22. 71 {1(1-2F1) / F1(1-f1) 4; -14 -9 -4 1 f12(F1-1) / F1(1-f1)2 Figure 21. Fineman-Ross plot of the rec-lactide and rec-phenyllactide copolymerization data. A least squares linear fit to the data gave y = 1.0x + 0.12 (ILA=1.0 and rpL =12). 72 0.8 0.4 P o 0.4 0.8 1.2 E Figure 22. Kelen-Tudos plot for the copolymerization of rec-lactide and rac- phenyllactide. Extrapolation to E = 0 and E = 1 gave reactivity ratio values ILA = 1.0 and ”BL = 1.2. In addition, copolymers of lactide and mandelide, a phenyl-substituted glycolide, also form random copolymers. Mandelide should be even more sterically congested as the phenyl ring is closer to the carbonyls. However, copolymers of lactide and mandelide were homogeneous materials and DSC scans showed single TgS for each copolymer (Figure 23). The Tgs ranged from 48 °C to 100 °C, the Tgs of polylactide and polymandelide, respectively (Table 10).106 Furthermore, copolymerizations with L-lactide did not show any crystallinity, even up to 90 °/o L- lactide content, suggesting random placement of the mandelide within the polymer backbone. 73 Polymandelide (MD) I ”W .9 g MD/LA = 45/55 .5 *5 U C 3 E MID/My- :6 OJ .: MD/LAM— 0 25 50 75 100 125 T (°C) Figure 23. Glass transition temperatures of mandelide (MD) and lactide (LA) copolymers as a function of polymer composition. DSC samples were run under a nitrogen atmosphere at a heating rate of 10 °C/min. 74 Table 10. T93 of poly(Iactide-co-mandelide) copolymers. Polymer Mn (kma PDI Ia (°C)b poly(LA) 35.4 1 .78 56 poly(LA/MD) (9/1) 98 1 .65 61 poly(LA/MD) (75/25) 80 1.65 67 poly(LA/MD) (55/45) 58 1.47 82 poly(LA/MD) (25/75) 42 1.60 90 poly(LA/MD) (1/9) 37 1.45 95 pcflWD) 68 1 .63 100 (a) determined by GPC using polystyrene standards. (b) determined by DSC, heating rate 10 °C/min. The synthesis of functionalized lactide derivatives generated both racemic and meso isomers, and their separation is often challenging. To determine whether different isomers affect the copolymerization, we calculated the reactivity ratios for the copolymerization of a mixture of meso and rec-ethylglycolide with rac- lactide. From the analysis of the copolymerization data, we concluded that the mesa isomer may be slightly less reactive (rac-EG and rac-LA: ILA = 1.29, ’EG = 0.72, r1 r2 = 0.92; rac- and meso-EG with rac-LA: ILA = 1.48, res = 0.65, r1 r2 = 0.96), however, the difference was not significant (Figure 24).106 Therefore our design rules can be applied to any mixture of isomers within a homologous series and will be valuable when designing functionalized lactide monomers. 75 O O «Q 0 8 .0 . 0‘. 0 0.6 I 00 “F 2 0.4 I <9 0.2 - O 0 0 l L 4 A 0 0 2 0.4 0 6 0 8 1 Figure 24. Instantaneous copolymer composition (F1) as a function of monomer feed ratio (f1) for the copolymerization of rec-ethylglycolide and rec-lactide (r156I = 0.72, rLA = 1.29, r1 r2 = 0.92) (e) and copolymerization of rac- and meso- ethylglycolide with rec-lactide (’EG = 0.65, rLA = 1.48, r1 r2 = 0.96) (0). Conclusion In conclusion, the reactivity ratios of five disubstituted lactide derivatives were analyzed with respect to rec-lactide using three methods of analysis: the nonlinear least squares, Fineman-Ross, and Kelen-Tudos methods. The reactivity ratios provided information regarding the microstructure of the copolymers and related well to homopolymerization rates. The information allowed elucidation of structure-reactivity relationships that can be applied to the synthesis of future lactide derivatives. 76 Experimental Section Unless otherwise specified, ACS reagent grade starting materials and solvents were used as received from commercial suppliers without further purification. Rae-lactide was purchased from Aldrich and sublimed prior to use. 1H NMR (300 or 500 MHz) and 13c NMR (75 or 125 MHz) spectra were acquired using either a Varian Gemini 300 spectrometer or a Varian UnityPlus 500 spectrometer with the residual proton signal from the CDCI3 solvent used as chemical shift standard. Mass spectral analyses were carried out on a VG Trio-1 Benchtop GC-MS. IR spectra were taken with Mattson Galaxy 3000 FT-lR. Elemental Analyses were determined using a Perkin-Elmer 2400 CHNS/O Analyzer. Melting points were taken on an Electrothermal capillary melting point apparatus and are uncorrected. Differential Scanning Calorimetry (DSC) analyses of the polymers were obtained using a TA DSC 0100. Samples were run under a nitrogen atmosphere at a heating rate of 10 °C/min, with the temperature calibrated with an indium standard. Copolymerization data for the NLS method were fit to the copolymer equation (equation 8) using Origin (OriginLab Massachusetts, USA). Substituted Lactide Derivatives O A»... OH 2-Hydroxybutanolc acid (2). To a cooled (0 °C) solution of propanal (20.0 g, 340 mmol) dissolved in 20 mL THF and 400 mL saturated aqueous NaHSOall-I2O (3:5) 77 was added a solution of potassium cyanide (45.2 g, 690 mmol) in H20 (250 mL) dropwise over 1 hour. The reaction was stirred for 5.5 hours at 0 °C and then extracted with ether (4 x 200 mL). The combined organic layers were washed with saturated aqueous NaCl (2 x 50 mL), dried over M9804 and concentrated to give 2-hydroxybutanenitrile (19.1 g, 225 mmol, 65 %) as a colorless liquid, which was used without further purification. 1H NMR (300 MHz) 0 4.42, (t, 1H, J = 7.32 Hz), 1.87 (m, 2H), 1.08 (dd, 3H, J: 7.42, 7.69 Hz). Concentrated HCI (45 mL) was added to the cyanohydrin (19.1 g, 221 mmol) and the solution was stirred at room temperature for 18 hours and at reflux for 2.5 hours. Once cooled, the reaction was quenched with H20 (100 mL) and extracted with EtOAc (3 x 100 mL). The combined organic layers were washed with saturated aqueous NaCl (1 x 100 mL), dried over M9304, and concentrated. Purification by recrystallization (toluene at -30 °C) provided 2-hydroxybutanoic acid (16.7 g, 161 mmol, 63 %) as colorless crystals. 1H NMR (300 MHz) 6 4.24 (m, 1H), 2.0-1.8 (m, 1H), 1.8-1.6 (m, 1H), 0.99 (t, 3H, J: 7.32 Hz). ”C NMR (75 MHz) 6179.84, 71.19, 27.24, 8.89. mp = 40-41 °C. 0 OH 2-Hydroxy-3-methyl butyric acid (5). A solution of NaNO2 (72 g, 1.0 mol) in water (50 mL) was added dropwise to a cooled (0 °C) solution of DL-valine (30 g, 78 250 mmol) in 1 M H2SO4 (530 mL). The solution was stirred at 0 °C overnight and then extracted with ether (4 x 300 mL). The organic layers were dried over M9804 and the ether evaporated. Recrystallization from toluene yielded white crystals that were collected by filtration and dried under vacuum to give 2-hydroxy-3-methyl butyric acid (23 g, 195 mmol, 75 °/o) as a white solid. 1H NMR (300 MHz) 6 4.13 (d, 1H, J: 3.42 Hz), 1820 (m, 1H), 1.06-1.04 (d, 3H, J: 6.84), 0.92-0.90 (d, 3H, .1: 6.84 Hz). 13C NMR (75 MHz) 0 178.94, 74.78, 32.01, 18.76, 15.82. mp 169-170 °C. OH VWKn/OH o 2-Hydroxyoctanolc acid (8). A solution of 1-bromohexane (43 g, 260 mmol) in anhydrous THF (250 mL) was stirred with magnesium turnings (9.4 g, 390 mmol) at 0 °C. The resulting Grignard reagent was transferred to an addition funnel and the solution was added dropwise under nitrogen to a mechanically stirred solution of diethyl oxalate (29.5 g, 200 mmol) in anhydrous THF (100 mL) at -80 °C.‘°7 After stirring for 30 minutes at -80 °C, the reaction was quenched with 2 M HCI (200 mL). The aqueous layer was extracted with ether (3 x 300 mL) and the combined organic layers were washed with saturated aqueous NaCl (1 x 200 mL), dried over M9804, and the ether evaporated to give a light brown oil. The oil was poured into chilled methanol (300 mL, 0 °C), filtered, and the methanol was removed by rotary evaporation. Without further purification, a solution of the crude ketoester in acetic acid (250 mL) and 0.5 g of 5 % Pt/AI2O3 (Aldrich) was hydrogenated at 1500 psi 79 for 5 days at room temperature. The solution was filtered, poured into 500 mL ether and the acetic acid was removed by washing with water (3 x 200 mL). The ether was removed under reduced pressure and the colorless oil was hydrolyzed in a refluxing mixture of KOH in 70 % aqueous ethanol (15 wat %, 500 mL). After 2 days, the reaction was quenched with 2 M HCI and extracted with CH2CI2 (3 x 400 mL). The combined organic layers were dried over MgSO4 and the methylene chloride was evaporated. The crude a-hydroxy acid was recrystallized from hexanes to give 2-hydroxyoctanoic acid, isolated as white crystals (25.6 9.160 mmol, 60 %). 1H NMR (500 MHz) 0 4.27-4.23 (dd, 1H), 1.87-1.77 (m, 1H), 1.72- 1.62 (m, 1H), 1.49-1.36 (br, 2H), 1.35-1.21 (br, 6H), 0.94-0.77 (t, 3H, J: 6.84 Hz). 130 NMR (125 MHz) 6 180.04, 70.28, 34.09, 31.58, 28.88, 24.66, 22.52, 14.00. MS (El, m/z): 161.4 (M*). mp 67-69 °C (m. 70 0C).108 OH OH ' O 2-Cyclohexyl-2-hydroxyacetlc acid (11). The synthesis of 2-cyclohexyl-2- hydroxyacetic acid was carried out according to Jing and Baker.100 Mandelic acid (38.0 g, 250 mmol) was added to a mixture of MeOH (200 mL), acetic acid (2.5 mL), and 5 % Rh/C. The mixture was sealed in a Parr bomb, purged with nitrogen, and then pressurized to 1400 psi. After 8 hours, the reaction mixture was filtered, and evaporated to dryness. The resulting product was recrystallized from toluene to yield 2-cyclohexyl-2-hydroxyacetic acid (31.8 g, 203 mmol, 81 °/o) as colorless 80 crystals which were collected via vacuum filtration. 1H NMR (500 MHz, DMSO-ds) 8 3.71 (d, 1H, J: 4.4 Hz), 1.66-1.55 (m, 6H), 1.16-1.09 (m, 5H). ”C NMR (125 MHZ, DMSO-da) 6 175.27, 74.19, 41.15, 28.86, 26.74, 25.84, 25.76, 25.57. Anal. Calcd. for C3H1403: C, 60.74; H, 8.92. Found: C, 61.12; H, 9.28. mp 135-136 °C. General Procedure for the Synthesis of Symmetrical Substituted Lactides. A solution of 160 mmol of the appropriate a-hydroxy acid and p-toluenesulfonic acid (1.0 g) in toluene (1600 mL) was refluxed 3-6 days, removing the water azeotropically via a Dean-Stark trap. The toluene was evaporated and the solid was dissolved in ethyl acetate, washed with a saturated aqueous NaHCOs (3 x 100 mL), dried over MgSO4. The ethyl acetate was evaporated and the crude product was purified via recrystallization. Exceptions are noted below. 0 0 /~., 0 o /OIOL oioji/ Rec-3,6-Dlethyl-1,4-dioxane-2,5-dlone (3, EG).10 A solution of 2- hydroxybutanoic acid (10 g, 100 mmol) and p-toluenesulfonic acid (0.2 g) in toluene (700 mL) was refluxed 3 days. After concentrating the solution, 0.1 g ZnO was added and the residue was distilled under reduced pressure using a Kugelrohr distillation apparatus. Ethylglycolide was collected at 140 °C (3 mtorr) as a mixture of diastereomers (4.3 g, 25 mmol, 56 %). Rae-ethylglycolide was separated from mesa-ethylglycolide by flash chromatography on silica gel (1 :2 ether/hexanes eluent) to provide 1.2 g of rac-3,6-diethyl-1,4-dioxane-2,5-dione (7 mmol, 56 °/c). 81 1H NMR (300 MHz) 6 4.8 (dd, 1H, J: 7.32, 5.37 Hz), 4.8 (dd, 1H, J: 7.32, 4.88 Hz), 2.2-2.0 (m, 2H), 2.0-1.9 (m, 2H), 1.2-1.0 (dt, 3H, J: 7.32, 4.88 Hz). "‘6 NMR (125 MHz) 6 166.7, 76.5, 23.5, 8.83. Anal. Calcd. for C3H12O4: C, 55.81; H, 6.98. Found: C, 55.99; H, 6.98. MS (EI, m/z): 173.1 (M*). mp 51.5-52.5 °C. Jr ° J’I‘I’ 0 OK/ 0 o ”l/ Rec-3,6-DiisopropyI-1,4-dloxane-2,5-dlone (6, iPG).1o Rae-isopropylglycolide was recrystallized from toluene three times removing the mesa isomer to provide rac-3,6-diisopropyl-1,4-dioxane-2,5-dione (1.5 g, 7.5 mmol, 16 % yield from 10.7 g rac-2-hydroxy-3-methylbutyric acid). 1H NMR (500 MHz) 6 4.68 (2, 2H, J: 3.17 Hz), 2.46-2.57 (m, 2H), 1.15-1.13 (d, 6H, J = 6.96 Hz), 1.05-1.04 (d, 6H, J = 6.96 Hz). ”C NMR (125 MHz) 8 166.3, 79.6, 29.4, 18.6. Anal. Calcd. for C1oH1eO4: c, 59.98; H, 8.05. FoundzC, 59.22;H, 7.60. MS (El, m/z):200.1 (M*). mp137.5- 138.0 °C. 0 0 /W 0 O WOLW OLOIINV Rec-3,6-DlhexyI-1,4-dioxane-2,5-dlone (9, nHG).‘° n-Hexylglycolide was recrystallized from hexanes and isolated as white crystals to provide 3,6-dihexyl- 1,4-dioxane-2,5-dione (5.9 g, 22.3 mmol, 38 °/o). 1H NMR (500 MHz) 8 4.88 (dd, J: 8.05 Hz and J = 4.88 Hz) and 4.83 (dd, J = 7.81 Hz and J = 4.39 Hz) (2H total for the signals at 4.88 and 4.83), 2.10-1.97 (m, 2H), 1.97-1.88 (m, 2H), 1.56-1.40 (br 82 m, 4H), 1.36-1.22 (br m, 12H), 0.91-0.75 (t, 6H J: 6.84 Hz). ”C NMR (125 MHz) 8 166.99, 165.84, 76.36, 75.56, 31.89, 31.41, 30.36, 30.06, 28.69, 28.52, 24.45, 24.28, 22.44, 22.41, 13.96, 13.93. MS (EI, m/z): 285.2 (M*). mp 77-80 °c (lit. 78-80 °C).“) Rac-G,6-chyclohexyI-1,4-dloxane-2,5-dlone (12, cHG).‘°° Rac- cyclohexylglycolide was recrystallized from toluene and then cyclohexane. The white crystals were collected by filtration and dried under vacuum to give rec-3,6- dicyclohexyl-1,4-dioxane-2,5-dione (4.8 g, 17.1 mmol, 21 %). 1H NMR (500 MHz, DMSO-ds) 8 4.66 (d, 2H, J: 2.9 Hz), 2.20-2.12 (m, 2H), 1.86-1.64 (m, 10 H), 1.54- 1.43 (dq, 2H, J: 3.5, 12.8 Hz), 1.37-1.24 (m, 6H), 1.22-1.11 (m, 2H). 13C NMR (125 MHz, DMSO-da) 6 166.24, 79.33, 38.82, 28.94, 26.11, 25.75, 25.73 (peaks at 25.75 and 25.73 ppm were resolved to three separate peaks by adding NMR shift reagent europium(lll) tris(3-trifluoromethylhydroxymethylene)-d-camphorate)). MS (El, m/z)198.1,(M“). mp 183-185 °C. 0 o O o (DEVICE CEIIQ 3,6-Diphenyl-1,4-dloxane-2.5-dlone (14, PL).16 Phenyllactide was synthesized according to Simmons and Baker; except, DL-phenyllactic acid was used instead of L-phenyllactic acid to generate rec-phenyllactide. After recrystallization from 83 ethyl acetate and hexanes the colorless crystals were collected by vacuum filtration to provide phenyllactide (2.48 g from 10 g phenyllactic acid, 8.5 mmol, 31 %) as a mixture of diastereomers. Rec-phenyllactide was isolated from meso- phenyllactide by recrystallization from ethyl acetate and hexanes to provide rac- 3,6-diphenyl-1,4-dioxane—2,5-dione (700 mg, 2.4 mmol, 49 % from 1.42 g rac- and meso- mixture): 1H NMR (500 MHz) 8 7.31 -7.23 (m, 10H), 5.03-4.99 (dd, 2H, J: 3.85, 7.97 Hz), 3.37-3.31 (dd, 2H, J: 3.85, 14.83 Hz), 3.01-2.93 (dd, 2H, J: 8.24, 14.83 Hz). 13C NMR (125 MHz) 8 165.4, 134.6, 129.7, 128.7, 127.5, 36,7. Meso- 3,6—diphenyI-1,4-dioxane—2,5—dione: 1H NMR (500 MHz) 8 7.2-7.1 (m, 10H), 4.2 (t, 2H, J: 4.95 Hz), 3.2 (d, 4H, J: 4.95 Hz), 13C NMR (125 MHz) 6 164.8, 133.7, 129.9, 129.0, 127.9, 38.4. MS (EI, m/z) 296, (M") mp 165 °C. General Procedure for the Copolymerization of rac-Dlsubstltuted Lactides with rec-Lactide. Monomers and a small magnetic stir bar were added to ampoules prepared from 3/8 in. diameter glass tubing. The ampoule was connected via a Cajon® fitting to a T-shaped vacuum adapter fitted with a stopcock and an air-free Teflon valve. The apparatus was attached to a vacuum line and evacuated through the Teflon valve. The ampoule was backfilled with argon, and a syringe was used to add predetermined amounts of the Sn(2-ethylhexanoate)2 and 4-tert-butylbenzyl alcohol solutions (both 38.1 mM in toluene) ([monomer]:[catalyst]:[initiator] = 400:1 :1) to the ampoule through the stopcock. After removing solvent in vacuo, the ampoule was flame-sealed and immersed in a preheated oil bath for 45 seconds (timing started once the monomers were 84 melted). The polymerization was quenched in ice at low conversion (<10 °/c), opened, and the monomer/copolymer mixture dissolved in CH2CI2. General Procedure for the Determination of Copolymer Composition. The monomer/copolymer CH2CI2 solution was passed through a short column of silica gel prepared from a 9-inch long Pasteur pipette to separate the polymer from the monomers. The eluent was collected in 20 mL vials, evaporated to dryness, and analyzed to confirm that the recovered material was free from polymer. After the separation was complete, the copolymer absorbed onto the silica was transferred into a 25 mL round bottom flask containing 20 mL MeOH. The mixture was stirred for 4 hours and then centrifuged 30 minutes. The colorless solution was decanted, evaporated to dryness, and the composition of the resulting copolymers was analyzed by 1H NMR using a VarianUnity Plus-500 Spectrometer at 500 MHz. The molar fraction of rac-LA in the copolymer (F1) was calculated according to equation 7 (see text): I F1'I ”I (7) LA+IM2 where lLA and IM2 are the total integration of the methine protons of LA and the comonomer, respectively. Representative 500 MHz 1H NMR spectra are available in the appendix. 85 Determination of F1 via 500 MHz ‘H NMR Copolymerization of ethylglycolide and lactide: Ethylglycolide and lactide were polymerized at 130 °C. The integration of the overlapping methine protons of EG and LA (5.1 -4.9 ppm and 5.3-5.1 ppm, respectively) was resolved by deconvolution using Varian software. Copolymerization of isopropylglycolide and lactide: Isoproplyglycolide and lactide were polymerized at 165 °C. The methine protons of iPG (5.0 ppm and 2.4-2.3 ppm) and the methine protons of LA (5.3-5.1 ppm) were used to determine F1. Copolymerization of n-hexvlglvcolide and lactide: n-Hexylglycolide and lactide were polymerized at 130 °C. The integration of the overlapping methine protons of nHG and LA were resolved using the 3:2:1 ratio between the methyl protons (CH3, 191.8 ppm), methylene protons (CH2, 2.0-1.7 ppm), and methine protons of nHG (see text). Once the number of methine protons of nHG was determined using the integration ratio of the methyl and methylene protons, the value was subtracted from the total methine proton integration (5.0-5.3 ppm), giving the integration of the methine protons of LA. Copolymerization of cyglohexylglycollde and lggtide: Cyclohexylglycolide and lactide were polymerized at 185 °C. The methine protons of cHG (5.0-4.9 ppm and 2.0-1.9 ppm) and the methine protons of LA (5.3-5.1 ppm) were used to determine F1. 86 Copolymerization of phenyllactide and lactide: Phenyllactide and lactide were polymerized at 185 °C. The integration of the overlapping methine protons of PL and LA were resolved using the 2:1 ratio between the methylene protons (CH2, 2.7-3.4 ppm) and methine protons of PL. Once the number of methine protons of PL was determined using the integration of methylene protons the value was subtracted from the total methine proton integration (5.4-5.0 ppm), giving the integration of the methine protons of LA. 87 Chapter 3 Branched Poly(lactide)s Introduction Linear poly(Iactide) has many pharmaceutical and biomedical applications including orthopedic fixtures,“'5 sutures,6 tissue engineering,7 and drug delivery.109 However, PLA’s hydrophobicity and lack of functional groups limit its utility. Functional polylactides, i.e. polylactides that contain pendant functional groups that permit further chemical elaboration, increase the versatility of conventional polylactide.“°'111 Several examples of functional polylactides have been reported 18,23,112 where functional groups were used to enhance hydrophilicity, provide 58"” and serve as branching attachment sites for biologically relevant molecules, sites in the synthesis of complex polymer architectures.‘*9'50 Since functional groups such as pendent amines, carboxylic acids, and alcohols often interfere with lactide polymerizations, they are usually introduced in a protected form. For example, groups have reported the synthesis and polymerization of side-chain 25-27,114,115 and polyester protected L-amino acid morpholine-2,5—dione monomers derivatives‘7‘19'75"‘2'115 to provide functionalized biomaterials. Most of these examples use serine, lysine, and other amino groups to introduce the functional groups, placing the group in close proximity to the polymer backbone, which may complicate deprotection and limit post-polymerization modifications. Complex polylactide architectures can be accessed using pendent functional groups to add branches, either by coupling reactions or initiation of d ’40-42 lactide polymerization. Hyperbranche star-shaped,“""45 and dendritic4648 88 polymers are more compact than linear polymers and have useful rheological and mechanical properties. In general, long-chain branches decrease viscosity and increase elasticity, whereas short-chain branches affect crystallinity. These properties provide control over degradation rates and drug release kinetics.“9'52 Branched architectures also provide multiple functional groups on polymer surfaces. With a goal of simplifying deprotection and chemical elaboration of functional groups in polylactides, we synthesized 3-benzyloxyoctyl glycolide, BOG (1), an AB glycolide monomer derived from oleic acid, an inexpensive and renewable resource found in soybean oil. Moving the functional group farther from the polymer backbone allowed facile deprotection of benzyl-protected primary alcohols, as exemplified by the initiation of L-lactide polymerization from the sites to generate PLA comb polymers. This chapter describes the preparation, characterization, and derivatization of alcohol-functionalized polylactides. Results and Discussion Monomer Synthesis The synthesis of 3-(8-(benzyloxy)octyl)-1,4-dioxane-2,5-dione (BOG) from oleic acid is shown in Scheme 32. This approach places a benzyl-protected primary alcohol eight methylene units from the polymer backbone, which should minimize any steric effects associated with the polymer backbone during chemical transformations at the pendent hydroxy group. Ironically, in this synthesis the carboxylic acid of the a-hydroxy acid is derived from the oleic acid double bond, 89 while the primary alcohol is generated by reduction of the oleic acid carboxylic acid. OH 1. 03, -78 0C 2. M828 H300 OCH3 O 3 MeOH ' ' ' OCH3 O TsOH — 99% 0'9“: Add 1. LiAIH4,THF,0°C, 2. NaOH, H20 95% H3COWOV© : NaH HSCOWOH OCH3 4 gm OCH3 3 98% AcOH/HzO 98% H\n/\/W\/O Op ——> NNCW/OQ NaHSO3, H20 0 6 97% HCI then NaOH 94% Oi:fimmom© ;B'§:’BrH 0W0 0\/© 3'(8'(benZY'0XY)Octyl)-1,4-dioxane-2,5—dione 2 Et3':H8:°e‘°"° O 1 Scheme 32. Synthesis of the AB glycolide 3-(8-(benzyloxy)octyl)-1,4-dioxane-2,5- dione. Ozonolysis of oleic acid in a 1:4 mixture of MeOH/CH2CI2 at —78 °C followed by a reductive workup with dimethyl sulfide and aldehyde protection provided 1,1-dimethoxynonane and methyl 9,9-dimethoxynonanoate (2), which were separated via vacuum distillation. LiAlH4 reduction of ester 2 provided 9,9- dimethoxynonanol (3) as a colorless liquid; Soxhlet extraction of 9,9- 90 dimethoxynonanol from the aluminum salts increased the yield to 95 %. The primary alcohol was benzyl-protected to prevent it from acting as a competing polymerization initiator.33'116 Treating acetal 4 with acetic acid deprotected the aldehyde, and reaction with KCN gave 10-(benzyloxy)~2-hydroxydecanenitrile (6). Acidic hydrolysis of the cyanohydrin gave an amide intermediate that was converted to ct-hydroxy acid 7 by basic hydrolysis. Finally, we used a previously I17 to condense the ct-hydroxy acid with bromoacetyl bromide to reported protoco form the unsymmetrical glycolide. The monomer was purified via flash chromatography using ethyl acetate as the eluent and crystallized from ether at 0 °C to give BOG in 38 % yield. The 1H NMR spectrum of BOG is shown in Figure 25. The methine and methylene protons of the substituted glycolide appear at 4.7 ppm and 4.8 ppm, respectively. 91 3 5 2 \ l 1 A 1% #_2 AL 2and6 4\ 3 5 \ . \U 1 B J L 22 Lz F3 T i ' T H 4 ' H l T ' H H l . r r r l ’1 7 6 5 4 3 2 ppm Figure 25. 500 MHz 1H NMR of BOG (A) and poly(BOG-co-LLA) (B). Bulk Polymerization Glycolide and lactide derivatives can be polymerized via ring opening polymerization in bulk or solution polymerization. Bulk polymerizations, or solvent- free polymerizations, are usually run at a temperature slightly higher than the melting point of the monomer whereas solution polymerizations are run at lower temperatures with the monomer dissolved in an organic solvent. We initially investigated bulk polymerization since solution polymerization rates for substituted glycolides are often slow. Melt polymerizations were run in sealed ampoules at 130 °C using SnOct2 as the catalyst and BBA as the initiator. High molecular weight polymers were obtained in ~45-60 minutes. The polymers were purified by 92 precipitation from cold methanol and analyzed by 1H NMR, 13C NMR, GPC, and DSC. It is interesting to note the effect of the pendant carbon chains on the T9 of poly(BOG) and poly(BOG-co-LLA). Long pendent groups decrease the dipole- dipole interactions between the ester groups in the polymer backbone and degrease the glass transition temperature.10 Accordingly, the T9 of BOG-L-lactide copolymers decreased as the mole fraction of BOG in c0polymers increased, reaching a limiting T9 of -32 °C for the BOG homopolymer. In addition, all copolymers were amorphous, suggesting that BOG is distributed along the polymer backbone. The T9 of poly(BOG) (-32 °C) is relatively low when compared to poly(3-benzyloxymethyl-1,4-dioxane-2,5-dione), which has only a two carbon tether (34 °C, Chapter 1, Scheme (5).17 93 Table 11. GPC and DSC characterization of BOG homo- and co-polymers before and after benzyl ether deprotection.‘ poly(BOG-co-LLA) poly(HOG-co-LLA) -3 M x10'3 Mnx10'3 T ent m I%b c n M" x 10 PDI T9 " PDI 9 W 0 0 V (g/mol)° (°C)d (9/m0')c (glmo|)° (°C)d theor. actual 1 100 45 °/o 22.9 1.2 -32 16.6 14.2 1.3 -5 2 50 95 °/o 20.6 1.1 -15 17.3 17.4 1.1 -3 3 20 93 °/o 51.0 1.3 26 45.8 42.2 1.4 39 4 10 96 °/o 36.0 1.2 35 34.0 33.7 1.3 45 5 5 97 % 32.6 1.2 46 31.6 31.5 1.2 54 (a) Polymerization Conditions: 130 °C, [SnOct2]:[BBA] = 1:1, 45 minutes. (b) mol % BOG in BOG-co-LLA copolymer; determined by NMR (c) measured by GPC using poly(styrene) standards. ((1) Determined by DSC, second heating scan; heating rate of 10 °C/min. Solution Polymerization Solution polymerizations are more controlled than bulk polymerization due to lower polymerization temperatures and more efficient mixing during polymerization. The ring-opening polymerization of lactides and lactones follow first order kinetics, which are expressed by equation 12 R=_d[M] at - lellll (12) where [M] and [I] are the concentration of the monomer and initiator and kp is the rate constant for propagation. In living polymerizations, [I] is constant and integration of equation 12 gives equation 13 -Ir( [Mk )= kpn10t (13) [M10 94 where [M]t is the concentration of the monomer at time t; [M]o is the initial monomer concentration at t = 0, and [Ho is the concentration of the initiator. According to equation 13, a plot of -ln([M]t/[M]o) versus time is linear for a polymerization with first order kinetics. We briefly examined the solution polymerization kinetics of BOG in toluene at 90 °C using SnOct2 and BBA as the catalyst and initiator, respectively. The polymerizations were relatively slow because of the 8-carbon tether, reaching 90 °/o conversion in 48 h. Figure 26 shows the evolution of molecular weight with conversion for the polymerization of BOG. The data show a linear relationship between Mn and conversion, although the data do not extrapolate to zero as expected. The polymerization of BOG is 1““t order in monomer concentration and initiator at low conversion; however, the polymerization rate drops at high conversion, never reaching 100 percent conversion (Figure 27). 95 —-L A 12 _ a} A / g 10 - /// E: 8 - /§$ ‘2; ,HK x 6 ’ // E: :3. 4 . 2 . 0 . ' . H . 0 20 40 60 80 100 conversion 0%) Figure 26. Molecular weight versus conversion for the polymerization of BOG. Solution polymerization conditions: 90 °C, [SnOct2]:[BBA]:[monomer] = 1:1:100. Mn determined via GPO using poly(styrene) standards. Each point corresponds to an average of three samples and the error bars represent the standard deviation. 96 '5 2' % E T. E -'.= 1- “,9 {13¢ 0 6 Oé l 1 1 I 0 1O 20 30 4O 50 time (hr) Figure 27. Solution polymerization of BOG with SnOct2 at 90 °C. All reactions were carried out in a 0.2 M BOG solution in toluene with a [SnOct2]:[BBA]: [monomer] ratio of 1 :1 :100. Each point corresponds to an average of three samples and the error bars represent the standard deviation. Failure to reach full conversion in Iactide polymerizations is often due to an equilibrium between polymerization and depolymerization. In these cases, lactide polymerization kinetics must be analyzed as a series of equilibrium reactions and account for the equilibrium monomer concentration ([Meq]).‘°'“6'118 Incorporating [Meq] into equation 13 gives equation 14, which describes the effect of the equilibrium between polymerization and depolymerization on the observed first order kinetics. 97 [Mlt '- [Mleq -| —— = k l 14 {WHO _ [Mleq] obs“ ( ) In equation 14 [Mleq is the monomer concentration at equilibrium, [M10 is the initial monomer concentration, kobs is the observed propagation rate, and [I] is the initiator concentration. The data of Figure 27 were plotted according to equation 1 4 and Figure 28 shows that a 0.2 M polymerization of BOG followed first order kinetics, assuming an equilibrium monomer concentration ([Mleq) 0.02 M, comparible with the [Meq] seen in other substituted glycolides.”16 Also, the polydispersity remained narrow throughout the polymerization, ~1.2-1.3, consistent with minimal transesterification during the entire 50 hour polymerization. 98 E] / / s: / 4 _ SS // . 2.5 e. 3- \ .. '23 d 2 E r— / E 2 ' // lo // E. ‘ FF - 15 .‘e’ 1 ' / 2F” 0 0 O 1 1 J 1 1 0 10 20 30 40 50 time(hr) Figure 28. Solution polymerization kinetics of BOG with SnOct2 at 90 °C. All reactions were carried out at 0.2 M BOG solution in toluene with a [SnOct2]: [BBA]:[monomer] ratio of 1:1:100. The equilibrium monomer concentration, [M]eq, is assumed to be 0.020 M. The line is a least squares fit to the data. Copolymerization One advantage of copolymerizations is the large number of materials that can be generated from few starting monomers. As mentioned in chapter 2, monomer structure affects the relative reactivity of the two monomers, which in turn determines the copolymer architecture, ranging from random or alternate to blocky structures. In a random copolymerization, the growing chain shows no preference between monomers and a homogeneous material is formed. In contrast, block copolymers are formed when one monomer preferentially polymerizes over the 99 other, generating multi-phase materials where each block has the properties of its corresponding homopolymer.102 When n-hexylglycolide is copolymerized with Iactide, the polymer tends to be blocky (r1 = 1.30, r2 = 0.15). Since the polymerization of a symmetrical disubstituted 3-benzyloxyoctyl glycolide monomer should result in a more highly blocky architecture, we synthesized an AB glycolide to increase the reactivity of BOG relative to lactide and provide access to random copolymers. Copolymers of BOG and L-Iactide were synthesized in bulk using SnOct2 and BBA as the catalyst and initiator, respectively (Scheme 33). The results are summarized in Table 11 (entries 2-4). The crude polymers were purified by precipitation into cold methanol to remove residual monomer, washed with dilute hydrochloric acid to remove catalyst impurities, and dried under vacuum. The copolymers were colorless solids with molecular weights ranging from 15 to 60 kD. 100 3..., Scheme 33. Synthesis of BOG-co-LLA copolymers. Polymerizations were run at 130 °C (bulk) or 90 °C (toluene solution) using SnOct2 and BBA as catalyst and initiator, respectively. Physical Properties of poly(BOG-co-LLA) We measured the thermal properties of various poly(BOG-co-LLA) copolymers using differential scanning calorimetry. DSC analysis showed that increasing the mol % of BOG in the copolymer decreased the T9 (Table 11). Furthermore, the lack of crystallinity suggested random incorporation of the comonomers. A nonrandom distribution would result in crystalline poly(LLA) segments, whose melting temperature would be detectable by DSC.19 Benzyl Deprotection Post-polymerization modification strategies require complete and efficient deprotection of pendant functional groups. We examined a variety of solvents and conditions for removal of the benzyl groups from poly(BOG) and poly(BOG-co-LLA) 101 copolymers and found that hydrogenolysis using 10 % Pd/C in MeOH/l’ HF effected quantitative debenzylation with no detectable polymer degradation to give poly(3- hydroxyoxyoctyl glycolide) (poly(HOG)) and poly(HOG-co-LLA). In a typical deprotection reaction, we dissolved the polymer in CH20I2 (~ 0.5 mL) and added it to a solution of MeOHfT HF (1 :1 vol °/o) containing 10 % Pd/C. The polymer was subjected to 1000-1500 psi hydrogen gas for 2-3 days. After filtration to remove the catalyst, the deprotected polymers were dissolved in CH2CI2 and precipitated from cold MeOH (copolymers) or dissolved in EtOH and precipitated from cold hexanes (homopolymer). Polymer samples were analyzed by 1H NMR, "’0 NMR, GPC, and DSC. The absence of peaks at 72-74 and 4.5 ppm in the 1H NMR spectrum shows quantitative removal of the benzyl ether protecting groups (Figure 29). This was confirmed by 13C NMR. The success of this strategy was previously demonstrated by the quantitative removal of the benzyl groups from the amphiphilic PEG/alkyl-grafted polymers synthesized in our lab.98 GPC analysis of the deprotected polymers showed minimal molecular weight degradation and changes in the PDl (Table 11). For example, the measured molecular weight of a deprotected poly(HOG-co-LLA) (5 % HOG) was 31.5 kg/mol, closely matching the theoretical molecular weight of 31.6 kg/mol (Table 11, entry 5). Branched polymers are more compact than linear polymers the same molecular weight and therefore have different sizes. When measured against linear polystyrene standards, the more branched copolymers (high BOG content) showed an apparent decrease in molecular weight upon deprotection; 102 degradation of the polylactide backbone was ruled out by control experiment that was run using linear polylactide. Thermal analysis showed an increased T9 of the material upon benzyl deprotection, as expected for hydrogen bonding between alcohol groups of the deprotected polymer. Poly(HOG) was completely soluble in methanol and swelled in aqueous solution. 95 O OH 3 3 A B O O 0 1,1' 8 L L: i JLJ 5 1 4 ‘ 3 -. A , _. /U\ , _ __L/ Kiwi- -______. .JL _ A...) k/FKW _ _____.._L {—f j f Y fii Y r T r f7 fifi T T V T Y I T l r r Y r Y 1 ‘r Y Y Y Y T Y Y ' T 1’ 8 7 6 5 4 3 2 1 ppm Figure 29. 1H NMR spectrum for poly(BOG-co-LLA) (5 mol% BOG) (top) and poly(HOG-co-LLA) (bottom) taken in CDCI3 at 500 MHz. Absence of peaks at 7.2- 7.4 ppm in the spectra indicates complete removal of the benzyl ether protecting groups. Degradation of poly(HOG-co-LLA) Crystalline poly(L-lactide) may take years to degrade by hydrolysis at body temperatures, however, drug delivery and other in vivo applications require a more rapid and controlled degradation. Introducing pendent alcohol groups to PLLA by 103 using BOG as a comonomer should lower the polymer Tg, suppress crystallization, and increase the polymer’s hydrophilicity, all of which should accelerate the hydrolytic degradation rate. However, the hydrophobic 8-carbon tether should slow degradation. We compared the accelerated degradation rates of poly(rac-lactide), PLLA and poly(HOG-co-LLA) at 55 °C in pH 7.4 phosphate buffered saline (PBS) solution. Although the temperature of the experiment was above the T9 for all three samples, PLLA showed no appreciable degradation, while both poly(rac-lactide) and poly(HOG-co-LLA) degraded within 25 days (Figure 30). The degradation rate of poly(HOG-co-LLA) was similar to poly(rac-lactide) suggesting that the alcohol groups compensate for the hydrophobicity of the pendent alkyl chains. 104 0.07 D 0.06 0.05 - / 0.04 F 0.03 - x 0.02 - 6335? 0.01 0 10 20 30 time (days) (Mn(0) / Mn(t) - 1) / Pn(0) \ b \\ Figure 30. Normalized molecular weight decrease for poly(HOG-co-LLA) (50 mol % HOG) (A), poly(rao-LA) (El) and PLLA (0) during hydrolytic degradation at 55 °C in pH = 7.4 phosphate buffer saline solution. Each point corresponds to the average of 3 samples and the line is a least squares fit to the poly(HOG-co-LLA) data. Graft Polymerizations Branched polyesters are popular drug delivery vehicles because the crystallinity, T9, and degradability can be controlled by varying the number and length of polymer branches.49 To demonstrate the accessibility of the pendant alcohol groups, we polymerized L-lactide branches from poly(HOG-co-LLA) using SnOct2 as the catalyst. The branch lengths ranged from 5-60 lactide units, and representative polymerization and characterization data are shown in Table 12. 1H NMR confirmed initiation from all alcohol groups; the 1H NMR signal at 3.6 ppm 105 from the protons adjacent to the alcohol in poly(HOG-co-LLA) shifted completely to 4.1 ppm (Figure 31). Since GPC analysis using poly(styrene) standards underestimates the molecular weight of branched polymers due to their smaller hydrodynamic volume, we used GPC with multiangle laser light scattering (GPC- MALLS) to determine absolute molecular weights. Representative polymerization data is given in Table 12. Table 12. Macroinitiation of L-lactide from poly(HOG-co-LLA) containing 5 mol % HOG.“ polymer DPb Mn(10'3)° Mw(10'3)° PDI lnl (mL/g)d a° Tg(°C)' 5 0 31.6 36.0 1.19 _ 0.74 46 5a 4.9 39.3 78.3 1.99 22.4 0.45 45 5b 15 55.2 113 2.05 25.5 0.42 47 5c 30 79.8 174 2.18 36.3 0.45 51 5d 41 96.4 226 2.34 37.8 0.39 49 5e 63 131 255 1.95 40.6 0.38 51 (a). Bulk polymerization of L-lactide from poly(HOG-co-LLA (5 mol% HOG) at 130 °C catalyzed by SnOct2, [SnOct2]:[alcohol initiator] = 1 :1. (b) Degree of polymerization of lactide side chains determined by GPC-MALLS. (c) Determined by GPC-MALLS. (d) Determined by viscometry. (e) Mark-Houwink a value determined by viscosity measurements. (f) Determined by DSC second heating scans, at a heating rate of 10 °C/min. 106 A onflj #WA A 1,1“ 4 2 3 B r—fi r....r....,14..,....,....,...4,....,..TTT.4.. 5.0 4.5 4.0 3.5 3.0 2.5 2.0 1.5 ppm Figure 31. 500 MHz 1H NMR spectrum of poly(HOG-co-LLA) (5 mol% HOG) (top) and poly((HOG-co-LLA)-g-LLA) (bottom) taken in CDCI3. Branching in poly((HOG-co-LLA)-g-LLA) also was confirmed by analyzing the light scattering and viscosity data from GPC-MALLS. Linear and branched polymers of the same molecular weight can be discriminated by their size; branched polymers are more compact than linear polymers and their different size/molecular weight relationships can be probed by intrinsic viscosity measurements and by evaluating the relationship between the root-mean squared radius of gyration and polymer molecular weight. According to equation 15, a log- log plot of the radius of gyration and polymer molecular weight (M.) allows 107 evaluation of the slope a, which contains information concerning the molecular architecture of the polymer. 5‘ logri=k+alog Mi ' (15) Branched polymers have smaller a values than random coils due to their compact structure. Figure 32 shows the rms radius of gyration extracted from the light scattering data of two GPC runs, linear PLA and poly((HOG-co-LLA)-g-LLA) (HOG = 5 %, LLA = 95 %; branch DP = 60). The data for PLA yielded a = 0.81, a typical value for extended random coil, and a = 0.24 for poly((HOG-co-LLA)-g-LLA) consistent with a compact, branched structure. 50 - 40 I 309 20- rms radius (nm) 10 1 [1111111 I Llllllll I [11111 104 105 106 107 molar mass (g/mol) Figure 32. Double logarithmic plot of rms radius of gyration versus molecular mass of linear PLA (broken red line) and poly((HOG-co-LLA)-g-LLA) (solid blue line) in THF at 35 °C. The smaller slope for poly((HOG-co-LLA)-g-LLA), a = 0.24, compared to linear PLA, .3 = 0.81, is consistent with the branched architecture of poly((HOG-co-LLA)-g-LLA). The HOG content in poly((HOG-co-LLA)-g-LLA) sample was 5 mol %, and the degree of polymerization of the LLA branches was 60. 108 We also confirmed the molecular architecture using the Mark-Houwink relationship, in] - KM? (16) where [n] is the intrinsic viscosity, M, is the viscosity average molecular weight, and K and a are constants for a particular polymer-solvent pair at a defined temperature. Branched polymers are more compact and have smaller a values than linear polymers of comparable molecular weight.119 Figure 33 shows intrinsic viscosity data obtained for the two polymers of Figure 32. The log-log plot of the data shows a lower a value (slope of log-log plot) expected for a branched polymer. Data for the series of poly((HOG-co-LLA)-g-LLA) polymers are shown in Table 12, which show a steady decrease in a as the polymers transform from linear polymers to a comb architecture with long branches. 109 200 _ 3 :1 100» E 80 - S .. 60 - 40 ~ 20 *- l llllllll L I lLlllLl J lllllll 104 105 106 107 molar mass (g/mol) Figure 33. Mark-Houwink plots for linear (broken red line) and branched (solid blue line) PLA in THF at 35 °C using light scattering and dilute-solution viscometry data, extracted from GPC-MALLS. The smaller slope for poly((HOG-co-LLA)-g-LLA) is consistent with the branched architecture of poly((HOG-co-LLA)-g-LLA). The HOG content in poly((HOG-co-LLA)-g-LLA) was 5 mol%, and the degree of polymerization of the LLA branches was 60. Linear PLA, a = 0.74; branched PLA, 3:034 Side Chain Crystallinity of poly((HOG-co-LLA)-g-LLA) DSC scans of the branched polymers did not detect crystallization of the branches despite degrees of polymerization of 60. We surmised that the lack of crystallization of the PLLA branches was related to the hydroxy groups at the termini of the branches, which could inhibit chain mobility by forming hydrogen bonds with adjacent branches. To test this hypothesis, we converted each terminal alcohol to an ester derived from either acetic or stearic acid using DCC coupling (Figure 34). Consistent with our hypothesis, esterification immediately led to 110 crystallization of the branches, as confirmed by DSC (Flgure 35) and polarized optical microscopy (not shown). Interestingly, the stearic acid derivative also showed a weak melting transition at 40 °C, which is consistent with melting of the C18 alkyl chains. Furthermore, the AHrus associated with melting was linearly related to the branch length, confirming the notion that the crystallinity was limited to the PLLA branches (Flgure 36). The near-zero intercept also suggests that the branches were highly crystalline, and only a few lactide repeat units, presumably adjacent to the alkyl chains, were in the amorphous phase. 0 /U\0H » . N DCC, DMAP MeOH. RT 1 5 040K oi Oi OH OH OH Figure 34. Acylation of poly((HOG-co-LLA)-g-LLA) using DCC coupling. 111 ==0£5MUg F heat flow (endotherm —>) f T I I 0 50 1 00 1 50 200 Temperature (°C) Figure 35. DSC second heating scans of poly((HOG-co-LLA)-g—LLA) at a heating rate of 10 °C/min. The degree of polymerization of the PLLA branches is 40; (OH terminated (— bottom), acetate terminated (— middle), and stearate terminated (— t013))- 112 25 20- f1 / g 15 - 121/ <1 10 If / // 5 - / [2| 0 l l 1 0 20 40 60 80 DPLLA Figure 36. Relationship between AHms and the branch length for poly((HOG-co- LLA)-g-LLA). The terminal —CH20H groups of the branches were acetylated. Conclusion We designed and synthesized a functionalized lactide derivative that places a benzyl protected alcohol eight methylene units away from the polyester backbone. Copolymerization with L-lactide yielded high molecular weight polymers with low PDls. Quantitative hydrogenolysis of the benzyl ether groups provided functionalized lactides with pendant primary alcohols distributed along the polymer backbone. Initiation of L-lactide from the alcohol groups provided high molecular weight polylactide comb polymers, confirmed by viscosity and light scattering data. Capping the hydroxy groups at the end of the PLLA branches provided crystalline comb polymers with melting points of 150-160 °C. ' 113 Experimental Section General. Unless otherwise specified, ACS reagent grade starting materials and solvents were used as received from commercial suppliers without further purification. L-lactide was purchased from Aldrich and sublimed prior to use. 1H NMR (300 or 500 MHz) and 13C NMR (75 or 125 MHz) spectra were acquired using either a Varian Gemini 300 spectrometer or a Varian UnityPlus 500 spectrometer with the residual proton signal from the CDCl3 solvent used as chemical shift standard. Mass spectral analyses were carried out on a VG Trio—1 Benchtop GC-MS. IR spectra were taken with Mattson Galaxy 3000 FT-IR. Elemental Analyses were determined using a Perkin-Elmer 2400 CHNSIO Analyzer. Melting points were taken on an Electrothermal capillary melting point apparatus and are uncorrected. Polymer Characterization. Polymer molecular weights were determined by GPC- MALLS at 35 °C using two PLgel 10p mixed-B columns in series (manufacturer- stated linear molecular weight range of 500-10x106 g/mol). The eluting solvent was THF at a flow rate of 1 mL/min, and the concentration of polymer solutions used for GPC was 1 mg/mL. An Optilab rEX (Wyatt Technology Co.) and a DAWN EOS 18-angle light scattering detector (Wyatt Technology Co.) with a laser wavelength of 684 nm were used to calculate absolute molecular weights. Differential scanning calorimetry analyses of the polymers were obtained using a TA DSC 0100. Samples were run under a nitrogen atmosphere at a heating rate of 10 °C/min, with the temperature calibrated with an indium standard. Polarized optical micrographs 114 were taken on a Nikon Optiphot2-POL polarizing optical microscope equipped with a Mettler FP82 hot stage and a CCD camera. H3CO OCH3 OCH3 0 Methyl 9,9-dlmethyoxynonanoate. Oleic acid (90 °/o, 25.71 g, 82 mmol) was dissolved in 450 mL of dichloromethane/methanol (4:1) and the solution was cooled to —-78 °C in a dry ice/acetone bath. Ozone was bubbled through the solution until a blue color remained. The resulting ozonide solution was purged with nitrogen, quenched with dimethyl sulfide (365 mmol, 27 mL) and stirred for 5 hours at room temperature. After the solvent was evaporated in vacuo, the residue was dissolved in ether (250 mL), washed with saturated aqueous NaCl (2 x 100 mL), dried over M9804 and the ether evaporated. The residue was dissolved in methanol (400 mL), p-toluenesulfonic acid (3.0 g) was added, and the solution was stirred overnight at room temperature. Sodium bicarbonate was added until the pH reached 10 and then the methanol was removed in vacuo. The residue was redissolved in ether (200 mL), washed with saturated aqueous NaHCOa (2 x 100 mL) and saturated aqueous NaCl (1 x 100 mL), dried over M9804, and the ether evaporated. Vacuum distillation (90 °C, 10 mTorr) yielded methyl 9,9- dimethoxynonanoate (18.3 g, 79 mmol, 96 %) as a colorless liquid. 1H NMR (300 MHz) 6 4.29 (t, 1H, J: 5.62 Hz), 3.60 (s, 3H), 3.25 (s, 6H), 2.24 (t, 2H, J: 7.57 Hz), 1.45-1.62 (m, 4H), 1.20-1.40 (m, 8H). "‘0 NMR (75 MHz) 6 174.2, 104.4, 52.5, 51.3, 34.0, 32.3, 29.2, 29.1, 28.9, 24.8, 24.4. 115 H3CO\'/\/\/\/\/OH OCH3 9,9-Dlmethoxynonanol. Under a nitrogen atmosphere, a solution of methyl 9,9- dimethoxynonanoate (17.0 g, 73.2 mmol) in anhydrous THF (250 mL) was added dropwise to a cooled (0 °C) suspension of LiAlH4 (54.9 mmol, 2.1 g) in 150 mL of anhydrous THF. After the addition was complete, the solution was stirred for 4.5 h at 0 °C, and then carefully quenched by the sequential addition of H20 (2.5 mL), 1 M NaOH (2.5 mL), and H20 (5 mL). After stirring for 30 minutes, the product was isolated from the salts via Soxhlet extraction (THF, 36 h). The THF was evaporated and the residue was dissolved in ether, washed with saturated NaCl solution (1 x 100 mL), and dried over MgSO4 to yield 9,9-dimethyoxynonanol (14.2 g, 69.6 mmol, 95 %) as a colorless liquid. 1H NMR (300 MHz) 6 4.32 (dd, 1H, J: 5.86 and 5.62), 3.58 (t, 2H, J: 6.59 Hz), 3.23 (s, 6H), 1.45-1.50 (m, 4H), 1 .2-1.4 (m, 10H). "*0 NMR (75 MHz) 6 104.5, 62.9, 52.5, 32.7, 32.4, 29.4, 29.33, 29.26, 25.7, 24.5. “Wop O (9-benzyloxy)nonanal. Under a nitrogen atmosphere, a solution of 9,9- dimethoxynonanol (14.2 g, 69.5 mmol) dissolved in anhydrous THF (100 mL) was added to a suspension of sodium hydride (60 % in mineral oil, washed with hexanes, 3.2 g, 139 mmol) in anhydrous THF (300 mL). The mixture was stirred at room temperature for 5 minutes and a solution of benzyl chloride (90.4 mmol, 10.4 116 mL) in 10 mL of anhydrous THF was added. The reaction was heated to reflux for 24 h under nitrogen, cooled to room temperature and poured into 100 mL saturated aqueous NaCI. The THF was removed in vacuo, and the aqueous phase extracted with EtOAc (3 x 100 mL). The combined organic layers were washed with a saturated NaCl solution (1 x 100 mL), dried over M9804, and the ethyl acetate was evaporated to give 1,1-dimethoxy-Q-benzyloxynonane (20.0 g, 68.0 mmol, 98 %) as a pale yellow liquid, which was used without further purification. 1H NMR (300 MHz) 8 7.20-7.33 (m, 5H), 4.50 (s, 2H), 4.38 (t, 1H, J = 5.77 Hz), 3.46 (dd, 2H, J: 6.59, 6.87), 3.31 (s, 9H), 1.55-1.65 (m, 4H), 1.20-1.41 (m, 10H). ‘30 NMR (75 MHz) 0 138.7, 128.3, 127.6, 127.4, 104.5, 72.8, 70.4, 52.5, 32.4, 29.7, 29.5, 29.4, 29.3, 26.1, 24.5. 1,1-Dimethoxy-9—benzyloxynonane (20.0 g, 68 mmol) was dissolved in a solution of THF (30 mL) and 50 % acetic acid (200 mL) and stirred at room temperature overnight. The reaction was quenched at 0 °C by the addition of sodium bicarbonate until the pH = 12. The combined organic layers were extracted with ether (3 x 100 mL), washed with saturated NaHCOa (2 x 75 mL) and saturated aqueous NaCl (1 x 75 mL), and then dried over MgSO4. The ether was evaporated to give (9-benzyloxy)nonanal (16.5 g, 66.5 mmol, 98 %) as a pale yellow liquid. 1H NMR (300 MHz) 0 9.7 (t, 1H, J = 1.95 Hz), 7.2-7.4 (m, 5H), 4.5 (s, 2H), 3.435 (t, 2H, J: 6.59 Hz), 2325 (td, 2H, J: 7.32 and 1.95 Hz), 1.5-1.8 (m, 117 4H), 1.2-1.5 (m, 10H). 130 NMR (75 MHz) 6 202.9, 136.6, 126.3, 127.6, 127.4, 72.6, 70.4, 43.8, 29.7, 29.2, 292,290, 26.1, 22.0. HOJOJ\'/\/\/\/\/O\/© OH 10-(Benzyloxy)-2-hydroxydecanolc acid. Saturated aqueous NaHSOa (30 mL) and H20 (50 mL) were added to a solution of (9-benzyloxy)nonanal (16.5 g, 66.5 mmol) dissolved in THF (30 mL). The reaction was purged with nitrogen and stirred at room temperature until a white slurry formed. The mixture was cooled to 0 °C, and a solution of KCN (13 g, 200 mmol) in H20 (70 mL) was added. After stirring for 6 hours at 0 °C, the aqueous phase was extracted with ether (4 x 100 mL) and the combined organic phases were washed with saturated aqueous NaCl (1 x 100 mL) and dried over MgSO4. Evaporation of the ether yielded 10- (benzyloxy)-2-hydroxydecanenitrile (17.7 g, 64.4 mmol, 97 %) as a pale yellow liquid, which was used without further purification. 1H NMR (300 MHz) 8 7.2-7.4 (m, 5H), 4.5 (s, 2H), 4.3 (dd, 1H, J = 6.59, 6.87 Hz), 3.4 (t, 2H, J = 6.59 Hz), 1.8 (m, 2H), 1.6 (m, 2H), 1.4 (m, 2H), 1.2-1.4 (m, 6H). 130 NMR (75 MHz) 6 136.3, 126.3, 127.7, 127.6, 120.0, 72.8, 70.4, 61.1, 35.1, 29.6, 29.2, 28.7, 26.0, 24.4. Concentrated HCI (30 mL) was added to the cyanohydrin (17.5 g, 64 mmol) and the solution was stirred at room temperature overnight. The solution pH was then adjusted to 12 with 6 M NaOH and the resulting amide was collected via vacuum filtration, washed with hexanes, and added to 250 mL 2.5 M NaOH. The 118 mixture was refluxed for 36 h and then acidified with 6 M HCI. The aqueous phase was extracted with ether (4 x 100 mL) and the combined organic layers were washed with saturated aqueous NaCl (1 x 100 mL), and then dried over MgSO4. After filtration, evaporation of the ether gave 17.5 g (60.2 mmol, 94 %) of the crude a-hydroxy acid. Recrystallization from hexanes/EtOAc (1:9) provided 9.9 g (33.7 mmol, 56 %) of 10-(benzyloxy)-2—hydroxydecanoic acid as white crystals. 1H NMR (500 MHz) 0 7.2-7.2 (m, 5H), 4.5 (s, 2H), 4.2 (dd, 1H, J = 7.32, 7.57 Hz), 3.5 (t, 2H, J = 6.59 Hz), 1.9 (m, 1H), 1.7 (m, 1H), 1.6 (m, 2H), 1.4 (m, 4H), 1.2-1.4 (m, 10H). 13C NMR (125 MHZ) 0 178.5, 138.5, 128.4, 127.7, 127.5, 72.8, 70.5, 70.2, 34.1, 29.6, 29.2, 29.2, 29.1, 26.0, 24.6. mp 51-52 °C. 10% o o o’\© 3-(8-(benzyloxy)octyl)-1,4-dloxane-2,5-dlone (BOG). 10-(Benzyloxy)-2- hydroxydecanoic acid (9.5 g, 33 mmol), 2-bromoacetyl bromide (3.3 mL 39 mmol) and 300 mL of ether were added to a 500 mL 3-neck flask. The flask was purged with nitrogen, cooled to 0 °C, and then a solution of Et3N (8.7 mL) in ether (50 mL) was added dropwise. After stirring for 6 hours at 0 °C, the mixture was filtered to remove a white precipitate and the filtrate was washed with 2 N HCI (2 x 50 mL) and saturated aqueous NaCI (1 x 100 mL), and then dried over MgSO4. Filtration and removal of the solvent in vacuo gave the crude linear acid ester as a viscous liquid (11.5 g, 27.7 mmol). The ester was dissolved in a mixture of acetone (700 119 mL) and EmN (44.5 mmol, 6.3 mL), and refluxed for 2 hours. The solids were removed by filtration and the filtrate was concentrated in vacuo. The residue was dissolved in ethyl acetate (100 mL), washed with 2 M HCI (2 x 50 mL), saturated NaHC03 (2 x 50 mL) and saturated NaCl (1 x 50 mL), and then dried over MgSO4. The product was purified by flash chromatography (EtOAc) and dried under vacuum. Recrystallization from ether (0 °C) provided 2.7 g (8.08 mmol, 38 %) of 3- (8-(benzyloxy)octyl)-1,4-dioxane-2,5-dione as colorless crystals. 1H NMR (500 MHz) 8 7.3 (m, 5H), 4.9 (m, 2H), 4.9 (m, 1 H), 4.5 (s, 2H), 3.5 (t, 2H, .1: 6.59 Hz), 2.1 (rn, 1H), 1.9 (m, 1H), 1.6 (m, 2H), 1.5 (m, 1H), 1.5 (m, 1H), 1.2-1.4 (m, 8H). ”’0 NMR (125 MHz,) 0 165.6, 164.3, 138.6, 128.3, 127.6, 127.4, 75.6, 72.8, 70.4, 65.2, 30.8, 29.7, 29.2, 29.1, 28.8, 26.0, 24.3. lR (KBr) v (cm") 2928, 2909, 2845, 1752, 1297, 1325, 1122, 1081. Anal. Calcd. For C19H2605: C, 68.24; H, 7.84. Found: C, 68.09; H, 7.99. MS (El), m/z 334 (M*), 316, 107, 91, 45, 40. mp 65.5-66.5 °C. Solution Polymerization Kinetics of BOG. A Schlenk flask loaded with BOG (0.669 g, 2.0 mmol) and a magnetic stir bar was fitted with a rubber septum and evacuated under vacuum. Sn(2-ethylhexanoate)2 (525 11L of a 38.1 mM solution in anhydrous toluene) and 4-tert-butylbenzyl alcohol (525 11L of a 38.1 mM solution in anhydrous toluene) were added to the flask through the septum ([BOG]:[I]:[catalyst] = 100:1 :1). Anhydrous toluene (8.95 mL) was added (monomer concentration = 0.2 M) and the Schlenk flask was immersed in a preheated oil bath at 90 °C. 120 Aliquots of the reaction solution were removed via syringe at specific intervals and analyzed via NMR to determine the conversion of monomer to polymer. Bulk Polymerization and Copolymerization of BOG. Monomer(s) and a small magnetic stir bar were added to ampoules prepared from 3/8 in. diameter glass tubing. The ampoule was connected via a Cajon® fitting to a T-shaped vacuum adapter fitted with a stopcock and an air-free Teflon valve. The apparatus was attached to a vacuum line and evacuated through the Teflon valve. The ampoule was backfilled with argon, and a syringe was used to add predetermined amounts of the SnOct2 and BBA solutions (both 38.1 mM in toluene) to the ampoule through the stopcock. After removing solvent in vacuo, the ampoule was flame-sealed and immersed in an oil bath at 130 °C for the desired polymerization time. The polymerization was quenched in ice and a portion of the polymer was analyzed by 1H NMR to evaluate conversion. The remaining polymer was dissolved in CH20I2, precipitated into cold methanol, and dried under vacuum (4 mTorr) at 40 °C for 24 h and analyzed by GPC to determine the molecular weight. Poly(BOG). 1H NMR (500 MHz) 6 7.2-7.4 (m, 5H), 5.1-5.2 (m, 1H), 4.6-4.9 (m, 2H), 4.5 (s, 2H), 3.4 (t, 2H, .1 = 6.59 Hz), 1.8-2.0 (m, 2H), 1.5 (m, 2H), 1.2-1.4 (m, 10H). 13C NMR (125 MHz) 6 (note: minor chemical shifts due to polymer tacticity not included) 168.9, 166.6, 138.7, 128.3, 127.6, 127.4, 75.6, 72.8, 70.4, 65.2, 60. 7, 30.9, 29.7, 29.3, 29.0, 28.8, 26.1, 24.9. Poly(BOG-co-LLA). 1H NMR (500 MHz) 6 (assignments based on a 50:50 feed ratio of BOG and LLA) 7.2-7.4 (m, 5H), 5.1-5.3 (m, 3H), 4.6-4.9 (m, 2H), 4.5 (s, 121 2H), 3.4 (t, 2H, J = 6.34 Hz), 1.8-2.0 (m, 2H), 1.4-1.6 (m, 8H), 1.2-1.4 (m, 6H). ”C NMR (125 MHz, CDCI3) 6 (note: minor chemical shifts due to polymer tacticity not included) 169.5, 166.5, 138.7, 128.3, 127.6, 127.4, 72.8, 75.6, 70.4, 69.0, 65.2, 60.7, 30.9, 29.7, 29.3, 29.0, 28.8, 26.1, 24.9, 16.6, 15.8. Removal of Benzyl Ether Protecting Group from poly(BOG).120 Benzyl ether protected poly(BOG) or poly(BOG-co-LLA) was dissolved in CH20I2 (1 mL) and added to a mixture of THF/MeOH (100 mL, 1:1 vol%) containing 20 mg of 10 % Pd/C. The heterogeneous solution was sealed in a Parr bomb, purged with hydrogen, and then pressurized with Hz to 1000-1500 psi. After 2-3 days at room temperature, the reaction mixture was filtered through a borosilicate microfiber filter, and evaporated to dryness. The crude product was dissolved in CH2C|2 and precipitated into cold MeOH (copolymers) or dissolved in EtOH and precipitated into cold hexanes (homopolymer). The product was isolated as a white solid (copolymers) or a viscous oil (homopolymer). Poly(HOG). 1H NMR (500 MHz, DMSO-da) 6 5.1-5.2 (m, 1H), 4.8-5.0 (m, 2H), 4.3 (s, 0H), 3.3-3.4 (dd, 2H, J = 11.46, 6.10 Hz), 1.7-1.9 (m, 2H), 1.3-1.5 (m, 2H), 1.2- 1.3 (m, 10H). 130 NMR (125 MHz, DMSO-da) 6 (note: minor chemical shifts due to polymer tacticity not included) 168.6, 166.8, 75.6, 72.2, 60.7, 32.5, 30.5, 28.8, 28.5, 25.5, 24.7, 24.1. 122 Poly(HOG-co-LLA). 1H NMR (500 MHz) 6 (assignments based on a 50:50 feed ratio of BOGzLLA in the original copolymer) 5.1-5.3 (m, 3H), 4.6-4.9 (m, 2H), 3.6 (t, 2H, J = 6.59 Hz), 1.8-2.0 (m, 2H), 1.4-1.6 (m, 8H), 1.2-1.4 (m, 8H). 130 NMR (125 MHz) 6 (note: minor chemical shifts due to tacticity not included) 169.6, 166.5, 72.8, 70.4, 69.0, 66.7, 60.7, 32.7, 30.8, 29.1, 28.8, 28.5, 25.6, 24.7, 16.6. Degradation Studies. Polymer samples (10 mg) were placed in 20 mL culture tubes and 10 mL PBS solution (pH = 7.4) was added. The tubes were sealed and placed in a degradation chamber at 55 °C. At allotted times samples were removed and the solvent was decanted from the samples. Once the residue was dried the molecular weight was determined by GPC-MALLS. Graft Polymerization of L-Lactlde for Branched Polylactides. Solvent-free graft polymerizations were carried out in sealed ampoules as described above. Bulk polymerization of L-Iactide from poly(HOG) and poly(HOG-oo-LLA) was catalyzed by SnOct2 ([OH groups]:[catalyst] = 1 :1). Poly(HOG) or poly(HOG-co-LLA) (100 mg) was dissolved in anhydrous THF (500 (1L) and introduced into the ampoule, followed by the predetermined amount of LLA. The contents were stirred for 30 minutes and the solvent removed in vacuo. SnOct2 (38.1 mM solution in toluene) was added under nitrogen, the solvent evaporated, and the ampoule was sealed as described above. The sealed tubes were placed in an oil bath at 170 °C for 90 seconds and transferred to an oil bath at 130 °C for 45 minutes. Purification was carried out as described above. 123 Poly((HOG-co-LLA)-g-LLA): Poly(HOG-co-LLA) (50 mg, 5 mol % HOG, M. = 31.6 kg/mol, PDI = 1.19) gave 51 mg of poly((HOG-co-LLA)-g-LLA) (96 %, M. = 39.3 kg/mol, PDI = 1.99) after purification. 1H NMR (500 MHz) 6 5.0-5.3 (m, COCH(CH3)O), 4.8-4.9 (m, COCHzO), 4.3 (dd, J = 13.67 and 6.84 Hz, COCH(CH3)OH), 4.1 (m, (CH2)6CH20CO), 1.8-2.0 (m, CHCH2(CH2)5O), 1.4-1.8 (m, COCHCH3), 1.2-1.4 (m, CHCH2(CH2)GOCO). “’0 NMR (125 MHz) 6 (note: minor chemical shifts due to polymer tacticity not included) 175.1, 169.6, 72.4, 69.0, 66.7, 65.6, 30.8, 30.2, 29.7, 29.1, 29.0, 28.4, 25.6, 20.5, 16.6, 16.5. Esterlflcatlon of Poly((HOG-co-LLA)-g-LLA). The procedure for the acetate synthesis is described; the stearate derivative was prepared using the same procedure. A small round bottom flask containing poly((HOG-co-LLA)-g-LLA) (5 mol% HOG, 50 mg, 5.2 x 10‘7 mol, 6.2 umol OH groups, M. = 96.4 kg/mol) and DMAP (1.3 (mol, 0.2 mg) were dried under vacuum for four hours. Anhydrous CH2CI2 (10 mL) was added via syringe under nitrogen and the solution was stirred until completely dissolved. Acetic acid (19 (mol, 107 11L of a 1 mM solution) and DCC (13 (mol, 195 uL of 64 mM solution in CH20I2) were added via syringe and the reaction was stirred under nitrogen overnight. The reaction mixture was filtered to remove the insoluble dicyclohexyl urea byproduct, further concentrated, and filtered again to fully remove the urea. The CH2CI2 polymer solution was further 124 concentrated to a 10 wt % solution and precipitated from cold methanol. The polymer was dried under vacuum to yield 40 mg (80 %) of the acetylated polymer (M. = 97.0 kg/mol, PDI = 1.75). 125 Chapter 4 Post-Polymerization Modification of Polyglycolides via “Click” Chemistry Introduction The term “synthetic biopolymers” refers to a class of polymers that includes 121-123 124127 polymeric drugs, polymer-drug conjugates, encapsulated drugs within “3:128 and multicomponent polyplexes that are being developed polymer micelles, as non-viral vectors for delivery of siRNA and DNA.‘29'131 All of these polymers are water-soluble and are intrinsically bioactive or are an inert carrier of the active drug, protein, or gene. Functionalized polyesters are particularly suited for these applications because they are biodegradable and can contain a variety of functional groups, which impart a wide range of properties. Recent examples of polylactide l10,33 l 16,105 I ! derivatives with pendant alky ary allyl,34 and poly(ethylene glycol)33 have been reported. In addition, polylactides with pendant functional groups such as alcohols, amines, and carboxylic acids have been described by others and us.“"‘9'2‘5'2‘3-76 Incorporation of functional groups affects polymer physical properties such as hydrophilicity, degradability, and the glass transition temperature. Water- soluble polyesters containing hydrophilic poly(ethylene glycol) segments are especially interesting targets because they enhance a polymer’s in vivo delivery capabilities by preventing uptake by the reticuloendothelial system (RES).62'7°'72'132 The RES eliminates old cells, debris, foreign substances and pathogens from the blood stream and can complicate controlled delivery.133 126 The synthesis of functionalized polyesters can be accomplished via two routes: the functional monomer approach or the post-polymerization modification approach. The functional monomer approach adds the desired functionality directly to the monomer (Scheme 1, Chapter 1). However, since many functional groups are incompatible with polymerization conditions they must be protected prior to polymerization and deprotected in a second step. The deprotection steps must be highly regulated to ensure complete removal of the protecting groups and to minimize backbone degradation. Although several groups have reported 17-20,22 successful deprotection strategies, simple methods for the introduction functional groups are still desired. in addition, any small polymer modification requires the synthesis and characterization of new monomers. A second functionalization route is post-polymerization modification, which introduces functionality after polymerization. This approach minimizes the number of protection-deprotection steps, and perhaps more importantly, allows a wide range of polymers to be synthesized from a single monomer. Examples of strategies for post-polymerization modification of polyesters include direct grafting,35'38 DCC-coupling,34'57ketoximine coupling?“51 Michael addition,”134 67,69,72 atomic transfer radical addition, and click chemistry.”72 Probably the most successful of these approaches is “click” chemistry, specifically the copper(l)- mediated 1,3-dipolar cycloaddition of azides and alkynes (Scheme 34). Originally 135,136 introduced in polymer science by Sharpless, click chemistry has now been 127 used in the synthesis of dendrimers,‘37‘140 block copolymers,““'142 dendronized linear polymers,“""‘“"145 and branched polymers.7°'72'146 R‘ : H 1 _ f MEN-.. ’\ .. R, [LnCU]+ (‘9 R2 R‘ : rlsuLn.2 / , 2 C9 N EN R NeN’ ‘92 1 R CULn-1 H R1 AC N\\ N ~|ULn2 Scheme 34. Mechanism for the copper(l)-mediated 1,3-dipolar cycloaddition between alkynes and azides, also known as “click” chemistry. Recently we reported the synthesis of an acetylene functionalized glycolide derivative, 3,6—dipropargyl-1,4-dioxane-2,5-dione (propargylglycolide, PGL), and its post-polymerization via click chemistry.73 We demonstrated appropriate conditions to apply click chemistry to polyglycolides, which are more hydrolytically sensitive than other polyesters, such as .s-caprolactone.7°'72 Generating Cu(l) in situ via copper sulfate pentahydrate and sodium ascorbate in DMF at room temperature prevented degradation of the polyglycolide backbone. We applied these techniques to the synthesis of water-soluble comb polyglycolides by the addition of short poly(ethylene glycol) segments to the poly(PGL) (PPGL) backbone. The polymers self-assemble in aqueous solution 128 into small, uniform unimolecular micelles (Figure 37) that are unaffected by a critical micelle concentration (cmc).147 In contrast, traditional micelles are often formed from amphiphilic block copolymers and exist as single molecules below the cmc and as dynamic aggregates above the crnc.““3“52 Although there has been some progress towards unimolecular micelles, including dendrimers and a few aliphatic graft-copolymer systems, synthetic challenges have impeded their wide spread use.‘53'158 Dendrimers are structurally well-defined molecules with high functional group densities; however many dendrimer syntheses are tedious and low yielding due to their iterative nature. Aliphatic graft-copolymers could be useful, however simple attachment of hydrophilic groups to hydrophobic polymer backbones currently remains a challenge. This chapter describes the synthesis, characterization, and applications of functionalized PPGL polymers. \N’Nwoméovfio/ Figure 37. The comb architecture of poly(propargylgcholide)-graff-poly(ethylene glycol monomethyl ether) and formation of a unimolecular micelle in aqueous solution. The hydrophobic backbone and hydrophilic side chains are illustrated in red and blue, respectively. 129 Results and Discussion Post-Polymerization Modification Scheme 35 shows the synthesis and post-polymerization functionalization of poly(propargylglycolide) by click chemistry.159 The synthesis of propargylglycolide began with a Reformansky—type reaction of propargyl bromide with ethyl glyoxalate, in the presence of activated zinc, to generate ethyl 2-hydroxy- 3-butynoate. Hydrolysis of the ester in refluxing water yielded 2-hydroxy-3- butanoic acid, which was dimerized in refluxing toluene using p-toluenesulfonic acid as a catalyst, giving propargylglycolide in 34 % yield. The monomer was polymerized via ring opening polymerization using SnOct2 and BBA as the catalyst and initiator, respectively. An mPEG azide suitable for click chemistry was synthesized by tosylation of 2,5,8,1 1 ,14-pentaoxahexadecan-16-ol (tetraethylene glycol monomethyl ether, mPEG) followed by displacement of the tosyl group by sodium azide, giving 16- azido-2,5,8,1 1,14-pentaoxahexadecane (tetraethylene glycol monomethyl ether azide, mPEG-azide). The azide group is easily identified by its characteristic IR absorbance at 2105 cm'1 and the resonances from the methylene Otto the azide at 3.38 ppm in its 1H NMR spectrum and 50.5 ppm in its 13C NMR spectrum. The click reaction was performed under the reported conditions, generating Cu(l) in situ (5 mol °/o copper(ll) sulfate pentahydrate and 12 mol % sodium ascorbate) in DMF at room temperature. 130 O '1)le \/ o o 0 H20 __ ‘ /\ —\— r N0 MGR Br / ll é Z" OH reflux OH TH Flether 84 0/0 stOH 0 0 C toluene 51 % reflux 34 % ll % 0 Sn" / 0 o H A R0 0 ° ‘ a :E/ O " 0 O Q BBA CuSO4-5H20 Propargyl glycolide dium ascorbate N3 WONF, \ DMF, RT N\ \N’N \/\0/\é0\/\)p0/ Scheme 35. Synthesis, polymerization, and functionalization of poly(propargylglycolide). The 1H NMR spectra of the polymer before and after click functionalization with mPEG-azide are shown in Figure 38. The reaction went to 100 % completion as indicated by the disappearance of the resonances at 2.85 ppm (-CI-_12-CCH) and 2.05 ppm (-CH200_f-_I) and the appearance of a new resonance at 7.6 ppm corresponding to the triazole proton. 131 C a Mk A W, I ‘l' I I I I I I I I I I I I T T I I Ifi I I I I I I I I I I I I I I l I I I I l l l l l l l l l l l 7.5 7.0 6.5 6.0 5.5 5.0 4.5 4.0 3.5 3.0 2.5 ppm Figure 38. 500 MHz 1H NMR spectra of poly(propargylglycolide) and its click functionalization with mPEG-azide. Integration of a = c = a’ = c’ = 1. The click reaction can also be characterized via infrared (IR) spectroscopy. The spectra of unfunctionalized PPGL (bottom) and PPGL-graft—poly(ethylene glycol) monomethyl ether (PPGL-g-mPEG) (top), is shown in Figure 39. Notably, the alkyne-stretching mode at 3200 cm" completely disappears upon click functionalization and a new band appears at 3300-2500 cm'1 from the triazole N-H. CH2 stretching at 3000 cm'1 also increased in accordance with the addition of PEG branches. The absence of a band at 2105 cm'1 suggests that the polymer is free of residual azides. 132 J .‘J 4000 3000 20001 1 000 wavenumbers (cm' ) Figure 39. IR spectra of unfunctionalized PPGL (bottom) and PPGL-g—mPEG (top). Notably, the alkyne stretch at 3200 cm'1 completely disappears upon click functionalization and a new stretch at 3300-2500 cm'1 from the N-H triazole stretch appears. Although the click reaction is highly selective and functional group tolerant, it suffers from residual copper contamination. Triazoles coordinate multivalent ions such as Cu2‘“, which tainted the functionalized polymers light green and increased the polymer’s toxicity in in vivo applications. While polymers are often extensively dialyzed against Milli-Q water to remove residual copper, polylactides degrade significantly under the same conditions. As an alternative, we used ion exchange resin beads (Amberlite® lRC-748 ion exchange resin) to remove residual copper 133 from the polymers. Since exposure to the slightly basic beads in Milli-Q water degraded the polymers, we added the beads directly to DMF polymer solutions after the “click” reaction was complete. When exposed to the beads in DMF for 24 hours the copper concentration was reduced to < 10 ppb (determined by flame atomic absorption) without appreciable degradation (initial polymer: M. = 56,200, PDl = 1.07; after 24 hour exposure to ion exchange resin in DMF: M. = 55,900, PDI = 1.12) (Figure 40). Alternatively, polymer-supported Cu(l) catalysts could be used, which do not require additional steps to remove the copper from the final product.160 l I 1 6 17 1 8 l 9 elution time (min) Figure 40. GPC traces of PPGL-g-mPEG polymers before and after exposure to ion exchange resin beads. The Cu(ll) concentration was reduced to < 10 ppb. Original polymer (Mn = 56,200; PDI = 1.07; solid blue line); dialysis solution without ion exchange resin (Mn = 56,600; PDI = 1.11; dashed green line); polymer after 24 hours exposure to ion exchange resin in DMF (Mn = 55,900; PDl = 1.12; dotted red line). 134 Formation and Characterization of Unimolecular Micelles Previous work on PEG-substituted polylactides suggested that grafting PEG chains onto PPGL should provide hydrophilic and perhaps water-soluble polylactides.161 Using PPGL polymers with molecular weights ranging from 7-60 kD, we covalently linked oligo(ethylene glycol) segments as the hydrophilic branches. PEG is important for in vivo applications because it increases blood circulation time and decreases polymer-protein interactions?” There have been many examples of PEG block copolymers that organize into micelles on the order of 20 to 60 nm in aqueous solution.“’3"63'167 The micelles are dynamic aggregates above the cmc and single molecules below the cmc. Typically the micelles are formed using an oil-in-water process. We applied this technique to the PPGL-g-mPEG polymers. The polymer was dissolved in acetone (< 0.5 mL) and added dropwise to Milli-Q water. The acetone was removed by passing a stream of nitrogen through the solution or in vacuo, which caused micellization. Micelle characterization often includes determination the critical micelle concentration (cmc) by pyrene encapsulation,"5"'169 dynamic light scattering (DLS),170 or surface tension measurements.168 For example, when pyrene is added to a solution of polymer micelles, the pyrene is preferentially solubilized in the hydrophobic regions due to its low water solubility, and the pyrene 333 nm peak in the pyrene emission spectra shifts to 338.5 nm. The cmc for amphiphilic molecules can be determined by monitoring the pyrene emission spectra over a range of concentrations. Below the cmc the equilibrium shifts from micelles to 135 individual surfactant molecules and the pyrene emission band is at 333 nm.171 To determine the cmc of PPGL-g-mPEG micelles we prepared polymer concentrations (0.05 - 100 (lg/mL) containing 1 x 10" M pyrene. When excited, all solutions emitted at 338 nm suggesting that pyrene was sequestered within the polymers at all the polymer concentrations. Dynamic light scattering (DLS) measures the hydrodynamic radius of a particle in solution by monitoring its time-dependent fluctuation in scattering intensity due to Brownian motion. At short time delays, the correlation is high because the particles cannot move very far away; as time increases, the correlation decays to zero. Since larger particles move more slowly than smaller particles, the decay is related to the size of the particles through the Stokes- Einstein equation, equation 17, which calculates the size of a particle based on its velocity due to Brownian motion, where D is the diffusion constant, k3 is the Boltzmann’s constant, T is the temperature, T) is the viscosity of the solution and r is the radius of a spherical particle. KBT 6m7r D (17) When traditional micelles reach their cmc, there is a dramatic shift in the hydrodynamic radius (Rh), indicating the formation of micelles. We analyzed the PPGL-g-mPEG micelles at various polymer concentrations using 6 Protein Solutions Dyna Pro-MS/X system at 25 °C. Prior to analysis, the polymer was dissolved in Milli-Q water and filtered through 0.2 um Whatman PTFE 136 (poly(tetrafluoroethylene)) filter. Notably, the hydrodynamic radius the PPGL-g- mPEG polymers was unaffected by polymer concentration, suggesting the formation of unimolecular micelles. The DLS data are compiled in Table 13 and a representative unimodal particle distribution can be seen in Figure 41. Table 13. Properties and size of PPGL-g-mPEG unimolecular micelles. Mn lea Mn (PPGL- Fin Have Dave (PPGL)“ g-mPEG)a (nm)b (nm)° (mm)d 1a 10,000 1.19 31,300 35:1:2 3.2:0.1 1011 Polymer 2a 23,800 1.30 69,000 35 :1: 2 3.0 :1: 1.0 11 :1: 1 38 31,000 1.28 90,000 37 :1: 2 3.0 :1: 0.1 14 :i: 1 (a) determined by GPC-MALLS using THF as the eluting solvent at 35 °C. (b) determined by DLS at 25 °C. (0) determined by AFM; 0.20 mg/mL solution spin coated in Si wafer. ((1) determined by TEM; 0.20 mg/mL solution dropped on Formvar-coated nickel grid; stained with 2 % uranyl acetate solution. 137 O 50 100 150 Rh (nm) Figure 41 . Hydrodynamic radius of PPGL-g-mPEG micelles as determined by dynamic light scattering (Rh = 31 nm; 2 mg/mL solution in Milli-Q water). Analysis of the structural differences between block and comb copolymers sheds light on micelle formation. In dilute aqueous solution, a single diblock amphiphilic copolymer cannot assemble properly to shield the hydrophobic block from water so it aligns itself at the air-water interface. Once the interface becomes saturated with polymer, the addition of more polymer chains causes the spontaneous formation of micelles; the hydrophilic exterior surrounding the hydrophobic domain. In contrast, when short PEG segments were grafted to PPGL, a comb architecture was formed that naturally shields the hydrophobic backbone by the hydrophilic PEG shell (Figure 42). 138 ll OOOOOOOOOOOOOOO block copolymer PPGL-g-mPEG unimolecular micelle Figure 42. Organization of traditional micelles and unimolecular micelles from block copolymers and PPGL-g-mPEG, respectively. The red and blue spheres indicate hydrophobic and hydrophilic polymer segments, respectively. We further characterized the unimolecular micelles via tapping mode atomic force microscopy (AFM) and transmission electron microscopy (TEM). AFM measures the height of particles mounted on a flat surface via an oscillating probe. AFM samples were prepared by spin coating a 200 ug/mL polymer solution on a freshly cut silicon wafer. Figure 43 shows a representative AFM height image of PPGL-g—mPEG on a silicon surface. The image shows uniformly sized particles and the narrow height distribution provides evidence that only single polymer chains are present without particle agglomeration. AFM heights measurements are reported in Table 13. To rule out dust contamination, we analyzed a series of polymer solutions and determined that the number of particles scaled according to the polymer concentration. 139 7.25 nm 0.00 nm 0.00 pm 1.75 pm 3.50 pm Figure 43. AFM height image of PPGL-g—mPEG on a silicon wafer substrate; height = 3.0 :1: 0.1 nm. Finally, we analyzed the particles via transmission electron microscopy. TEM samples are examined by passing a beam of electrons though an ultra thin specimen. The image generated by the transmitted electrons is focused by objective lens and projected onto a fluorescent screen. Staining allows the sample to be seen more clearly because high-Z metals such as osmium, lead, or uranium have high scattering cross-sections. In negative staining, the heavy atoms are deposited around the sample causing the background around the sample to appear dark in the images. Samples were prepared for TEM by dropping an aqueous polymer solution on a Formvar-coated nickel grid. After two minutes, the excess solution was removed and the sample was negatively stained 140 with a 2 % uranyl acetate solution. As seen in Figure 44, the particles are uniformly dispersed and have a uniform size distribution. TEM diameter measurements are reported in Table 13. Figure 44. TEM image of PPGL-g-mPEG stained with uranyl acetate; diameter = 10 :t 2 nm. Image taken at 200,000x; scale bar represents 100 nm. Hydrophobic Guest Encapsulation Self-assembling block copolymer micelles have long been explored as drug carriers, delivering both encapsulated and conjugated drug molecules to appropriate cells. Hydrophobic drugs are sequestered within polymeric carriers using an oil-in-water emulsion approach?”174 The encapsulated material is released by permeation through the polymeric carrier or by in vivo degradation of a biodegradable polymeric carrier. Unimolecular micelles from comb polymers seem particularly promising as drug carriers because they can be prepared by controlled polymerization methods, 141 the functional group density can be very high, and they are not subject to a cmc. We dissolved azobenzene, a UV-active surrogate for hydrophobic drugs, and PPGL-g—mPEG polymer (Mn = 116,500 g/mol) in acetone (<1 mL) and dispersed the solution in Milli-Q water. The acetone was removed and the solution was filtered to remove any unencapsulated azobenzene. The homogenous solution appeared yellow, an indication of azobenzene within the polymeric micelles. We confirmed its encapsulation by its characteristic absorbance at 320 nm (Figure 45). The UV absorbance in the polymer spectra can be attributed to the polyester backbone (260 nm) and pendant triazoles (< 250 nm). 1 absorbance .0 .0 O) on p .5 .0 ro o ‘ - 200 250 300 350 400 450 wavelength (rim) Figure 45. UV-vis spectra of azobenzene loaded polymeric micelles (top, blue), polymeric micelles (middle, green), and azobenzene (bottom, red) in Milli-Q water at room temperature (polymer concentration = 1 mg/mL). 142 To track the release of azobenzene from the unimolecular micelles, we placed the azobenzene-loaded micelles in dialysis tubing (molecular weight cutoff (MWCO) = 12-14 kD; concentration 1 mg/mL) and dialyzed with Milli-Q water. Removing aliquots of the micelle solution from the dialysis bag and measuring their UV-vis spectrum allowed us to monitor the continual release of azobenzene over 10 hours (Figure 46). After analysis, the micelle solution was returned to the dialysis bag and the surrounding Milli-Q water was replaced. The high molecular weight of the polymer and its relatively slow degradation prevented the polymer from escaping the dialysis membrane. normalized absorbance at 311 nm IJAAA l A A I All 10 100 1000 time (min) Figure 46. Continual release of azobenzene from loaded polymeric micelles suspended in Milli-Q water at room temperature over 10 hours. At each time point, the azobenzene remaining in the micelles was determined (at 311 nm) by UV-vis spectroscopy. 143 Although the nanoparticles effectively encapsulated azobenzene, many therapeutic molecules, such as proteins, are much larger. As a proof of principle, for encapsulating large molecules, we encapsulated poly(methyl methacrylate) (PMMA) in micelles. The PMMA was tagged with azobenzene by initiating polymerization of MMA from an azobenzene ATRP initiator as shown in Scheme 36. Following the same encapsulation procedure used for azobenzene, we encapsulated azobenzene-initiated PMMA (azobenzene-PMMA) in PPGL-g-mPEG micelles. PMMA polymers with molecular weights of up to 10 kD were successfully encapsulated (Figure 47). The lower azobenzene UV absorbance and higher ester UV absorbance is consistent with the azobenzene-PMMA structure. ON’NO 3 ‘ ”<1 ° 2. OH Br\lI/L 0&3 Br C) \N/ \N/ ,2” C) CuBr II C) nBr C) C) \ Scheme 36. Synthesis of azobenzene PMMA initiator and ATRP polymerization of poly(methyl methacrylate) using CuBr/2,2’-bipyridine in anisole at room temperature. CL 144 .0 .o O) (D absorbance .0 A 250 300 350 400 450 wavelength (nm) Figure 47. UV-Vis spectra of azobenzene-PMMA (M. = 10 kD) loaded polymeric micelles (top, blue), polymeric micelles (middle, green), and azobenzene-PMMA (bottom, red) in Milli-Q water at room temperature. Prodrug Conjugation Conjugation of biologically relevant molecules is another important aspect of controlled drug delivery. Common water-soluble molecules that cannot be encapsulated are directly conjugated to the polymer or to a prodrug linker. Prodrugs are inactive until released from the polymer, minimizing potential toxic side effects. ”6'17“" Uhrich and co-workers‘wm182 synthesized polymerizable prodrugs that released the active drug as the polymer degrades. For example, degradation of salicylic acid-derived poly(anhydride esters), released up to 62 wt% of active drug. 145 Active drugs can also be tethered to polymer through a site that does not compromise drug activity. Emrick and co-workers reported the synthesis of a polyester-grafi-drug copolymer conjugate that attached camptothecin to a water- soluble polyester. The water solubility of the conjugate and the biodegradability of the polyester backbone allowed the polymer conjugate to be used in polymer therapies.113 In collaboration with Christina Chan in the Chemical Engineering Department at Michigan State University, we investigated the delivery and release of L-cycloserine (LCS), a small molecule inhibitor in the ceramide synthesis pathway relevant to Alzheimer’s disease. Preliminary results suggested that inhibiting de novo ceramide synthesis in primary astrocytes with L-cycloserine prevented tau hyperphosphorylation and decreased the oxidative-stress effects seen in neurons.""°"186 However, achieving appropriate in vivo levels of LCS required impractical injection frequencies. Polymeric carriers that provide sustained controlled release can overcome this limitation. To explore the possibility of LCS delivery, we appended the LCS to a hydrolysable methyl actetoacetate derived linker.187 The linker and mPEG was then attached to PPGL via click chemistry (Scheme 37). The short oligo(ethylene glycol) segments impart water-solubility and increase in vivo circulation time. 146 O O NaN ,DMF Mo“ HOMCI 3 o ; HOMN3 A, MOMN3 100 C DMAP, cyclohexane. 76 0/0 reflux, 24 hr 77% O H N O 2 H2N 'NH NH ” O I 15h O r L-cycloserine O :32 1“ 0 m ' 3 < . MOMNS sodium ascorbate CuSO4-5H20 Scheme 37. Synthesis of L-cycloserine hydrolyzable linker and conjugation to PPGL-g-mPEG. We monitored the release of LCS from the LCS functionalized polymer at pH 6.0 and pH 7.4. The functionalized polymer was dissolved in water (10 mg/mL), placed in dialysis membrane (MWCO = 1 kD) and dialyzed with Milli-Q water (pH = 6.0) or PBS solution (pH = 7.4). At allotted times, the exterior solution was removed, frozen, and replaced. Thawed samples were analyzed for released LCS using liquid chromatography-mass spectrometry (LC-MS). Unfortunately, quantification of the LCS released was severely complicated by the dynamic equilibrium of LCS between its monomeric and multiple dimeric forms.188 To avoid this problem, we synthesized L-homoserine Iactone 147 hydrobromide (LHS), a structural mimic of LCS, in 97 % yield from L-methionine for kinetic studies. lts syntheSls and conjugatlon to PPGL is shown in Scheme 38.189 o o G) 0 / OH 7 O IPrOH/HZO/ACOH NH2 . 2. HCl,dloxane 56 °’° Et3N o o O M9304 MGM N RN 3 CH20l2 O 72 °/o L-homoserine Iactone ll 0 O PPGL mPEG-N3 IN 0 sodium ascorbate MOM N3 CuSO4-5H20 Scheme 38. Synthesis of L-homoserine Iactone hydrobromide and conjugation to PPGL-g—mPEG. We repeated the previous experiment using L-homoserine Iactone at pH 6.0 (Milli-Q water) and 7.4 (PBS solution), quantifying the release of LHS using LC-MS based on external LHS standards. During the initial 11 hours of the experiment, we observed rapid LHS release at both pH conditions, most likely from the release of unbound, encapsulated LHS. However, analyses over two weeks showed controlled release of LHS as a function of pH (Flgure 48). 148 -ln([Ll-ls1t I [LHS]O) OD 1 r 0f 1 4L 0 50 100 150 time (hr) Figure 48. Release of L-homoserine Iactone from PPGL-g-mPEG polymer in Milli- 0 water (pH = 6.0; A) and PBS solution (pH = 7.4; x). Lower Critical Solution Temperature Materials One of the most interesting applications for functionalized biopolymers is as a “smart” material that responds to various environmental stimuli such as pH or temperature. For example, poly(N-isopropyl acrylamide) (lePAAm) and poly(acrylic acid) (pAAc) respond to the cell’s acidic endosome by releasing their cargo upon entry into the cell.“""192 Thermoresponsive behavior has also been well documented.‘93’195 For example, NIPAAm and PMMAs exhibit sharp transitions when the polymer is heated through its lower critical solution temperature. At the LCST, polymers undergo a solution-gel transition that corresponds to the entropically driven expulsion of solvating water molecules from the polymer. Such materials have a variety of interesting and potentially useful applications. For 149 example, a thermally responsive bioadhesive surface becomes resistant to protein adsorption below its LCST, enabling facile growth and harvesting of biomaterials from surfaces?”198 Recently there has been increasing emphasis on responsive LCST 2932499202 and current research is focusing on materials that contain PEG segments how chemical composition and structure influences the LCST. Recently, Lutz and co-workers synthesized a family of PEG-grafted methacrylate polymers that exhibit LCST behavior over a wide temperature range. Varying the hydrophilic/hydrophobic ratio of the two comonomers, poly(ethylene glycol) methyl ether methacrylate and 2-(2’-methoxyethoxy)ethyl methacrylate (less hydrophilic) provided accurate control over the LCST from 27 °C to 90 °C.3°'2°0 While Lutz’s work is an excellent example of control over LCST behavior, the methacrylate system has limitations; each polymer must be individually synthesized and characterized, and the polymers are not biodegradable, limiting in vivo applications. Following a similar approach, we synthesized biodegradable thermoresponsive polymers via click chemistry. However, instead of copolymerizing two monomers, we controlled the hydrophilic/hydrophobic ratio by attaching a mixture of diethylene glycol monomethyl ether-azide (mDEG-azide, hydrophilic) and decyl azide (hydrophobic) segments to PPGL (Scheme 39). Varying the ratio between the hydrophobic alkyl and hydrophilic PEG segments, we obtained LCST values ranging from 25-65 °C. To monitor the LCST, the 150 purified polymers were dissolved in Milli-Q water and their cloud point temperatures were determined by measuring the solution turbidity as a function of temperature. Choosing a wavelength where the polymers show ~95 % transmittance below their cloud point, 450 nm, we increased the temperature while monitoring the change in transmittance. At the LCST, the “apparent transmittance” decreased because gel formation caused a dramatic increase in scattering (Flgure 49). A representative plot in Figure 50 shows the transition to be relatively sharp, spanning a temperature range of ~ 3 °C. \\ O m N3/\/O\/\O/\/O\ R0 0 O H O n \\ 17m N3/\/\/\/\/\ ll CuSO4-5H20 sodium ascorbate Scheme 39. Synthesis of thermoresponsive polyglycolides. — LIJ‘HL-Zij'ég 13,232.32 1' _’1 below above LCST LCST Figure 49. Visualization of thermoresponsive poly(glycolide) (59 % mDEG; 41 °/o decyl) below (left) and above (right) its LCST (26 °C). 151 100 00.? 30 - c Q 5 so - E 3 4o . 5 -. 32 Q 20 . 1. 500000. O 1 n 1 20 25 30 35 40 Temperature (° C) Figure 50. Cloud point determination at 450 nm for PPGL “clicked” with a mixture of hydrophobic alkyl and hydrophilic PEG chains. The composition of the “clicked” polymer, (59 % mDEG chains, 41 % decyl chains) was determined by 1H NMR. Significantly, the relationship between the cloud point temperature and the mole fraction of mPEG chains in the polymer is linear, as shown in Figure 51. 152 \l O 65 . x13 60 . ’5 A55 L 0'... «:50 , .1: 152.45 i 9" 64° - .4 '635 - g so - 1 25 . 5 20 4 . 50 60 70 80 90 100 mol % peg azide In copolymer Figure 51. Relationship between the cloud point observed at 450 nm and the mole percentage of PEG chains “clicked” onto the PPGL homopolymer. The broken line is a least squares fit to the data. We also observed polymer aggregation above the LCST via TEM (Flgure 52). Two TEM samples were prepared by dropping an aqueous polymer solution onto Formvar-coated nickel grids below the LCST of the polymer. Before removing the excess water, we heated one sample above the LCST using a hot plate and kept the other sample at room temperature (below its LCST). After two minutes the excess water was removed and the samples were stained with a 2 °/o uranyl acetate solution. Images of the two samples were taken at 270,000x. Clearly the sample heated and dried above the LCST showed polymer agglomeration, confirming the LCST transition. 153 Figure 52. TEM images of thermoresponsive poly(glycolide) (59 % mDEG; 41 % decyl) below (top) and above (bottom) its LCST (26 °C). Degradation of Poly(propargylglycollde)-graph-poly(ethylene glycol) PLA and related polymers are susceptible to hydrolytic degradation, degrading into natural metabolites. We ran degradation experiments on PPGL-g- mPEG polymers at 36 °C in Milli-O water to assess the polymer degradation rate. Shown in Figure 53 is the evolution in molecular weight during degradation, where 154 Mn(0) and Mn(t) are the initial and partially degraded number average molecular weight at time t, respectively. Since the PPGL-g-mPEG polymer was completely water soluble, we collected and analyzed both the polymer and its degradation products at each time point. 100% 901 80- 7o- 60-<> 50- 40- 0 30- ° 20- °<> 10- 0 O 1 J 0 5 1O 15 tlme (days) Mn (1) / Mn (0) We) Figure 53. Molecular weight change of PPGL-g-mPEG during hydrolytic degradation in Milli-Q water at 36 °C. Each point represents an average of three samples. To further evaluate the degradation of PPGL-g-mPEG, we analyzed the decrease in molecular weight according to the random chain scission model, (Mn(0)/Mn(t)-1)/Pn(0), where Pn(0) is the initial degree of polymerization.203 Data plotted from polymers that undergo random chain scission are linear and the degradation rates (slopes) can be easily compared to other degradable polymers. 155 As shown in Figure 54, PPGL-g-mPEG degrades more rapidly than rec-lactide run under accelerated degradation conditions (55 °C), as expected. 0.08 0.07 - [A 0.06 . l 0.05 . 0.04 - ,U 0.03 - d [.25 0.021 1"A (Mnlol / Mn“) '1) / l"’n(0) 0.01 - o 024. . .. 0 5 10 1 5 20 25 time (days) Figure 54. Molecular weight change of PPGL-g—mPEG (l3) and rao-PLA (A) during hydrolytic degradation. Degradation experiments were run at 36 °C for PPGL-g-mPEG and 50 °C for rac-PLA. The lines are least squares fits to a random chain scission model. The data points represent an average of 3 samples. Conclusion Using “click” chemistry we prepared poly(ethylene glycol) grafted polyglycolides. In aqueous solution the polymers self-assembled into unimolecular micelles on the order of a few nanometers as demonstrated by DLS, AFM and TEM. We showed the encapsulation and release of hydrophobic molecules, conjugation and release of water-soluble, biologically active molecules, and the hydrolytic degradation of PPGL-g-mPEG polymers. In addition, we prepared 156 functionalized biodegradable polymeric materials that show lower critical solution temperature behavior over a wide and physiological temperature range. Experimental Section General. Unless otherwise specified, ACS reagent grade starting materials and solvents were used as received from commercial suppliers without further purification. Ethyl glyoxalate (Alfa Aesar, 50 wt % in toluene) was distilled prior to use. THF was dried over sodium and benzophenone, methylene chloride was dried over CaHz, DMF was dried over activated 4A molecular sieves and methyl methacrylate was purified by passage through activated basic alumina. Zinc (Spectrum, 20 mesh) was pretreated with 2M HCI, and then washed sequentially with distilled water and absolute ethanol and dried under vacuum at 60 °C. Propargyl bromide (Alfa Aesar, 80 wt % in toluene) was used as received. Ion exchange resin beads (Amberlite® lRC-748) were purchased from Aldrich and washed with DMF prior to use. 1H NMR (300 or 500 MHz) and “*0 NMR (75 or 125 MHz) spectra were acquired using either a Varian Gemini 300 spectrometer or a Varian UnityPlus 500 spectrometer with the residual proton signal from the CDCI3 solvent used as chemical shift standard. Mass spectral analyses were carried out on a VG Trio-1 Benchtop GC-MS. IR spectra were taken with Mattson Galaxy 3000 FT-IR. Elemental Analyses were determined using a Perkin-Elmer 2400 CHNSIO Analyzer. Melting points were taken on an Electrothermal capillary melting point apparatus and are uncorrected. 157 Polymer Characterization. Polymer molecular weights were determined by GPC- MALLS 35 °C using two PLgel 10p mixed-B columns in series (manufacturer- stated linear molecular weight range of 500-10x106 g/mol). The eluting solvent was THF at a flow rate of 1 mUmin, and the concentration of polymer solutions used for GPC was 1 mg/mL. A Waters 2410 differential refractometer was used as the detector and an Optilab rEX (Wyatt Technology Co.) and a DAWN EOS 18-angle light scattering detector (Wyatt Technology Co.) with a laser wavelength of 684 nm were used to calculate absolute molecular weights. Flame atomic absorption analyses were carried out on a Hitachi Z-9000 Graphite Furnace Atomic Absorption Spectrophotometer. Liquid chromatography-mass spectrometry analyses were carried out on a Quattro micro API (Waters Corporation, Milford, MA). UV-Vis spectra were recorded with a Cary 300 Bio WinUV, Varian spectrophotometer. Fluorescence was recorded using a Fluorolog-3 (Instruments S.A., Inc.) fluorometer. All samples were diluted to 1 X 10'6 M using Milli-Q water, excited at 16x = 440 nm, and the fluorescence emission spectra were observed from 450-500 nm. Dynamic light scattering (DLS) data were obtained using a temperature-controlled Protein Solutions Dyna Pro-MS/X system. All samples were filtered through a 0.2 urn Whatman PTFE syringe filter and allowed to equilibrate in the instrument for 15 minutes at 25 °C before measurements were taken. The uniformity of the particle sizes were determined by a monomodal curve fit, which assumes a single particle size with a Gaussian distribution. Surface profile measurements were performed with a Pacific Nanotechnology Nano-R 158 atomic force microscope in close contact (oscillating) mode to generate height images that were not altered other than a simple leveling procedure. Silicon tips with a spring constant of 36 N/m, tip curvature of 10-20 nm, and a resonance frequency of 286-339 kHz were used for all experiments. The polymeric micelles were dispersed in Milli-Q water and the solutions (typical concentration, 20-300 pig/mL) were spin coated at 5000 rpm for 40 sec on freshly cut silicon substrates. Heights were determined using AFM by taking the average of 20 particle heights. The lateral size could not be determined due to convolution effects created by the 60201205 TEM micrographs were collected at 270,000-370,000 x magnification on a JEOL-iOOCX transmission electron microscope. TEM samples (typical concentration, 20-300 ug/mL) were dropped onto nickel grids pretreated with Formvar and allowed to settle for 2 min before removing the excess solution. The sample was negatively stained with either a 1 °/o phosphotungstic acid (PTA) stain or 2 % uranyl acetate (UA) stain prepared with deionized water. 0 o/\ OH \\ Synthesis of 2-hydroxy-4-pentynolc acld ethyl ester. Propargyl bromide (277 g, 2.35 mol) was added under a blanket of N2 to a 3 L round bottom flask containing anhydrous THF (350 mL) and activated Zn (230 g, 3.5 mol). The mixture was stirred at room temperature for 30 min and then cooled in an ice bath. A toluene solution of ethyl glyoxalate (51 wt %, determined by NMR, 473 g, 2.36 mol) and a toluene solution of propargyl bromide (80 wt %, 352 g, 2.36 mol) were 159 combined with a mixture of 500 mL anhydrous THF and 700 mL anhydrous ether and added dropwise to the stirred slurry. After the addition was complete, the mixture was stirred at 0 °C overnight. The reaction mixture was then poured into a 4 L Erlenmeyer flask containing 1 L of ice-cold 3 M HCI. After separation of organic layer, the aqueous layer was extracted with ether (3 x 300 mL) and the combined organic layers were dried over MgSO4. Filtration and removal of the solvents by rotary evaporation gave a dark blue oil, which was washed with EtOAc/hexanes (30/70, 3 x 200 mL) as the eluent. Vacuum distillation (50-55 °C/100 mTorr) gave 170 g of 1 as a colorless oil (1.20 mol, 51 %). 1H NMR (500 MHz) 6 4.25 (m, 3H), 3.11 (d, 1H, J: 6.35 Hz), 2.65 (m, 2H), 2.03 (t, 1H, J: 2.66 Hz), 1.26 (t, 3H, J: 7.20 Hz). “’0 NMR (125 MHz) 6 172.99, 76.53, 71.25, 66.64, 62.11, 24.61, 14.13. CH \\ 1.. 0H Synthesis of 2-hydroxy-4-pentynolc acid. The ethyl ester (170 g, 1.20 mol) was added to distilled water (800 mL) and heated to reflux for three days. After cooling to room temperature, the solution was acidified by the addition of concentrated HCI (100 mL) and continuously extracted with ether for two days. The ether solution was diluted to 1.5 L with additional ether and dried over MgSO4 for two hours. After filtration, the solution was concentrated by rotary evaporation and dried under vacuum to give a light brown solid, which was purified by recrystallization from CH2CI2 at 0 °C, followed by sublimation at 58 °C and a second recrystallization 160 from CH20I2 at 0 °C to give 115 g of 2-hydroxy-4-pentynoic acid as white crystals (1.01 mol, 64 %). 1H NMR (500 MHz) 6 4.42 (t, 1H, J: 5.00 Hz), 2.75 (m, 2H), 2.10 (t, 1H, J: 2.56 Hz). ”C NMR (125 MHz) 6 177.26, 77.97, 71.96, 66.51, 24.66. MS (m/z) 115.3 (M+1), mp 61-63 °C. / o 0 a I/ 0 0 Synthesis of 3,6-dl-2-propynyl-1,4-dloxane-2,5-dlone. 2-Hydroxy-4-pentynoic acid (18 g, 173 mmol) and p-toluenesulfonic acid monohydrate (1.5 g) were added to a 2 L round bottom flask charged with 1.8 L of toluene. The flask was heated to reflux for 3 days, removing water azeotropically using a Dean-Stark trap. After cooling to room temperature, the toluene was removed by rotary evaporation, and the residue was dissolved in 500 mL CH2CI2, washed with saturated aqueous NaH003 (3 x 150 mL) and dried over MgSO4. Filtration and removal of the CH20I2 gave the product as a light brown solid which was washed with diethyl ether (3 x 50 mL), recrystallized from toluene and sublimed at 75 °C to give 6.1 g propargylglycolide as white crystals (31.8 mmol, 34 %). 1H NMR (300 MHz) 6 5.29 (t, J = 4.64 Hz), 5.05 (dd, J = 7.08 Hz, J = 4.39 Hz), (1 H total for the signals at 5.29 and 5.05 ppm), 2.95 (m, 2H), 2.17 (t, J: 2.56 Hz), 2.11 (t, J = 2.69 Hz), (1 H total for the signals at 2.17 and 2.11 ppm). ‘30 NMR (75 MHz) 6 164.26, 163.44, 76.77, 76.67, 74.82, 74.15, 73.34, 72.02, 23.94, 21.24. Anal. Calcd. for C1oH304: C, 161 62.50; H, 4.17 Found: C, 62.80; H, 4.01. MS (CI, m/z) 193.2 (M+1), mp 103-106 °C. General procedure for bulk polymerizations. Monomer(s) and a small magnetic stir bar were added to ampoules prepared from 3/8 in. diameter glass tubing. The ampoule was connected via a Cajon® fitting to a T-shaped vacuum adapter fitted with a stopcock and an air-free Teflon valve. The apparatus was attached to a vacuum line and evacuated through the Teflon valve. The ampoule was backfilled with argon, and a syringe was used to add predetermined amounts of the SnOct2 and BBA solutions (both 38.1 mM in toluene) to the ampoule through the stopcock. After removing solvent in vacuo, the ampoule was flame-sealed and immersed in an oil bath at 130 °C for the desired polymerization time. The polymerization was quenched in ice and a portion of the polymer was analyzed 1H NMR to evaluate conversion. The remaining polymer was dissolved in CH2CI2, precipitated into cold methanol, and dried under vacuum (4 mTorr) at 40 °C for 24 h. % O O n % Representative polymerization of propargylglycolide. Propargylglycolide (0.65 g) was polymerized for 45 min of a [Ml/[l] = 150. The conversion of monomer to polymer calculated from 1H NMR was 97 %. Precipitation and drying under vacuum gave 0.64 g poly(propargylglycolide) as a light brown solid (98 %). 1H NMR (300 162 MHZ) 6 5.31-5.46 (br, 1H), 2.79-3.03 (br m, 2H), 2.01-2.18 (br, 1H). Mn, GPC- MALLS = 31.0 kg/mol, PDI = 1.28. N /\/0\/\O/\/0\/\O/\/0\ 3 Synthesis of 16-azldo-2,5,8,11,14-pentaoxahexadecane (mPEG-azide). Pentaethylene glycol (20.0 g, 80 mmol) in 60 mL THF was added dropwise to a solution of NaOH (10.0 g, 250 mol) in a water/T HF mixture (3:2 vol %, 200 mL) at 0 °C. The mixture was stirred for 30 minutes and then p-toluenesulfonyl chloride (16 g, 170 mmol) in 100 mL THF was added dropwise. The mixture was stirred at 0 °C for 3 hours and then for 6 hours at room temperature. The mixture was poured into 30 mL ice water and the water layer was extracted with ether (3 x 250 mL). The combined organic layers were washed with a saturated aqueous NaCl solution and dried over MgSO4. The ether was evaporated and the tosylate was used without further purification (29.5 g, 74.9 mmol, 94 %). The tosyl PEG (29.0 g, 74 mmol) and sodium azide (9.6 g, 148 mmol) were dissolved in 200 mL DMF and the solution was heated at 60 °C for 15 hours. The solution was cooled and added to 100 mL water. The solution was extracted with ether (4 x 100 mL) and the combined organic layers washed with a saturated aqueous NaCl solution (2 x 100 mL) and dried over MgSO4. The ether was evaporated and the residue was purified by vacuum distillation at 130 °C (50 mtorr) to give 16-azido-2,5,8,11,14-pentaoxahexadecane as a colorless liquid (15.4 g, 55.6 mmol, 77 %). 1H NMR (500 MHz) 6 3.66-3.63 (m, 16H), 3.55-3.54 (m, 2H), 163 3.39-3.37 (m, 5H). ‘30 NMR (125 MHz) 6 71.88, 70.65, 70.61, 70.56, 70.56, 70.54, 70.52, 70.46, 69.96, 56.96, 50.63. /NO\/\O/\/O\/\O/\/O\/\ N pg} NN\/\O/\/O\/\O/\/O\/\O/ General procedure for “click” functionalization. The desired amount of acetylene functionalized polymer, 100 mol % of PEG-azide, and 12 mol % sodium ascorbate powder (all mole percent values are in reference to the mole percent of alkynyl groups on the polymer) were dissolved in 5 mL DMF in a Schlenk flask. The solution was deoxygenated by three freeze-pump-thaw cycles. After the solution warmed to room temperature a 0.1 M solution of CuSO4-5H20 in deoxygenated DMF (5 mol %) was added under nitrogen; the reaction mixture was stirred at room temperature for 3-16 hours. At the end of the reaction, the solids in the reaction mixture were removed by filtration and ion exchange resin beads (Amberlite® lRC-748 ion exchange resin) were added to the solution for 6-24 hours to remove residual copper. The beads were removed by filtration and the DMF removed in vacuo. The polymer was purified via dialysis (MWCO = 12-14 kD) in acetone/water (1 :1) mixture overnight and then dried under vacuum. PPGL-g-mPEG. PPGL (50 mg) (Mn, GPC-MALLS = 31.0 kg/mol, PDI = 1.28) and 140 mg PEG-azide (100 mol °/o) were dissolved in 5 mL DMF for the click reaction. Copper was removed from the polymer solution via ion exchange resin, the DMF 164 was evaporated and the polymer was purified via dialysis (MWCO 12-14 kD) (180 mg, 95 %). 1H NMR (500 MHz) 6 7.8-7.4 (b, 1H), 5.6-5.2 (b, 1H), 4.5-4.4 (b, 2H), 3.9-3.7 (b, 2H), 3.6-3.4 (b, 16H), 3.3 (b, 3H), 3.3-3.2 (b, 2H). “’0 NMR (125 MHz) 6 167.37, 141.19, 123.87, 71.78, 70.37, 69.21, 58.86, 50.06, 27.31, 21.89. Formation of unimolecular micelles (1 mg/mL solution). A solution of PPGL-g- mPEG (20 mg) in 0.5 mL acetone was slowly added dropwise to stirred ice-cold Milli-Q water (20 mL) in a Schlenk flask. The solution was allowed to stir for 30 minutes before the acetone was removed under a stream of nitrogen or in vacuo. Azobenzene encapsulation and release from unimolecular micelles. Encapsulation: to encapsulate azobenzene or azobenzene-initiated PMMA, the polymer and azobenzene were dissolved in acetone (<1 mL) and the acetone solution was added dropwise to Milli-Q water. The acetone was removed and the solution was filtered to remove any unencapsulated azobenzene. We confirmed its encapsulation by it characteristic absorbance at 320 nm, seen in UV-vis spectroscopy. Release: azobenzene-loaded polymeric micelles were placed in dialysis membrane (MWCO 12-14 kD) and dialyzed with Milli-Q water. At allotted times, aliquots of the polymer solution was removed from the dialysis bag, filtered to remove any insoluble released azobenzene, and the azobenzene remaining was determined using UV-vis. 165 Synthesis of 2-bromoproplonlc acId-2-(4-phenylazophenyl)ethyl ester.206 Triethylamine (7.6 mmol, 0.765 g) was added to a solution of 4-hydroxy azobenzene (3.8 mmol, 0.75 9) dissolved in 60 mL anhydrous CHZClg. The solution was stirred at 0 °C and 2-bromopropionyl bromide (7.6 mmol, 1.65 9) dissolved in 10 mL anhydrous CH20I2 was added dropwise over 30 min under a nitrogen atmosphere. The mixture was stirred at room temperature overnight, filtered and washed 3x with deionized water. The organic layer was dried over M9804, and the methylene chloride evaporated. The crude product was recrystallized from ethanol 2x and dried under vacuum 4 hours to yield an orange solid (0.30 g, 0.95 mmol, 25 % after two recrystallizations). 1H NMR (500 MHz, DMSO-da) 6 7.97-7.96 (dd, 2H, J = 6.84 and 2.07), 7.91 -7.89 (dd, 2H, J = 6.84 and 1.53), 7.52-7.47 (m, 3H), 7.29-7.27 (dd, 2H, J: 6.64 and 2.20), 4.60 (q. 1H, J: 6.96), 1.97-1.95 (d, 3H, J: 6.84). 13c NMR (125 MHz, DMSO-da) 6 201.87, 168.46, 152.31, 131.17, 129.11, 124.13, 122.89, 121.73, 39.45, 21.42. / °MO .. 0 .. O Synthesis of azobenzene-Initiated PMMA.206 A 3.5 M solution of methyl methacrylate (monomer, 106 mmol) in anisole, 2-bromopropionic acid-2-(4- phenylazophenyl)ethyl ester (initiator, 21.2 mmol, 70 mg) and 2,2’-bipyridine 166 (ligand, 21 mmol, 33.1 mg) (monomer: initiator: catalyst: ligand = 50:1 :1 :1) were added to a Schlenk flask and deoxygenated by three freeze-pump-cycles. The solution was refrozen and copper(l) bromide (catalyst, 21 mmol, 31 mg) was added under nitrogen. After an additional freeze-pump-thaw cycle, the polymerization was stirred at room temperature for 16 hr and then quenched. The polymer solution was purified by precipitation from cold methanol (yield = 0.124 g, 15 %, Mn, GPC-MALLS = 10.0 kg/mol, PDI = 1.12). Ho/\/\ N3 Synthesis of 3-azldopropan-1-ol.2°7 3-chloropropan-1-ol (17.3 g, 183 mmol) and sodium azide (366 mmol, 17.8 g) were dissolved in 200 mL deionized water and stirred at 70 °C overnight. The solution was cooled and extracted with ether (3 x 75 mL). The organic layers dried over MgS04, filtered and concentrated. The liquid was purified via vacuum distillation (25 °C at 10 mTorr) to give 14.0 g (139 mmol, 76 %) of 3-azidopropan-1-ol as a colorless liquid. 1H NMR (500 MHz) 6 3.72-3.70 (t, 2H, J: 5.96), 3.43-3.40 (t, 2H, J: 6.59), 2.01 (b, 1H, OH), 1.82-1.77 (dt, 2H, J: 5.96 and 5.69). “’0 NMR (125 MHz) 6 59.7, 46.4, 31.4. 0 0 Mom N3 Synthesis of 3-azldopropyl-3-oxobutanoate.2°° A mixture of methyl aoetoacetate (26 mmol, 3.0 g). 3-azidopropan-1-ol (28.4 mmol, 2.87 9). and DMAP (0.05 equiv., 1.3 mmol, 0.15 g) were dissolved in cyclohexanol (30 mL, 1 mL/mmol alcohol) and heated to reflux overnight. The water was removed azeotropically via a Dean- 167 Stark trap. The solution was concentrated, dissolved in ethyl acetate and washed with 1 M HCI (2 x 60 mL). The organic layer was dried over MgSO4. filtered and concentrated. The 3-aziodpropyI-3-oxobutanoate was purified by vacuum distillation at 80 °C (5 mTorr) to give 3.6 g (19.5 mmol, 77 %) as a colorless liquid. 1H NMR (500 MHz) 6 4.22-4.20 (t, 2H, J= 6.16), 3.45 (s, 2H), 3393.37 (t, 2H, J: 6.65), 2.24 (s, 3H), 1.93-1.89 (m, 2H). ”C NMR (125 MHz) 6 200.4, 166.9, 49.9, 48.0, 30.2, 28.0. H’N O o N o MGM N3 Synthesis of 3-azldopropyl 3-(3-oxolsoxazolldln-4-yllmlno)butanoate.187 L- cycloserine (50 mg, 0.49 mmol) and 3-azidopropyl 3-oxobutanoate (45 mg, 0.25 mmol) were placed in a 25 mL round bottom flask and stirred overnight. Methylene chloride (10 mL) was added and the solution filtered to remove any free and insoluble L-cycloserine. The methylene chloride solution was concentrated and dried under vacuum to give 69 mg (100 % recovery of 3-aziodpropyI-3- oxobutanoate, 72 % conversion to imine) of 3-azidopropyl 3-(3-oxoisoxazolidin-4- ylimino)butanoate as a colorless liquid. To determine the percent conversion to the imine, we compared the integration of the methyl protons from the aoetoacetate derivative with and without LCS bound. For example, the methyl protons shift from 2.2 ppm (singlet) to 2.0 ppm (singlet) upon LCS attachment (see Appendix 43). 168 69 o Br H3N 3 1i, Synthesis of L-homoserlne Iactone hydrobromide.209 A mixture of bromoacetic acid (1.54 g, 11.0 mmol) and L-methionine (1.52 g, 10 mmol) in 14.4 mL of H202- propanol/AcOH (5:5:2 v/v) was refluxed for 8 hours. The reaction was cooled and the solvents evaporated. The orange solid was dissolved in 6 mL dioxane/HCI (2:1), heated at 50 °C for 10 min and then stirred at room temperature for 5 hours. The mixture was placed on ice 1 hr to evoke precipitation. The colorless solid was isolated by filtration and washed with cold isopropanol to give 1.00 g (5.6 mmol, 56 %). 1H NMR (500 MHz, DMSO-da) 6 4.46-4.42 (dt, 1H, J: 6.91 and 1.22), 4.35- 4-24 (m. 2H). 2.58-2.52 (m. 1H). 2.34-2.26 (m. 1H). 13c NMR (125 MHz, DMSO-da) 6 173.7, 66.6, 46.1, 27.4. mp = 230 °C (lit 233-236 °C)2‘° (’1 )1 o MOMNs Synthesis of 3-azldopropyl 3-(2-oxo-tetrahydrofuran-3-yllmlno)butanoate.‘°° Homoserine Iactone hydrobromide (0.55 mmol, 100 mg), triethylamine (0.66 mmol, 0.09 mL), and M9804 (0.5 g) were stirred in 10 mL anhydrous CH20|2 at room temperature for 1.5 hours. 3-azidopropyl 3-oxobutanoate (0.86 mmol, 0.17 g) was then added and the heterogeneous mixture stirred overnight. The reaction was filtered to remove the M9804 and concentrated. The white solid was triturated in 169 ether to precipitate the amine salt and filtered. The ether was evaporated giving 105 mg (72 % recovery of 3-azidopropyl-3-oxobutanoate, 100 % conversion to imine) of the imine as a colorless liquid. 1H NMR (500 MHz) 6 4.6 (s, 3H), 4.4-4.3 (m, 1H), 4.3-4.3 (m, 1H), 4.2-4.1 (m, 1H), 4.1 (t, 2H, J: 6.10), 3.736 (t, 2H, J: 6.75), 2.7-2.6 (m, 1H), 2.4-2.3 (m, 1H), 2.02 (s, 3H), 1.9-1.8 (m, 2H). 13c NMR (125 MHz) 6 175.6, 170.1, 160.6, 85.5, 65.0, 59.5, 51.5, 48.3, 30.7, 28.4, 19.6. Experimental procedure for continuous release of L-cycloserlne or L- homoserlne Iactone from PPGL-g-mPEG polymers. PPGL-g—(mPEG-co-LCS) (Mn = 90.0 kg/mol, 120 mg) containing LCS (0.12 mmol) was dissolved in 10 mL Milli-Q water or PBS solution and placed in a dialysis membrane (MWCO = 1,000 Da). The membrane was placed in 100 mL Milli-Q water (pH = 6.0) or 100 mL PBS solution (pH = 7.4) at 37 °C and kept in the dark. At predetermined times, a portion of the solution surrounding the dialysis bag was removed, placed in a 1.5 mL vial, and stored at -20 °C. The remaining solvent was discarded and replaced by fresh Milli-Q water or PBS solution (1 OOmL). The thawed samples were analyzed using LC-MS to determine the concentration of released L-cycloserine in solution. The same procedure was used for determining the release kinetics of LHS. N 3/\/O\/\O/\/O\ Synthesis of 1-azldo-2-(2-(2-methoxyethoxy)ethoxy)ethane (mDEG-azide). Triethylene glycol (20.0 g, 122 mmol) in 60 mL THF was added dropwise to a solution of NaOH (14.6 g, 366 mmol) in a water/T HF mixture (6:4 vol %, 200 mL) at 170 0 °C. The mixture was stirred for 30 minute and then p-toluenesulfonlc chloride (23 g, 122 mmol) in 100 mL THF was added dropwise. The mixture was stirred at 0 °C for 3 hours and then for 6 hours at room temperature. The mixture was poured into 30 mL ice water and the water layer was extracted with ether (3 x 250 mL). The combined organic layers were washed with a saturated aqueous NaCl solution and dried over MgSO4. The ether was evaporated and the tosylate was used without further purification (35.4 g, 111 mmol, 91 %). The tosyl PEG (35.4 g, 111 mol) and sodium azide (14.4 g, 222 mmol) were dissolved in 200 mL DMF and the solution was heated at 60 °C for 15 hours. The solution was cooled and added to 100 mL water. The solution was extracted with ether (4 x 100 mL) and the combined organic layers washed with a saturated aqueous NaCl solution (2 x 100 mL) and dried over MgSO4. The ether was evaporated and the product was purified via vacuum distillation at 130 °C (50 mtorr) to provide 16-azido-2,5,8,11,14-pentaoxahexadecane211 as a colorless liquid (16.7 g, 66.3 mmol, 80 %). ‘H NMR (500 MHz) 6 3.65-3.62 (m, 8H), 3.53- 3.51 (m, 2H), 3.36-3.34 (m, 5H). “*0 NMR (125 MHz) 6 71.88, 70.66, 70.61, 70.57, 69.98, 58.98, 50.63. N3/\/\/\/\/\ Synthesis of 1-azldodecane.212 1-Bromodecane (10.0 g, 45.2 mmol) and sodium azide (5.88 g, 90.4 mmol) were dissolved in 50 mL DMF and stirred at 60 °C for 16 hours and cooled. The mixture was added to 50 mL water and extracted with ether 171 (4 x 50 mL). The combined organic layers were dried over MgSO4, filtered, and the ether evaporated. 1-Azidodecane was collected as a colorless liquid via bulb- to-bulb distillation (10 mtorr, 35 °C) (7.70 g, 42.1 mmol, 93 %). 1H NMR (500 MHz) 6 3.25-3.22 (t, 2H, J = 6.96), 1.61 -1 .55 (m, 2H), 1.36-1.25 (m, 14H), 0.66-0.65 (dd, 2H, J: 6.84, 7.06). 13c NMR (125 MHz) 6 51.49, 31.66, 29.49, 29.47, 29.27, 29.14, 28.82, 26.71, 22.66, 14.08. LCST measurements. The side chain compositions of PPGL-g-(n—decyl-co- mDEG) were determined using 1H NMR comparing the integration of the methylene protons adjacent to the triazole (DEG functionalized = 4.6-4.4 ppm and alkyl functionalized = 4.4-4.2 ppm, see Appendix 52). The polymers isolated and purified by dialysis (MWCO = 14-16 kD; Milli-Q water, 16 hr). LCST behavior was determined by dissolving polymer samples (10 mg/mL) in Milli-Q water and monitoring solution turbidity by measuring transmittance at 450 nm as a function of temperature. PPGL-g-(decyI-co-mDEG). PPGL (122 mg, Mn, GPC = 35.5 kg/mol, PDI = 1.44), 1-azido-2-(2-(2-methoxyethoxy)ethoxy)ethane (480 mg, 2.6 mmol), and 1- azidodecane (240 mg, 1.3 mmol) were dissolved in 5 mL of DMF for the “click” reaction. The product was isolated via dialysis (320 mg, 88 %). 1H NMR (500 MHz) (70:30 ratio of mDEGzn-decyl) 6 7.8-7.4 (b, 1H), 5.6-5.3 (b, 1H), 4.6-4.4 (b, 1.4H), 4.4-4.2 (b, 0.6H), 3.8 (b, 1.4H), 3.6-3.4 (b, 2.6H), 3.4-3.3 (b, 2.1H), 3.3-3.1 (b, 2H), 1.9-1.8 (b, 0.6H), 1.3-1.2 (b, 4.2H), 084-082 (D, 0.3H). ‘30 NMR(125 MHz)6 172 167, 141, 123, 122, 71.85, 70.41, 69.30, 59.92, 50.08, 31.84, 30.24, 29.50, 29.44, 29.26, 29.03, 26.51, 22.63, 14.09. Degradation of PPGL-g-mPEG. 10 mg samples of PPGL-g-mPEG were placed in 20 mL culture tubes and 10 mL of Milli-Q water was added. The tubes were capped and placed in a degradation chamber at 36 °C. Samples were removed at predetermined times, dried, and the molecular weight was determined by GPC- MALLS. For comparison, the accelerated degradation of rac-LA was determined at 50 °C. The results are normalized to account for the degree of polymerization (Pn(0)) of each polymer. 173 Chapter 5 Biodegradable Cross-Linked Nanoparticles Introduction Inorganic nanoparticles have a long and rich history, having benefited from favorable physical properties such as mechanical, thermal, and chemical stability. While comparable developments in organic nanoparticles are more recent, the structural diversity of organic materials and their application to complex problems in medicine has made the synthesis and characterization of organic nanoparticles one of the most active topics in polymer science.213 Organic nanoparticles are important in many applications, but especially for the encapsulation of small molecules for the delivery of therapeutic agents, personal care products, and colorants.214 Until recently, nearly all applications were based on nanoparticles prepared by spray drying or an oil-in-water emulsion approach, using homopolymers as the host. The need for “smart” particles in medical applications, to control release rates, target specific sites in the body, and to transport large biomolecules such as small double stranded RNA for gene therapy, instigated interest in more complicated polymer architectures and function.‘2"2‘5'216 A variety of architectural motifs were synthesized including many varieties of linear homopolymers and block copolymers, and branched polymers 43-45 d40-42 such as stars, combs?”220 dendrimers, “’48 and hyper-branche structures that provide multiple sites (usually at chain ends) for tethering molecules. 174 Although many of these polymers assemble into traditional micelles in aqueous solution, their aggregation numbers vary and the critical micelle concentration defines their lower concentration limit. Many applications, however, require stable nanoparticles that maintain their shape and structure under a variety of conditions. Therefore significant progress has been made by cross-linking the 221222 or shellm'229 domains of amphiphilic polymer micelles. core Cross-linking was first introduced in 1979 by Prochaska and Baloch who photochemically irradiated a polybutadiene micellar assembly, cross-linking its core.230 Following this initial account, many others have utilized cross-linking strategies to stabilize polymeric micelles. For example, Ding and Liu exposed poly(styrene)-block-poly(2-cinnamoylethyl methacrylate) (PS-b-PCEMA) to UV light inducing the cycloaddition of the cinnamoyl groups.22“'23"236 Hawker237 and Harth238 provided a route to well defined nanoparticles by thermal cross-linking of well-defined styrene-vinylbenzocyclobutene copolymers. Using a dilute solution to promote intramolecular cross-linking, the well-dispersed benzocyclobutene units were activated in hot benzyl ether forming cycloaddition cross-links. The robust particles behaved similarly to polystyrene, however their T9 was approximately 20° higher than polystyrene, reflecting the nature of cross-linked materials. Alternatively, cross-linking can be accomplished through a difunctional 2272394“ and others used water-soluble diamines to cross-link cross-linker. Wooley the poly(carboxylic acid) shell of p0Iy(styrene)-block-poly(acrylic acid) (PS-b-PAAc) micelles via DCC coupling. The cross-linked micelles were unaffected by solution 175 concentration. Armes and co-workers242 reported the shell cross-linking of thermoresponsive poly[2-(dimethylamino) ethyl methacrylate-block-2-(N— (morpholino)ethyl methacrylate)] (PDMA-b-PMEMA) using 1,2-bis(2-iodoethoxy)- ethane (BIEE) as a cross-linking agent. The difunctional reagent selectively quarternized unreacted tertiary amines on the PDMA blocks, leaving the thermoresponsive PMEMA core unaffected. The cross-linked polymeric micelles showed a lower critical solution temperature of 25 °C. Armes also used divinyl sulfone (DVS) to covalently cross-link poly[ethylene oxide-block-glycerol monomethacrylate-block-2-(diethylamino) ethyl methacrylate] (PEO-b-PGMA-b- PDMA) and poly[ethylene oxide-block-2-hydroxylethyl methacrylate-block-Z- (dimethylamino) ethyl methacrylate) (PEO-b-PHEMA-b-PDMA) pH responsive micelles.243 Using DVS as the cross-linker provides a convenient method to determine the cross-linking efficiency by sulfur microanalysis. More recently Wooley and co-workers reported “click cross-linking” within the shell and core of polymeric nanoparticles.229 Azido-terminated dendrimers were used to cross-link the shell domain of acetylene-functionalized poly(acrylic acid)- block-poly(styrene) (PAAc-b-PS).244 The alkyne groups, located at the corona reacted with a G(1.0) - G(3.0) azide-terminated dendrimer cross-linking the shell. Moving the alkynyl groups to the poly(styrene) block allowed core click cross-linked (CCL) nanoparticles to be synthesized?” The efficiency of click cross-linking was determined by 1H NMR; upon cross-linking, new triazole proton signals appear at 176 7.6 ppm. Furthermore, it has been suggested that the triazole chromophore could be monitored by UV spectroscopy.245 Unfortunately many cross-linking reaction require high dilution to promote intramolecular cross-linking and avoid inter-micelle cross-linking, which leads to micelle agglomeration. Many shell cross-linking reactions are run below 0.5 % solids, presenting problems for scale-up and industrial use of these materials. To circumvent this problem, Armes and co-workers synthesized a triblock copolymer that could be cross-linked under higher polymer concentration.246 Steric stabilization by the PEO cornea, allowed poly[ethylene oxide-block-2- (dimethylamino) ethyl methacrylate-block-2-(N-morpholino) ethyl methacrylate] (PEO-b—PDMA-b-PMEMA) to be cross-linked using BIEE, at 10 °/o solids.247 Cross-linked nanoparticles have many applications, one being the conjugation of biologically important molecules. For example, Wooley and co- workers conjugated fluorescently labeled shell cross-linked micelles with the protein transduction domain (PTD) peptide sequence derived from HIV.241 The biologically active nanoparticle easily penetrated cell membranes. In addition, cancer cells have also been targeted by cross-linked nanoparticles bearing folate receptor ligands.248 PAAc-b-PS cross-linked nanoparticles were also functionalized with tritrpticin, an antimicrobial agent?”250 This chapter describes degradable comb polymers that individually self- assemble in aqueous solution and can be chemically cross-linked to from organic nanoparticles that retain their chemical functionality. Forming unimolecular 177 micelles from a single polymer chain is key to the formation of these nanoparticles. Analysis of these new materials revealed robust spherical nanoparticles with diameters ranging from 4-8 nm. The nanoparticles showed superior dimensional stability and were used for controlled release. Results and Discussion Synthesis of Cross-Linked Nanoparticles The synthesis, characterization, and applications of the unimolecular micelles based on functionalized poly(propargylglycolide) were presented in chapter 4. To move from the amphiphilic comb polymers to cross-linked nanoparticles, we first appended short oligo(ethylene glycol) segments to ~75 % of available alkynes on the PPGL backbone (for example, poly(PGL[EG75X25] describes a poly(propargylglycolide) polymer functionalized such that 75 % of the pendent alkynes were reacted with the short oligo(ethylene glycol) segments and the other 25 % of alkynes are unreacted). The partially functionalized polymer was dissolved in a minimal amount of acetone followed by dropwise addition of the solution to stirred Milli-Q water (1 % solids) containing sodium ascorbate (1 equivalent with respect to the alkyne groups) and copper(ll) sulfate (0.25 equivalents with respect to the alkyne groups). The acetone was removed by passing a stream of nitrogen through the solution and then 1,5-diazidopentane (1 mg/mL in acetone) was added dropwise. The reaction was stirred 2-3 days at room temperature, cross-linking the residual alkyne groups on the PPGL base polymer (poly(PGL[EG750L25]) where CL25 denotes that the remaining 25 % of 178 the pendent alkynes were cross-linked) (Scheme 40). Initially we carried out the cross-linking reactions in dilute solution to prevent any intermolecular cross-linking reaction resulting in insoluble polymeric material; however, we determined that we could run the reaction at much higher polymer concentration (10 % solids) because of the steric stabilization of the PEG corona. We cross-linked 10-100 °/o of available alkynes within the preformed unimolecular micelles. The cross-linked nanoparticles were purified by dialysis to remove unreacted cross-linker and by Amberlite® beads to remove the residual copper. The progress of the cross- Iinking reaction was monitored by following the disappearance of the alkyne- stretching band in the IR spectrum (Figure 55). JELW m~~01mr 01A,, :VO‘MN-NNQ CuSO4- 5H20 sodium ascorbate DMF, RT C)1-m N o 0 3w VOW WN’N CuSO4 5H20 sodium scorbae 0)\::\0 y}: H25, RT t Nadgg% :66: 0000 Scheme 40. Cross-linking of poly(PGL[EG75X25]) unimolecular micelles by 1,5- diazidopentane. 179 absorbance Figure 55. IR spectra of poly(PGL (bottom), poly(PGL[EG75X25] (middle), and cross-linked poly(PGL[EG750L25] (top). Upon PEG functionalization and cross- linking, the alkyne stretching band (3340-3270 nm) was replaced by the triazole stretching band (N-H) (3300-2500 nm). In addition, no residual azide stetching bands were observed at 2100 nm. We determined the number of remaining alkynyl groups via 1H NMR comparing the integration of the methine protons on the polymer backbone (5.4 ppm) to the integration of the triazole protons (7.6 ppm) and the integration of the methylene protons adjacent to the triazole (4.5 ppm) provides the degree of click 11 I a“ _ .-/ I 2 ‘ l 4000 functionalization (Flgure 56). 3000 wavenumbers 2000 180 1000 bl bl b c \,_J ‘\/\h__hj\ _-_ _ __~ _ ,--A_ - IV I T I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I r Tfl I 7 fir I I I I I I T 7.5 7.0 6.5 6.0 5.5 5.0 4.5 4.0 3.5 3.0 2.5 ppm Figure 56. 500 MHz 1H NMR of poly(PGL[EG75X25]) before (bottom) and after (top) cross-linking with 1,5-diazidopentane (integration of signal czc’ = 1:0.75 (bottom) and 1:1 (top). Characterizatlon of Cross-Linked Nanoparticles Representative examples of unimolecular micelle solutions cross-linked at 10, 50, and 100 percent of available cross-linking sites are shown in Table 14. The percent cross-linking was determined using 1H NMR because the number of triazole protons increases upon cross-linking. We also confirmed cross-linking within the nanoparticle cores using DLS, AFM, and TEM. Compared with noncross-linked material, cross-linked nanoparticles are more robust and dense as evidenced by their smaller hydrodynamic radius, measured by DLS, and increased height profile, measured by AFM, and smaller diameter, measured by TEM (Table 181 14). An example of the 50 °/o cross-linked nanoparticles as analyzed by AFM is shown in Figure 57. Table 14. Characterization of cross-linked nanoparticles by DLS and AFM. Mn % cross- Fih H Dave ave e P°'Yme' (po|Y(PGL[EG75X25])) linked” mm). mm)“ (nm) a 1a 90,000 10 29 :2 3.5 e 0.8 7 a 2 2a 69mm 50 18:2 45:03 6:2 2b 69,000 100 12 e 2 6.0 1 1 4 = 1 (a) determined by GPC-MALLS using THF as the eluting solvent at 35 °C. (b) percent cross-linking of available cross-linking sites; determined by 1H NMR. (0) determined by dynamic light scattering at 25 °C. (d) determined by AFM; 200 pig/mL solution spin coated in Si wafer. 19.58 pm ooopm 0.00 pm 2.45 pm 4.90 pm Figure 57. Representative AFM image of cross-linked nanoparticles (polymer 2a, Table 14, 200 ug/mL solution spin-coated on a Si wafer substrate). 182 TEM further confirmed cross-linking within the nanoparticle cores. A representative TEM image of 50 % cross-linked nanoparticles shows uniform spheres of 6 a 1 nm (Figure 58). Depending on the cross-linking density, the nanoparticles had diameters of 4-8 nm. Samples were prepared for TEM by dropping an aqueous solution of the cross-linked nanoparticles on a Formvar- coated nickel grid. After two minutes, the excess water was removed and the sample was negatively stained with a 1 % phosphotungstic acid solution or 2 % uranyl acetate solution in deionized water. Figure 58. Representative TEM image of cross-linked nanoparticles (polymer 2a, Table 14, 100 ug/mL solution; 370,000x magnification; scale bar = 50 nm). Estimation of Cross-linked Nanoparticle Slze We estimated the number of polymer chains in the cross-linked nanoparticles using basic algebra. For example, one polymer chain with Mn = 183 72,000 g/mol (from GPC) weighs approximately 1.19 x 10499 (Mn x 1 mol/ 6.02 x 10?3 chains). If we assume the polymer has a 3 nm radius (6 nm diameter measured by TEM) and that it can be approximated as a sphere of density (p) = 1 g/cm3 then the mass of a dense particle would be 1.13 x 10"9 g (m = V x p = (4/3)rrr3 x 1 g/cma). Thus the mass of a 6 nm sphere (1 .13 x 10‘19 9) ~ the mass of a single polymer chain (1.19 x 10199). Conjugation to Cross-linked Nanoparticles To explore the possibility of loading cross-linked unimolecular micelles with covalently attached biologically relevant drugs and to confirm the core cross- Iinking, we used 3-azido-7-hydroxycoumarin (Scheme 41), which transforms into a fluorescent dye after undergoing a 1,3-dipolar cycloaddition reaction?51 HO OH O O O O + NaOAc, Ac 0 \fl/ 0 U AuACOOH ——2> o / Jk CHO IZ /0<\/\o)’\/O\/\ N’N, 1. HCletOH (2:1) 3 \ N 2. NaN02 O 3. NaN3 O H “0 1-m 0 O n % = HO 0 o CuSO4-5H20 m N3 sodium ascorbate fluorescent nanoparticle Scheme 41. Synthesis of 3-azido7-hydroxycoumarin and click attachment to poly(PGL[EG75X25). 184 Figure 59. Reaction of 3-azido-7-hydroxycoumarin with polymer micelles. Vial 1: polymeric micelles and dye in the absence of “click” catalyst; Vial 2: polymer micelles and dye after the “click" reaction; Vial 3: dye and components for “click” reaction without polymer micelles. Using poly(PGLjEG75X251) as the substrate we carried out the click reaction with an equimolar mixture of the dye precursor and 1,5-diazid0pentane. The resulting particles fluoresced brightly indicating successful reaction of the dye precursor. We monitored the reaction of nonfluorescent 3-azido-7— hydroxycoumarin with polymer micelles using a hand held UV lamp (Figure 59). The ability to load previously characterized cross-linked particles may be advantageous, especially in standardizing drug delivery. Therefore, we investigated a 2-step process where we initially cross-linked 5, 10, or 100 % of the available alkynes on poly(PGL[EG75X25]) with 1,5-diazidopentane, and then carried out a second click reaction using an excess of the dye precursor. Reaction of the dye precursor with residual alkynes was evident from the luminescence of nanoparticles (Figure 60), providing proof of principle for loading cross-linked particles. We measured the fluorescence intensities of 1 x 1045 M nanoparticle 185 solutions to assess the relative dye loadings, and as expected, particles prepared by the 2-step process had lower loadings. The dye precursor could also provide a method to count the number of available alkyne groups, which is important for drug delivery because each represents a site of drug attachment. 2 , cross-linker plus dye precursor 1.5 intensity x 10-5 0.5 5 % cross-linker then dye precursor control 10 °/o cross-linker then dye precursor 450 470 490 510 530 550 wavelength (nm) Figure 60. Fluorescence intensity of 1-step and 2-step click reactions on poly(PGL[EG75X25]) substrates. (From top to bottom): a single click step using an equimolar mixture of the dye precursor and 1,5-diazidopentane (blue), a two step process where 5 % of the remaining alkynes were cross-linked, followed by a second click reaction using an excess of the dye precursor (red); the same protocol, with 10 % cross-linking (orange), and the same protocol, with 100 % cross-linking (green). 186 Controlled Release from Cross-linked Nanoparticles Controlled drug release improves drug efficacy, reduces toxicity and improves patient compliance and convenience.123 Cross-linked nanoparticles are attractive for long-term controlled drug delivery since the release rate should be tunable by controlling the degree of cross-linking in the particles. We dissolved azobenzene, a surrogate for hydrophobic drugs, and poly(PGL[EG75X25]) in a minimal amount of acetone. After dispersing the solution in Milli-Q water, and removal of the acetone in vacuo, we confirmed encapsulation of azobenzene by its characteristic absorbance at 320 nm. We then repeated the experiment, using cross-linked poly(PGL[EG75CL25]). To track the release of azobenzene from the cross-linked and control nanoparticles, we placed the nanoparticle solutions in dialysis tubing (MWCO = 12-14 kD) and dialyzed with Milli-Q water. Removing aliquots of the nanoparticle solution from dialysis bag and measuring their UV-vis spectrum allowed us to monitor the continuous release of encapsulated azobenzene over 10 hours. As shown in Figure 61, the azobenzene release rate from the cross-linked nanoparticles was ~1/2 the rate measured for the noncross- linked control. These data are consistent with successful cross-linking, and proof of principle for long-term controlled drug delivery using a mixture of cross-linked nanoparticles with different release rates. 187 normalized absorbance at 311 nm 10 100 1000 time (min) Figure 61 . Release of azobenzene from nanoparticles suspended in Milli-Q water at room temperature (1 mg/mL). Cross-linked nanoparticles (I) show 2x slower release than noncross-Iinked nanoparticles (D). The loss of dye from the nanoparticles was monitored at 311 nm by UV spectroscopy. Biocompatibility For in vivo applications, nanoparticles must be biocompatible and degradable. In a single click reaction, we cross-linked the poly(PGL[EG75X25]) and appended a rhodamine dye to the nanoparticles. After dialysis to remove residual dye and click reagents, we cultured cortical astrocytes in a 10 pg/mL solution of dye-labeled nanoparticles. After 24 hours, the cells were washed to remove particles that had not entered the cells and then imaged by confocal microscopy. Figure 62 shows that the nanoparticles readily entered the cells, probably due to their small size and their PEG exterior, which screens particle/protein interactions. 188 The cells remained healthy for more than 30 days and showed no signs of nanoparticle toxicity. These data are in agreement with the viability assays on various cross-linked nanoparticles, which show no detectable toxicity for nanoparticle concentrations as up to 3000 ug/mL (Figure 63). Figure 62. Cortical astrocytes cultured in media containing 10 ug/mL of cross- linked nanoparticles for 24 hours. The red color emanates from rhodamine dye covalently linked to the nanoparticles that have entered cells. 189 1.2 21- m 30... «‘3 EGA- : 30.2» 0 \Vofi\°¢V§’@\’§’&V®V6‘V§/¢‘V§ *0 {y}- <83. 06 Q ‘9 ‘9 \Q Q 0°} 5% 6‘5 ‘9‘ 6%VVQQ§%%§Q°§¢P@\°§? +§$$~§~§ »3 ‘W" (O’oVO’ 00’0’ 000 6060 Figure 63. Evaluation of nanoparticle toxicity on cell viability reported in relative fluorescence units (RFU). Noncross-linked nanoparticles (poly(PGL[EG75X25]), NonCL NP) and cross-linked nanoparticles (poly(PGL[EG75CL25]), CL NP) correspond to polymers mentioned in the text. LipoFectamine (LF2k), linear poly(ethylene imine) (Mn = 25,000) (LPEI), and poly(L-lysine) (Mn = 150,000) (PLL) were used as controls. LF2k is reported as dilutions of the purchased solution. Conclusion Combining click chemistry and the unimolecular micelle approach provides a simple route to functional organic nanoparticles. The robust, stable particles encapsulate and release hydrophobic molecules and contain pendant reactive sites that can be used for cross-linking or covalent drug attachment. The nanoparticles are biodegradable and can be used in biomedical applications. 190 Experimental Section General. Unless otherwise specified, ACS reagent grade starting materials and solvents were used as received from commercial suppliers without further purification. DMF was dried over activated 4A molecular sieves. Ion exchange resin beads (Amberlite® IRC—748) were purchased from Aldrich and washed with DMF prior to use. 1H NMR (300 or 500 MHz) and ‘30 NMR (75 or 125 MHz) spectra were acquired using either a Varian Gemini 300 spectrometer or a Varian UnityPlus 500 spectrometer with the residual proton signal from the CDCI3 solvent used as chemical shift standard. IR spectra were taken with Mattson Galaxy 3000 FT-IR. Polymer Characterizatlon. Polymer molecular weights were determined by using GPC-MALLS at 35 °C using two PLgel 10p mixed-B columns in series (manufacturer-stated linear molecular weight range of 500-10x106 g/mol). The eluting solvent was THF at a flow rate of 1 mL/min, and the concentration of polymer solutions used for GPC was 1 mg/mL. An Optilab rEX (Wyatt Technology Co.) and a DAWN EOS 18-angle light scattering detector (Wyatt Technology Co.) with a laser wavelength of 684 nm were used to calculate absolute molecular weights. UV-Vis spectra were recorded with a Cary 300 Bio WinUV, Varian spectrophotometer. Fluorescence was recorded using a Fluorolog-S (Instruments S.A., Inc.) fluorometer. All samples were diluted to 1 X 10“6 M using Milli-Q water, excited at Lax = 440 nm, and the fluorescence emission spectra were observed from 450-500 nm. Dynamic light scattering (DLS) data were obtained using a 191 temperature-controlled Protein Solutions Dyna Pro-MS/X system. All samples were filtered through a 0.2 pm Whatman PTFE syringe filter and allowed to equilibrate in the instrument for 15 minutes at 25 °C before measurements were taken. The uniformity of the particle sizes were determined by a monomodal curve fit, which assumes a single particle size with a Gaussian distribution. Surface profile measurements were performed with a Pacific Nanotechnology Nano-Fl atomic force microscope in close contact (oscillating) mode to generate height images that were not altered other than a simple leveling procedure. Silicon tips with a spring constant of 36 N/m, tip curvature of 10-20 nm, and a resonance frequency of 286-339 kHz were used for all experiments. The polymeric micelles were dispersed in Milli-Q water and the solutions (typical concentration, 20-300 rig/mL) were spin coated at 5000 rpm for 40 sec on freshly cut silicon substrates. Heights were determined using AFM by taking the average of 20 particle heights. The lateral size could not be determined due to convolution effects created by the tip?“205 TEM micrographs were collected at 270,000-370,000 x magnification on a JEOL-1OOCX transmission electron microscope. TEM samples (typical concentration, 20-300 rig/mL) were dropped onto nickel grids pretreated with Formvar and allowed to settle for 2 min before removing the excess solution. The sample was negatively stained with either a 1 % phosphotungstic acid (PTA) stain or 2 % uranyl acetate (UA) stain prepared with deionized water. 192 W N3 N3 Synthesis of 1,5-dlazldopentane. Sodium azide (1 .65 mol, 107 g) was added to a solution of 1,5-dibromopentane (165 g, 720 mmol) in 600 mL DMF. The solution was stirred at 60 °C for 9 hours and cooled. The mixture was poured into 200 mL water and extracted with ether (3 x 75 mL). The combined organic layers were dried over M9804 and the ether removed. The 1,5—diazidopentane252 was purified by vacuum distillation (60 °C, 5 mtorr) to yield a colorless liquid (105 g, 680 mmol, 95 %). 1H NMR (500 MHz) 6 3.29-3.26 (dd, 4H, J: 6.84, 6.71), 1.64-1.53 (m, 4H), 1.48-1.43 (m, 2H). 13c NMR (125 MHz) 6 51.21, 28.42, 23.936. 3 \N O A O H O 1-mo n N Ne, N R0 General procedure for cross-linking. The desired amount of PPGL-g—mPEG functionalized polymer dissolved in acetone (< 0.5 mL) was slowly added dropwise to stirred ice-cold Milli-Q water in a round bottom flask containing sodium ascorbate (1 eq with respect to the remaining acetylene groups) and CuSO4-5H20 (0.5 eq with respect to the remaining acetylene groups). The solution was stirred 5 min before a solution of 1,5-diazidopentane (10-100 mol% with respect to the remaining acetylene groups) dissolved in 1 mL acetone was added dropwise. The solution was allowed to stir under nitrogen 2 d and was then transferred to 193 presoaked dialysis tubing (MWCO = 12-14 kD), and dialyzed against a 4:1 solution of Milli-Q water and acetone overnight. To remove the copper, the nanoparticle solution was concentrated, redissolved in DMF and ion exchange resin was added (400 mg) for 24 hours. The nanoparticles were resuspended in Milli-Q water following DMF removal. Cross-llnked PPGL-g-mPEG nanoparticles (poly(PGL[EGaoCLzo])). 1H NMR (500 MHz) 6 7.8-7.4 (b, 1H), 5.652 (b, 1H), 4.5-4.4 (b, 1.6H), 4.3 (b, 3.2H), 3.9- 3.7 (b, 1.6H), 3.6-3.4 (b, 9.6H), 3.3 (b, 2.4H), 3.3-3.2 (b, 2H), 1.4-1.2 (b, 1.2H). HO 0 0 (11,, Synthesis of 3-azldo-7-hydrow-coumarin.” A mixture of 2,4- dihydroxybenzaldehyde (2.76 g, 20 mmol), N-acetyl glycine (2.34 g, 20 mmol), anhydrous sodium acetate (4.92 g, 60 mmol) in freshly distilled acetic anhydride (100 mL) was refluxed under nitrogen for 4 hours. The mixture was cooled and poured onto ice to provoke a yellow precipitate. After filtration, the yellow solid was washed with ice water and added to a solution of concentrated HCI and ethanol (2:1 vol %, 30 mL). The solution was refluxed for 1 hour, cooled to 0 ° C and ice (60 g) was added. After sodium nitrate (76 g, 40 mmol) was added in portions, the solution was stirred 10 minutes and sodium azide (3.91 g, 60 mmol) was added. The mixture was stirred an additional 30 minutes and filtered. The collected brown solid was washed with water and dried under vacuum to give 0.5 g (2.8 mmol, 14 °/o) of 3-azido-7-hydroxy coumarin. 1H NMR (500 MHz) 6 7.57 (s, 1H), 7.46 (d, 1H, 194 J= 8.55), 6.81 (d, 1H, J: 8.55), 6.75 (s, 1H). “*0 NMR (125 MHz) 6 160.3, 157.3, 152.7,129.1, 127.8,121.1,113.7,111.3, 102.0. Covalent attachment of 3-azldo-7-hydroxycoumarln. Using PPGL-g-mPEG and the “click” conditions described earlier, we carried out two types of click reactions: (1) a single click step using an equimolar mixture of the dye precursor and 1,5- diazidopentane; and (2) a two step process where 5, 10, or 100 °/o of the remaining alkynes of PPGL-g—mPEG were first cross-linked with 1,5-diazidopentane, followed by a second click reaction using an excess of the dye precursor. Azobenzene release from cross-linked nanoparticles. Encapsulation: to encapsulate azobenzene the polymer, azobenzene, 1,5-diazidopentane, sodium ascorbate, and copper(ll) sulfate were dissolved in acetone (<1 mL) and the acetone solution was stirred for 2 days and then was added dropwise to Milli-Q water. The acetone was removed and the solution was filtered to remove any unencapsulated azobenzene. We confirmed its encapsulation by it characteristic absorbance at 320 nm using UV-vis spectroscopy. Release: azobenzene-loaded nanoparticles were placed in dialysis membrane (MWCO 12-14 kD) in stirred Milli- Q water. At allotted time intervals, the azobenzene solution was removed, filtered to remove any insoluble released azobenzene, and monitored using UV-Vis. The same procedure was followed for the noncross-linked micelles, except, 1,5- diazidopentane was not added to the reaction mixture. 195 Synthesis of Rhodamlne Derivative. DCC (459 (mol, 95 mg) in 2 mL anhydrous CH20I2 was added to a mixture of Rhodamine B (418 (mol, 0.20 g). 6- azidohexanol253 (835 (mol, 0.20 g), and DMAP (84 pmol, 10 mg) in 10 mL anhydrous CH2CI2. The mixture was stirred at room temperature for 8 hours under nitrogen and the insoluble urea was filtered from the solution. The CH20I2 was evaporated and the dye was triturated in hexanes to remove any unreacted alcohol and then dried under vacuum to give (0.231 g, 385 pmol, 92 %) of N-(9-(2-((6- azidohexyloxy)carbonyl)phenyl)-6-(diethylamino)-3I-f-xanthen-3-ylidene)-N— ethylethanaminium chloride. 1H NMR (500 MHz) 6 8.28 (dd, 1H, J: 8.0 and 1.5), 7.62 (dt, 1H, J: 8.0 and 1.5), 7.75 (dt, 1H, J: 8.0 and 1.5), 7.31 (dd, 1H, J: 8.0 and 1.5), 7.10 (d, 2H, J = 9.5), 6.94 (dd, 2H, J: 9.5 and 2.5), 6.87 (d, 2H, J = 2.5), 3.62-3.60 (d, 2H, .1 = 6.7), 3.26-3.24 (t, 2H, J = 6.9), 1.32-1.30 (m, 8H). “*0 NMR (125 MHz) 6 165.2, 158.7, 157.7, 155.5, 133.3, 132.9, 131.9, 131.2, 131.2, 130.3, 130.2, 130.1, 114.2, 113.4, 96.3, 65.5, 51.4, 46.0, 32.5, 28,8, 26,5, 25.3, 12.5. Toxicity Studies. HeLa cells were plated in 96-well black-clear bottom plates at 10,000 cells/well in 100 (1L of culture medium lacking antibiotic and grown for 22 hr at 37 °C. Cells were ~70—80 % confluent based on microscopy. Nanoparticle 196 solutions (50 [.11 total volume, when required dilutions were prepared in OptiMEM, Invitrogen) were added to the 100 pL of culture medium. The cells were incubated for an additional 22 hr and then 20 pL Cell Titer Blue (CTB) reagent was added. Fluorescence was quantified in a fluorescence plate reader 2 hr after CTB reagent was added, according to the manufacturer’s instructions. LipoFectamine (LF2k), linear poly(ethylene imine) (M. = 25,000) (LPEI), and poly(L-lysine) (Mn = 150,000) (PLL) were used as controls. LF2k is reported as dilutions of the purchased solution. 197 Appendix NMR SPECTRA 198 see 3... 88 mod afim mfim ON.m mud om.m mm.m Sim mend P _ PL b h n p h — — — n w — — — — p — — p b _ — h — — p — p h n p — F p _ — — h - h b — .— h P b — _ p n P — p h h _ b Ed co. mom and x _ 5n— Om_ <.__ (d o9 n o.m-mfim mom a fiméh . E59 Em 2: Beam. .82.:sz 2000 E028 I, 320:2 0.. mm 9293th ~15. com 5:639”. oom man—3E: cmtm> EoEasz :x; O x O Ow CE I On On /_\F0m O 0 Appendix 1. 1H NMR spectra for the determination of F1 for the copolymerization of rac-lactide and rac-ethylglycolide. 199 o o c x; O x O o e o s 0m 0 O o 5.0 cow 08 $6 (Al On:— <1__ <1; 6.000 Emzom cm? 2 NM- Ya I. 86.82 2: n 3.58 0.. mm oSfiEanh 08 n 5.0“.“ N 6:339 com gigs: cm_._w> EmEame 638020 J —— Appendix 2. 1H NMR spectra for the determination of F1 for the copolymerization of rac-Iactide and rac-isopropylglycolide. 200 n .w 5% or me om mm on mm 3. me on m P _ _ p F P L p _ _ p _ _ _ _ _ FL » — p _ p _ — p b _ p _ _ e .L _ p p . b — P _ p _ _ _ _ _ L .m fir/Nifr . r 13. 11 r J. 1 m 0 \2 1|. 9 o n m ..m .m _ _ 1 ._ -..I. a new m F e - - .m C xiv O o x O W I1. 0 D O m ‘0”. calla. O a O m .m r o O O r m a d e ..m .m e m £8 2: mam 58 m D. Su— w::. 5. 5* M M N 58 E628 com a 3-3 H I. 63282 mom 6 3-3 3 0° mm QBEoQEoP mac n o.m-m.m W ~12 com 5:269“. - n E .296 £53 Em m. com wedge: cmtm> E06265 .mo_Eoco M. of rac-Iactide and rac-n-hexylglycolide. 201 ex; 0 x O Om CE I on on flow/3% O 0 mod cow omm 05.0 _ 5“. 6:6. S. 5. m.000 Eozow o S o m. Ta.— I. 8282 2: e 8.4-6.4 06 mm mSEEanP omm a F.m-m.m N 35:09 :5. 8m n. .E_ .296 €53 :Em oom 3.33:5 catw> EoEasz .mo_Emco Appendix 4. 1H NMR spectra for the determination of F1 for the copolymerization of rac-lactide and rac-cyclohexylglycolide. 202 can mm 3.. m8 «.8 on ad 3. we 3. 3. Wm F|. o c x; O o x O O n O m Om. I O a O m o O O omd cow mmm omd Sn. 42 (d (d 1|. some 62.8 6 new a I_ 25.032 com a mdéd 00 mm meamcanmP mmow n m.v-v.m NIS. com 33:63.09“. .E. .mcm.m E53 Ecm _8_Eezo com 3.13:5 amt—m5 EoEEfiE Appendix 5. 1H NMR spectra for the determination of F1 for the copolymerization of rac-lactide and rec-phenyllactide. 203 E& F N m x. m - p - _ — _ b b _ _ _ p - p b _ - _ _ _ _ _ _ _ _ _ P n .5 Im E om. 79... :4 E me. 7%.. SR IN 3 5m Io w mud e In 6 8.8 New I. . mm... . Ea... £5 .NI. 3 E. 65...... .6255 M £00 £000 EmZom mIOO\:/\/>>¥Q/OOmI I. 2.20:2 o m 06 mm 9396aEmH NIS. com 5:039“. Dom _c_Emmv :w_..m> “cm-teams. Appendix 6. 1H NMR spectra of 9,9-dimethoxynonanoate. 204 Eda 0w 0% 8 cm 00.. our 0:. 00.. _»_L_.-b____b_—Pp-____~b__.pFL~_pppp..___-_~_.b__—__p.__P__—___rb_..__~____.-p_—._____plri_1 11151; )3 0 £00 fooL./\/\/\/L/oof N.¢N mfim QwN Raw Nam me 0.3 m. ..m mNm View N445. Eng :20 6269.0 £000 09 06 mm ~Is. mu com E58 8.55 .5200 96.032 9298th 55:09”. EmEame Appendix 7. 13C NMR spectra of 9,9-dimethoxynonanoate. 205 IO In: E N44... Iv E 3.70m; 2 I0 m mud mmd IN 3 wmd «06.006 IF no «we . €3.55 .~I< .5 65.6. 3.520 m m IOO s)\/\/\/\_/ a Com: m o n.000 .5200 T... $39032 0.. mm ocamcanmk NIS. com 33:023.”. com _:_Emw :m_.m> EoEEfic. Appendix 8. 1H NMR spectra of 9,9-dimethoxynonanol. 206 End ON 0* 00 mIoo IO§OOmI 00 00—. our 0.3. 00—. .__prpp.—»__.___ppfirp___r»»p—_p_»____hr._L._..ppfFLFFFLb».—_.p_ph»._—.____3P__F_pr.__... mém NmN omdm mmdm v.0N Yum 5N0 m.Nm 0N0 mayor .58.. £5 .8565 M".000 0? 06 mm NI5. mm 000 .5960 catm> 59.00 96.032 9393.23 .6539... EoEazmc. Appendlx 9. 13C NMR spectra of 9,9—dimethoxynonanol. 207 can. F N m 0 n m —1 p p _ p — p p b — — . . p — r r FF — — _ _ - — _ _ P L P P — ~ g m 8 0I00 \U/\\—/OOmI 805620 «.000 Eo>_om 09 26.032 0° mm 9398th ~15. mm 55:09”. com EEmO cmtm> EoEzsz Appendlx 11. ”C NMR spectra of ((9,9-dimethoxynonyloxy)methyl)benzene. 209 5% F N m w m o x. m m 210 P r p _ _ p p b h h — p p — — _ ...1, 2 iiijv 7 I. . m o O m U \< EmEEfiE Appendlx 12. 1H NMR spectra of (9-benzyloxy)nonanal. :0 m. a o o In E N.....¢.—. m”.000 Emzom IN E v; . m IF 95.022 IN E mw 0.. mm 9.39.886... IN E mw ~Is.oom .6539". mad IO u mN com .c.Emo cacw> .coEzsz . mmd IN . ed 39%... I. 8 m... IO m u m . 33.5.5 .NI... .5 mega .3596 . 211 Appendlx 13. 1H NMR spectra of 10-(benzyloxy)-2-hydroxydecanenitrile. Eng F N m V m c h p — p p p L — p p p p — L p b — .— b n - — _ PL! — — n b .m m 1.. a v m n O\<_om I w n Nd I. 320:2 Im E NEE 0.. mm 9398:.8. ANI. 5 .E. .956 Eng. Ecm NIS. com 35:09“. . . .mo_EwIO oom .c.Ec0 chm> EmEasz Appendix 14. 1H NMR spectra of 10-(benzyloxy)-2-hydroxydecanamide. 212 . E... N N v m o N — .1 p p F _ p h _ — p _ p — p h _ p t— - h _ — — . — p h p _ _ _ . 1.. a o u ”Goo .828 E& «N... NNE I, 8282 __________________________1:_____ I. _ __ _ 0° mN 92.93%... Io E Nil -1). _. .. 1-11 ~IS. com 55:09”. I... E 3 .. S. < 8m E_Eoo cm_.m> EmEaaE IN E o. n I. E E m I. E 3 mm... IN . mm B N NN N I. on N v I O IN .0. m... o o OI 1.1 Im E Ni; O a . .EE. :5 o o .~I¥ E_ Rama _8_Eoco Appendix 15. 1H NMR spectra of 10-(benzyloxy)-2-hydroxydecanoic acid. 213 Eng hprLP—pbr___-h1rrrbp_—r:P-bp___-_-_[L1~1—bpnh—-——n—-—F——_—_._-—-—_—_~—1P1nP-—PLp~—~Php—-—-r— Camp mméN mo.©N oodN NNdN MNdN N®.mN 00.3” mwdw 9.05 atom VNNN vaNF moNNw mmdNF 9.de mod: ON 3 Do «.000 09 0° mN I: N. 08 SEES 55> Eczom $0.022 SENEQEmE 5:309”. EmEssz .EE. 5% _8_Em_.o ow cow ONw 03. 8.. IO OI O\<< EmEEfiE IF w a... 1 I. E 2 O 1 Im E 3-: O a O O . . .EEZEN _ .NI... E. mama _8_Emco O O m Appendlx 17. 1H NMR spectra 3-(8-(benzyloxy)octyI-1,4-dioxane-2,5-dione (BOG). 215 3 8 8 3 /\/>>\OHM/H\O 8.. de odN QwN P.mN NdN NmN mdm N60 v.05 QNx. 9mm vNNF @NNF 0me 0.8.. ado? odmw .58. ...cm _8_E20 ON? cc? 8' EF1EL_»___rrbp_.._b_»r»~—_E_p—PL1PPPE»_E—___p___.__h____»b.._b._pr»pp__..ppL»____»pp_prp__ £000 09 Do mN I: N. 08 853:: cans Emzom 2.20:2 SENEmQEmH 5:032“. EmEEEE Appendix 18. 13C NMR spectra 3-(8—(benzyloxy)octyl-1,4-dioxane-2,5-dione (BOG). 216 u C o I Io. E N..-¢.. m IN E m4 IN E m..-o.N mm... IN . .8 IN m m... IN E mééé 0.000 Em>_om I P E F.m-N.m I. 32032 Im E N. 5v. .2. 0° mN Qaumcmanh NIS. com 3:309“. 31. a ..E 3.66 .Mmmwwhwm com SEES cmcm> EmEEfiE Appendlx 19. 1H NMR spectra of poly(BOG). 217 Ema .__..___~....__._.VP__._EEEF~_.___#.....ppr_er__..___.p..___._bh.___._~__p..#.._.__.___._prrL 0N QVN ...0N 0.0N o.mN mdN NaN 0.00 N00 Yon 0.Nn 0.0m vKNP 0NN.. 0.0N.. 5.00.. 0.00.. 0.00P .EE. 5...... _8_Eoco av 00 o0 «.000 0n. 0.. 0N NI: .N. 08 952.5 Em_a> .5200 320:2 mcamcmaEmh 5:33.“. EmEzme oow ONF ovp 00.. Appendix 20. 13C NMR spectra of poly(BOG). 218 u m - EVE... - _ c O . O 0 .. O a .0m I\A O I O m - _ o o - IN E N..-.... o IN E 3.... IN E N..-o.N p .3 IN . ...N O 0 IN w m... IN E 3-..... IN E ..m-m.m In E N5... . ..ccwo . . €3.55 #5.. E_ .29.... 525.6 (Jr—”wom *0 030.. 000* omnom 0 CO 00000 wEQEngmm .600 I. o. N Is. 8m 8m maize: cm_a> .5200 96.032 9:8.maEm... 33:52“. EmEEfic. Appendix 21. 1H NMR spectra of poly(BOG-co-LLA). 219 EQQ 0N CV 00 00 00 p ON F ov .. 00 w _.r.-_r..____.___».——»...—.pP.—_._rhph_.—..~.b...._._b_..b.._...k__...—».E.F#p..—_._EPEPE. 1:. Z... - - . 0.0.. 0.09 méN ...0N 0.0N CNN mdN NaN 0.00 r I No... 0 O 0.”. . o . N.mp I \AO/_\../O\_/.L.A OJ; v.2. N,NN Em. ...NN. 3N. N.NN. .68 .528 O Ndflw 0... 000-002 98. 0.00. on 0N QBSSENE NIS. 0N. .6539“. Eng. :20 com 855...: cm..m> EmEEfiE _8_Em..o 220 Appendlx 22. 13C NMR spectra of poly(BOG-co-LLA). c O I09 Om Om. O o 0 IO En... .. N _ . _ _ _ _ _ . . _ . trill-IISIJ 11 1 . Io. E N..-¢.. IN E 0.-.... IN E 5.-.... 0.0.3... IN up 3-..... IO 0 06 IN E o.v-m.v-o.m I. E ..m-N.m NI... E. .95. .mmmwfim 0020 .5200 I. 390:2 0o 0N 93.9883 NIS. com 35:09.“. 000 2.52:5 cm_.m> E9552. Appendlx 23. 1H NMR spectra of poly(HOG). 221 Eaa .F..—pp»._.p.b_._FL—_...__..._Lpr__h__._._.P_._p.______.._p__..-_.p.___..h—...-F._..__-.r ON wém New m.mN mdm Qmm mdm mNm 5.8 You NNN ohm 0.00—- odor Eng 55 _8_Em;o 0* cm cm 00.. ON? 0»; 00—- 0 Or. 0920 09 0o mN NI: mNF 8m 8:35 cm_.m> szom 26.032 I O 9 39383 35:09“. 25:5me Appendix 24. ”C NMR spectra of poly(HOG). 222 O 0: On - _ O _ Enema In E Nil IN E 3-3 IN E mfoN m3 IN I cm IN E ova...- Im E ..m-m.m Ecmo II:- .E_ .955 fimwwfim (Jr—”00m *0 02m; Umww Dmuom m :0 wawn mar—mECQWmm O \Ji-bm O m 0 IO gone I. oo mN NI: 0% 8m 9:35 ENE-S .528 30.022 9388th 35:09“. EmEPzwc. Appendlx 25. 1H NMR spectra of poly(HOG-co-LLA). 223 600 ON 9. ...pr.|___—.p________.._r______.__._...___._.—_._._..___pp_p~..___b_.._._p_—.__E_.__0___.p 0.09 NVN Qmw 9mm mdm 2mm wdm 5N0 Now Now Odo You QNN 9mm mémw mam? .EE. :5 $28020 om ow cop our 0.3 O O O O R fife/1. o - O 3000 .0028 02 030.032 00 mm 0.3.0.00E0E ~15. mm? 0.0.5309”. com 03.03.03 cm..m> ENEEEE 0 IO ow? .Om Appendix 26. 13C NMR spectra of poly(HOG-co-LLA). 224 E00 . N n v m o n w p P L p — p E r p — _ h F p — .r Fb F — - — _ — — h p _ p F P — P — h b h p _ p p h p 31.-r - l .- 4,- 0 a m ..c: 02.00. BEE-.0. IN. E «.3... :8 E 170.. 1 I c _ IN E w..-o.N O O IN E ..v 0 .. O a .0m vwo mm 0 m I\m O .. O\m/=\Y . .h . F F ..V I 0 .- _ O I IN E mas-mi . INN E o.m-m.m . .3355 3:... E. Econ _8_Eo..0 o E u 0.000 Eczom I. 030.032 0 O 00 mm oSEEaEm... E. NIs. com 5:03.09“. 0 _ com 0333...: cm_.m> EmEEfic. _ {f0 .. .050 :22 o. u 09.0.85 5.. .0 an. (...HGOm .6 0:0. 000. 8.8 a cc 0800 m.c0Ecm_wmm Appendix 27. 1H NMR of poly((HOG—co-LLA)-g—LLA). 225 E00 N N v m o \- 0 P — b p — _ p p p — _ p — p - _ p — _ p h P — r p P — — p - p p — _ - p p g .- . I. a ..c: 00000. .0:.E.0. In . 0.0.0.0 u I _ :3 E N.. c O . O IN. E N..-.-.. O E O a .0... I00 E .10.. I\ O .. O m 0 O IN E w..-0.N r _ 1 IN E ..v 0005.0. I. 00 00 IN E 0.0-0.... 0 INN E 0.0-Om E 0 . .3355 .NI... E. .29... _8_E0..0 O O c a :0>0 .000 . _ m 0 _ I. 030.03z _ 00 N 0.3.0.00E0E c 0 . NIS. com 500300.“. . .0000 com 03.03.03 :0_.0> E0E3:mc_ o. N 0000005 <._.. .0 n5 SHOOm .0 0:0. 000. 8.8 0 :0 00000 0.:0E09000 Appendlx 23. 1H NMR of poly((HOG-co-LLA)-g-LLA), stearic acid terminated. 226 E00 . N m 0 0 \- . p — F — p _ . _ — — L p p — E — h p — p P h h — p . _ _ p b . _ _ h P L T rm - a x 0 0 0 \ 0 \ fi 0 O moN I. . ...N mmN . 5N . 0.000 .0028 IN E mm N . . . . I. 030.032 mm 0. mo \- 00 00 m I. 0.. mN 0.3.0.00E0... 00.0 . 0N0 00 NIE 000 500300.“. E _ ANI. 0 .E. .0090 ..00_:..%....n.um com .EEmw :0_.0> 2.0:.ng Appendix 29. 1H NMR spectra of propargylglycolide. 227 Egg ON 0.. 00 00 8. 0N. 0.: 00. ...._...._...__..r.__..._....—._..._...._...E_.._.—7..._.rr..____._...._. __..._._..— _ ..._.... NN. .N \ O O 00.0N \ \\ NaN. 0 O 00.0. 0...: N060 00.0... 0.000 E0200 00.00. 00. 030.032 03.0. 0.. 0N 0.30.00th .E00. :20 I... 0. 500300.... .00.E0..o 000 .:_E00 :0_.0> E0E3=05 Appendix 30. 13C NMR spectra of propargylglycolide. 228 E00 . N 0 n 0 P p _ _ b P _ h!— _ p F p P — — — h E b — P L P _ 1 1‘ é 3W1 0 fl 0 O o Om O 0 fl I . 0 .o.N-0 ..N 0.000 .00200 IN E 03-8.0 I. 0:282 I. a 00.0.0.0 0°0N 920.00%. €080.00 NIE 000 08:080.”. ..0. .0090 . _8_Em._0 80 0200.5 00...; E3202 Appendix 31. 1H NMR spectra of poly(propargylglycolide). 229 .000 ON O0 O0 00 OO. ON . O0 . O0 . —....p....—.T.._.0.._...._...._....—...._....—....—...._....—......E..—._..F...__...._.... ' pits?! uh I. V r Irrbrb'll tr? I LID-Ir ‘5’ LI In, D,“ r 7| V psl [I'll'r F 'D 7 {bl {I}, Pl'b LL-r' I D 111 I!“ 1‘ A ddl#) {III .I J {I}! 1411 14.111I ‘14 11 441‘ All 41 n 1111' 1“! ‘ 4 4 ‘ A 11‘ 14 1‘ 00. .N 00. .N Rd. 0 00.0. I 00. K 00. E 00... 00... 00... .0... 00.00. 230 0.000 00020.0 00. 030.032 N000. N000.- V0.00. .EE. £00 .00_E0..0 000 0200.03 0000> 0.0.02.0... 0° 0N 0.3.0.00E0E NIS. 0N. 0.0003000“. Appendlx 32. ”C NMR spectra of poly(propargylglycolide). E00 . N 0 e 0 0 N 0 p P F E —r h b h . _- — p p p h . . . . _ p . _ . _ p _ . .- — . L n L — p h . P _ p _ — r _ 0 o o o m 0 0 0 I0 E 3.0-00.0 I 0000 .8200 IN E 00.0-00.0 I. 030.032 I0. E 00.0-00.0 0 0° 0N 0.3.0.00E0E N 5003000 ..0. .0090 €080.00 Is. 000 n. .00.E000 000 03.05.03 0000> .00E3..00. Appendlx 33. 1H NMR spectra of 16-azido-2,5,8,11,14-pentaoxahexadecane (mPEG). 231 E00 cm 3 8 8 8. om? 9: 8P FF».____..—_»~_—____—.P0»Lh...bhpkprppbp_p|rLb_ppL__.».p_0.____.__—__.r_...0—p_0._p#r__rppp_ Appendix 34. "‘0 NMR spectra of 16-azido-2,5,8,11,14-pentaoxahexadecane (mPEG). 00.00 00.00 0 mmdo /O\/\O/\/O\/\O/\/O\/\ Z 3.00 N000 V0.00 00.00 00.00 N000 0.000 Emzom 00.00 02 030.032 00. K 0° 0N 003000000000. E08 :20 ~12 09 50030000 .028ch 000 03.0b_:: 00:05 00053000. 232 :0 0 00.0.0 :0 0 0.0 0 :00 0 00.0.0 :0 0 0.0.00 :0 a 3.0.0. I0! T: a 0.0-0.0 T: 0 00.0.0 .2: .0090 fimmwwfiw m.000 60200 I— mam—0:2 0° 0m 00300000800 ~15. 000 5003000“. 233 000 0203:: :0_.0> 0:083:00. Appendlx 35. 1H NMR spectra of PPGL-g-mPEG. E00 00 00 8 00 8. 8: 03 000 __.________.._—.__p______.________Prp.___.rr__.b__~__._._.__~__F___0»—___._...___rh____.»_ 33 m0. FN ..mNN 00.00 00.00 5:00 50.05 05. ..N 00.0mw 0 F. S; nmfiop 0:000 030 02520 0000 50200 02 0:282 00 mm 0030000600 ~12 03 5003000.... 000 03.03:: :0_00> 000E300... 234 Appendlx 36. 13C NMR spectra of PPGL-g-mPEG. 800 P N 0 0 F _ p L p h L b p p — h p _ _ _ _ — p r 0 _ p p p - {ll-gi- 0 00.0 I0 0 00. 700. F 00.0 I. a 00.... 0N.N €00 IN 00 swan-0N.N :0 E 00.0-me 00.0 $0.0 IN 00 00.0.5.» SN .00 IN 00 00.0.3.» . £000 030 3.0 \- E_ .0090 096000 0 0 0 O 0000 20200 .0 0 O 0 0 IF 030.032 0 0 Z A-Z oo 0N 00300000600 0 O— ~I_2 000 500300.“. 0 0 000 03.0bEa :0_00> «00:03:00. Appendlx 37. 1H NMR spectra of 2—bromopropionic acid-2-(4- phenylazophenyl)ethyl ester. 235 E00 00 9. 00 8 Cor our 03. 00p 00.. com —__F___..._prrp_—_____»_~_.___Pp_~L—~_.______—_p»»—___»—.:._—__.._._._—_..0—0~_phpppppph.__.._._____FL»E___0__ NV. FN mvdm mu. 5.. mwdmr mwémp 50$. h F. R: «0.0m: 5&0? 00.000 50. EN E03 =20 _8_E20 0500 00. 0.. 0m ~15. 000 80 0200503 :0_.0> E0200 030.032 0020000000.? 50030000 00053000. Appendix 38. 13C NMR spectra of 2-bromopropionic acid-2-(4- phenylazophenyl)ethyl ester. 236 E00 F N h p h .t — p - p — b - _ b m 0 o 00.0 .000 IN .0 R. 700; :0 I. 0 5.0 00.0 IN 0 000.000 00.0 :0 . 00.0-00.0 . .E00. 020 3.. 0 E. .0000 _8_E000 0500 IF 0o 0N NI: com 000 03.03.03 cm...0> E0200 030.032 00300000E00- 50030000 E0E3000. Appendix 39. 1H NMR spectra of 3-azidopropanol. 237 can 00 00 00 cm 0%? 00.. —~»».__.__b»p_.—_pL~—._h#bprpp—h|rr_—b_._—.p_»—h_.»—___.__.p_—-»0»L»|PL~PPLT#P~F..--.____._ éiiii v. 5 v.9. 5mm :83 :20 .025ch 0.000 00. 0o mm N15. 02 80 0203:: :0_.0> Emzow 030.032 939883 5:00:69“. E95505 Appendlx 40. 13C NMR spectra of 3-azid0propanol. 238 Egg 0. m n r h P P P p h P h ~ h b F p p — p w p p P _ h tr 4 1 v n 02 o o m /\./\ : u n O IN E 00. 700., In 0 VNN 00.0 IN a 5.0-00.0 0.000 Em>_ow IN 0 med I. 020.032 05 :0 a 8.0-00.0 9. 00 0.20.850» ANIV 5 .E_ 20:90 E53 55 ~22 com 58:09”. $28050 oom SEES cmtm> EmEEEE Appendix 41. 1H NMR spectra of 3-azidopropyI-S-oxobutanoate. 239 Ema ON De ow cm 00 _. ON _. 0v F ow _. oar oow —_._.____.—..-__Trrr___P._..bpp».p.—_bpp—..__—_____...r—Tp_____»_____.—-____p.____»_._~__._-_p»~_ppr__bpbrpukr O O QwN Ndm adv 0.00. I 3N0 0.000 Emzow ab? 00. 030.022 «.80 0° 00 0.30.888 358 5:5 NIS— me 5:00:59“. .mo_Emco com mange: cmcm> 0:053:05 240 Appendix 42. 130 NMR spectra of 3-azidopropyl 3-oxobutanoate. .o 9.0 o .U 9.0 u m b 9.0 n .0 new 0 o m . .0 L .U b o z 0 “0 2r 0 M50.0 E0200 0 o 0 I I. 030.032 z/>\OJ_\/=\ 0° 0N 9205025... u a O O NIs. com 55:09”. com 02¢ch cmtm> EmEasz Appendlx 43. 1H NMR spectra of 3-azidopropyl 3-(2—oxo-tetrahydrofuran)-3- ylimino)butanoate. 241 242 San _. N 0 w m P P h _ _ p p _ n p P r r p _ — b L! P t— F L b _ p b a 0 IF E ©N.N-¢0.N IF E Nm.N-mm.N IN E vméémé b .n N; dd IF 6 —¢.v-®v.v ® :0 n 00.0 0 20: .m . £53 020 m0 0 3.: a E .200 .5505 0920 Em>_om I? 030.032 0.. mN 0.20.0060... ~I_2 com 55:09“. 000 02¢ch cmtm> 0:053:05 Appendix 44. 1H NMR spectra of L-homoserine Iactone hydrobromide. meN 5.3 8.00 05m: .Eaa. Ezm $0.520 0min Em>_om 00— 039022 0° mN 930.853 NIS. mNP 5:03.09“. 80 maniac: catm> EmEabmc. 8.. 2.32:...71 . . . 20: .m 9 . 1 Appendix 45. 130 NMR spectra of L-homoserine Iactone hydrobromide. 243 Eaa v m o h m — p p t— n p b — p p _ — — b p p — — p u p P P p n a c m o o m w m Z/\/\O/=\/__\ U a c O 2 g _ O O 0600 Em>.om IF m30_032 0° mN 920.0053 NIS. com 3:03.09”. com 00:03.5 cmtm> EmEEEE Appendlx 46. 1H NMR spectra of 3-azidopropyl-3-(3-oxoisoxazolidin-4— ylimino)butanoate. 244 E00 ON Ofi OO OO OOF ON .. OE. OO .. —_FrLFbpb__._._____F_.__p_..pp—brpp_ppb|r—.pp..»..____.—_.__.___.-...__—___.___Pk_rrpr....__ Effigy; No.00 EtON 00.00 mm. Fm mZ/>\O a... : . 0 Z 00 00 0030 5 00.00 0 00.00 2.00 00.00 00.00. 0.000 50200 00.05 09 030.032 00.30 0.. 0N 0.20.00E0... .600. £20 ~15. 0N0 300309.... 00.5000 80 0203.03 :0..0> 50520:. Appendlx 47. ”C NMR spectra of 3-azidopropyl-3-(3-oxoisoxazolidin-4- ylimino)butanoate. 245 E00 0 0 0 x. 0 — h b P LP — h p p - P _ — p p _ ~ - _ p F L! P h p .7 by up w - p 0 o 02 /O\/\ /\/O\/\ 0 0 I” E Vm:mr®mm £000 wCQ>_om IN E Fm.@nmm.m IF m=®_032 I0 E N0.0-00.0 0° 0N 0.3.0.00E00. .E. .0090 .E00. :20 ~I.2 000 500300.“. .00_E0co 000 03.03.03 cm_.0> E0E3=0c_ Appendix 48. 1H NMR spectra of 1-azido-2-(2-(2-methoxyethoxy)ethoxy)ethane (mDEG). 246 can cm 8 8 8 8’ cm. 9: 8. —...ppppr_.pFLh._..__pp_—PPPPL»_..__r.__..hp—bLL____..___..—_pp__.__.—_____phkp—______._. 0 m2 8.0m wmdm made hmdh m”Run—O Em>_ow 5.0% 09 320:2 00.05 0° mm 92988:. no.2. a8 ~12 mmw 3:039“. E 58 326ch com maniac: cmtm> EmEEfiE Appendlx 49. 130 NMR spectra of 1-azido-2-(2-(2-methoxyethoxy)ethoxy)ethane (mDEG). 247 lle I! 1%. _ n n m 0 8.58.8 IN 8 8.986 :3 E 8.78? IN E 8.76.. 8.8 IN . 85-88 v . E5358 fix: E_ .953 .mo_Eo.._0 £000 I. Go mm NIE Com com 952E: cmtm> Eo>_ow 320:2 929369 55:09“. E06285 Appendix 50. 1H NMR spectra of 1-azidodecane. 248 Tl. can 8 9. 8 8 8. cm. oi 8F ______.___...Pc__.__P_pp___.____L.______P__.r++Lc______b_r~_.____.____p_.___._._.__._r___ ii}; no.3 coda Kdm mmdm /\/\/\/\/\m2 :a 31mm tram mvdm n“.000 Emzom mm. 5 09 26.032 21m 00 mm ESSmQEm... €53 Ecm NI: mm— 55:09”. 328050 com gigs: cmtw> Emezzms Appendix 51. ”C NMR spectra of 1-azidodecane. 249 5% F N N v m o N — b h p p — p b — p — P p _ p — p b b P r F p p P — F p _ — p _ p F p h r h _ _ j 4 2 J» 33 n z o o m _ n.08 59.8. 950: Inc a N386 I, a. z o 9: 9m Em INe 9 NI! 0 mN a“ a a. N com cm: 9 Iod a 8.1.: Is. u. oom Sign: cm=m> EmEgms IN a :38 N IIN m main _IN.N a 8.8...” O O o \ /\/O\/\ /\/Z\Z I: n a...“ m m ,,2 IQo 9 N31. . I: a 3.3 c x; O T: a 3.88 I 3.3 893% . . 88 IF 9 «E: E .28 $5358 2 / m z _ $2.:ch a . Appendlx 52. 1H NMR spectra of PPGL-g-n-decyl-co-mDEG. 250 T Eaa ON 0* ow cm 00 _. ON F 0% w om P —_P-__br+*t——hpk__P»_—_-»PP—_—L—--PLFFpt_rb-———»_~.——p~p~——_—L—~_bbbhbbh_—_——__PLF_——-»—p-— jig; No.3 8.NN 6.8 8.8 0N.NN 0.08 .828 Exam 02 390:2 omdm 0.. mm 929,3th vmdm ~15. mmp 35:09". cm. 5 com "Sign: cmtm> Ewe—Ema. 8.8 8.8 0% \O/\/O\/\O/\/ Z\7,_I 3.2 / z 8. E. c NNr NNP E R: Eng 55 8.520 251 Appendlx 53. 130 NMR spectra of PPGL-g-n—decyl-co—mDEG. Eng r m n v m o n m — r .— p p — h p p b — p p .r P _ p p b p — _ n p p — P b p P — u p p h — - _ b h F p h w L, 4 4 t .11 - J? o n n mZ/\/>\oZ m o m 9 IN E 91.3: m.000 Emzom Iq E 8.38; IF 320:2 Kdémé Iv up mmdémd 00 mm 929363 N. ~12 oom 5:309... 33. E .955 $5355 :8?sz com 353E: cmtm> E06355 252 Appendix 54. 1H NMR spectra of 1,5-diazidopentane. _.i.n.1 I EQQ ON 0% 00 00 00.. our 03. 09. p»_..__h_._~p___rpkr.pp_.____pp________.r___~FLL___.___»_.____~_.._~—.____...___..__hp~r MUZ NZ N Em> o N.NN />\/\ .08 _ m cam 09 £5.32 N. B 0.. mm SSSBEmH E53 :Ew NIS. mmw 5:032“. 52,550 oom 2.53:5 cmzm> EmEasz Appendix 55. 130 NMR spectra of 1,5-diaziopentane. 253 N500 Eczom I, 3282 00 mm 935389.. ~15. com 55:02". o8 maize: 88> Emeaas m IN. F n N. 3. P IN a N.m-m.m 1v.m a On 19m a «6.08 10. P p ~.m-m.m 1N8 n 0v 1w. P a v.v-m.v 1F a «8.08 1? n v.5: .E_ .956 Afimmwwfimw Appendlx 56. 1H NMR spectra of cross-linked PPGL nanoparticle by 1,5- diazidopentane. 254 5% P N N e o m m S — p L r L — P — — — h p n — r p — F p b \— h F p u p — — b b L F — [F u n p F b h n h r I? {\11 13a J81 o u m IF m mg 2 o mmd I F U 5.0 O 0920 Em>_om 88 IF u o: O N 01 I. 8282 IF m EN m 0.. mm 929389.. ~ >ocoacm. ANIV a ..5 686 £53 58 Is. o8 u. _wo_Eoco com man—3E: ENE) 89::sz Appendix 57. 1H NMR spectra of 3-azido-7-hydroxycoumarin. 255 58 ON 2. 8 8 8F 8. o: o9 __F__p_.___p_._rFr__...___p.»_...p__h___._p____._—..p.—_php_.—__—_.__—~.___.».»PL».._____ QNS mé: Em: :Np mfimp ENF m”z szo Eo>_om \ Emmw 09 30.32 m um? 0° mm 939383 0 O OI 00? ~15. 99 55:09.... 25358 62520 08 maize: 88> smegma. Appendlx 58. 130 NMR spectra of 3-azido-7-hydroxycoumarin. 256 References 257 (1) (2) (4) (5) (7) (8) (9) (10) (11) (12) (13) (14) Baker, G. L.; Vogel, E. B.; Smith, M. R. Polym. Flev. 2008, 48, 64-84. Dechy-Cabaret, 0.; Martin-Vaca, B.; Bourissou, D. Chem. Flev. 2004, 104, 61 47-61 76. Chiellini, E.; Solaro, R. Adv. Mater. 1996, 8, 305-313. Leenslag, J. W.; Pennings, A. J.; 803, R. R. M.; Rozema, F. FL; Boering, G. Biomaterials 1987, 8, 70-73. Ashammakhi, N.; Flokkanen, P. Biomaterials 1997, 18, 3-9. Albertsson, A. C.; Varma, I. K. Biomacromolecules 2003, 4, 1466-1486. Agrawal, C. M. ; Ray, R. B. J. Biomed. Mater. Res. 2001, 55, 141-150. Martin, W. FL; Jackson, M; Wideburg, N. E.; Denison, F. W.; Coleman, W. H.; Cantrell, R. Biochimica El Biophysica Acta 1962, 62, 165-&. Li, X. F.; Jiang, 3.; Pan, B. L. Biotech. Lett. 2007, 29, 593-597. Yin, M.; Baker, G. L. Macromolecules 1999, 32, 7711-7718. Jing, F.; Smith, M. R; Baker, G. L. Abstr. Pap. Am. Chem. Soc. 2005, 230, U4090-U4090. Liu, T. 0.; Simmons, T. L.; Bohnsack, D. A.; Mackay, M. E.; Smith, M. R; Baker, G. L. Macromolecules 2007, 40, 6040-6047. Trimaille, T.; Gurny, R; Moller, M. Chimia 2005, 59, 348-352. Snell, K. D.; Draths, K. M.; Frost, J. W. J. Am. Chem. Soc. 1996, 118, 5605- 5614. 258 (15) (16) (17) (18) (19) (20) (EU (22) (23) (24) (25) (26) (27) Trimaille, T.; Moller, M.; Gurny, R. J. Polym. Sci. Part A: Polym. Chem. 2004, 42, 4379-4391. Simmons, T. L.; Baker, G. L. Biomacromolecules 2001, 2, 658-663. Yang, J. Y.; Yu, J.; Pan, H. 2.; Gu, Z. W.; Cao, W. X.; Feng, X. D. Chin. J. Polym. Sci. 2001, 19, 509-516. Gerhardt, W. W.; Noga, D. E.; Hardcastle, K. |.; Garcia, A. J.; Collard, D. M.; Weck, M. Biomacromolecules 2006, 7, 1735-1742. Leemhuis, M.; van Nostrum, C. F.; Kruijtzer, J. A. W.; Zhong, Z. Y.; ten Breteler, M. FL; Dijkstra, P. J.; Feijen, J.; Hennink, W. E. Macromolecules 2006, 39, 3500-3508. Marcincinova-Benabdillah, K.; Boustta, M.; Coudane, J.; Vert, M. Biomacromolecules 2001 , 2, 1279-1284. Benabdillah, K. M.; Coudane, J.; Boustta, M.; Engel, FL; Vert, M. Macromolecules 1999, 32, 8774-8780. Taguchi, K. Y., S.; Hiratani, K.; Norikio, M.; Okahata, Y. Macromolecules 1988, 21, 3338-3340. Ouchi, T.; Fujino, A. Makromol. Chem. 1989, 190, 1523-1530. Barrera, D. A.; Zylstra, E.; Lansbury, P. T.; Langer, Fl. J. Am. Chem. Soc. 1993, 115, 11010-11011. Barrera, D. A.; Zylstra, E.; Lansbury, P. T.; Langer, R. Macromolecules 1995, 28, 425-432. Feng, Y.; Klee, D.; Hocker, H. Macromol. Chem. Phys. 2002, 203, 819-824. Veld, P.; Dijkstra, P. J.; Feijen, J. Makromol. Chem. 1992, 193, 2713-2730. 259 (23) (29) (30) (31) (32) (33) (34) (35) (36) (37) (38) (39) (40) (41) Aoshima, S.; Oda, H.; Kobayashi, E. Journal of Polymer Science Part a- Polymer Chemistry 1992, 30, 2407-2413. Huh, K. M.; Bae, Y. H. Polymer1999, 40, 6147-6155. Lutz, J. F.; Hoth, A. Macromolecules 2006, 39, 893-896. Zhong, Z. Y.; Dijkstra, P. J.; Jan, F. J.; Kwon, Y. M.; Bae, Y. H.; Kim, S. W. Macromol. Chem. Phys. 2002, 203, 1797-1803. Porjazoska, A.; Dimitrov, P.; Dimitrov, l.; Cvetkovska, M.; Tsvetanov, C. B. Macromol. Symp. 2004, 210, 427-436. Jiang, X., Michigan State University, 2006. Radano, C. P., Michigan State University, 2003. Ponsart, 8.; Coudane, J.; Vert, M. Biomacromolecules 2000, 1, 275-281. Nottelet, 8; El Ghzaoui, A.; Coudane, J.; Vert, M. Biomacromolecules 2007, 8, 2594-2601. Huang, M. H.; Coudane, J.; Li, S. M.; Vert, M. J. Poly. Chem. Part A-Poly. Chem. 2005, 43, 4196-4205. Ponsart, S.; Coudane, J.; McGrath, J.; Vert, M. J. Bioact. Compat. Polym. 2002, 17, 417-432. Pan, J.; Wang, Y. L.; Qin, S. H.; Zhang, B. B.; Luo, Y. F. J. Biomed. Mater. Res. Part 82005, 748, 476-480. Tasaka, F.; Ohya, Y.; Ouchi, T. Macromol. Rapid Comm. 2001, 22, 820- 824. Garlotta, D. J. Polym. Environ. 2001, 9, 63-84. 260 (42) (43) (44) (45) (45) (47) (48) (49) (50) (51) (52) (53) (54) (55) Trollsas, M.; Atthofi, B.; Claesson, H.; Hedrick, J. L. Macromolecules 1998, 31, 3439-3445. Hao, O. H.; Li, F. X.; Li, Q. 8; Li, Y.; Jia, L.; Yang, J.; Fang, Q.; Cao, A. M. Biomacromolecules 2005, 6, 2236-2247. Zhao, Y. L.; Qing, C.; Jing, J.; Shuai, X. T.; Bei, J. 2.; Chen, C. F.; Fu, X. Polymer 2002, 43, 5819-5825. Kricheldorf, H. FL; Ahrensdort, K.; Rost, S. Macromol. Chem. Phys. 2004, 205, 1602-1610. Hay, G.; Mackay, M. E.; Hawker, C. J. J. Polym. Sci. Pt. B-Polym. Phys. 2001, 39, 1766-1777. Caminade, A. M.; Laurent, FL; Majoral, J. P. Adv. Drug Deliv. Rev. 2005, 57, 2130-2146. Zhao, Y. L.; Shuai, X. T.; Chen, C. F.; Xi, F. Macromolecules 2004, 37, 8854-8862. Tasaka, F .; Ohya, Y.; Ouchi, T. Macromolecules 2001, 34, 5494-5500. Tasaka, F.; Miyazaki, H.; Ohya, Y.; Ouchi, T. Macromolecules 1999, 32, 6386-6389. Wittmar, M. ; Unger, F.; Kissel, T. Macromolecules 2006, 39, 1417-1424. Dailey, L. A.; Wittmar, M.; Kissel, T. J. Control. Release 2005, 101, 137-149. Pistel, K. F.; Breitenbach, A.; Zange-Volland, R.; Kissel, T. J. Control. Release 2001, 73, 7-20. Breitenbach, A.; Pistel, K. F.; Kissel, T. Polymer 2000, 41, 4781-4792. Breitenbach, A.; Li, Y. X.; Kissel, T. J. Control. Release 2000, 64, 167-178. 261 (55) (57) (58) (59) (50) (51) (52) (53) (54) (55) (55) (57) (58) (59) Breitenbach, A.; Kissel, T. Polymer 1 998, 39, 3261 -3271 . Parrish, B.; Emrick, T. Macromolecules 2004, 37, 5863-5865. Porter, K.; Hossain, M.; Wang, M.; Radano, C. P.; Baker, G.; Smith, M. R.; McCabe, L. R. Mol. Biol. Rep. 2006, 33, 1-12. Tian, D.; Dubois, P.; Jerome, R. Macromolecules 1997, 30, 2575-2581. Tian, D.; Dubois, P.; Jerome, R. Macromolecules 1997, 30, 1947-1954. Tian, D.; Dubois, P.; Grandfils, 0.; Jerome, R. Macromolecules 1997, 30, 406-409. Taniguchi, I.; Mayes, A. M.; Chan, E. W. L.; Griffith, L. G. Macromolecules 2005, 38, 216-219. Lecomte, P.; Riva, R.; Schmeits, S.; Rieger, J.; Van Butsele, K.; Jerome, C.; Jerome, R. Macromol. Symp. 2006, 240, 157-165. Van Horn, B. A.; Wooley, K. L. Soft Matter 2007, 3, 1032-1040. Mecerreyes, D.; Humes, J.; Miller, R. D.; Hedrick, J. L.; Detrembleur, C.; Lecomte, P.; Jerome, R.; San Roman, J. Macromol. Rapid Comm. 2000, 21 , 779-784. Rieger, J. V. B., K.; Lecomte, P.; Detrembleur, 0.; Jerome, R.; Jerome, C. Chem. Commun. 2005, 274, 274-276. Lenoir, S.; Riva, R.; Lou, X.; Detrembleur, 0.; Jerome, R.; Lecomte, P. Macromolecules 2004, 37, 4055-4061. Riva, R.; Lenoir, 8.; Jerome, R.; Lecomte, P. Polymer 2005, 46, 8511-8518. Riva, R.; Rieger, J.; Jerome, R.; Lecomte, P. J. Poly. Chem. Part A-Poly. Chem. 2006, 44, 6015-6024. 262 (70) (71) (72) (73) (74) (75) (75) (77) (73) (79) (30) (81) Parrish, B.; Breitenkamp, R. B.; Emrick, T. J. Am. Chem. Soc. 2005, 127, 7404-7410. Riva, R.; Schmeits, 3.; Jerome, C.; Jerome, R.; Lecomte, P. Macromolecules 2007, 40, 796-803. Riva, R.; Schmeits, P.; Stoffelbach, F.; Jerome, 0.; Jerome, R.; Lecomte, P. Chem. Commun. 2005, 5334-5336. Jiang, X.; Vogel, E. B.; Smith, M. R.; Baker, G. L. Macromolecules 2008, 41, 1 937-1944. Kimura, Y.; Shirotani, K.; Yamane, H.; Kitao, T. Macromolecules 1988, 21, 3338-3340. Kimura, Y.; Shirotani, K.; Yamane, H.; Kitao, T. Polymer1993, 34, 1741- 1748. Leemhuis, M; van Steenis, J. H.; van Uxem, M J.; van Nostrum, C. F.; Hennink, W. E. Eur. J. Org. Chem. 2003, 3344-3349. MacDonald, R. T.; McCarthy, S. P.; Gross, R. A. Macromolecules 1996, 29, 7356-7361. Ovitt, T. M.; Coates, G. W. J. Am. Chem. Soc. 1999, 121, 4072-4073. Thakur, K. A. M.; Kean, R. T.; Hall, E. S.; Kolstad, J. J.; Munson, E. J. Macromolecules 1998, 31, 1487-1494. Jacobs, C.; Dubois, P.; Jerome, R.; Teyssie, P. Macromolecules 1991, 24, 3027-3034. Kumar, M; Ravikumar, M. N. V.; Domb, A. J. Adv. Drug Deliv. Rev. 2001, 53, 23-44. 263 (32) (83) (84) (85) (85) (87) (88) (89) (90) (91) (92) (93) (94) (95) Ouchi, T.; Miyazaki, H.; Arimura, H.; Tasaka, F.; Hamada, A.; Ohya, Y. J. Polym. Sci. Part A: Polym. Chem. 2002, 40, 1218-1225. Veld, P.; Velner, E. M.; VanDeWitte, P.; Hamhuis, J.; Dijkstra, P. J.; Feijen, J. J. Polym. Sci. Part A: Polym. Chem. 1997, 35, 219-226. Ouchi, T.; Ohya, Y. J. Polym. Sci. Part A: Polym. Chem. 2004, 42, 453-462. Shen, Y. 0.; Zhu, K. J.; Shen, Z. 0.; Yao, K. M. J. Polym. Sci. Part A: Polym. Chem. 1996. 34, 1799-1805. Vanhoorne, P.; Dubois, P.; Jerome, R.; Teyssie, P. Macromolecules 1992, 25, 37-44. Li, X. R; Yuan, X. Y. Prog. Chem. 2007, 19, 973-981. Tu, C. F.; Cai, Q.; Yang, J.; Wan, Y. Q.; Bei, J. 2.; Wang, S. Polym. Adv. Technol. 2003, 14, 565-573. Li, S. M.; Garreau, H.; Vert, M. J. Mater. Sci-Mater. Med. 1990, 1, 131-139. Grijpma, D. W.; Pennings, A. J. Macromol. Chem. Phys. 1994, 195, 1633- 1647. Liu, G.; Fang, Y. E.; Shi, T. Y. Chem. J. Chin. Univ-Chin. 1997, 18, 486- 488. Fineman, M.; Ross, S. D. J. Polym. Sci. 1950, 5, 259-262. Tudos, F.; Kelen, T. J. Macromol. Sci: Chem. 1981, A16, 1283-1297. Tidwell, P. W.; Mortimer, G. A. J. Polym. Sci. PtA-Gen. Pap. 1965, 3, 369- 387. Dong, C. M.; Qiu, K. Y.; Gu, Z. W.; Feng, X. D. Chin. Chem. Lett. 2000, 11, 815-818. 264 _‘ (96) (97) (98) (99) (100) (101) (102) (103) (1 O4) (105) (106) (107) (108) (109) (110) Saulnier, B.; Coudane, J.; Garreau, H.; Vert, M. Polymer 2006, 47, 1921- 1929. Yang, J. Y.; Yu, J.; Li, M.; Gu, Z. W.; Feng, X. D. Chin. J. Polym. Sci. 2002, 20, 413-417. Jiang, X.; Vogel, E. B.; Smith III, M. R.; Baker, G. L. J. Poly. Chem. Part A- Poly. Chem. 2007, 45, 5227-5236. Jiang, X. W.; Smith, M. R.; Baker, G. L. Macromolecules 2008, 41, 318-324. Jing, F.; Smith, M. R.; Baker, G. L. Macromolecules 2007, 40, 9304-9312. Mori, K.; Takaishi, H. Tetrahedron 1989, 45, 1639-1646. Odian, G. Principles of Polymerization; 4th ed.; John Wiley & Sons: New York, 2004. Yin, M., Michigan State University, 2000. Wang, 0., Michigan State University, 2001. Jing, F., Michigan State University, 2006. Liu, T., Michigan State, 2002. Rozen, S.; Ben-David, I. J. Org. Chem. 2001, 66, 496-500. Guest, H. H. J. Am. Chem. Soc. 1947, 69, 300-302. Soppimath, K. S.; Aminabhavi, T. M.; Kulkarni, A. R.; Rudzinski, W. E. J. Control. Release 2001, 70, 1-20. Bourissou, D.; Moebs-Sanchez, 8.; Martin-Vaca, B. C. R. Chim. 2007, 10, 775-794. 265 “ (111) (112) (113) (114) (115) (116) (117) (118) (119) (120) (121) (122) (123) (124) Williams, C. K. Chem. Soc. Rev. 2007, 36, 1573-1580. Yang, J. Y.; Yu, J.; Li, M.; Pan, H. 2.; Gu, Z. W.; Cao, W. X.; Feng, X. D. Acta Chimica Sinica 2001, 59, 1809-1812. Parrish, B.; Emrick, T. Bioconjugate Chem. 2007, 18, 263-267. Lee, R. S.; Yang, J. M. J. Appl. Polym. Sci. 2001, 81, 1581-1587. Lee, R. S.; Yang, J. M. J. Poly. Chem. Part A-Poly. Chem. 2001, 39, 724- 731. Nijenhuis, A. J.; Grijpma, D. W.; Pennings, A. J. Macromolecules 1992, 25, 6419-6424. Witzke, D. R.; Narayan, R.; Kolstad, J. J. Macromolecules 1997, 30, 7075- 7085. Du, Y. J.; Lemstra, P. J.; Nijenhuis, A. J.; Vanaert, H. A. M.; Bastiaansen, C. Macromolecules 1995, 28, 2124-2132. Sperling, L. H. Introduction to Physical Polymer Science; 3rd ed.; Wiley- lnterscienoe, 2001. Jiang, X. V., E. B; Smith III, M. R.; Baker, G. L. Journal of Polymer Science Part a-Polymer Chemistry 2007, 45, 5227-5236. Duncan, R. Nat. Rev. Drug Discov. 2003, 2, 347-360. Uhrich, K. E.; Anastasiou, T. J.; Beaton, M. L. Abstr. Pap. Am. Chem. Soc. 2001, 222, U317-U317. Uhrich, K. E.; Cannizzaro, S. M.; Langer, R. S.; Shakeshefi, K. M. Chem. Rev. 1999, 99, 3181-3198. Vert, M. Prog. Polym. Sci. 2007, 32, 755-761. 266 (125) (126) (127) (128) (129) (130) (131) (132) (133) (134) (135) (136) (137) Nair, L. S.; Laurencin, C. T. Prog. Polym. Sci. 2007, 32, 762-798. Khandare, J.; Minko, T. Prog. Polym. Sci. 2006, 31, 359-397. Duncan, R.; Ringsdorf, H.; Satchi-Fainaro, R. In Polymer Therapeutics I: Polymers as Drugs, Conjugates and Gene Delivery Systems 2006; Vol. 192, p 1-8. Yokoyama, M.; Miyauchi, M.; Yamada, N.; Okano, T.; Sakurai, Y.; Kataoka, K.; lnoue, S. J. Control. Release 1990, 11, 269-278. Jeong, J. H.; Kim, S. W.; Park, T. G. Prog. Polym. Sci. 2007, 32, 1239-1274. Wong, S. Y.; Pelet, J. M.; Putnam, D. Prog. Polym. Sci. 2007, 32, 799-837. Zhu, s. G.; Li, G. Y. Prog. Biochem. Biophys. 2002, 29, 868-871. Rieger, J.; Bernaerts, K. V.; Du Prez, F. E.; Jerome, R.; Jerome, C. Macromolecules 2004, 37, 9738-9745. Adams, M. L.; Lavasanifar, A.; Kwon, G. S. J. Pharm. Sci. 2003, 92, 1343- 1355. Mecerreyes, D.; Miller, R. D.; Hedrick, J. L.; Detrembleur, C.; Jerome, R. J. Poly. Chem. Part A-Poly. Chem. 2000, 38, 870-875. Kolb, H. C.; Finn, M. G.; Sharpless, K. B. Angew. Chem-Int. Edit. 2001, 40, 2004-+. Rostovtsev, V. V.; Green, L. G.; Fokin, V. V.; Sharpless, K. B. Angew. Chem-Int. Edit. 2002, 41 , 2596-+. Wu, P.; Feldman, A. K.; Nugent, A. K.; Hawker, C. J.; Scheel, A.; Volt, 8; Pyun, J.; Frechet, J. M. J.; Sharpless, K. B.; Fokin, V. V. Angew. Chem-Int. Edit. 2004, 43, 3928-3932. 267 (138) Wu, P.; Malkoch, M.; Hunt, J. N.; Vestberg, R.; Kaltgrad, E.; Finn, M. G.; Fokin, V. V.; Sharpless, K. B.; Hawker, C. J. Chem. Commun. 2005, 5775- 5777. (139) Joralemon, M. J.; O'Reilly, R. K.; Matson, J. B.; Nugent, A. K.; Hawker, C. J.; Wooley, K. L. Macromolecules 2005, 38, 5436-5443. (140) Fernandez-Megia, E.; Correa, J.; Rodriguez-Meizoso, l.; Riguera, R. Macromolecules 2006, 39, 21 13-2120. (141) Tsarevsky, N. V.; Sumerlin, B. S.; Matyjaszewski, K. Macromolecules 2005, 38. 3558-3561. (142) Opsteen, J. A.; van Hest, J. C. M. Chem. Commun. 2005, 57-59. (143) Sumerlin, B. S.; Tsarevsky, N. V. ; Louche, G.; Lee, R. Y.; Matyjaszewski, K. Macromolecules 2005, 38, 7540-7545. (144) Helms, B.; Mynar, J. L.; Hawker, C. J.; Frechet, J. M. J. J. Am. Chem. Soc. 2004, 126, 15020-15021. (145) Mynar, J. L.; Choi, T. L.; Yoshida, M.; Kim, V.; Hawker, C. J.; Frechet, J. M. J. Chem. Commun. 2005, 5169-5171. (146) Malkoch, M.; Thibault, R. J.; Drockenmuller, E.; Messerschmidt, M.; Voit, 8; Russell, T. P.; Hawker, C. J. J. Am. Chem. Soc. 2005, 127, 14942-14949. (147) Newkome, G. R.; Moorefield, C. M; Baker, G. R.; Saunders, M. J.; Grossman, S. H. Angew. Chem-Int. Edit. Engl. 1991, 30, 1178-1180. (148) Lawrence, M. J. Chemical Society Reviews 1994, 23, 417-424. (149) Kakizawa, Y.; Kataoka, K. Advanced Drug Delivery Reviews 2002, 54, 203- 222. 268 (150) (151) (152) (153) (154) (155) (156) (157) (158) Kataoka, K.; Harada, A.; Nagasaki, Y. Advanced Drug Delivery Reviews 2001,47113-131. Becker, M. L.; Remsen, E. E.; Wooley, K. L. Journal of Polymer Science Part a-Polymer Chemistry 2001 , 39, 4152-4166. Bhattarai, N.; Bhattarai, S. R.; YI, H. K; Lee, J. C. Khil, M. S.; Hwang, P. H.; Kim, H. Y. Pharmaceutical Research 2003,20, 2021-2027. Tomalia, D. A.; Berry, V.; Hall, M.; Hedstrand, D. M. Macromolecules 1987, 20, 1164-1167. Vanhest, J. C. M.; Delnoye, D. A. P.; Baars, M.; Vangenderen, M. H. P.; Meijer, E. W. Science 1995, 268, 1592-1595. vanHest, J. C. M.; Delnoye, D. A. P.; Baars, M.; ElissenRoman, C.; vanGenderen, M. H. P; Meijer, E. W. Chemistry-a European Journal 1996, 2, 1616-1626. Chapman, T. M.; Hillyer, G. L.; Mahan, E. J.; Shaffer, K. A. Journal of the American Chemical Society1994, 116, 11195-11196. Frey, H.; Lorenz, K.; Mulhaupt, R.; Rapp, U.; MayerPosner, F. J. Macromolecular Symposia 1996, 102, 19-26. Liu, H. B.; Jiang, A.; Guo, J. A.; Uhrich, K. E. Journal of Polymer Science Part a-Polymer Chemistry 1999, 37, 703-71 1 . (159) Jiang, X. V., E. B; Smith III, M. R.; Baker, G. L. Macromolecules 2008, 41, (160) (161) 1 937-1 944. Chan, T. R.; Fokin, V. V. QSAR Comb. Sci. 2007, 26, 1274-1279. Skwarczynski, M.; Hayashi, Y.; Kiso, Y. J. Med. Chem. 2006, 49, 7253- 7269. 269 (162) (163) (164) (165) (166) (167) (168) (169) (170) (171) (172) (173) Gref, R.; Minamitake, Y.; Peracchia, M. T.; Trubetskoy, V.; Torchilin, V.; Langer, R. Science 1994, 263, 1600-1603. Osada, K.; Kataoka, K. In Peptide Hybrid Polymers 2006; Vol. 202, p 113- 153. Jette, K. K.; Law, D.; Schmitt, E. A.; Kwon, G. S. Pharm. Res. 2004, 21, 1184-1191. Otsuka, H.; Nagasaki, Y.; Kataoka, K. Adv. Drug Deliv. Rev. 2003, 55, 403- 41 9. Yasugi, K.; Nagasaki, Y.; Kato, M.; Kataoka, K. J. Control. Release 1999, 62, 89-100. Hagan, S. A.; Coombes, A. G. A.; Garnett, M. C.; Dunn, S. E.; Davis, M. C.; lllum, L.; Davis, 8. S.; Harding, 8. E.; Purkiss, S.; Gellert, P. R. Langmuir 1996, 12, 2153-2161. Hussain, H.; Busse, K.; Kressler, J. Macromol. Chem. Phys. 2003, 204, 936-946. Wilhelm, M.; Zhao, C. L.; Wang, Y. C.; Xu, R. L; Winnik, M. A.; Mura, J. L.; Riess, G.; Croucher, M. D. Macromolecules 1991, 24, 1033-1040. Castro, E.; Taboada, P.; Barbosa, S.; Mosquera, V. Biomacromolecules 2005, 6, 1438-1447. Astafieva, I.; Zhong, X. F.; Eisenberg, A. Macromolecules 1993, 26, 7339- 7352. Liu, J.; Lee, H.; Allen, C. Curr. Pharm. Design 2006, 12, 4685-4701. Panyam, J.; Williams, 0.; Dash, A.; Leslie-Pelecky, D.; Labhasetwar, V. J. Pharm. Sci. 2004, 93, 1804-1814. 270 (174) (175) (176) (177) (178) (179) (180) (181) (182) (183) (184) (185) (186) (187) Xu, J. P.; Ji, J.; Chen, W. D.; Shen, J. C. J. Control. Release 2005, 107, 502-512. Kondo, S.; Hosaka, S.; Sasai, Y.; Kuzuya, M. Chem. Pharm. Bull. 2004, 52, 1302-1306. Wang, L. F.; Chiang, H. N.; Chen, W. B. J. Pharm. Pharmacol. 2002, 54, 1129-1135. Cho, S. H.; Kim, K. S. Macromol. Res. 2003, 11, 317-321. Erdmann, L.; Uhrich, K. E. Biomaterials 2000, 21, 1941-1946. Schmeltzer, R. C.; Anastasiou, T. J.; Uhrich, K. E. Polym. Bull. 2003, 49, 441 -448. Schmeltzer, R. C.; Uhrich, K. E. J. Bioact. Compat. Polym. 2006, 21, 123- 133. Anastasiou, T. J.; Uhrich, K. E. J. Poly. Chem. Part A-Poly. Chem. 2003, 41, 3667-3679. Prudencio, A.; Schmeltzer, R. C.; Uhrich, K. E. Macromolecules 2005, 38, 6895-6901. Patil, 8.; Chan, C. Neurosci. Lett. 2005, 384, 288-293. Patil, 8.; Sheng, L. F.; Masserang, A.; Chan, C. Neurosci. Lett. 2006, 406, 55-59. Patil, S. P.; Chan, C. Faseb J. 2006, 20, A1420-A1421. Patil, S.; Melrose, J.; Chan, C. Eur. J. Neurosci. 2007, 26, 2131-2141. Jensen, N. P.; Friedman, J. J.; Kropp, H.; Kahan, F. M. J. Med. Chem. 1980, 23, 6-8. 271 (188) (189) (190) (191) (192) (193) (194) (195) (196) Tsuda, Y.; Kikuchi, A.; Yamato, M.; Nakao, A.; Sakurai, Y.; Umezu, M.; (197) (198) (199) (200) Lassen, F. 0.; Stammer, C. H. J. Org. Chem. 1971, 36, 2631-8. Grigg, R.; Sarker, M. A. B. Tetrahedron 2006, 62, 10332-10343. Yamamoto, K.; Serizawa, T.; Muraoka, Y.; Akashi, M. Macromolecules 2001, 34, 8014-8020. Schilli, C. M.; Zhang, M. F.; Rizzardo, E.; Thang, S. H.; Chong, Y. K.; Edwards, K.; Karlsson, G.; Muller, A. H. E. Macromolecules 2004, 37, 7861- 7866. Jeong, B.; Gutowska, A. Trends Biotechnol. 2002, 20, 305-311. Gil, E. 8.; Hudson, 8. A. Prog. Polym. Sci. 2004, 29, 1173-1222. Yoo, M. K.; Seok, W. K.; Sung, Y. K. Macromol. Symp. 2004, 207, 173-186. lijima, M.; Nagasaki, Y. J. Poly. Chem. Part A-Poly. Chem. 2006, 44, 1457- 1469. Okano, T. Biomaterials 2005, 26, 1885-1893. Yang, J.; Yamato, M.; Kohno, C.; Nishimoto, A.; Sekine, H.; Fukai, F.; Okano, T. Biomaterials 2005, 26, 6415-6422. Nakayama, M.; Okano, T.; Miyazaki, T.; Kohori, F.; Sakai, K.; Yokoyama, M. J. Control. Release 2006, 115, 4656. Yang, H.; Kao, W. Y. J. Pharmaceutical Research 2006, 23, 205-214. Lutz, J. F.; Akdemir, 0.; Hoth, A. J. Am. Chem. Soc. 2006, 128, 13046- 13047. (201) Zhao, B.; Li, D. J.; Hua, F. J.; Green, D. R. Macromolecules 2005, 38, 9509- 9517. 272 (202) (203) (204) (205) (206) (207) (208) (209) (210) (211) (212) (213) (214) (215) (216) Han, S.; Hagiwara, M.; Ishizone, T. Macromolecules 2003, 36, 8312-8319. Shih, C. Pharm. Res. 1995, 12, 2036-2040. Markiewicz, P.; Goh, M. C. Langmuir 1994, 10, 5-7. Ebenstein, Y.; Nahum, E.; Banin, U. Nano Lett. 2002, 2, 945-950. 1 Wang, G.; Zhu, X. L.; Cheng, Z. P.; Zhu, J. J. Poly. Chem. Part A-Poly. Chem. 2005, 43, 2358-2367. Boyer, J. H.; Hamer, J. J. Am. Chem. Soc. 1955, 77, 951-954. Christoffers, J.; Onal, N. Eur. J. Org. Chem. 2000, 1633-1635. Angle, S. R.; Henry, R. M. Journal of Organic Chemistry/1998, 63, 7490- 7497. Natelson, S.; Natelson, E. A. Microchem J. 1989, 40, 226-232. Schmidt, M; Amstutz, R.; Crass, G.; Seebach, D. Chemische Berichte- Recueil 1980, 113, 1691 -1707. Lieber, E.; Chao, T. S.; Rao, C. N. R. J. Org. Chem. 1957, 22, 238-240. Liu, Y. Y.; Miyoshi, H.; Nakamura, M. Int. J. Cancer 2007, 120, 2527-2537. Couvreur, P.; Puisieux, F. Adv. Drug Deliv. Rev. 1993, 10, 141-162. Jin, 8.; Ye, K. M. Biotechnol. Prog. 2007, 23, 32-41. Hawker, C. J.; Wooley, K. L. Science 2005, 309, 1200-1205. 273 (217) (218) (219) (220) (221) (222) (223) (224) (225) (226) (227) Westedt, U.; Kalinowski, M.; Wittmar, M.; Merdan, T.; Unger, F.; Fuchs, J.; Schaller, S.; Bakowsky, U.; Kissel, T. J. Control. Release 2007, 119, 41-51. Liu, F.; Liu, W. M.; Xue, Q. J. J. Macromol. Sci-Pure Appl. Chem. 2004, A41, 1001-1010. Dailey, L. A.; Kleemann, E.; Wittmar, M; Gessler, T.; Schmehl, T.; Roberts, 0.; Seeger, W.; Kissel, T. Pharm. Res. 2003, 20, 2011-2020. Breitenbach, A.; Jung, T.; Kamm, W.; Kissel, T. Polym. Adv. Technol. 2002, 13, 938-950. Kakizawa, Y.; Harada, A.; Kataoka, K. J. Am. Chem. Soc. 1999, 121, 11247-11248. O'Reilly, R. K.; Joralemon, M. J.; Hawker, C. J.; Wooley, K. L. New J. Chem. 2007, 31, 718-724. Butun, V.; Lowe, A. B.; Billingham, N. C.; Armes, S. P. J. Am. Chem. Soc. 1999, 121, 4288-4289. Ding, J. F .; Liu, G. J. Macromolecules 1998, 31 , 6554-6558. l-lales, M.; Barner-Kowollik, C.; Davis, T. P.; Stenzel, M. H. Langmuir 2004, 20, 10809-10817. Sanji, T.; Nakatsuka, Y.; Kitayama, F.; Sakurai, H. Chem. Commun. 1999, 2201-2202. Huang, H. Y.; Remsen, E. E.; Wooley, K. L. Chem. Commun. 1998, 1415- 1416. (228) Joralemon, M. J.; Murthy, K. 8.; Remsen, E. E.; Becker, M. L.; Wooley, K. L. Biomacromolecules 2004, 5, 903-913. 274 (229) (230) (231) (232) (233) (234) (235) (236) (237) (238) (239) (240) (241) (242) Joralemon, M. J.; O'Reilly, R. K.; Hawker, C. J.; Wooley, K. L. J. Am. Chem. Soc. 2005, 127, 16892-16899. Prochazka, K.; Baloch, M. K.; Tuzar, Z. Makromol. Chem. 1979, 180, 2521- 2523. Ding, J. F.; Liu, G. J. J. Phys. Chem. B 1998, 102, 6107-6113. Henselwood, F .; Liu, G. J. Macromolecules 1 997, 30, 488-493. Henselwood, F.; Wang, G. C.; Liu, G. J. J. Appl. Polym. Sci. 1998, 70, 397- 408. Liu, G. J. Curr. Opin. Colloid Interface Sci. 1998, 3, 200-208. Wang, G. C.; Henselwood, F.; Liu, G. J. Langmuir1998, 14, 1554-1559. Stewart, 8.; Liu, G. J. Chem. Mater. 1999, 11, 1048-1054. Harth, E.; Van Horn, 8.; Lee, V. Y.; Germack, D. S.; Gonzales, C. P.; Miller, R. D.; Hawker, C. J. J. Am. Chem. Soc. 2002, 124, 8653-8660. Croce, T. A.; Hamilton, S. K.; Chen, M. L.; Muchalski, H.; Harth, E. Macromolecules 2007, 40, 6028-6031. Huang, H. Y.; Kowalewski, T.; Remsen, E. E.; Gertzmann, R.; Wooley, K. L. J. Am. Chem. Soc. 1997, 119, 11653-11659. Remsen, E. E.; Thurmond, K. B.; Wooley, K. L. Macromolecules 1999, 32, 3685-3689. Wooley, K. L. J. Poly. Chem. Part A-Po/y. Chem. 2000, 38, 1397-1407. Butun, V.; Billingham, N. C.; Armes, S. P. J. Am. Chem. Soc. 1998, 120, 12135-12136. 275 (243) (244) (245) (246) (247) (248) (249) (250) (251) (252) (253) Liu, S. Y.; Weaver, J. V. M.; Save, M.; Armes, S. P. Langmuir 2002, 18, 8350-8357. O'Reilly, R. K.; Joralemon, M. J.; Hawker, C. J.; Wooley, K. L. Chem-Eur. J. 2006, 12, 6776-6786. Read, E. S.; Armes, S. P. Chem. Commun. 2007, 3021-3035. Butun, V.; Wang, X. S.; Banez, M. V. D.; Robinson, K. L.; Billingham, N. C.; Armes, S. P.; Tuzar, Z. Macromolecules 2000, 33, 1-3. Napper, D. H. Polymeric Stabilization of Colloidal Dispersions; Acadmic Press: London, 1983. Pan, D.; Turner, J. L.; Wooley, K. L. Chem. Commun. 2003, 2400-2401. Becker, M. L.; Liu, J. 0.; Wooley, K. L. Biomacromolecules 2005, 6, 220- 228. Angot, S.; Ayres, N.; Bon, S. A. F.; Haddleton, D. M. Macromolecules 2001, 34, 768-774. Sivakumar, K.; Xie, F.; Cash, B. M.; Long, S.; Barnhill, H. M; Wang, Q. Organic Letters 2004, 6, 4603-4606. Thomas, J. R.; Liu, X. J.; Hergenrother, P. J. J. Am. Chem. Soc. 2005, 127, 12434-12435. Mantovani, G.; Ladmiral, V.; Tao, L.; Haddleton, D. M. Chem. Commun. 2005, 2089-2091. 276 Illjllljljlflljlijjljljjll(1(1)le