CHEMISTRY TECHNIQUES a Dissertation Presented to the Graduate Faculty Of

Home , Polymer chemistry

PREPARING POLYMERIC BIOMATERIALS USING “CLICK” CHEMISTRY

TECHNIQUES

A Dissertation

Presented to

The Graduate Faculty of The University of Akron

In Partial Fulfillment

of the Requirements for the Degree

Doctor of Philosophy

Fei Lin

May, 2014

PREPARING POLYMERIC BIOMATERIALS USING “CLICK” CHEMISTRY

TECHNIQUES

Fei Lin

Dissertation

Approved: Accepted:

______Advisor Department Chair Dr. Matthew L. Becker Dr. Coleen Pugh

______Committee Member Dean of the College Dr. Abraham Joy Dr. Stephen Z. D. Cheng

______Committee Member Dean of the Graduate School Dr. Chrys Wesdemiotis Dr. George R. Newkome

______Committee Member Date Dr. Shi-Qing Wang

______Committee Member Dr. Robert A. Weiss

ABSTRACT

Significant efforts have been focused on preparing degradable polymeric biomaterials with controllable properties, which have the potential to stimulate specific cellular responses at the molecular level. “Click” reactions provide a universal tool box to achieve that goal through molecular level design and modification. This dissertation demonstrates multiple methodologies and techniques to develop advanced biomaterials through combining degradable polymers and “click” chemistry.

In my initial work, a novel class of amino acid-based poly(ester urea)s (PEU) materials was designed and prepared for potential applications in bone defect treatment.

PEUs were synthesized via interfacial polycondensation, and showed degradability in vivo and possessed mechanical strength superior to conventionally used polyesters. Further mechanical enhancement was achieved after covalent crosslinking with a short peptide crosslinker derived from osteogenic growth peptide (OGP). The in vitro and in an in vivo subcutaneous rat model demonstrated that the OGP-based crosslinkers promoted proliferative activity of cells and accelerated degradation properties of PEUs.

As a continuous study, extra efforts were focused on the development of PEUs with functional pendant groups, including alkyne, azide, alkene, tyrosine–phenol, and ketone groups. PEUs with Mw exceeding to 100K Da were obtained via interfacial polycondensation, and the concentration of pendent groups was varied using a copolymerization strategy. Electrospinning was used to fabricate PEU nanofiber matrices iii with mechanical strengths suitable for tissue engineering. A series of biomolecules were conjugated to nanofiber surface following electrospinning using “click” reactions in aqueous media. The ability to derivatize PEUs with biological motifs using high efficient chemical reactions will significantly expand their use in vitro and in vivo.

Based on similar principles, a series of mono- and multifunctionalized polycaprolactone (PCL) bearing various “clickable” groups, including ketone, alkyne, azide, and methyl acrylate (MA), were synthesized via ring opening polymerization. A quartz crystal microbalance (QCM) was used to quantify the rate and extent of surface conjugation between RGD peptides and polymer thin films. The successful conjugation was further confirmed by static contact angle and NMR measurements. QCM results also verified and quantified the sequential immobilization of peptides onto polymer films.

Besides polymer functionalization, “click” reactions were also utilized for hydrogel fabrication and post-gelation modification. Polyethylene glycol-based hydrogels were formed via oxime ligation. The gelation process and final mechanical strength of the hydrogels can be tuned using pH and the catalyst concentration. The time scale to reach the gel point and complete gelation, and the storage modulus of hydrogels can be tuned in two orders of magnitude. Azide- and alkene-functionalized hydrogels were also fabricated, and further post-gelation functionalization was achieved via alkyne-azide cycloaddition and thiol-ene radical addition for spatially defined peptide incorporation. These materials with tunable mechanical regimes and biomolecule patterns were attractive for soft tissue engineering.

ACKNOWLEDGEMENTS

This dissertation would not have been possible without the help of so many people in so many ways. It’s also a product of tremendous serendipity and encounters with people who have significant effect on my career.

First of all, a very special thank you should be given to my advisor, Dr. Matthew L.

Becker, for all his support during the past years. He was very suggestive and supportive during my researches and it was my great honor to work with him. I am also very grateful to Dr. Abraham Joy, Dr. Avraam I. Isayev, Dr. Chrys Wesdemiotis, Dr. Robert Weiss, and

Dr. Shi-Qing Wang for their cooperation, help, discussion, advice and comment during my research.

I also want to say thanks to students in the College of Polymer Science and

Engineering, in particular Jing Zhou, Xueyuan Wang, Yiwen Li, Kai Guo, Shiwang Cheng,

Tian Liang, Hao Su, Kan Yue, Yin Xu, Panpan Lin, all my group members and friends.

They have been a critical source of support and encouragement. I can’t thank them enough for their kind friendship, trust, and support.

I am also very grateful for my parents and grandparents, who give me all the love and encouragement for ever. They have instilled in me the importance of education, working hard, and caring for others.

TABLE OF CONTENTS

Page

LIST OF TABLES………………………………………………………………………ix

LIST OF FIGURES……………………………………………………………………..x

LIST OF SCHEMES………………………………………………………………….xxii

CHAPTER

I. INTRODUCION………………………………………………………………………..1

1.1 Degradable Polymers……………………………………………………..………..3

1.2 “Click” Chemistry and Its Applications in Biopolymers………………………….42

1.3 Conclusion………………………………………………………………………...72

II. EXPERIMENTAL SECTION………………………………………………………...73

2.1 Materials…………………………………………………………………………..73

2.2 Instruments………………………………………………………………………..75

III. RESORBABLE, AMINO ACID-BASED POLY (ESTER UREA)S CROSSLINKED WITH OGP PEPTIDE WITH ENHANCED MEAHCNICAL PROPERTIES AND BIOACTIVITY…………………………………………………………………………..81

3.1 Outline..……………………………………………………………………………81

3.2 Introduction…………………………………………………………………….….82

3.3 Experimental Section…………………………………………………………..….85

3.4 Results……………………………………………………………………………..96

3.5 Discussion…………………………………………………………………..……105

3.6 Conclusion……………………………………………………………………….108

3.7 Acknowledgement……………………………………………………………….109

IV. POST-ELECTROSPINNING “CLICK” MODIFICATION OF DEGRADABLE AMINO ACID-BASED POLY(ESTER UREA) NANOFIBERS…………………...…110

4.1 Outline..…………………………………………………………………………. 110

4.2 Introduction………………………………………………………………………111

4.3 Experimental Section………………………………………………………...…..114

4.4 Results and Discussion…………………………………………………………..135

4.5 Conclusion……………………………………………………………………….152

4.6 Acknowledgment…………………………………………………………….…..153

V. PEPTIDE-FUNCTIONALIZED OXIME HYDROGELS WITH TUNABLE MECHANICAL PROPERTIES AND GELATION BEHAVIOR……………………...154

5.1 Outline..…………………………………………………………………………. 154

5.2 Introduction………………………………………………………………………155

5.3 Experimental Section………………………………………………………...…..158

5.4 Results and Discussion…………………………………………………………..168

5.5 Conclusion……………………………………………………………………….181

5.6 Acknowledgment…………………………………………………………….…..182

VI. CASCADING “TRI-CLICK”FUNCTIONALIZATION OF POLYCAPROLACTONE THIN FILMS QUANTIFIED VIA QCM……………………………………………....183

6.1 Outline..…………………………………………………………………………. 183

6.2 Introduction………………………………………………………………………184

6.3 Experimental Section………………………………………………………...…..186

vii

6.4 Results and Discussion…………………………………………………………..201

6.5 Conclusion……………………………………………………………………….215

6.6 Acknowledgment…………………………………………………………….…..216

REFERENCES…………………………………………………….…………………...238

APPENDIX…………………………………………………………………………….271

viii

LIST OF TABLES

Table Page

1.1. Structure of cyclic lactones and corresponding polymers…………………………….5

3.1. Characterization data summary for the amino acid-based poly(ester ureas)…………88

3.2. Summary of the mechanical properties of peptide-crosslinked PEU………………..99

4.1. Reaction formulation summary of experiment groups and control groups for nanofiber surface modification. …………………………………………………………...….133

4.2. Molecular mass data of PEUs. aMass average molecular weight (g/mol); bDistribution of molecular mass (Dm, Mw/Mn) of polymers after precipitation in water, which significantly narrows the molecular mass distribution from what is expected in a step growth polymerization……………………………………………………..………139

4.3. Comparison of tensile properties of nanofiber matrices: PEU, PCL, PLGA, gelatin, cartilage and skin. Nanofibers of PEU, PCL, PLGA, and gelatin are randomly oriented. PEU fibers’ diameter: 350-500 nm; PCL and gelatin fibers’ diameter: 10-1000 nm; PLGA fibers’ diameter: 500-800 nm………………………………………………148

5.1. Molecular mass data of polymers. Mn: number average molecular weight; PDI: polydispersity index. Mn and PDI were calculated from polystyrene standards………………………………………………………………………..…170

6.1. Thickness of Spinning coated polyester films…………………………………….199

6.2. Summary of (a) QCM frequency shift (△f) after peptide conjugation to mono- functionalized polymers surface, (b) Amount of covalently bound RGD calculated from △f based on Sauerbrey model, (c) reaction efficiency was calculated by the ratio between reacted functional groups and the whole amount in entire film; (d) change of contact angle (△θ) after peptide conjugation……………………………………..206

LIST OF FIGURES

Figure Page

1.1. Shear strength of SR-PGA rods after implantation in subcutis of rabbits. Rod sizes: (○) 1.5 x 50 mm, (△) 2.0 x 50 mm, (□) 3.2 x 50 mm, and (◇) 4.5 X 50 mm……...... 8

1.2. (a) SEM micrograph of the PGA fiber scaffolds coated with 45S5 Bioglass particles. (b) Excised PGA-Bioglass composite at 42 days viewed enface. The implanted meshes were completely encapsulated by new tissue……………………………...... 9

1.3. Photomicrograph of the distal femur at 36 months of follow-up. PLLA was still seen in the screw channel (asterisk), and a bone rim (arrow) has formed around the screw……………………………………………………………………………..…11

1.4. SEM micrographs of C17-2 cells on the PLLA nano-fibrous scaffold (5% wt/v) cultured for 1 day: (a) small magnification (1000); (b) magnified view of a differentiated cell with short neurite (2000)………….…………………………….12

1.5. Modeled in vivo release profiles for 50:50, 65:35, 75:25 and 85:15 PLGA (PLA:PGA). A biphasic release profile with an initial zero release period followed by a rapid drug release has been observed. The profiles correspond to the degradation process of PLGA……………………………………………………………………………....14

1.6. Decrease in molecular weight (MW) of PCL capsules with time after implanted in rats. The results indicated a linear relationship between logarithm of MW and time…………………………………………………………………………………17

1.7. (a) Implantation of the PCL-TCP scaffold into the pig cranium. (b) SEM analysis using backscattered electron mode. Visualisation of the calcium content represented in grey. (c) Extensive mineralization is evident throughout the scaffold in u-CT analysis. (d) Histological analysis showing new bone formation (nb) scaffold struts (s) osteoblasts (ob) and osteocytes (oc)…………………………….………………………………18

1.8. Synthesis scheme of PPF: (a) step-growth polycondensation (b) chain growth polymerization…………………………………………………………………...…20

1.9. (a) Mean pore sizes of PPF scaffolds compared to computer-aided design (CAD) models in two directions. The dashed line indicates exact matching between CAD

models and fabricated PPF scaffolds. Values represent mean ± SD (n = 5). (b) Mean pore sizes of PPF scaffolds with various sized square pores in two directions. Scale bars 400 μm……………………………………………….………………………..21

1.10. Compressive modulus of PPF/PPF-DA networks as a function of degradation time for the four sample groups: A (◇, PPF 33%, PPF-DA 67%, pH 7.4 ), B (□, PPF 30%, PPF-DA 60%, TCP 10%, pH 7.4), C (△, PPF 67, PPF-DA 33%, pH 7.4)), and D (○, PPF 33%, PPF-DA 67%, pH 5.0) (all in weight percentage). Bars represent means ± standard deviation for n = 5………………………………………………………..23

1.11. Schematic representation of amino acid containing degradable polymers and possible specific functions...... 25

1.12. (a) Molecular model of a single KLD-12 self-assembling peptide. (b) Photo picture of chondrocyte-seeded peptide hydrogel. (c) Light microscope image of chondrocytes encapsulated in peptide hydrogel. The alternating hydrophobic and hydrophilic residues on the backbone promote β-sheet formation. The positively charged lysines (K) and negatively charged aspartic acids (D) are on the lower side of the β-sheet, and the hydrophobic leucines (L) are on the upper side. This molecular structure facilitates self-assembly through intermolecular interactions………………………………...26

1.13. Synthesis scheme of poly (amino acid)s via NCA polymerization. “A” and “B” chain groups depend on catalysis systems..………………………………………………28

1.14. (a) Scheme for covalent coupling GRGDY-I2 peptides to PLA-PLL via amine residues in PLL units. SEM pictures of bovine aortic endothelial (BAE) cells attached to PLA (b) and RGD modified PLA-PLL (c). Cell spreading was enhanced in peptide modified polymers…………………………………….……………………………30

1.15. (a) Synthesis scheme of poly (amino acid)s containing DOPA (blue) and L-lysine. (b) Influence of polymer concentration on adhesive strength. (c) Adhesive strength of copolymers on different substrates. (CBZ = carbobenzyloxy, PS = polystyrene, PE = polyethylene, PMMA = poly(methyl methacrylate………………………………...32

1.16. Synthesis scheme of PEAs via solution polycondensation, interfacial polymerization and melt polycondensation…………………………………………………………33

1.17. Molecular weight (Mw) changes of PEAs incubated in (a) PBS at 37 °C for up to 28 days and, (b) chymprtypsin at 37 °C for up to 72 h. As to the nomination, Ph48 means the amino acid is L-phenylaniline, number 4 means the ester unit (Y in Figure 1.16) is -(CH2)4-, and number 8 means the amide unit (X in Figure 1.16) is - (CH2)8-……………………………………………………………………………..35

1.18. Micrographs of Hela cells after 48 hours culture. (a) Cells cultured on the surface of a pure HA hydrogel; (b) cells cultured on the surface of the HA-PEA hybrid hydrogel; (c) wet hydrogel after staining and washing as a blank control; (d) MTT assay for the xi

Hela cells after 48 hours' culture in a Dulbecco's minimal essential medium (DMEM) medium on HA and HA-PEA hybrid hydrogel surfaces. Hybrid hydrogels exhibited a significant enhancement of cell attachment and proliferation…………………….37

1.19. Chemical structures of tyrosine-derived polycarbonates and corresponding monomers with different alkane pendent chains………………………………..38

1.20. In vivo degradation profiles of poly(DTE carbonate), poly(DTH carbonate) and PLLA in canine bone chamber model………………………………………………40

1.21. (a) Average percent area occupied by bone in the PLLA, poly(DTE carbonate), and poly(DTH carbonate) test chambers in canine bone model. (b) Interface between pins of poly(DTE carbonate) and bone after 3 years implantation. Mineralized bone was stained red, fibrous tissue was blue, and osteoid was stained in a green hue. High frequency (73%) of bone apposition responses was observed...... 42

1.22. Schemes of six popular “click” reactions………………………………………….44

1.23. Simplified representation of CuAAC mechanism…………………………………46

1.24. (a) Bioinspired modular synthesis of elastin-mimic polymers via CuAAC (top). Structures of peptide oligomer monomers 1 (VPGVG), 2 (GVGVP), and 3 (VDPGVG) with the corresponding elastin-mimic polymers P-1, P-2, and P-3 (bottom) (V=Valine, P=Proline, G=Glycine). (b) True stress−strain curves for the elastin-mimic polymers in hydrated forms (13 wt% water): P-1, blue; P-2, cyan; P-3, magenta. (c) LCSTs of elastin-mimic polymers at a concentration of 3.0 mg/mL in water (Wavelength 550 nm): P-1, black; P-2, red; P-3, green……………………………………………..…47

1.25. (a) Synthesis scheme of (a) alkyne derived polyesters via ring opening polymerization (ROP) and their functionalization with PEG oligomers or RGD peptides via CuAAC. (b) Mannose-poly (amino acid) conjugates via CuAAC post-polymerization technique……………………………………………………...... 49

1.26. (a) Fabrication scheme of polysaccharide-based hydrogels via CuAAC. (b) Photographs of hydrogels (b) before the removal of copper and (c) after the removal of copper via dialysis against H2O/EDTA. (d) SEM images of hydrogels containing S.cerevisiae cells……………………………………………………………………50

1.27. Activated alkynes for copper free azide-alkynecycloaddition reactions…………..52

1.28. Click-functionalized macromolecular precursors react via the SPAAC to form an end- linked hydrogel network; By varying the molecular weight of the PEG, well-defined networks of differing cross-linking density are formed……………………...... 53

1.29. (a) Scheme of thiol-ene photo-initiated radical addition. (b) Scheme of thiol-ene Michael addition. TEA = triethylamine, and EWG = electron-withdrawing groups..55 xii

1.30. Computational and kinetic analysis of the influence of alkene functionality on the reactivity of thiol-ene photo-initiated addition……………………………………...56

1.31. Chemical structures of carbon-carbon double bond containing polymers (a) and thiol derivates (b) for thiol-ene reaction………………………………………………….58

1.32. (a) Preparation scheme of Fe3O4-based cysteine (Cys) functionalized magnetic water-soluble nanoparticles (NPs). (b) T2-weighted magnetic resonance images of (1) Cys-Fe3O4 NP-labeled cells and (2) untreated cells as the control. A dark contrast appeared in b(1) marked by the arrow, indicating that Cys-Fe3O4 NPs can be employed as an MRI contrast agent………………………………………………...60

1.33. Synthesis scheme of poly (thioether) based dendrimers and the functionalization of the chain ends using thiol-ene addition……………………………………………..61

1.34. Schematic of (a) Chain polymerized (CP) hydrogels via radical polymerization and (b) Step polymerization (SP) hydrogels via thiol-ene addition. (c) Shear toughness of two types of hydrogels. SP hydrogels were tougher than CP hydrogels upon stretching. (d) The erosion depth of photopatterned channels as a function of irradiation time. SP hydrogels degraded faster than CP hydrogels………………………………………61

1.35. (a) Schematic of photo-patterning process of C=C containing hydrogels via thiol-ene “click” chemistry. Fluorescent images of hydrogels after photo-patterning: (b) Three different fluorescently labeled biological cues were incorporated to hydrogels at user- defined times and spatial locations sequentially. (c) By controlling the focal point of the laser light in three dimensions using a confocal microscope, micrometer-scale spatial patterning resolution was achieved………………………………………….62

1.36. (a) Mechanism of imine bond formation. (b) Transformation of the protonated aniline Schiff base to the oxime/hydrazone. (c) Commercially available chemicals as superior catalysts for oxime/hydrazone formation. At pH 7.0 and 7.4, in the catalyst sequence from the left to the right, the reaction speed increases………………………………65

1.37. (a) Quantitative attachment of alkoxyamine derived PEG to ketone-PCL.308,309 (b) Bioconjugation between alkoxyamine terminated poly(N-isopropylacrylamide) (pNIPAAm) and bovine serum albumin (BSA) proteins derived with ketone……………………………………………………………………………….66

1.38. Synthesis scheme of (a) ketone derived Hyaluronic acid (HA) via oxidation and (b) hydrazide derived HA via carbodiimide coupling…………………………………..67

1.39. (a) Chemical structures of two precursors for self-healable hydrazone hydrogels, and schematic of resulted hydrazone bond. (b) Self-healing properties of hydrazone hydrogels in different pH……………………………………………………………68 xiii

1.40. (a) Chemical structure of Doxorubicin (DOX) grafted PEO. (b) pH-dependent release at 37 °C from PEO-based prodrugs (drug loading 2.9%)…………………………..70

1.41. (a) Fabrication of micropatterned aminooxy surfaces on glass substrates by photo- deprotection, and subsequent ligand immobilization via oxime ligation. (b), (c), and (d) Patterns with different morphology were obtained via using different photomask………………………………………………………………………………...71

3.1. The two-step general synthetic route of amino acid-based poly(ester urea)s (PEU)s. First 1,6-hexane diol was condensed with 2 equivalents of α-amino acids catalyzed and protonated with p-toluenesulfonic acid. Following the diamine salt synthesis, interfacial polycondensation with triphosgene yielded the PEU homopolymer. To both enhance the mechanical properties and impact specific biological signaling to the polymer, highly reactive vinyl groups were coupled to a short fragment of OGP peptide and the construct was crosslinked photochemically. Small mole fractions of crosslinker and relatively high molecular masses enabled increased mechanical properties despite inducing radical-based chain scission…………………………………………….…86

3.2. The mitogenic OGP(10–14) peptide, YGFGG, is symmetrically functionalized with lysine (K) residues at both the N- and C-termini (KYGFGGK) with a highly reactive vinyl groups coupled on the side chain of each lysine residue……………….….….89

3.3. The elastic modulus and tensile properties of the poly(1-LEU-6) and the poly(1-PHE- 6) were measured using an Instron 3365 universal materials testing machine. The Young’s moduli of the poly(1-PHE-6) (black), 0.5% OGP poly(1-PHE-6) (red), and 1.0% OGP poly(1-PHE-6) (blue) data were determined using a TA Q800DMA instrument. Using small strains (<0.15%), the Young’s moduli were determined using the slope of the tangent line in the linear regime. Values for Young’s moduli and standard deviations were determined from four individual measurements……….…98

3.4. WST-1 proliferation assay of MC3T3-E1osteoblast and primary murine fibroblast cells. Consistent with the biphasic, concentration-dependent proliferative effect as previously reported371, the unfunctionalized homopolymers (PLLA, poly(1-LEU-6) or poly(1-PHE-6)) did not exhibit cell-type-dependent proliferative activity. However, the increasing proliferative trends in both the 0.5% and 1.0% OGP functionalized materials for the osteoblasts relative to the corresponding decreasing trends in the fibroblasts showed that the peptide was bioavailable to the receptor and that we were in a bioactive concentration regime. ∗P < 0.05 compared to polymer controls…………………………...... 100

3.5. Digital images of histology stained slides stained with Masson’s Trichrome at 100×. Tissues were removed at 4 weeks (row A) and 12 weeks (row B) post-implantation. Row C, serial sections of 12 week histology slides shown in row B stained with Alizarin Red………………………………………………………………………..103 xiv

3.6. Quantitative histological analysis of the respective measurements as collected from Masson’s Trichrome analysis at 4 weeks (black) and 12 weeks (grey). Analysis representative of biodegradation (A) by measuring area of tissue migration within the polymer space, cellular infiltration (B), capsule thickness (C), immune response (D) by number of giant cells present and vascularization (E) by counting the number of associated blood vessels. Statistical significance (P < 0.05) was indicated by 1 = compared to PLA; 2 = compared to base polymer; 3 = 4 week vs. 12 week results; 4 = 0.5% OGP vs. 1.0% OGP; 5 = poly(1-PHE-6) vs. poly(1-LEU- 6)…………………………………………………………………………………...104

4.1. 1H NMR of monomers (DMSO-d6). (a) L-phenylalanine based monomer for PEU-1 (M1); (b) Alkyne-monomer for PEU-2 (M2): alkyne signals 4.76 and 3.55 ppm; (c) Azide-monomer for PEU-3 (M3): azide signal 3.50 ppm; (d) Alkene-monomer for PEU-4 (M4): alkene signals 5.85 and 5.00 ppm; (e) Benzyl-protected tyrosine monomer for PEU-5 (M5): Bzl signal 5.07 ppm. Solvent residues are marked with asterisks…………………………………………………………………………....137

4.2. Size-exclusion chromatography (SEC) elution curves of the individual PEUs (RI signal, eluent DMF with 0.01M LiBr, flow rate 0.8 mL/min, temperature 50 oC, polystyrene standards)…… …………………………………………………………………….138

4.3. 1H NMR spectra of PEU (DMSO-d6): (a) PEU-1, without chemical functionalities; (b) Alkyne PEU-2; (c) Azide PEU-3; (d) Alkene PEU-4; (e) Benzyl protecting PEU-5; (f) Tyrosine PEU-6; (g) Ketone PEU-7. The content of alkyne functionalized units in PEU-2 is 2.5%; the number is 5% in other functionalized PEUs. Solvent residues are marked with asterisks…………………………………………………………..….141

4.4. Normalized UV-Vis Spectra of PEUs. The peak at 257 nm is assigned to phenylalanine absorption; the shoulder at 278 nm is assigned to tyrosine unit absorption. The content of alkyne derived tyrosine units in PEU-2 is 2.5%; in other PEUs, that number is 5%...... 142

4.5. 1H NMR spectra of PEUs following derivation with small molecule probes (DMSO- d6). (a) CuAAC between PEU-2 and 3-azidopropan-1-ol; (b) CuAAC product between PEU-3 and propargyl alcohol; (c) Thiol-ene addition product between PEU-4 and mercaptopropionic acid; (d) Oxime ligation product between PEU-7 and o-(pent-4-en- 1-yl) hydroxylamine…………………………………………………………….…145

4.6. SEM micrograph of PEU nanofibers (a) scale bar 1 um and (b) scale bar 100 nm; Nanofibers with a narrow diameter distribution (350 nm to 500 nm) are obtained. (c) Stress-strain curve of PEU nanofiber matrices, where the Young’s modulus is 300 ± 45 MPa, ultimate tensile stress is 8.5 ± 1.2 MPa, and the ultimate tensile strain is 65 ± 8 % (n=3). PEU nanofiber matrices have sufficient mechanical properties, which meet the requirements for some trabecular bones447, cartilage448 and skin449 defect repairs……………………………………………………………………………...147

4.7. Fluorescent images of nanofibers that were modified post-electrospinning (a-e) and the corresponding chemical reactions (a1-e1). (a) PEU-2 labeling with Chromeo 488 azide via CuAAC click methods; (a1) Chemical reaction site between PEU-2 and Chromeo 488 azide via CuAAC; (b) PEU-3 modified with alkyne-RGD-biotin via CuAAC methods, followed by labeling with TRITC-conjugated streptavidin; (b1) Chemical reaction site between PEU-3 and alkyne-RGD-biotin via CuAAC; (c) PEU- 4 labeling with FITC-RGD-thiol via thiol-ene reaction; (c1) Chemical reaction site between PEU-4 and FITC-RGD-thiol via thiol-ene reaction; (d) PEU-6 modified with alkyne-derivatized diazodicarboxamide (compound 4.21) via ene-type addition, followed by labeling with Chromeo 488 azide; (d1) Chemical reaction scheme between PEU-6 and alkyne derived diazodicarboxamide; (e) PEU-7 labeling with Alexa fluor 568 hydrazide fluorescence via hydrazone formation; (e1) Chemical reaction site between PEU-7 and Alexa fluor 568 hydrazide fluorescence via hydrazone formation. Fluorescent images of labeled PEU nanofibers (x20, scale bar 50 m)……………………………………………………………………………….151

4.8. Fluorescent images of nanofibers after physical absorption as control groups. (a) PEU- 1 nanofibers were soaked in PBS buffer solution containing Alexa Fluor 488 hydrazide; (b) PEU-6 nanofibers were directly incubated in PBS buffer solution containing Chromeo 488 azide without the step of “ene-type” modification. (x20, scale bar 50 m)…………………………………………………………………………………..152

5.1. MALDI-TOF mass spectra of (a) Bi-aldehyde functionalized PEG (PEG-bCHO); (b) Model reaction product between PEG-bCHO and o-(pent-4-en-1-yl) hydroxylamine (Intermediate 5.2) show monomodal distributions and complete reaction conversion………………………………………………………………………….169

5.2. SEC elution curves of PEG raw materials (black), PEG-bCHO (red), and PEG-b- alkene (blue). THF was used as eluent with a flow rate of 1.0 mL/min at 30 °C……………………………………………………………………………….170

5.3. (a) The chemical structure of bi-aldehyde-functionalized PEG (PEG-bCHO); (b) The chemical structure of 4-arm aminooxy crosslinker; (c) The general reaction scheme for oxime ligations. The reaction is pH and nucleophile catalyst dependent; (d) A schematic of hydrogel network formation via oxime bond ligation using a 4-arm aminooxy crosslinker; (e) Photos of hydrogels formed at pH 7.4 and hydrogels formed at pH 4.5 (f) show clearly the mechanical differences in the hydrogels with the fast crosslinking kinetics. Gels obtained at pH 4.5 are stronger than those at pH 7.4 at similar time intervals. (g) The FT-IR spectra of PEG-bCHO (black) and hydrogels (red) clearly show the extent of oxime formation in the hydrogels; (h) Raman spectra of phenyl and C=N transitions for PEG-bCHO (black) and hydrogels (red) compliment the FT-IR data showing the extent of reaction during the gel formation…………………………………………………………………………..171

xvi

5.4. The extent of crosslinking and mechanical properties are highly dependent on pH. (a) Modulus vs. Frequency behavior of hydrogels formed at pH 4.5 and 7.4 show that the gels formed at low pH are much stronger at comparable time intervals; (b) The storage modulus (G’)-pH plot (frequency 10 rad/s, strain 10%) show that the mechanical properties can be tuned via pH. The data are displayed as the mean and the error bars represent the standard deviation of three independent measurements for each condition…………………………………………………………………………..173

5.5. The gelation kinetics are highly dependent on pH. (a) Modulus-time behavior of hydrogels formed at pH 4.5; (b) The plot of pH vs time captures the gel point and the times to reach complete gelation. At low pH the gelation time is too fast to capture. The asterisks (*) in (b) represent time intervals that are too short to detect. The data represent the mean and the error bars represent the standard deviation of three independent samples for each condition…………………………………………..176

5.6. Influence of aniline catalyst (0.1 wt%) on hydrogels storage modulus (a), time to reach the gel point (b) and the complete gelation time (c) as a function of pH. The data represent the mean and the error bars represent the standard deviation of three independent samples for each condition…………………………………………..178

5.7. (a) Fabrication procedure of oxime hydrogels with “clickable” groups. PEG-bCHO was initially mixed together with Alkene/Azide-ONH2 for minutes, followed by the addition of 4-arm crosslinker. (b1) FT-IR spectra of azide functionalized hydrogels before (black) and after (red) “click” conjugation with alkyne-RGD-biotin peptides. (b2) Fluorescence microscopy image of azide hydrogels after bonding with TRITC labeled streptavidin. Red areas are from labeled hydrogels and black areas are the glass side background. (b3) Schematic of the “click” cycloaddition reaction and the binding sites for peptide or protein conjugation. (c1) Raman spectra of alkene hydrogels before (black) and after (red) the thiol-ene addition of 2-mercaptoethanol. (c2) Fluorescence microscopy image of alkene functionalized hydrogels after 3D patterning with FTIC- RGD-thiol. (c3) Schematic of thiol-ene addition sites. (Scale bar 100 um)………180

6.1. 1H NMR spectrum of mono-functionalized PCL in CDCl3: (a) PCL-Ketone, (b) PCL- Alkyne, (c) PCL-Azide, (d) PCL-MA. The oxime ligation is quantitative and proceeds under reaction conditions that minimize the degradation of the polymer. Peaks of CDCl3 and water are marked with asterisk………………………………………..203

6.2. FT-IR spectra of mono-functionalized polymers indicate the presence of the functional groups. (a) PCL-Alkyne: -C≡C-H stretch 3264 cm-1; (b) PCL-Azide: -N3 stretch 2096 cm-1; (c) PCL-MA: C=C-H stretch 3077 cm-1 and C=C stretch 1642 cm- 1…………………………………………………………………………………....204

6.3. 1H NMR spectra of functionalized PCL polymers in CDCl3. Peaks between 5.75 ppm and 6.50 ppm are assigned to alkene group of MA marked within red dash line. (a) PCL-MA-alkyne: characteristic peak of alkyne 2.45 ppm marked within green dash

xvii

line; (b) PCL-MA-alkyne-azide: characteristic peak of methyl group bonded to azide 3.25 ppm marked within blue dash line. (c) PCL-MA-azide………………………205

6.4. QCM frequency shift: (a) RGD-aminooxy and PCL-Ketone; (b) RGD-thiol and PCL- MA; (c) RGD-azide and PCL-Alkyne, (d) RGD-alkyne and PCL-Azide via CuAAC; (e) RGD-cyclooctyne and PCL-Azide via SPAAC…………………………………207

6.5. Static contact angle of PCL-alkyne thin film before (a) and after (b) RGD-azide peptide conjugation. Contact angle drops after reaction indicates the increase of surface hydrophylicity due to water-soluble RGD attachment to film……………………..210

6.6. 1H NMR spectra of solution cast films in CDCl3 before (red) and after (black) surface reactions: (a) Keto-PCL with o-(prop-2-yn-1-yl)hydroxylamine (intermediate 6.3); (b) PCL-alkyne with 3-azidopropan-1-ol; (c) PCL-azide with propargyl alcohol; (4) PCL- MA with 2-hydroxyl-1-ethanethiol…………………………………………………212

6.7. The shift in QCM frequency for the cascading conjugation between peptide and surface of multi-functionalized polymer thin film. (a) PCL-MA-alkyne: Michael addition between RGD-thiol and MA units first, CuAAC between RGD-azide and alkyne units second; (b) PCL-MA-azide: Michael addition between RGD-thiol and MA units first, SPAAC between RGD-cyclooctyne and azide units second; (c) PCL-MA-alkyne-azide: Michael addition between RGD-thiol and MA units first, SPAAC between RGD- cyclooctyne and azide units second, CuAAC between RGD-azide and alkyne units third………………………………………………………………………………...214

6.8. The shift in QCM frequency in the PCL-MA-azide system. In the first step (black line) RGD-cyclooctyne for SPAAC was introduced. A significant decrease of frequency indicates the extent of a chemical reaction. In the second step (red line) RGD-thiol for Michael Addition was introduced. A slight drop of frequency indicates that little to no chemical reaction occurred………………………………215

A. 1. 1H NMR (DMSO-d6) spectrum of PEU-1 monomer 4.1 (M1)…………………...250

A. 2. 13C NMR (DMSO-d6) spectrum of PEU-1 monomer 4.2(M1)………………...... 251

A. 3. 1H NMR (DMSO-d6) spectrum of PEU-2 monomer 4.2(M2)…………………….251

A. 4. 13C NMR (DMSO-d6) spectrum of PEU-2 monomer 4.2(M2)……………..…...252

A. 5. 1H NMR (CDCl3) spectrum of intermediate 4.3……………..…………………..252

A. 6. 1H NMR (CDCl3) spectrum of intermediate 4.4………………..………………..253

A. 7. 1H NMR (CDCl3) spectrum of intermediate 4.5…………………………..……..253

A. 8. 1H NMR (DMSO-d6) spectrum of PEU-4 monomer 4.6 (M4)………………..…254

xviii

A. 9. 13C NMR (DMSO-d6) spectrum of PEU-4 monomer 4.6 (M4)…………………254

A. 10. 1H NMR (CDCl3) spectrum of intermediate 4.7……………………...... 255

A. 11. 1H NMR (CDCl3) spectrum of intermediate 4.8…………………...…………..255

A. 12. 1H NMR (CDCl3) spectrum of intermediate 4.9………..……………………...256

A. 13. 1H NMR (DMSO-d6) spectrum of PEU-3 monomer 4.10 (M3)…………..…..256

A. 14. 13C NMR (DMSO-d6) spectrum of PEU-3 monomer 4.10 (M3)……………..257

A. 15. 1H NMR (CDCl3) spectrum of intermediate 4.11………………………….….257

A. 16. 1H NMR (DMSO-d6) spectrum of PEU-5 monomer 4.12 (M5)…………..…..258

A. 17. 13C NMR (DMSO-d6) spectrum of PEU-5 monomer 4.12 (M5)…………...... 259

A. 18. 1H NMR (CDCl3) spectrum of intermediate 4.13………………………….….259

A. 19. 1H NMR (DMSO-d6) spectrum of intermediate 4.14……………………..…..260

A. 20. 13C NMR (DMSO-d6) spectrum of intermediate 4.14…………………..…....260

A. 21. 1H NMR (CDCl3) spectrum of intermediate 4.15………………………..…...261

A. 22. 1H NMR (DMSO-d6) spectrum of intermediate 4.16…………………..….…261

A. 23. 13C NMR (DMSO-d6) spectrum of intermediate 4.16………………..……...262

A. 24. 1H NMR (DMSO-d6) spectrum of intermediate 4.17…………………..….…262

A. 25. 1H NMR (DMSO-d6) spectrum of intermediate 4.18………………..…….…263

A. 26. 1H NMR (DMSO-d6) spectrum of intermediate 4.19………………..…….…263

A. 27. 1H NMR (DMSO-d6) spectrum of intermediate 4.20……………….…….…264

A. 28. 13C NMR (DMSO-d6) spectrum of intermediate 4.20……………….……...264

A. 29. FT-IR spectrum of PEU-3 (azide-PEU) before and after CuAAC……………265

A. 30. Optical microscope image of PEU nanofibers (x20)……………….………...266

A. 31. Optical microscope image of PEU nanofibers (x100)…………………..…....266

A. 32. UV-Vis absorption of PEU-4 nanofibers after thiol-ene reaction with Fmoc-RGD- thiol………………………………………………………………………………267 xix

A. 33. Fluorescent microscope image of alkyne derived PEU-7 (keto-PEU) nanofibers labeled with Chroemo 488 azide (x20). Scale bar 50 um………………………..267

A. 34. Fluorescent images of nanofibers in controls (x20, 50 um). PEU-2, without copper catalyst. PEU-3, without copper catalyst. PEU-4, without UV irradiation. PEU-6, directly coupled with Chromeo 488 azide without intermediate 4.21 PEU-7, using PEU-1 nanofibers as control. Alkyne derived PEU-7 nanofibers, directly coupled with Chromeo 488 azide without intermediate 4.14…………………………………….268

A. 35. ESI spectrum of alkyne-RGD-biotin…………………………………………….269

A. 36. MALDI spectrum of FITC-RGD-thiol…………………………………………..270

A. 37. MALDI spectrum of Fmoc-RGD-thiol……………………………………….….271

A. 38. 1H NMR of PEU-2 nanofibers…………………………………………………..271

A. 39. 1H NMR of PEU-3 nanofibers…………………………………………………..272

A. 40. 1H NMR of PEU-4 nanofibers.………………………………………………….272

A. 41. 1H NMR of PEU-6 nanofibers.……………………………………….………….273

A. 42. 1H NMR of PEU-7 nanofibers. …………………………………………………273

A. 43. ESI spectrum of 4-arm aminooxy crosslinker. ……………………….………….274

A. 44. ESI spectrum of Alkyne-RGD-Biotin……………………………………………275

A. 45. MALDI-TOF spectrum of FTIC-RGD-thiol…………………………………….276

A. 46. 1H NMR spectrum of 2-(2-(aminooxy)acetoxy)ethyl acrylate (intermediate 5.1)…………………………………………………………….…………….…..…277

A. 47. 1H NMR spectrum of O-(6-azidohexyl) hydroxylammonium chloride (intermediate 5.2)…… ………………………………………………………………………...…277

A. 48. 1H NMR spectrum of Propargyl hydroxylammonium chloride (intermediate 5.3)…………… ……………………………………………………….………….278

A. 49. ESI spectrum of RGD-cyclooctyne……………………………………………..278

A. 50. FT-IR spectra of PCL-azide solution casting film before (red) and after (black) reaction with propargyl alcohol…………………………………………………...279

A. 51. FT-IR spectra of PCL-alkyne solution casting film before (red) and after (black) reaction with 3-azidopropan-1-ol…………………………………………….……280 xx

A. 52: MS Spectrum (Na+ as ion source) of PCL-MA with propargyl alcohol as initiator…………………………………………………………………………….281

xxi

LIST OF SCHEMES

Scheme Page

4.1. The synthetic route of monomer 4.1 (M1) for PEU-1……………………………...115

4.2. Chemical structure of monomer 4.2 (M2) for PEU-2………………………...... 116

4.3. The synthetic route of PEU-4 monomer 4.6 (M4)………………………………….118

4.4. The synthetic route of PEU-3 monomer 3.10 (M3)………………………………...119

4.5. The synthetic route of PEU-5 monomer 4.12 (M5)…………………………………120

4.6. The synthetic route of compound 4.14……………………………………………..121

4.7. The synthetic route of compound 4.16……………………………………………..122

4.8. The synthetic route of alkyne derived cyclic diazodicarboxamide (4.21)………….123

4.9. The synthetic route for PEUs using interfacial polycondensation methods. PEU (1) no chemical functionalities; PEUs possessing functional alkyne (2), azide (3), alkene (4), benzyl-protected tyrosine (5), unprotected tyrosine (6), and ketone (7) can be synthesized using a copolymerization approach that is highly versatile…………………………………………………………………………....126

5.1. (a) Synthesis route of bi-aldehyde functionalized PEG Precursor (PEG-bCHO, n=104). (b) Synthesis route of 4-arm aminooxy crosslinker. (c) Chemical structure of azide- aminooxy extender (azide-ONH2). (d) Chemical structure of alkene-aminooxy extender (alkene-ONH2). (e) Model reaction of oxime ligation between PEG-bCHO and o-(pent-4-en-1-yl) hydroxylamine (Intermediate 5.2)………………………..160

5.2. Synthesis route of azide-aminooxy extender (azide-ONH2)………………..……162

5.3. Synthesis route of alkene-aminooxy extender (alkene-ONH2)…………………..163

5.4. Model reaction of oxime ligation between PEG-bCHO and Intermediate 5.2…………………………………………………………………………………165

6.1. Synthesis route for 2-(2-(aminooxy)acetoxy)ethyl acrylate (6.1)……………...…188

xxii

6.2. Synthesis route for O-(6-azidohexyl)hydroxylammonium (6.2)………………….190

6.3. Synthesis route for propargyl hydroxylammonium chloride (6.3)………………..191

6.4. The synthetic route for the ring opening polymerization of keto-derivatized poly(caprolactone). The use of multiple functionalized amino-oxy reactive groups enables the post polymerization modification of multiple “clickable” groups on PCL- based copolymer films…………………………………………………………….193

6.5. The synthesis route of RGD-cyclooctyne…………………………………………198

xxiii

CHAPTER I

INTRODUCTION

During the last several decades, significant efforts have focused on the development of advanced materials for biomedical applications. The development and use of novel biomaterials are expected to dramatically increase in the coming decades. By definition, a biomaterial is a material that interfaces with a biological system to evaluate, treat, augment or replace any tissues, organ or function of the body.1

Conceptually, there are three generations of biomaterials.2 During the 1960s and the 1970s, the first generation of biomaterials was used inside the human body, including metals, ceramics, polyethylene (PE), and so on. Those materials arose from available sources and are generally inert when implanted in the body. The main goal is to reduce or minimize the immune response to the foreign materials.3 The second generation of biomaterials development occurred in 1980s. They were designed specifically to be bioactive, biodegradable, and resorbable. Those materials were designed to interact with the biological environment in a way that enhanced the biological response, including the tissue/surface bonding and the tissue regeneration or healing. The third generation biomaterials have the potential to stimulate specific cellular responses at the molecular level, directing functions such as cell proliferation, differentiation, and extracellular matrix

(ECM) production and organization. The molecular modification of biomaterials with specific biomaterials provides a promising method to meet those goals. Two strategies are 1 now available for functional repair with the help of third-generation biomaterials: tissue engineering and in situ tissue regeneration. It’s worth noting that the materials in each new generation do not necessarily override the use of those of a previous one.4

Biodegradable polymers provide unique material libraries to meet those goals. They have the characteristics of controllable preparation, processing, diversity, and versatility.

Due to the nearly unlimited pool of potential starting materials, the chemical, physical, and biological properties of the resulting polymers can be tuned precisely from the molecular level. Degradable polymers are rapidly replacing other material classes, especially those used for permanently implanted prostheses such as metals and non-degradable engineering plastics.5 Currently many types of degradable polymers are being investigated or commercialized for biomedical applications, including polyesters, poly-(amino acid)s, poly (ester amide)s, polyurethanes, polycarbonates and polysaccharides, and their blends or copolymers.5 Each of these polymers has found applications based on what properties they possess and how they are functionalized and manufactured.

Discovery and development of advanced polymer-based biomaterials require the combination of chemistry, physics, engineering, biology, simulation, etc. First, novel polymers are continuously developed with unique chemical and physical behavior, in particular the degradation timeline and mechanical properties. Ideally the degradation profile of material scaffolds should match the regeneration process of new tissues, and mechanical properties of those scaffolds should be similar to those of surrounding tissues.

Second, polymer-biomolecule hybrids have emerged as a new strategy to control the interfacial behavior of polymeric materials and the surrounding biological environment.

Third, much attention has also been paid to fabricate well-designed 2D or 3D macro- 2

/micro- structures, mimicking the complexity and versatility of natural biological structures and environment.5-10

Chemistry plays an important role in the history of biomaterial development and improvement. As a practical approach, chemistry is used to transfer virtual materials by combinatorial and computational methods into reality. Often older chemistry finds new life in biomedical fields, and new chemistry is also developed to mitigate the harsh requirements related to biological environments. The most well-know technique is “Click” chemistry, which is defined generally as robust, efficient, and orthogonal covalent chemical reactions. “Click” chemistry has made extensive contributions to prepare materials with well-defined structures, as a result, with desired properties, including mechanical, degradation, physical, biological, and chemical properties.

In the following chapters, we will primarily focus on two approaches : 1) the preparation, properties and applications of degradable polymers, including aliphatic polyesters and amino acid containing polymers; 2) introduction of three highly efficient chemical reactions: copper-catalyzed/free azide-alkyne cycloaddition, thiol-ene addition and oxime ligation, and their applications in development of polymeric biomaterials.

1. 1 Degradable Polymers

Two groups of degradable polymers will be discussed: aliphatic polyesters

(including PGA, PLA, PCL, and PPF) and amino acid-containing polymers (including poly-(amino acid), poly (ester amide), and tyrosine-derived polycarbonate).

1.1.1 Degradable Aliphatic Polyesters

Aliphatic polyesters are some of the most popular degradable polymers in medicine.11 They are generally semi-crystalline polymers with a wide range of physical, mechanical and degradation properties depending on their chemical structures and architectures. Two methods are conventionally used to prepare aliphatic polyesters: polycondensation and ring opening polymerization (ROP).11,12 With regard to polycondensation, the biggest advantage is that it enables the combination of almost any kinds of monomers in a single system and result in polyesters with highly tunable properties.

However, polycondensation always requires more elaborate reaction conditions (high temperature and often vacuum) and longer reaction times to obtain high molecular weight

(MW) polymers. The molecular mass distribution of the resulting polymers is always large

(~2.0), which influences the physical and degradation behaviors.13 Most polyesters used clinically are obtained from ROP that are generated using mild reaction conditions, short reaction times, and the absence of byproducts. ROP of polyesters are living polymerizations. The molecular mass of the final polymer can be controlled easily, and the mass distribution is generally small (<1.2). This method is widely used to prepare commercialized polyesters, including poly (glycolic acid) (PGA), poly (lactic acid) (PLA), poly (caprolactone) (PCL), and any number of respective copolymers. The limitation of this strategy is that monomers are generally limited to six or seven membered lactones.12,14,15

Aliphatic polyesters undergo bulk degradation mechanisms. In vitro this is a hydrolytic process, which takes place randomly in the molecular chain or at the chain end.

Ester hydrolysis is self-catalyzed due to the carboxylic acid byproducts. In physiological 4 environments polyesters are also subjected to enzymatic hydrolysis degradation processes.

Many factors influence the degradation profile, including molecular mass and its distribution, crystallinity, composition of repeat units, implanted location, morphology, surface characteristics, etc. Regardless of the non-enzymatic or enzymatic degradation mechanisms, polyesters degrade into monomers or oligomeric subunits, which are generally soluble in aqueous environments. Those byproducts will be resorbed by the body, excreted in the urine, or metabolized into carbon dioxide and water.16,17

Table 1.1. Structure of cyclic lactones and corresponding polymers.5

Cyclic lactone Linear homopolymer

Poly (glycolic acid) (PGA) Glycolide

Lactide Poly (lactic acid) (PLA)

Dioxanone Poly(dioxanone) (PDS)

Caprolactone Poly (capropactone) (PCL)

Lactide Glycolide Poly (lactic-co-glycolic acid) (PLGA) 5

Among the class of aliphatic polyesters, the most extensively investigated polymers in biomedical field are poly (glycolic acid) (PGA), poly (lactic acid) (PLA), poly

(caprolactone) (PCL), poly-(propylene fumarate) (PPF) and their blends or copolymers.18

The chemical structures of those polymers and corresponding lactone monomers are listed in Table 1.1.5 Due to their physical properties and the ease of industrial level manufacturing, they are wildly used in medical devices and implants, tissue engineering scaffolds, drug/gene delivery vehicles, etc. Many of those materials and devices have been commercialized.19

1.1.1.1 Poly (glycolic acid) (PGA)

PGA is the simplest linear aliphatic polyester. PGA-based sutures (Dexon) have been commercially available since the 1960s. High molecular mass PGA is obtained via the ROP of glycolide, a cyclic dimer of glycolic acid as shown in Table 1.1.20 PGA is a semi-crystalline polymer with crystallinity near 50%. Due to its high crystallinity and short alkane chain length in the monomer units, PGA is very hard and strong. Its tensile yield strength approaches 100 MPa, and the rigidity is around 3 GPa. The melting temperature of PGA is near 230 oC and the glass transition temperature is about 35 oC. The density of

PGA is around 1.6 g/cm3 and is dependent on the crystallinity.5,21 Conventional manufacturing techniques including extrusion, injection, and compression molding have been successfully used to process PGA into well defined structures for biomedical applications. It is very important to control the processing conditions carefully because of

PGA’s high sensitivity to hydrolytic degradation at high temperature.22

As a biodegradable and resorbable polymer, the degradation byproduct of PGA is glycolic acid, which is also produced during normal metabolic processes in the human body.23,24 Hence, after a certain period PGA will eventually be totally resorbed in the body.

Although the glycolic acid can be metabolized, at high concentration it causes a localized acidification that results in tissue damage and further accelerating the degradation process.

PGA degrades relatively quickly in biological environments. As discussed above, this degradation process is influenced by many factors, such as initial molecular mass, surface and 3-dimensional (3D) morphology, implantation location, and crystallinity.25-27 In the in vitro degradation study of Dexon sutures, the degradation of amorphous regions happened in 21 days with the diffusion of buffer solution into those regions, and a further 28 days for the degradation of the crystalline regions. After 49 days, the weight loss was near 40%, and the sutures lost the mechanical strength due to the bulk degradation.28 Tormala and coworkers performed in vitro and in vivo studies of self-reinforced polyglycolide (SR-PGA) composite rods for internal fixation of bone fractures. They found that rods with diameter

1.5 mm lost their mechanical strength in 4-5 weeks following implantation in the subcutis of rabbits, while it took 7-8 weeks for rods with diameter 3.2 mm to lose their mechanical strength (Figure 1.1).29 In a rabbit model in vivo, PGA screws were found to vanish in cancellous bone at 250 days30 to 48 weeks31.

Figure 1.1. Shear strength of SR-PGA rods after implantation in subcutis of rabbits. Rod sizes: (○) 1.5 x 50 mm, (△) 2.0 x 50 mm, (□) 3.2 x 50 mm, and (◇) 4.5 X 50 mm.29

While the use of PGA has limitations due to the acidic degradation byproducts and high degradation rates, there are still many commercial PGA-based devices for regenerative medicine due to its excellent mechanical properties and non-toxic degradation byproducts. The most well-know product are PGA-based sutures.32 Commercial names of

PGA-based sutures include Dexon, Vicry, Maxon, etc. PGA has also been successfully employed as internal fixation devices, including pins, plates, screws and rods.23 For example, self-reinforced PGA composite rods with its fibers exhibited a modulus up to 15

GPa, and the initial shear modulus was up to 250 MPa.29 The ultra-high strength rods were concluded to be suitable for fixation of cancellous bone fractures, osteotomies, or epiphyseal plate fractures.24,33 Besides self-reinforced materials, PGA fibers are also used

8 for the mechanical enhancement of other biomaterials. The incorporation of PGA fibers can increase the compressive modulus of collagen sponges by several orders of magnitude without impairing biocompatibility.34 PGA-based nanofiber scaffolds have also been investigated for tissue engineering.35-37 Day and coworkers found that composites of PGA- fiber and bioglass were biocompatible and were able to stimulate neovascularization, which was beneficial for tissue regeneration and repair, ensuring an adequate blood supply for delivery of nutrients and oxygen into the new tissue.36 As shown in Figure 1.2, PGA- bioglass composites were completely encapsulated by new tissue, demonstrating its bioactivity for tissue engineering applications.36 Langer and coworkers prepared PGA- based nanofiber tubes, which were further stabilized by PLA or PLGA coatings. They found that those devices maintained their structure during fibrovascular tissue ingrowth and drove the formation of tubular tissues with appropriate distribution of smooth muscle cells and endothelial cells.37

Figure 1.2. (a) SEM micrograph of the PGA fiber scaffolds coated with 45S5 Bioglass particles. (b) Excised PGA-Bioglass composite at 42 days viewed enface. The implanted meshes were completely encapsulated by new tissue.36 9

1.1.1.2 Poly (lactic acid) (PLA)

Poly (lactic acid) (PLA) has also been extensively explored for biomedical applications. It is commonly used as the control material to evaluate the behavior of other novel biomaterials. Similar to PGA synthetic methods, high molecular mass PLA is also obtained from ROP.38 The monomer is a six-member cyclic lactides, the di-ester of lactic acid. The starting materials are renewable and obtained from the fermentation of sugar feed stocks. Commercial PLA grades are poly(L-lactic acid) (PLLA) and poly(D, L-lactic acid)

(PDLLA), which are produced from L-lactides and D, L-lactides, respectively. The ratio of

L- to D, L-enantiomers is known to affect the properties of PLA significantly, such as the melting temperature, degree of crystallinity, mechanical properties and degradation behavior. The stereoisomer composition provide a very good method to control PLA properties.39

PLLA is a semi-crystalline polymer, and is used more widely in biomedical application than PDLLA. Its crystallinity depends on the molecular mass, degree of tacticity, and processing parameters. It has a glass transition temperature near 65 oC, and the melting temperature is near 170 oC.39 PLLA has very good thermal stability, and can be processed using traditional thermal forming techniques.39,40 Compared to PGA, PLLA degrades much slower, though their chemical structure difference is only the methyl group.41 In a long-term histological study in the Walton research group, PLLA scaffolds

(rods or screws) were implanted in sheep as orthopaedic fixation devices. PLLA implants lost the mechanical strength around 6 months, but they persisted for years after their surgical role ended.42 Similar results were also observed in the work of Pihlajamäki. As shown in Figure 1.3, after 3 years implantation, PLLA screws were still obvious in rabbit 10 femora.43 Therefore, significant effort has been focused on controlling the degradation of

PLLA using alternative means. Those methods include copolymers, polymer blends, additives, irradiations, and so on.41,44 In the work of Hadjiargyrou, it was demonstrated that the degradation of PLA-based nanofibers can be accelerated via blending with PLA-PEO block polymers or lactide as the hydrolytic catalyst.45 During the degradation process of

PLLA, the hydrolysis byproduct is the L-lactic acid, which is ultimately broken down into water and carbon dioxide.

Figure 1.3. Photomicrograph of the distal femur at 36 months of follow-up. PLLA was still seen in the screw channel (asterisk), and a bone rim (arrow) has formed around the screw.43

PLLA has a tensile ultimate strength up to 60 MPa, and the elastic modulus between

3-4 GPa.39 As such PLLA has been widely used as a load-bearing material for orthopedic applications, such as rods, pins, and screws.18,43,46 However, there are some limitations associated with its use, especially the slow degradation process.42,43 As discussed above, in long-term in vivo degradation of PLLA in bone, it was found that after 3 years much of the 11 polymer was still present, although as isolated fragments with high crystallinity. After 6 years of implantation, these fragments were still not fully resorbed.47 PLLA is also capable of forming fibers, and was Food and Drug Administration (FDA) approved in 1971 for use in solutions to improve the properties of PGA sutures.48,49 Porous scaffolds and hydrogels based on PLLA find applications in tissue engineering for vascular implants, bone defects repairs, and the controlled release of drugs.50-52 Wang and coworkers fabricated nano- structured PLLA scaffolds using liquid-liquid phase separation, and found that those scaffolds acted as a positive cue to support the differentiation of nerve stem cells and the outgrowth of neurites, as shown in Figure 1.4.50 Aligned PLLA nanofibers surfaced functionalized with YIGSR peptides were also under research for nerve tissue engineering.53

Figure 1.4. SEM micrographs of C17-2 cells on the PLLA nano-fibrous scaffold (5% wt/v) cultured for 1 day: (a) small magnification (1000); (b) magnified view of a differentiated cell with short neurite (2000).50

As to PDLLA, it is an amorphous polymer due to its steric-irregularity. PDLLA degrades faster than PLLA. It loses its mechanical integrity in 2 months and the mass loss happens after 1 year.23 PDLLA has an elastic modulus of 1.9 GPa, and is not suitable for load-bearing materials. Instead PDLLA has been commonly used for drug delivery devices.

Piskin and coworkers prepared micelles using PDLLA-PEG block polymers, and found that these micelles were degraded in phosphate buffer at 37 oC in about 5-6 weeks.54

Adriamycin as a model drug was physically encapsulated in these micelles and then released. This releasing process was mainly controlled by micelle degradation. PDLLA also found applications as a drug delivery coating on inorganic implants such as metals and bioglass.55,56

1.1.1.3 Poly (lactic-co-glycolic acid) (PLGA)

PLGA is the random copolymer of lactide and glycolide, and has being investigated extensively. The copolymerization methodology is a very practical tool to tailor physical properties and degradation behavior of PLGA via varying the monomer ratio between lactide and glycolide. It has been reported that the glass transition temperature of PLGA decreases with an increase of PGA content in the copolymer composition.57 Higher content of PGA also leads to faster degradation with an exception of 50:50 ratio of PLA/PGA, which exhibits the fastest degradation. This is reported in the degradation times of 50:50

PLGA, 75:25 PLGA, and 85:15 PLGA being 1–2 months, 4–5 months and 5–6 months, respectively.41

One of the most successful applications of PLGA is are delivery devices.58-60

PLGA-based devices are used for the controlled delivery of proteins, small molecular drugs, 13

DNA or other bioactive molecules. Some PLGA-based delivery devices have been approved by FDA and commercialized, including, including Zoladex, Enantone, Lupron and so on. PLGA-based delivery devices have many types of structures, including micro- and nano-particles, films and membranes, porous scaffolds, micelles, etc.58-60 Many factors influence the release profiles, including drug type, carrier morphology, PLGA composition, molecular mass and distribution, in vivo locations and so on. Therefore, the release pattern is often unpredictable. In Figure 1.5, it describes a biphasic curve for drug release as a result of PLGA biodegradation, which is a collective process of bulk diffusion, surface diffusion, bulk erosion and surface erosion.58

Figure 1.5. Modeled in vivo release profiles for 50:50, 65:35, 75:25 and 85:15 PLGA

(PLA:PGA). A biphasic release profile with an initial zero release period followed by a rapid drug release has been observed. The profiles correspond to the degradation process of PLGA.58

PLGA is also widely used to fabricate scaffolds for tissue engineering, repair and reconstruction.61,62 For example, Shin and coworkers fabricated PLGA nanofibers using electrospinning.63 Modulus and ultimate tensile stress/strain of PLGA nanofiber constructs were similar to those of the human skin and were slightly lower than those of the human cartilage. In the in vitro experiments using chondrocytes as a model cell, PLGA nanofiber scaffolds were non-toxic, and the cell proliferation and ECM formation in nanofiber constructs were superior to those in membrane-type scaffolds. Those results demonstrated that PLGA nanofiber scaffolds were promising for cartilage reconstruction.63 In another work, G. Khang and coworkers found that fibrin/PLGA hybrid scaffolds seeded with chondrocytes can promote the formation of cartilaginous tissue in vivo.64

1.1.1.4 Polycaprolactone (PCL)

PCL is another widely used aliphatic polyester. The most used method to produce high molecular mass PCL is also ROP.15 The monomer is e-caprolactone, which is produced from the oxidation of cyclohexanone by peracetic acid. Compared to PGA and

PLA, the industrial cost of PCL is much cheaper. The density of PCL is near 1.1 g/cm3, smaller than that of both PGA and PLA. Its glass transition temperature is down to -65 oC, and its melting temperature is about 65 oC. The thermal decomposition temperature of PCL is up to 350 oC. The crystallinity of PCL can reach 69%. PCL has elastic modulus between

0.2 GPa and 0.4 GPa, mechanically compatible to polyethylene (PE), polypropylene (PP) and some natural rubbers. Unlike PLA and PGA which have failure strains near 4%, the break strain of PCL is up to 1000%. Also, PCL has a very good solubility in conventional organic solvents and has exceptional blend-compatibility with many types of polymers.15,65 15

PCL has bulk degradation mechanism in vivo. Due to its high crystallinity and hydrophobicity, PCL degrades much slower than PLA. PCL has a total degradation timeline of 2–4 years (depending of the starting molecular mass and implant location).18,28 There are two stages during PCL implant degradation in vivo. In the first stage, hydrolysis happened in amorphous regions, and PCL implants gradually lost mechanical strength and broke into oligomeric pieces. The degradation rate of PCL in the first stage is almost identical to that of the in vitro hydrolysis at 37 oC. Molecular weight (MW) of PCL deceased with time and followed a linear relationship between log MW and time as shown in Figure 1.6. In the second stage degradation of PCL implants in vivo, low molecular mass

PCL pieces were metabolized and ultimately completely excreted from the body through urine and feces.66-68 Copolymerization of CL with lactides or glycolide enables the acceleration of PCL degradation and facilitates to fulfill the desired degradation behavior.69

With a high permeability to many types of hydrophobic drugs, PCL is widely used in drug delivery systems, especially the long-term system due to PCL slow degradation process. Within the last decades, PCL-based microspheres, nano particles or other structures have been the major area of interest to develop controlled delivery devices.66,70

For example, micelles prepared via the self-assembly of PCL-containing block polymers are extensively studied for the drug delivery and controlled release. PCL hydrophobic core is capable of improving the solubility of hydrophobic therapeutic compounds that can be incorporated via physical encapsulation, chemical attachment, or electrostatic interactions.71,72 PCL-based porous scaffolds are also explored as biomolecule delivery devices.73,74 K.H. Ho and coworkers fabricated PCL-based honeycomb scaffolds via fused deposition modeling (FDM) method. Recombinant bone morphogenetic protein-2 16

(rhBMP-2) was physically loaded in those structures with efficiency 70% or 37% based on the composition of scaffolds, which also resulted in different release profiles.74

Figure 1.6. Decrease in molecular weight (MW) of PCL capsules with time after implanted in rats. The results indicated a linear relationship between logarithm of MW and time.67

Besides drug delivery devices, PCL is also widely used for other biomedical applications. PCL-based medical devices include sutures, would dressing, fixation devices, adhesive barriers, contraceptive devices and dentistry.18,66,75-78 PCL-based 3D porous scaffolds have shown big potentials for the regeneration of bones, cartilage, blood vessels, skins, and so on.66,79,80 PCL nanofiber matrix enables the mineralization and collagen formation.81 PCL/inorganic composites such as the PCL-beta-tricalcium phosphate (PCL-

TCP) were developed to enhance scaffold mechanical strength, and improve hydrophilicity and osteoconductivity for load-bearing applications.82,83 Hutmacher and coworkers did the

2-year implantation of PCL-TCP porous scaffolds in pig model for critical-sized cranial

17 defect repair.84 Both constructs demonstrated extensive bone formation and bone remodeling were observed. As shown in Figure 1.7, extensive mineralization was found using X-ray microcomputed tomography (u-CT), and histological analysis clearly demonstrated the new bone formation. Bone research at the National University of

Singapore based on PCL composites was also commercialized after clinical approval in

2008 (OsteoporeTM).85,86

Figure 1.7. (a) Implantation of the PCL-TCP scaffold into the pig cranium. (b) SEM analysis using backscattered electron mode. Visualisation of the calcium content represented in grey. (c) Extensive mineralization is evident throughout the scaffold in u-

CT analysis. (d) Histological analysis showing new bone formation (nb) scaffold struts (s) osteoblasts (ob) and osteocytes (oc).66,84

1.1.1.5 Poly (propylene fumarate) (PPF)

PPF has attracted much attention in the last two decades. It is based on fumaric acid, a component of the Krebs cycle. PPF is amorphous and has good solubility in a wide range of organic solvents, including CH2Cl2, CHCl3, THF, DMF, acetone, alcohol, ethyl acetate, etc.28 Unlike PGA, PLA or PCL, the traditional method for PPF preparation is not ROP, but step-growth polycondensation.28,87-90 The common method for PPF synthesis follows a two-step procedure, as shown in Figure 1.8 (a).88 The starting chemicals are diethyl fumarate and propylene glycol, and zinc chloride is employed as the catalyst.

Bis(hydroxypropyl) fumarate is produced in the first step. PPF is obtained via the transestreification of bis(hydroxypropyl) fumarate in the second step, and propylene glycol is released as the byproduct. Many factors influence the molecular mass of the final polymers, including purity of reactants, water residue, temperature, catalyst, vacuum, and so on. Linear PPF with high molecular mass (>4000) is difficult to obtain by polycondensation technique because of possible side reactions, particularly due to the presence of double bonds. In addition, polycondensation for PPF needs high temperature and is very energy intensive. Recently, Coates and coworkers reported the synthesis of PPF via the chain-growth ring opening copolymerization of maleic anhydride and 2- methyloxirane catalyzed by a chromium(III) salen complex (Figure 1.8 (b)). High molecular mass PPF (>20,000) was achieved with a narrow distribution (<1.2).91

Low molecular mass PPF is a viscous liquid at room temperature. The glass transition temperature of linear PPF varies from -20 to 30 oC depending on molecular mass.

The entanglement molecular weight (Me) of PPF is 4900 confirmed both by experiments and calculation.92 As unsaturated polyester, PPF can be radically cured by itself, or with 19 various monomers, including N-vinyl pyrrolidone (NVP), PPF-diacrylate (PPF-DA), diethyl fumarate, methacrylate, etc.93-95 When combined with suitable crosslinking system,

PPF is valuable for developing injectable materials that could fill skeletal defects of varying size and shape.96 This crosslinking process is involved with volume shrinkage. Drugs, proteins or peptides can be encapsulated in the injectable PPF cement for controlled release study.97,98 Meanwhile, due to the ultraviolet-curable characteristic, PPF is widely used for the fabrication of 3D porous scaffolds with precisely controlled structures using stereolithography based on computer-aided design (CAD), as shown in Figure 1.9.99,100

Those scaffolds have applications in diverse tissue engineering. PPF is also used for the surface coating and modification of bone grafts due to its curable capability.101

Figure 1.8. Synthesis scheme of PPF: (a) step-growth polycondensation (b) chain growth polymerization.88,91

Figure 1.9. (a) Mean pore sizes of PPF scaffolds compared to computer-aided design (CAD) models in two directions. Dashed line indicates exact matching between CAD models and fabricated scaffolds. Values represent mean ± SD (n = 5). (b) Mean pore sizes of PPF scaffolds with various sized square pores in two directions. Scale bars 400 μm.99

Crosslinked PPF and its composites have been investigated for use in orthopedic applications and bone tissue engineering for many years.90,100,102,103 Mikos and coworkers have made significant contributions to this field. Mechanical properties and degradation behavior of PPF crosslinked networks depend on the choice of PPF oligomers, crosslinkers, initiators, and the crosslinking method. Some alternative protocols have also been

21 developed to enhance the mechanical strength of PPF constructs, especially the composites of PPF and inorganic compounds, including tricalcium phosphate (TCP), calcium carbonate, calcium sulfate and carbon nano-tube.104-107 For instance, a 74% increase was achieved for the compressive modulus and a 69% increase for the flexural modulus of PPF-

SWNT (single wall nano tube) nano-composites with SWNTs at a 0.05 wt% loading.104

Besides orthopedic applications, PPF-based materials are also used for drug delivery and cell transplantation devices.108-110

Crosslinked PPF is degradable in biological environments with bulk degradation mechanism. Experiments have shown that degradation byproducts of PPF in vitro are primarily fumaric acid and propylene glycol, upon hydrolysis of ester linkages.93,111,112

Crosslinked PPF exhibits a biphasic degradation behavior in physiological environments.93,94,112 At the first stage, modulus and yield strength of PPF implants both increase. This unexpected phenomenon is explained by the second-step crosslinking process in vivo. The 37 oC environment provides enough energy for the chemical reactions among the entrapped monomers and initiators. In the second stage, similar to other aliphatic polyesters, PPF implants lose mechanical strength, due to the weight loss and chain scission.

Yaszemski and coworkers recorded the mechanical behaviors of NVP crosslinked PPF during incubation in PBS buffer in 37 oC. Their compressive modulus was 113 (± 40) MPa initially, and then increased up to 691 (± 74) MPa at day 42, and then dropped to 421 (+82)

MPa at day 56. Similar profile was found in the compressive strength.94 Similar results were also recorded by Mikos and coworkers for the PPF-DA crosslinked PPF networks as shown in Figure 1.10.112

Figure 1.10. Compressive modulus of PPF/PPF-DA networks as a function of degradation time for the four sample groups: A (◇, PPF 33%, PPF-DA 67%, pH 7.4 ), B (□, PPF 30%,

PPF-DA 60%, TCP 10%, pH 7.4), C (△, PPF 67, PPF-DA 33%, pH 7.4)), and D (○, PPF

33%, PPF-DA 67%, pH 5.0) (all in weight percentage). Bars represent means ± standard deviation for n = 5.112

1.1.2 Amino-acid containing degradable polymers

Though polyesters have obtained instant success in biomedical fields, there are some limitations associated their use, including chemical inertness, lack of bioactivities, acidic degradation byproducts, and hydrophobic polymer chain.16 Those kinds of limitations lead to the fast development of other types of degradable polymers, especially natural amino acid containing degradable polymers. Amino acid-based polymers have diversity and versatility, due to the rich pool of starting materials. Amino acids bear amine and acid groups, both of which are active for various chemical reactions. Meanwhile, amino

23 acids contain chemically reactive side groups, including hydroxyl, amine, acid, thiol, and phenol. Those groups can be introduced to polymer constructs very easily when using amino acids as polymer building blocks. It enables biomolecule conjugation to those sites and also makes it possible to control physical properties of polymer constructs via a chemical pathway. Additionally, amino acids themselves would improve biological properties, including the enzymatic degradation of polymers and the interaction between biological environments and materials. Figure 1.11 describes the schematic representation of amino acid containing degradable polymers and possible specific functions.113 The review paper of Zhong summarized the synthesis and applications of various amino acid containing polymers in detail.113 Many types of amino acid containing degradable polymers are extensively studied for biomedical applications, including poly-(amino acid)s, poly

(ester amides), tyrosine-derived polycarbonates, polydepsipeptides, pseudo-poly-(amino acid)s, poly (ester ureas), and so on. Currently most of those studies are in the stage of benchside research. Significant effort is required to fulfill the translation from benchside to bedside. In the following part, we will focus on three primary amino acid containing polymers: poly-(amino acid)s, poly (ester amides) (PEAs), and tyrosine-derived polycarbonates.

Figure 1.11. Schematic representation of amino acid containing degradable polymers and possible specific functions.113

1.1.2.1 Poly-(amino acid)s

The most well-known amino acid containing polymers are natural biologically active poly-(amino acid)s, called proteins, such as enzymes, collagens, elastins, and so on.

Short chain poly-(amino acid)s are generally called peptides. These biological poly-(amino acid)s have specific properties and functions originating from their complex and well defined-structures, including the precisely controlled sequences and compositions of amino acids, as well as the 3D architectures. Biological poly-(amino acid)s are generally extracted and purified from natural products, prepared via DNA-assistant technology, or synthesized by and bacteria. Though those natural poly(amino acids) have unique biological advantages over synthesized materials, their complicated synthesis and limited supplies drive the development of chemically synthesized poly-(amino acid)s.114-116

Figure 1.12. (a) Molecular model of a single KLD-12 self-assembling peptide. (b) Photo picture of chondrocyte-seeded peptide hydrogel. (c) Light microscope image of chondrocytes encapsulated in peptide hydrogel. The alternating hydrophobic and hydrophilic residues on the backbone promote β-sheet formation. The positively charged lysines (K) and negatively charged aspartic acids (D) are on the lower side of the β-sheet, and the hydrophobic leucines (L) are on the upper side. This molecular structure facilitates self-assembly through intermolecular interactions.117

The most employed method for peptide synthesis is the Solid Phase Peptide

Synthesis (SPPS) developed by Merrifield.118 He was a winner of Nobel Prize in Chemistry in 1984 due to his contributions in innovating SPPS methodology and technique for peptide and protein synthesis. SPPS involves repeated cycles of coupling-wash- deprotection-wash. This method has been demonstrated a robust technique to precisely synthesize poly-(amino acid)s with designed sequence. Peptides with natural sequence have special bioactivities to control cell-material interactions.119 For example, RGD peptide with a sequence of RGD (R: arginine, G: glycine, D: aspartic acid) has the function 26 to mediate integrin-based cell adhesion in a number of cell types. It still remains bioavailable on surfaces and polymers and when tethered appropriately.120 One the other hand, peptides with well-designed sequences are capable of forming topological structures, including tubes, fibers, hydrogels, nano particles, etc.121,122 In the work of Grodzinsky and coworkers, the KLD-12 peptides, with a sequence of AcN-KLDLKLDLKLDL-CNH2 (K: lysine, L: Leucine, D: aspartic acid), were self-assembled into hydrogels in aqueous solution with a concentration of 0.5% (Figure 1.12). Those self-assembly hydrogels provided an appropriate environment for chondrocytes retention and had the potential for cartilage repair.117

As to long chain poly-(amino acid) synthesis, the most popular method is the polymerization of N-carboxyanhydrides (NCA), which are five-member cycle monomers prepared from α-amino acids (Figure 1.13). Though the resulted polymers lack the sequence specificity and monodispersity of natural proteins, NCA polymerization is currently the most common, economic and efficient technique used for large scale preparation of poly-(amino acid)s.123,124 Alkane amine is generally used as the initiator for

NCA polymerization. However, amine initiated NCA polymerization lacks control over the reactivity of the growing polymer chain end, and the amine end is free to undergo undesired side reactions. In order to eliminate those drawbacks, other methods, such as metal-initiated or coordination polymerizations, are extensively studied. Those methods are capable of giving extra-ordinary control over the molecular mass and distribution, end group chemistry, tactility and so on. Deming and his research group have developed a series of catalyst systems.123,124 NCA polymerization is also able to produce di- and tri-block poly-

(amino acid)s via sequential addition of different monomers. Poly-(amino acid)s with 27 pendant groups such as alkyne, alkene, azide or PEG oligomers are also prepared via

NCA.125-127 In addition, hybrid copolymers are also obtained via NCA polymerization, such as polyester-poly(amino acid) and PEO-poly-(amino acid) copolymers, where polyester or

PEO are used as the macro-initiators respectively.128,129 These side-chain functionalized poly-(amino acid)s and their hybrid systems enrich the pool of amino acid containing polymers, and largely broaden applications of poly-(amino acid)-based materials.

Figure 1.13. Synthesis scheme of poly-(amino acid)s via NCA polymerization. “A” and “B” chain groups depend on catalysis systems.

The most widely studied synthesized poly-(amino acid) are poly (L-glutamic acid)

(L-PGA), poly (aspartic acid) (PAA), and poly (L-lysine) (PLL). All of them are highly susceptible to degradation by enzymes such as lysosomal.130,131 Degradation byproducts are natural amino acids, which makes them ideal candidates as degradable biomaterials.

Several studies have focused on degradation behavior of poly-(amino acid)s using tissue isolated enzymes. Polymer compositions and side chains have a big effect of the degradation of poly-(amino acid)s.130,132,133 For example, L-PGA degrades faster than PAA.

Theincorporation of hydroxyl or amine side chains can promote the degradation of poly-

(amino acid)s. Due to enzymatic degradation, it is very difficult to predict or control the

28 degradation behavior of poly-(amino acid)s in vivo. The copolymer of polyester-poly-

(amino acid) generally shows a faster degradation than pure polyester, since the incorporation of amino acid units disrupts the crystallinity of polyesters, and also increases polymer hydrophylicity.

Poly-(amino acid)s are popular in biomedical field, due to their chemical functionalities and diversity, enzyme degradation, natural building blocks, and specific cell-material interactions.113 Poly-(amino acid)s are widely studied as drug and gene delivery devices.131,134,135 Micro-size porous particles based on PLA-co-PLL were explored as vehicles for pulmonary drug delivery targeting lungs.136 In vivo experiments with male

SpragueDawley rats found that porous PLA-co-PLL particles had similar delivery behaviors compared to those based on PLGA. Non-porous PLA particles as controls were more located in tracheas, while porous PLA-co-PLL particles were more deposited in lungs.

Drugs can also be covalently conjugated to poly-(amino acid) constructs via chemically reactive side chains, such as the carbonyl acid groups in L-PGA and amine groups in

PLL. .131,137 For example, Peter de Vries and coworkers attached cancer chemotherapeutic drug camptothecin (CPT) to L-PGA via ester bonds, and found that those conjugates enhanced the CPT stability and efficacy in vivo.138 Arginine and lysine based polymers were widely studied for DNA delivery.139-141 PLL-based crosslinked nano particles were explored to support the co-adsorption of a plasmid DNA and a peptide hormone for concurrent transfection and induction of a cellular function.142 Also L-PGA is highly charged in physiological environment, indicating that it is very promising for DNA delivery.143,144

Figure 1.14. (a) Scheme for covalent coupling GRGDY-I2 peptides to PLA-PLL via amine residues in PLL units. SEM pictures of bovine aortic endothelial (BAE) cells attached to

PLA (b) and RGD modified PLA-PLL (c). Cell spreading was enhanced in peptide modified polymers.145

Poly-(amino acid)s also have big potentials in tissue engineering fields. On one hand, amino acids themselves are able to provide specific biological functions. On the other hand, side chain residues of amino acids enable the tailoring of chemical and biological behaviors of scaffolds via a chemical pathway.113 There are several examples. PEG-b-

PLLA-b-PLL triblock copolymers can promote the osteoblast adhesion and proliferation as compared to pure PLLA and PEG-b-PLLA copolymers.146 It was explained that free

30 amine groups provided effective interactions with cell recognition motifs and growth factors. In another study, free amine residues in lysine were used to conjugate RGD peptides covalently to polymer construct, to further enhance interactions between materials and biological environments.145 As shown in Figure 1.14, the cell spreading on RGD modified poly (lactic acid-co-lysine) (PLA-PLL) was enhanced by 4 folds compared to unmodified PLA samples.145

Poly-(amino acid)s have been used in hydrogels, films, adhesives, antimicrobial materials, and so on.147 Deming’s group has done many studies in hydrogel making via self-assembly of amphiphilic poly-(amino acid) block copolymers.148-150 Other methods, such as chemical crosslinking or high energy radiation, are also applied for hydrogel fabrication.151,152 Poly-(amino acid)s are also explored as bio-adhesives, where hydrogen bonding and electrostatic force contribute a lot.153 L-PGA based adhesives were found to have better soft tissue binding abilities than fibrin glue and collagen glue in animal study.154

In the natural adhesive proteins found in mussels, there is a very high content of amino acid 3, 4-dihydroxyphenyl-L-alanine (DOPA).155,156 DOPA containing homopolymers- or copolymers with lysine, tyrosine, glutamic acid or some other natural amino acids, were designed and synthesized to mimic the structure, composition and function of natural adhesive proteins.157-160 For example, Deming and coworkers prepared random copolymers of DOPA and L-lysine using NCA polymerization, and then explored the influence of molecular mass, polymer concentration, curing time and conditions on the adhesive strength of poly-(amino acid)s-based adhesives (Figure 1.15).158 Those synthetic poly-

(amino acid)s have big potentials as moisture resistance medical adhesives.

Figure 1.15. (a) Synthesis scheme of poly-(amino acid)s containing DOPA (blue) and L- lysine. (b) Influence of polymer concentration on adhesive strength. (c) Adhesive strength of copolymers on different substrates. (CBZ = carbobenzyloxy, PS = polystyrene, PE = polyethylene, PMMA = poly(methyl methacrylate).158

1.1.2.2 Poly (ester amides) (PEAs)

Another popular amino acid containing degradable polymers are poly (ester amide)s (PEAs). Puiggali summarized PEAs’ synthesis and application in detail.161

Polydepsipeptides are not discussed here. PEAs are generally synthesized via a polycondensation process. There are three methods to prepare PEAs as shown in Figure

1.16: solution polycondensation, interfacial polymerization and melting polycondensation.113,161 In solution polymerization, di-p-toluenesulfonic acid salts of bis-

(amino acid) diesters react with di-p-nitrophenyl esters of diacids to generate desired polymers. Chu and coworkers did extensive studies in this field.162-164 High molecular weight PEAs can also be obtained via interfacial polymerization between the di-p- toluenesulfonic acid salts and diacid chlorides. Solvent free melt condensation method involves a transesterification process. It commonly requires high temperature. In the work of Puiggali and coworkers, melting polymerization was first carried out at 160-190 oC in a nitrogen atmosphere and then at 200-220 oC in high vacuum.165

Figure 1.16. Synthesis scheme of PEAs via solution polycondensation, interfacial polymerization and melt polycondensation. 33

Polycondensation can easily combine various kinds of amino acids in one construct, though sometimes protection-deprotection steps are necessary. Up to now, L-Phenylalanine

(L-Phe), L-Alanine (L-Ala), L-valine (L-Val), L-leucine (L-Leu), L-isoleucine (L-ILeu),

L-Norleucine (L-NLeu), D,L-methionine (DL-Met), and L-Arginine (L-Arg) based PEAs or their copolymers are successfully synthesized.161,162,166 The combination of different amino acid and diol units provides a universal method to tune PEA properties, such as the mechanical, degradation and thermal properties. For example, when changing the alkane chain structure of diol units in L-PHE-based PEAs, the glass transition temperature of L-

PHE-based PEAs varied from 20 oC to 110 oC, and the weight loss during 6 weeks enzymatic degradation varied from 10% to 90%.162,167 When changing the amino acids but keeping the diol units the same, it was found that L-PHE-based PEAs degraded slower than

L-methionine-based PEAs.166 On the other hand, chemical reactive groups can be incorporated into PEA constructs. Further chemical modification of polymers was conducted to control polymer properties. A good example was that biomolecules were successfully conjugated to the unsaturated PEA constructs via thiol-ene reaction.162

Hydrogels were also prepared via the radical crosslinking of unsaturated PEAs.168

Some studies have focused on the degradation of PEAs. In the work of Mequanint, hydrolytic and enzymatic degradation of PEAs were both explored in detail, and chymotrypsin was chosen as a model enzyme.166 There was a surface erosion process at the beginning, which was concluded from the observed small blisters in SEM studies. L- methionine-based PEAs experienced faster erosion than L-PHE-based PEAs due to a higher hydrophilicity. The addition of chymotrypsin enzyme accelerated the surface erosion process significantly. In SEC studies, polymers degraded in PBS for 28 days and 34 those degraded in chymotrypsin for 72 h had similar trace. In both cases, molecular mass dropped initially and then didn’t change for the rest of experimental period, as shown in

Figure 1.17. It implied that chymotrypsin had similar degradation mechanism as the hydrolytic degradation, and it can speed up this process significantly. Faster degradation in chymotrypsin was also confirmed in the weight loss experiment in other studies.162

Combining the SEM and SEC studies, it was concluded that PEAs were undergoing an initial bulk degradation followed by surface erosion for the remainder degradation times studied.

Figure 1.17. Molecular weight (Mw) changes of PEAs incubated in (a) PBS at 37 °C for up to 28 days and, (b) chymprtypsin at 37 °C for up to 72 h. As to the nomination, Ph48 means the amino acid is L-phenylaniline, number 4 means the ester unit (Y in Figure 1.16) is -(CH2)4-, and number 8 means the amide unit (X in Figure 1.16) is -(CH2)8-.

Similar to poly-(amino acid)s, PEAs have also received much attention for their potential medical applications, including drug/gene delivery devices, hydrogels, tissue engineering, adhesives, hybrids and smart materials.113,161 For example, arginine-based

PEAs were explored as non-viral transfection agents for gene delivery.169 Those polymers 35 showed good biocompatibility and comparable efficiency to commercial transfection reagent Superfect, but with a markedly lower cytotoxicity. In the study of Reinhart-King and coworkers, phenylalanine-based neutral PEAs and other two kinds of functionalized

PEAs (amino-functionalized positive and carboxylic acid functionalized negative) were all found to be noncytotoxic and noninflammatory in vitro to support endothelial cell viability, proliferation, and adhesion.163,170 In addition, the amino-functionalized PEAs best supported endothelial cell adhesion, growth, and monolayer formation. Those data supported that PEAs had potential for use in tissue engineering applications, promoting the growth of appropriate cells and supporting the tissue repair. Porous scaffolds and hydrogels based on PEAs are also studied for tissue engineering application.166,168,171 In recent work of Chu and coworkers, they prepared hybrid hydrogels containing arginine-based PEA and hyaluronic acid (HA).168 Resulted hydrogels showed 50 to 70% weight loss within 6 days in trypsin enzyme containing environments. Compared to pure HA hydrogels, hybrid hydrogels of HA-PEA exhibited a significant enhancement of the attachment and proliferation of Hela cells on hydrogel surfaces, as shown in Figure 1.18. In addition, hybrid hydrogels were more sustainable than pure HA hydrogels in controlled release study using bovine serum albumin (BSA) as a model drug.168

Figure 1.18. Micrographs of Hela cells after 48 hours culture. (a) Cells cultured on the surface of a pure HA hydrogel; (b) cells cultured on the surface of the HA-PEA hybrid hydrogel; (c) wet hydrogel after staining and washing as a blank control; (d) MTT assay for the Hela cells after 48 hours' culture in a Dulbecco's minimal essential medium (DMEM) medium on HA and HA-PEA hybrid hydrogel surfaces. Hybrid hydrogels exhibited a significant enhancement of cell attachment and proliferation.168

1.1.2.3 Tyrosine-derived Polycarbonates

Inspired from biphenol-based polycarbonates, tyrosine-derived polycarbonates were developed as a novel class of degradable polymers with good stiffness, mechanical strength and non-toxicity in the 1990s.172,173 Commonly used tyrosine-derived polycarbonates are based on 3-(4’-hydroxy phenyl) propionic acid and tyrosine alkyl esters

(Figure 1.19).173 Comparing to tyrosine bipeptides-based polycarbonates, the existence of 37

3-(4’-hydroxy phenyl) propionic acid units influence physical properties of tyrosine- derived polycarbonates significantly, including enhancement of solubility, ductility, and processibility.172,174 High molecular mass of tyrosine-derived polycarbonates (>100,000

Da) are obtained by polycondensation between diphenolic monomers and phosgene or triphosgene (Figure 1.19). Diphenolic monomers are synthesized from the carbodiimide- mediated coupling reactions between 3-(4’-hydroxy phenyl) propionic acid and tyrosine alkyl esters. Physical and biological properties of tyrosine-derived polycarbonates can be tuned via changing “R” pendent groups.175

Figure 1.19. Chemical structures of tyrosine-derived polycarbonates and corresponding monomers with different alkane pendent chains.173

Tyrosine-derived polycarbonates are amorphous. Their physical properties largely depend on pendent groups. The glass transition temperature varies from 50 to 90 oC.

Tyrosine-derived polycarbonates have high thermal stability with thermal degradation temperature near 300 oC. They can be processed using conventional polymer processing

38 techniques including extrusion, injection molding and compression molding. Tensile modulus of tyrosine-derived polycarbonate films varies from 1 GPa to 2 GPa, and the strength at break from 60 to 220 MPa. Water uptake of tyrosine-derived polycarbonates is generally less than 3%, much smaller than that of aliphatic polyesters.173,175,176

Degradation of tyrosine-derived polycarbonates involves the hydrolysis of carbonate bonds in polymer backbones and ester bonds in pendent chain bonds, and the final degradation products in vitro are desaminotyrosyl-tyrosine and the alcohol used to protect carboxylic acids.177,178 The cleavage of carbonate bonds is four times faster than that of pendent ester bonds. , It dominates the degradation at the initial stage and leads to a reduction in polymer molecular mass without any concomitant mass loss. Significant hydrolysis of pendent ester bonds only occurs in the final stage of degradation and results in hydrophilic polymer matrix and acid-mediated autocatalysis. This process is essential for the complete resorption of tyrosine-derived polycarbonates. Significant swelling is not observed during the whole degradation process, and only about 10% of carboxylic acid groups are produced compared to the degradation of poly (aliphatic esters). Increasing the length of the alkane pendent groups lowers hydrolysis rates of both carbonate and ester bonds, possibly by hindering the access of water molecules to those sites. In addition, the in vitro and in vivo degradation rate is very similar to each other, implying the absence of enzymatic involvement during the in vivo degradation process.179 According to the molecular mass change during the degradation process, tyrosine-derived polycarbonates have similar degradation rates to PLLA (Figure 1.20), but the mass change of tyrosine- derived polycarbonates only occurs at the end stage of degradation.179,180 Copolymerization with PEG or pre-deprotection of alkyl pendent groups is employed to accelerate 39 degradation rates of tyrosine-derived polycarbonates, though stiffness and mechanical strength of final polymers are lowered.175,181-183

Figure 1.20. In vivo degradation profiles of poly(DTE carbonate), poly(DTH carbonate) and PLLA in canine bone chamber model.180

Tyrosine-derived polycarbonates have found increased utility in numerous biomedical applications, including tissue engineering scaffolds, drug delivery vehicles, medical devices and coatings.175 Porous scaffolds based on tyrosine-derived polycarbonates were fabricated using types of techniques including electrospinning, particle-leaching and phase separation.175,181,184,185 Those scaffolds exhibited degradation behaviors, non-toxicity, and in vivo biocompatibility. Nanospheres prepared from

PEG/tyrosine-derived polycarbonate copolymers can form strong complexes with hydrophobic molecules including the fluorescent dye 5-dodecanoylaminofluorescein

(DAF) and the antitumor drug, paclitaxel. The nanosphere-paclitaxel complexes retain the

40 high antiproliferative activity of paclitaxel in vitro, implying that these nanospheres have potentials for delivery of hydrophobic therapeutic agents.186

Tyrosine-derived polycarbonates have high stiffness and mechanical strength, especially with induced molecular orientation structures.175,176 They are studied widely for orthopedic applications. Kohn and coworkers performed extensive studies in this field.

Orthopedic devices of poly-(DTH carbonate) were implanted in the metaphyseal proximal tibia and distal femur of White New Zealand Rabbits, and close bone apposition and new bone formation were observed throughout the 26-week test period for poly(DTH carbonate) implants.187 Similar results were also found in the canine bone chamber model studies.180

Both poly(DTE carbonate) and poly(DTH carbonate) were characterized by sustained bone ingrowth throughout the 48 weeks implantation period (Figure 1.21 (a)).180 In both studies, mild inflammatory response was observed throughout the whole study, which was different to that of poly (aliphatic esters) controls. Both PLLA and PDS control groups exhibited an increase in the intensity of inflammation as their implants began to lose weight, and fibrous tissue layer was always present between the PLLA or PDS devices and the bone.179,180,187

Kohn and coworkers also explored the influence of small changes in alkane pendent groups on biological behaviors of tyrosine-derived polycarbonates.188 Extruded pins of poly(DTE carbonate), poly(DTB carbonate), poly(DTH carbonate), and poly(DTO carbonate) were implanted transcortically in rabbits’ bone defects and followed over a period of 3 years. In a long term, frequency of direct bone apposition responses at the bone-implant interfaces decreased with the increase of pendent chains, while the frequency of encapsulation responses increased with longer pendent chains. Figure 1.21(b) described the interface histological studies between poly(DTE carbonate) pins and bones after 3 years 41 implantation. High frequency (73%) of bone apposition responses was observed. All of those four polymers were all osteocompatible since most obvious deleterious effects such as bone resorption or large concentrations of inflammatory cells were not observed at implant sites.

Figure 1.21. (a)180 Average percent area occupied by bone in the PLLA, poly(DTE carbonate), and poly(DTH carbonate) test chambers in canine bone model. (b)188 Interface between pins of poly(DTE carbonate) and bone after 3 years implantation. Mineralized bone was stained red, fibrous tissue was blue, and osteoid was stained in a green hue. High frequency (73%) of bone apposition responses was observed.

1.2. “Click” Chemistry and Its Applications in Biopolymers

In previous sections, we have discussed the chemical synthesis and physical properties of several types of commonly used degradable polymers, as well as their applications. The rapid development of biomedical technologies leads to the increasing demand of materials with precisely controlled structures and properties, including

42 architecture, functionality, reactivity, solubility, polarity, mechanics, and degradation.5 The emergence of “Click” chemistry has largely expanded the chemical tool box to meet those goals. The concept of “Click” chemistry was first introduced by Sharpless and co-workers in 2001.189 From their definition, the reaction termed with “click” must be modular wide in scope, result in high yields, and generate only inoffensive byproducts which can be removed easily. Generally speaking, the reaction should proceed rapidly in mild reaction conditions, tolerant to a wide variety of chemical groups, produce stable products, be simple to perform and insensitive to moisture and oxygen. Due to these characteristics,

“click” reactions are used widely in biomaterials.

The first reaction termed with “click” was copper (I)-catalyzed azide-alkyne cycloaddition (CuAAC).189 Currently besides CuAAC, many other thermal dynamic favored reactions are also included in “click” chemistry library, as long as they meet the requirements of being “simple”, “robust” and “efficient”. That list includes “thiol-ene” addition, Diels-Alder (DA) reaction, carbonyl-based reaction (especially oxime condensation), nucleophilic ring opening reaction of epoxides and aziridines, strain promoted cycloadditions, Michael addition and so on.190,191 Figure 1.22 shows schemes of several popular “click” reactions.190 They are widely used in both materials science and life science, including novel polymer synthesis, bioconjugation and functionalization, labeling, drug delivery and discovery, etc. In the following chapter, there is a brief summary of the respective mechanisms and bio-applications of three popular “click” reactions: azide-alkyne cycloaddition, thiol-ene reaction, and carbonyl-based reactions.

Figure 1.22. Schemes of six popular “click” reactions.190

1.2.1 Azide-Alkyne Cycloaddition

Huisgen cycloaddition happens between organic azide and alkyne groups and affords 1,2,3-triazoles.192 Both of azide and alkyne groups are present in many commercial organic reagents or can be incorporated into parent structures easily. Both of azide and alkyne groups are stable under a variety of conditions. However, traditional Huisgen azide- alkyne cycloadditions require elevated temperatures that are harmful to biological molecules. At the beginning of the 20th century, Meldal and Sharpless reported the copper

(I)-catalyzed azide-alkyne cycloaddition (CuAAC) independently, which really made this reaction comply the definition of “click” reaction.193,194 Currently, CuAAC is the most explored “click” reaction.

Based on Density Functional Theory (DFT) calculations, the activated energy in azide-alkyne cycloaddition decreases from 24 kcal/mol to 11 kcal/mol, after using Cu(I) catalysts.195 CuAAC has a stepwise cycloaddition mechanism.196,197 Figure 1.23, is a simplified representation of CuAAC mechanism.198 Briefly, the first step is the coordination between Cu(I) and terminal alkyne to form active copper acetylide species, followed by the azide displacement of one alkyne ligand to generate the copper acetylide- azide complex. The next step is the cyclization to form the metallocycle intermediate. After protonation and dissociation, the reaction ends to generate final products and Cu(I) catalysts. Besides tradition Cu(I) salts, Cu(I) species generated in situ from Cu(II) reduction or Cu metal oxidation is also able to catalyze the azide-alkyne cycloaddition.

Meanwhile, it is found that the addition of some heterocyclic chelates can further accelerate

CuAAC, and also could largely decrease the amount of Cu(I), which is very important in biological systems.196,199 For example, the stabilizing ligand tris-(benzyltriazolylmethyl) amine (TBTA) developed in Sharpless group can protect Cu(I) from oxidation and dispropotionation.200 As a result, the catalytic efficiency was accelerated by 106 fold.

Microwave techniques have also been used to activate CuAAC. Reaction time of CuAAC can be shortened from 12 hours to 1 hour, though considerable improvement of yield was not gained.196,201 There are dozens of reviews about applications of CuAAC, including in polymer and materials science, general organic synthesis, bioconjugation, drug discovery, and so on.190,191,197,202-206

Due to high reaction efficiency and speed and inertness to other groups, CuAAC is demonstrated as a powerful polymerization technique to prepare structure-unique functional polymers. It is termed as “click” polymerization.207,208 Guan and coworkers 45 employed CuAAC polycondensation to prepare elastin-mimic polymers.209 Monomers were peptide oligomers with designed sequences as shown in Figure 1.24 (a). The C- terminus was coupled with alkyne groups, and the N-terminus was azide-functionalized.

The chemical linkage in the final polymers was the triazole ring. The resulting polymers have secondary structures and possessed similar properties to native elastin: lower critical solution temperature (LCST) behaviors in aqueous solution (Figure 1.24 (c)) and high elasticity in bulk with partial hydration (Figure 1.24 (b)). Reineke and coworkers prepared a series of glycopolymers using “click” polymerization.210 CuAAC conjugated the trehalose moiety and oligoamine more efficiently than traditional aminolysis polycondensation. The resulting polymers were biocompatible in biological studies and enabled DNA delivery in serum. It was noted that trehalose units promote biocompatibility, amide-triazole groups can enhance DNA binding affinity, and oligoamine units were able to facilitate DNA encapsulation, phosphate neutralization, and interactions with cell surfaces. Other materials such as dendrimers were also achieved via “click” polymerization.211-213 CuAAC provides an efficient approach to prepare a wide range of novel materials for medical applications.

Figure 1.23. Simplified representation of CuAAC mechanism.198

Figure 1.24. (a) Bioinspired modular synthesis of elastin-mimic polymers via CuAAC (top).

Structures of peptide oligomer monomers 1 (VPGVG), 2 (GVGVP), and 3 (VDPGVG) with the corresponding elastin-mimic polymers P-1, P-2, and P-3 (bottom) (V=Valine,

P=Proline, G=Glycine). (b) True stress−strain curves for the elastin-mimic polymers in hydrated forms (13 wt% water): P-1, blue; P-2, cyan; P-3, magenta. (c) LCSTs of elastin- mimic polymers at a concentration of 3.0 mg/mL in water (Wavelength 550 nm): P-1, black;

P-2, red; P-3, green.209

CuAAC is also widely used for the conjugation between polymeric structures and bioactive molecules, to tailor the interactions between polymeric materials and biological environments. Azide or alkyne groups are incorporated into polymer constructs at chain ends or on the side chains, either by direct polymerization or post-polymerization modification.190,214,215 Emrick and coworkers explored the bioconjugation between polyesters and peptides via CUAAC.216 In Figure 1.25 (a), aliphatic polymers with pendent alkyne groups were synthesized using controlled ring opening polymerization (ROP), and then azide derived PEG oligomers or RGD peptides were grafted to the polyester via triazole bonds. Both PEG and RGD conjugated polymers were demonstrated to be biocompatible by in vitro cytotoxicity evaluation, suggesting their suitability for a range of biomaterial applications. In some other studies, azide groups were also introduced to polyester constructs, and then alkyne derived biomolecules are conjugated to polymers via

CuAAC.217,218 Besides degradable polyesters, other polymers such as polyurethanes, polycarbonates and poly-(amino acid)s with “clickable” groups are also extensively studied for functionalization via CuAAC.151,219-222 For instance, Zhang and coworkers synthesized poly (γ-chloropropyl-L-glutamate) (PCPLG) with chloride pendent groups via NCA polymerization, and then converted chloride into azide groups by substitution reaction

(Figure 1.25 (b)).223 Mannose was quantitatively grafted to polymer side chains via

CuAAC, yielding water-soluble mannose-poly-(amino acid) hybrids. Circular dichroism

(CD) analysis demonstrated that the mannose- poly-(amino acid) hybrids retained α-helical conformation in aqueous solution. Hennink and coworkers attached low molecular mass cationic polymers to poly(hydroxyethyl methacrylate) (pHEMA) via CuAAC, to modulate the transfection activity and toxicity profile of gene delivery pHEMA polymers.224 48

Sumerlin and coworkers prepared polymer-protein hybrids through coupling azide derived poly(N-isopropylacrylamide) (p(NIPAM)) and alkyne functionalized bovine serum albumin (BSA) proteins.225 Those hybrids were able to form stable nanoparticles above the lower critical solution temperature of p(NIPAM).

Figure 1.25. (a) Synthesis scheme of (a) alkyne derived polyesters via ring opening polymerization (ROP) and their functionalization with PEG oligomers or RGD peptides via CuAAC.216 (b) Mannose-poly-(amino acid) conjugates via CuAAC post- polymerization technique.223

Figure 1.26. (a) Fabrication scheme of polysaccharide-based hydrogels via CuAAC. (b)

Photographs of hydrogels (b) before the removal of copper and (c) after the removal of copper via dialysis against H2O/EDTA. (d) SEM images of hydrogels containing

S.cerevisiae cells.226

CuAAC is found fabricating crosslinked hydrogel networks powerfully. Hydrogels have big applications in biomedical fields due to their high water content, elasticity, and capability for solute transport.227 The first CuAAC resulted hydrogels were developed by

Hilborn and coworkers in 2006.228 Hydrogels were formed in minutes when mixing azide derived poly(vinyl alcohols) (PVA) or PEG, alkyne derived PVA, and Cu(I) catalysts all together. Mechanical properties of hydrogels were easily controlled by changing the functionality (number of functional groups) of crosslinkers. Besides PVA, many other polymers are also employed as building blocks for hydrogels, including PEG, hyaluronan, poly(N-isopropylacrylamide), polymer-peptide hybrids, and so on.226,229-231 Crescenzi and coworkers attached azide or alkyne short links to polysaccharide main chains separately.226

Hydrogels were formed in minutes when two polymer precursors and CuCl catalysts were 50 mixed together in aqueous environments. In the in vitro biological experiment, homogenous distribution of cells was found in those hydrogels, and cells exhibited proliferating activity inside hydrogels (Figure 1.26). As to CuAAC hydrogels, dialysis is commonly used to remove copper salts that are entrapped in the gel. During this process, ethylenediaminetetraacetic acid (EDTA) is used as the metal chelator. CuAAC hydrogels are found in several applications, including drug delivery devices, tissue engineering scaffolds, cell culture templates and so on.

As to CuAAC, the critical point is that copper catalysts accelerate reaction speed and increase efficiency of azide-alkyne cycloaddition significantly. However, the toxicity of copper has largely hindered the application of CuAAC in biological systems. Some studies showed that even the content of cooper was decreased to 100 um, considerable cell death still occurred.232 To improve the biobeaviors of azide-alkyne cycloaddition and to further broaden its applications in biomedical fields, much effort has been focused on the development of copper-free azide-alkyne cycloaddition, where alternative alkyne activation methods are being explored to replace metal catalysts. One method is applying the ring-strain, named as strain-promoted azide-alkyne cycloaddition (SPAAC).233-235 In this system, alkyne is located on the eight-member ring called cyclooctyne (Figure 1.27, alkynes 1, 2, and 3). Cyclooctyne are activated further by introducing electron- withdrawing fluorine atoms in the δ-position (Figure 1.27, alkyne 2)236, or fusing two aryl rings to the cyclooctyne scaffold (Figure 1.27 alkyne 3)237. Another strategy is directly functionalizing alkyne groups with electron-withdrawing groups in the α-position, such as carbonyl and sulfonyl groups (Figure 1.27, alkynes 4, 5, and 6).238,239 Both systems of cooper-free cycloaddition have similar reaction speeds and efficiencies compared to 51

CuAAC. The copper free reaction has unique advantages in biomedical fields, especially in the conjugation and labeling where living systems are involved.240 However, the drawback of this copper-free technology is still obvious. The preparation of those activated alkynes requires multi-step organic synthesis, and the total yield is low (~30%).233

Figure 1.27. Activated alkynes for copper free azide-alkynecycloaddition reactions.

Though currently copper-free technique is more used in relatively low molecular mass chemistry and related bioconjugation, increasing attention has been paid to the combination of copper free technique and macromolecular science and engineering. In many bio-related systems, copper free click reaction gradually takes the role of traditional

CuAAC, and works effectively.233,234 For example, cyclooctyne coupling with two aryl rings was used as the initiator for controlled ROP of PLLA.53 Azide derived peptide Tyr-

Ile-Gly-Ser-Arg (YIGSR) was then conjugated to the nanofiber of end-functionalized

PLLA via SPAAC. Those surface functionalized nanofiber scaffolds were found increasing the level of neurite extension and gene expression of mouse embryonic stem cells (mESC).

This strategy allowed us to precisely functionalize degradable polymers with bioactive molecules, which was able to capitalize on the advantages of both synthetic and natural

52 systems. Cooper-free azide-alkyne cycloaddition was also widely used on the functionalization of many other types of polymers, including PCL, poly-(amino acid)s, paracylophanes coatings and so on.241-243

Figure 1.28. Click-functionalized macromolecular precursors react via the SPAAC to form an end-linked hydrogel network; By varying the molecular weight of the PEG, well-defined networks of differing cross-linking density are formed.244

Similar to CuAAC, cooper-free reaction is also studied as the crosslinking method to prepare hydrogels for biomedical applications. Anseth and coworkers have done extensive work in this field.244-248 As show in Figure 1.28, hydrogels are formed via the

53 cycloaddition between bifunctionalized peptides with cyclooctyne and tetra-functionalized

PEG with azide groups. Through molecular design of peptides, photo-degradable or photopatterned hydrogels were successfully developed. Peptide building block also provided sites for specific enzyme degradation in biological environments. Those copper-free resulted hydrogels were demonstrated as promising scaffolds for cell culture and soft tissue engineering.

1.2.2 Thiol-ene “click” reactions

The thiol-ene reaction is a kind of hydrothiolation, which happens through the addition of thiol groups (SH) to carbon-carbon double bonds (C=C). The reaction has a long history and the thiol-ene reaction is used widely to prepare polymeric networks, coatings or films.249 Based on the mechanism, thiol-ene reaction is divided into two classes.

One is thiol-ene radical addition, in particular photo-initiated radical addition. The other one is thiol-ene Michael addition, a kind of nucleophilic addition. Both thiol-ene photo- initiated and Michael additions are termed with “click” reactions, as shown in Figure 1.22.

They can be conducted in mild conditions with high speed and efficiency, region-selectivity, and insensitivity to oxygen and moisture. Thiol-ene reaction has a large pool of raw materials, either commercially available or prepared via simple organic synthesis. There are several reviews focusing on thiol-ene reactions.250-252

In Figure 1.29 (a), it describes how it proceeds of the photo-initiated thiol-ene radical addition.251 With photoinitiator, thiyl radicals are formed upon UV irradiation. Thiyl radicals are then directly added to carbon-carbon double bonds to generate carbon-centered radicals, following the rule of anti-Markovnikov. Final products are yielded via hydrogen 54 transfer from thiols to carbon-centered radicals, and new formed thiyl radicals participate in the next cycle addition. Photo-initiated thiol-ene addition has radical step-growth mechanism. Compared to traditional radical additions, a photo-initiated process can be activated at specific times and locations under ambient temperature. It is rapid, quantitative and tolerant to oxygen. It is capable of generating more homogenous networks with less internal stress, shrinkages, and defects.253-255 In addition, thiol-ene radical additions proceed with almost all carbon-carbon double bonds, including both electron-efficient and deficient ones. The reactivity of different carbon-carbon double bonds varies from each other depending on their chemical structures.256 In Figure 1.30, it describes the alkene conversion of different structures in thiol-ene radical addition from the computational and kinetic analysis.256 The alkene reactivity follows the orders: norbornene > vinylether > propenyl > alkene z vinylester > N-vinylamide > allylether > acrylate > N-substituted maleimide > acrylonitrile z methacrylate > styrene > conjugated.254,257 One limitation for thiol-ene radical addition is that it requires excess thiols to prohibit crosslinking or dimer formation.

Figure 1.29. (a) Scheme of thiol-ene photo-initiated radical addition. (b) Scheme of thiol- ene Michael addition. TEA = triethylamine, and EWG = electron-withdrawing groups.251

Figure 1.30. Computational and kinetic analysis of the influence of alkene functionality on the reactivity of thiol-ene photo-initiated addition.256

The second type of thiol-ene “click” reaction is thiol-ene Michael addition.250,251

Thiol groups are nucleophilicly added to carbon-carbon double bonds activated with electron withdrawing groups (EWG), such as carbonyl and cyanogen groups. Addition products also follow the anti-Markovnikov rule. Thiol-ene Michael addition needs mild base or nucleophilic catalysis environments. The most popular catalysts include hexylamine, triethylamine, dimethylphenylphosphine and so on.252,258 As described in

Figure 1.29 (b) with triethylamine (TEA) as the catalyst, thiolate anions are formed via the

56 reaction between thiols and triethylamine.251 Thiolate anions are then added to the electron deficient C=C. Resulted carbon-centered anions are highly reactive, and they abstracts protons from thiols or ammonium cations to generate products. Newly formed thiolate anions participate in the next cycle reaction. When highly activated C=C are employed such as vinylsulfone or maleimide, thiol-ene Michael addition can process in highly polar solvents such as DMF, DMSO and water without catalysis, where spontaneous dissociation of thiols into thiolates occurs.

Significant efforts have been focused on the functionalization of polymers with bioactive molecules via thiol-ene technique.250-252,254,259,260 First, the pool of thiol containing molecules is rich. Thiol groups exist in biomolecules naturally such as the cysteine residue, or can be obtained via the reduction of disulfide bonds. Second, carbon- carbon double bonds can be incorporated into polymer skeletons easily, either by postpolymerization modification or direct polymerization of alkene-derived monomers.

For instance, Zhong and coworkers synthesized a series of electron-deficient C=C containing poly-(amino acid)s and polyesters, which were then functionalized with peptides via thiol-ene Michael addition.261,262 Dove and his research team have also been actively investigated the application of thiol-ene reaction in the modification and functionalization of degradable polymers.214,263-265 Additionally, alkyne groups are also active for photo-initiated radical addition with thiols.266-268 In this case, two thiols can be attached to one alkyne. This reaction is generally termed with thiol-yne addition. Wooley and coworkers prepared polyphosphoesters containing alkyne pendant groups using controlled ROP, and further functionalized those polymers via both CuAAC and thiol-ene reaction.269,270 The polymer library available for thiol-ene modification includes 57 polycarbonates, PEG and epoxy based polymers, glycopolymers, polysaccharides, polyoxazoline, polysiloxane, rubber, polyesters, polyurethanes, poly-(amino acid)s, and so on.271-276 Grafting or end functionalization via thiol-ene technique has been demonstrated to have a very high conversion, and is powerful to control and tune properties of polymeric based materials.250,251,260 In Figure 1.31, it shows some alkene/alkyne containing polymers and thiol derivates for thiol-ene reaction, either thiol-ene photo-initiated or Michael addition.

Figure 1.31. Chemical structures of carbon-carbon double bond containing polymers (a) and thiol derivates (b) for thiol-ene reaction.

Thiol-ene chemistry is also demonstrated a powerful strategy to prepare nanoparticles, dendrimers, hydrogels, polymers, etc.190,251,259,260 For example, Caruso and coworkers employed thiol-ene reaction to stabilize and functionalize polymer multilayer- coated silica particles and capsules, which had potential applications as drug delivery devices.277 PEG was further grafted the surface of capsules via thiol-ene chemistry, creating protein-resistant or protein (amine)-reactive particles. Yoga and coworkers developed cysteine modified Fe3O4 magnetic nanoparticles via thiol-ene reaction as shown in Figure

278 1.32 (a). Those particles exhibited a T2-weighted magnetic resonance imaging (MRI) contrast-enhancing effect (Figure 1.32 (b)). Nanoparticles based on polymers such as polyesters and polybutadiene are also developed with thiol-ene approach.279-282 Thiol-ene addition is also powerful in dendrimer synthesis and functionalization.283,284 In Figure 1.33,

Hawker and coworkers employed thiol-ene chemistry to construct poly(thioether) based dendrimers as well as modify chain ends with various functionalities.

Thiol-ene radical addition possesses unique advantages to prepare more homogenous crosslinked networks compared to traditional radical method, due to the chain step-growth mechanism.250,251 It provides a powerful tool to control crosslinked network properties such as hydrogels. Anseth and coworkers demonstrated that thiol-ene resulted

PEG-based hydrogels exhibited increased ductility, tensile toughness, and yield strain compared to radical chain-polymerized hydrogels due to the increase in network cooperativity and decrease in heterogeneity in the thiol-ene resulted hydrogels (Figure

1.34).285 In the photo-degradation testing as shown in Figure 1.34 (d), thiol-ene hydrogels exhibited faster degradation than radical chain-polymerized hydrogels due to the decreased network connectivity at the junction points inside thiol-ene hydrogels. Crosby and Tew 59 developed highly resilient PEG/PDMS-based hydrogels using thiol-ene radical addition.286,287 Physical properties of those hydrogels can be easily tuned by varying the ratio between PEG and PDMS. Thiol-ene Michael addition is also employed to prepare hydrogels for biomedical applications.288-291 The “ene” units are generally based on vinylsulfone or maleimide, both of which are highly electron deficient. They can react with thiol groups in neutral or pH 7.4 buffer environments.

Figure 1.32. (a) Preparation scheme of Fe3O4-based cysteine (Cys) functionalized magnetic water-soluble nanoparticles (NPs). (b) T2-weighted magnetic resonance images of (1) Cys-

Fe3O4 NP-labeled cells and (2) untreated cells as the control. A dark contrast appeared in b(1) marked by the arrow, indicating that Cys-Fe3O4 NPs can be employed as an MRI contrast agent.278 60

Figure 1.33. Synthesis scheme of poly (thioether) based dendrimers and the functionalization of the chain ends using thiol-ene addition.284

Figure 1.34. Schematic of (a) Chain polymerized (CP) hydrogels via radical polymerization and (b) Step polymerization (SP) hydrogels via thiol-ene addition. (c) Shear toughness of two types of hydrogels. SP hydrogels were tougher than CP hydrogels upon stretching. (d)

The erosion depth of photopatterned channels as a function of irradiation time. SP hydrogels degraded faster than CP hydrogels.285

Figure 1.35. (a) Schematic of photo-patterning process of C=C containing hydrogels via thiol-ene “click” chemistry. Fluorescent images of hydrogels after photo-patterning: (b)

Three different fluorescently labeled biological cues were incorporated to hydrogels at user-defined times and spatial locations sequentially. (c) By controlling the focal point of the laser light in three dimensions using a confocal microscope, micrometer-scale spatial patterning resolution was achieved.248

Besides fabricating hydrogels, thiol-ene radical addition is also widely used for hydrogel functionalization, since photo-chemistry can be conducted at specific sites at specific time. Anseth and her groups focused a lot in creating 3D patterns in the hydrogel

62 networks using thiol-ene photo chemistry.292 They incorporated carbon-carbon double bonds in the SPAAC resulted hydrogels (Figure 1.28).244,246 Those C=C functionalized as reactive sites for biomolecule conjugation. Thiol-derived biomolecules were attached to hydrogel constructs at UV-light exposed positions via thiol-ene photo-initiated addition.

Pattern properties were controlled by photo masks and the irradiation process (Figure 1.35).

Multi-functionalized hydrogels were also fabricated using sequential irradiation method.293

Hawker and coworkers developed high throughput spotted hydrogel microarrays using thiol-ene reaction, and then further functionalized the hydrogel surface with peptides, proteins, drugs or dyes using thiol-ene reaction.294

1.2.3 Carbonyl-Based Condensation

Another popular organic reaction termed with “click” is the condensation reaction between ketone or aldehyde groups and nucleophiles, with water as the byproduct.190 When hydrazine and alkoxyamine are employed as nucleophiles, the resulted bonds are called hydrazone and oxime, respectively. Both hydrazone and oxime are more stable than traditional imine bonds due to the δ-effect,295 and the carbonyl condensation favors the imine formation296. Additionally, oxime is more stable than hydrazone in physiological conditions.297

In order to obtain high reaction speed and efficiency, both hydrazone and oxime ligation are required to proceed in mildly acidic environments that largely limit their applications in bioconjugation.298 Significant efforts have been focused on the development of nucleophilic catalysis systems that allow hydrazone and oxime ligation to proceed rapidly and efficiently in neutral or physiological environment.298-301 Dawson and 63 coworkers did a thorough kinetics study on the aniline catalysis system.298 It was found that reaction rates were accelerated by two orders in peptide-peptide conjugation when using aniline. Figure 1.36 describes the mechanism of imine bond formation. In Figure

1.36 (a), the reactive specie is the weakly protonated carbonyl groups when catalysts are not used; while in Figure 1.36 (b) with aniline, it is the more highly populated protonated aniline Schiff base. The transformation of the protonated aniline Schiff base to the oxime proceeds extremely rapidly. Based on the structure of aniline, a series of other nucleophilic catalysts have been developed (Figure 1.36 (c)).300-303 For example, Manuel Baca and coworkers found that under neutral pH, p-phenylenediamine catalysts resulted in a 19-fold faster rate of protein conjugation compared to the equivalent aniline-catalyzed reaction.300

The reaction speed not only depends on catalyst structures, but also pH of reaction media.298 Whatever catalysts are used, the reaction favors light basic environments. In addition, Wooley and coworkers employed p-toluenesulfonic acid (p-TsOH) as the catalyst, which worked efficiently in organic environment such as THF.304,305 The oxime ligation can be completed in one hour with stoichiometric dosage.

The advantages of oxime ligation are obvious. It can be conducted at room temperature without any metal catalysts, heating and UV light irradiation. Carbonyl groups and alkoxyamine are stable and they can be easily introduced into precursors, either biomolecules or polymer constructs. Resulted oxime bonds are very stable even without reduction of C=N double bonds. Oxime ligation is insensitive to oxygen and moisture, and can proceed in an open system, and even in vivo.

Figure 1.36. (a) Mechanism of imine bond formation. (b) Transformation of the protonated aniline Schiff base to the oxime/hydrazone. (c) Commercially available chemicals as superior catalysts for oxime/hydrazone formation. At pH 7.0 and 7.4, in the catalyst sequence from the left to the right, the reaction speed increases.298-303

Similar to CuAAC and thiol-ene reactions, oxime ligation has been demonstrated powerful and efficient in preparing functional polymers and protein/peptide-polymer hybrids.190,306 Biomolecules or other motifs are attached to polymer side chains or chain ends post-polymerization. One example is the modification of PCL-based copolymers bearing ketone groups.307 PEO was successfully grafted to ketone-PCL, showing the utility of oxime ligation in polymer modification (Figure 1.34 (a)).308,309 Wooley and coworkers explored the quantitative multi-functionalization of ketone-PCL copolymers sequentially and even one-pot.305 Maynard and coworkers did much work in designing and preparing polymers with ketone or alkoxyamine groups via RAFT and ATRP.310-314 Specific ligands

65 such peptides and proteins were conjugated to polymers via oxime ligation for biomedical applications. For example, a series of alkoxyamine chain-end derived polymers including polystyrene and poly (N-isopropylacrylamide) (pNIPAAm) have been synthesized using controlled radical polymerization. The bioconjugation between polymers and proteins have been utilized to prepare smart materials (Figure 1.37 (b)).315 Kochendoerfer and coworkers did extensive work in preparing PEGylated proteins via oxime ligation.316-318 Both ketone and alkoxyamine groups were incorporated into protein constructs at amine lysine residues, and then PEG with corresponding groups was conjugated to proteins at those specific modified sties. Improved biological activities of PEG-protein hybrids were demonstrated in additional in vitro and in vivo studies. Oxime ligation was a powerful tool in the rational design of protein-polymer constructs with potential therapeutic applications.

Figure 1.37. (a) Quantitative attachment of alkoxyamine derived PEG to ketone-PCL.308,309

(b) Bioconjugation between alkoxyamine terminated poly(N-isopropylacrylamide)

(pNIPAAm) and bovine serum albumin (BSA) proteins derived with ketone.315

Carbonyl-based condensation is utilized to fabricate hydrogels, especially hyaluronic acid (HA)-based hydrogels, since aldehyde groups can be introduced into HA

66 building blocks via simple oxidation, and hydrazide can be introduced by carbodiimide- mediated coupling (Figure 1.38).319-321 When ketone-HA and hydrazide-HA were mixed together in aqueous solution, they reacted with each other to generate crosslinked networks in minutes. The hydrogels were studied for soft tissue engineering and drug delivery applications. Besides hydrazone bonds, oxime ligation has been popular for hydrogel fabrication. In 2012, Maynard first developed first oxime hydrogels based on PEO, and

RGD peptides were also covalently attached in situ.322 Mesenchymal stem cells (MSC) were encapsulated in situ. High cell viability and proliferation demonstrated the utility of the oxime hydrogels for in vitro experiments. Oxime hydrogels based on PEG and HA were also capable of injection through a catheter, and exhibited rapid gelation upon injection into tissue. 321

Figure 1.38. Synthesis scheme of (a) ketone derived Hyaluronic acid (HA) via oxidation and (b) hydrazide derived HA via carbodiimide coupling.319,320

Figure 1.39. (a) Chemical structures of two precursors for self-healable hydrazone hydrogels, and schematic of resulted hydrazone bond. (b) Self-healing properties of hydrazone hydrogels in different pH.323

As discussed previously, carbonyl-based condensation has tunable kinetics. The reaction speed and efficiency depends on pH and catalyst concentration, and the resulting linkages are reversible. These reactions provide a unique strategy for designing viscoelastic materials to study cellular responses to scaffold elasticity and self-healing smart materials.

Becker and coworkers developed PEG-based oxime hydrogels. Mechanical properties and

68 gelation time of those hydrogels can be tuned in two orders of magnitude via simply controlling the pH or aniline catalyst concentration.324 Similar results were also found by

Christman and coworkers.321 Anseth and involves fabricated hydrazone-crosslinked hydrogels.325 Unlike traditional covalently crosslinked hydrogels with predominant elastic properties, their hydrogels were able to mimic aspects of the viscoelasticity of native tissues due to the dynamic stress relaxing crosslinks. Chen and coworker developed PEG- based dynamic hydrogels via hydrazone formation (Figure 1.39).323 Those hydrogels were found in an environmental adaptive self-healing ability and the sol-gel transition. Those hydrogels were self-healable in acidic (pH 3 and 6) environments through hydrazone exchange. In neutral environments (pH 7), those hydrogels can only repair damages when catalytic aniline was added at preparation. The aniline catalyst can also accelerate the self- healing process under acidic environments. When adjusting pH, hydrogels showed a sol- gel transition behavior.

The reversibility of hydrazone and oxime bonds was also utilized for designing controlled release devices, in particular pH sensitive systems. Zhu and coworkers have synthesized a triblock polymer of PEG-PCL-PEG, where PEG units and PCL units were connected via oxime bonds.323 Doxorubicin (DOX) was utilized as a model drug that was physically encapsulated inside PEG-PCL-PEG micelles. DOX release was accelerated when decreasing the pH from 7.4 to 5.0, implying that those micelles based on oxime- connected polymers were promising for pH-triggered drug release. Drugs can also be covalently attached to polymeric delivery constructs via hydrazone or oxime bonds.326-331

Drug loading and release can be precisely controlled by tuning the pH as well as the introduction of hydrophobic units. In Figure 1.40, Zhong and coworkers synthesized 69 hydrazine containing PEO copolymers first, and then covalently conjugated DOX drug to polymer side chains via hydrazone bonds. In vitro release studies demonstrated much faster release of DOX at pH 5.6 and 6.0 than at pH 7.4. The in vivo pharmacokinetics and biodistribution studies in mice showed that the DOX-grafted PEO prodrugs significantly prolonged circulation time and enhanced drug accumulation in the tumor as compared with free DOX.326

Figure 1.40. (a) Chemical structure of Doxorubicin (DOX) grafted PEO. (b) pH-dependent release at 37 °C from PEO-based prodrugs (drug loading 2.9%).326

Oxime ligation is also used for the surface modification and functionalization and preparing microarrays of bioactive motifs.306 Surfaces arrays are used in the fields of biomedical engineering, regenerative medicine, and cell engineering. Yousaf and coworkers did extensive work in this field. They fabricated functional surfaces on different substrates, including polymers, gold, glass, indium tin oxide (ITO), and even cell membranes.332-336 For example, they prepared polyesters substrate containing ketone groups for post-curing modification. Using microfluidic lithography (μFL) or photo-

70 chemistry followed by oxime ligation, they created patterns of various biomolecules, including peptides, proteins, DNA and other growth factors.335,337-339 Similar work has also been done on poly(methacrylate)-based substrates by Maynard and coworkers.340-343

Micropatterns of alkoxyamine groups were fabricated using photolithography.

Biomolecules were then tethered to well-designed sites via oxime ligation to study or control cell behaviors. Yousaf and coworkers focused on creating self assembled mono- layers (SAMs) on inorganic substrates, where the outside tails were either alkoxyamine or ketone groups. Dynamic surfaces including patterns, gradients, or morphologies were generated via photo-chemistry, micro-contact printing, dip pen nanolithography, etc.333,336,344,345 As shown in Figure 1.41, different patterns were achieved via the combination of oxime ligation and photo-chemistry.346

Figure 1.41. (a) Fabrication of micropatterned aminooxy surfaces on glass substrates by photo-deprotection, and subsequent ligand immobilization via oxime ligation. (b), (c), and

(d) Patterns with different morphology were obtained via using different photo-mask.346

1.3. Conclusion

Aliphatic polyesters and amino acid-containing polymers have received considerable use in biomedical fields, due to their biological tolerance, degradation and resorbable properties.Tey are used in a number of applications and devices that have been approved for use in humans by the United Sttes Food and Drug Administration. In the coming decades, combinatorial and computational approaches will continuously enrich and improve virtual biomaterial libraries, providing a substantial amount of information for novel biomaterial design and preparation. “Click” chemistry has been demonstrated to be a powerful and practical technique to create novel polymers, tailor polymer properties in molecular level, and optimize their biological behaviors. Therefore, the combination of

“click” chemistry and polymers largely broadens biomedical applications of polymer- based biomaterials. It has contributed a lot and will continue to enhance p third generation biomaterials and beyond.

CHAPTER II

MATERIALS AND INSTRUMENTS

2.1 Materials

All reagents were used as received without additional purification unless specifically noted. Fluorenylmethyloxycarbonyl (FMOC)-protected amino acids and preloaded Wang-resins were purchased from CEM Corp. Chromeo 488 azide and Alexa

Fluor 568 hydrazide were purchased from Life Technologies (Carlsbad, CA). Rhodamine

(TRITC)-conjugated streptavidin was purchased from Fisher Scientific. Silica gels were purchased from Sorbent (Norcross, GA), with porosity 60 Å, size 200x400 mesh, surface area 450-550 m2/g, bulk density 0.5 g/ml, and pH 6.0-7.0.

All organic solvents were purchased from Sigma-Aldrich as ACS Grade and used as received without additional purification unless specifically noted. Anhydrous tetrahydrofuran (THF) and N,N-dimethylformamide (DMF) were purchased from Sigma-

Aldrich and used as received. Anhydrous dichloromethane (CH2Cl2) and chloroform

(CHCl3) were prepared by distillation with CaH2 from ACS Grade dichloromethane and chloroform, respectively.

L-leucine: Sigma-Aldrich, ≥98.0%.

L-phenylalanine: Alfa Aesar, ≥99.0%.

P-toluenesulfonic acid: Fisher Scientific, ≥99%.

Triphosgene: Alfa Aesar, ≥98.0%. 73

Hydroxybenzotriazole (HOBt): Fisher Scientific, ≥98%.

N,N,N’,N’-Tetramethyl-O-(1H-benzotriazol-1-yl)uronium hexafluorophosphate

(HBTU): Sigma-Aldrich, ≥98.0%.

Diisopropylethylamine (DIPEA): Sigma-Aldrich, ≥99.0%.

Triisopropylsilane (TIPS): Sigma-Aldrich, ≥99.0%.

Trifluoroacetic acid (TFA): Sigma-Aldrich, ≥99.0%.

Sodium carbonate (NaCO3): Fisher Scientific, ≥99.5%.

Stannous octoate: Sigma-Aldrich, ≈95.0%.

2-Hydroxyethyl acrylate: Sigma-Aldrich, ≈96.0%.

(Boc-aminooxy) acetic acid: Sigma-Aldrich, ≥98.0%.

4-Dimethylaminopyridine (DMAP): Sigma-Aldrich, ≥99.0%.

N, N’ -Diisopropylcarbodiimide (DIPC): Sigma-Aldrich, ≥98.0%.

Sodium bicarbonate (NaHCO3): Sigma-Aldrich, 99.7-100.3%

1,6-dibromohexane: Fisher Scientific, ≥98%.

N-hydroxyphthalimide: Alfa Aesar, ≥98.0%.

Triethylamine (TEA): Sigma-Aldrich, ≥99%

Sodium azide (NaN3): Sigma-Aldrich, ≥99.5%

Propargyl bromide: Sigma-Aldrich, solution 80 wt. % in toluene.

Hydrazine monohydrate: Sigma-Aldrich, 64%-65%.

5-hexynoic acid: Sigma-Aldrich, 97%.

Bromohexanoic acid: Sigma-Aldrich, 97%.

Triphenylcarbenium tetrafluoroborate: Sigma-Aldrich.

1,4-cyclohexanedione monoethylene acetal: Sigma-Aldrich, 97%. 74

3-chloroperoxybenzoic acid: Sigma-Aldrich, ≤77%.

N-Boc-4-hydroxyaniline: Sigma-Aldrich, 97%.

Ethyl carbazate: Sigma-Aldrich, 97%.

1,1'-carbonyldiimidazole (CDI): Sigma-Aldrich, Reagent grade.

Potassium carbonate (K2CO3): Sigma-Aldrich, ≥99%.

5-bromo-1-pentene: Fisher Scientific, 96%.

4-formylbenzoic acid: Sigma-Aldrich, 97%.

Poly(ethylene glycol) (PEG): Sigma-Aldrich, Average Mn 4600.

Pentaerythritol: Sigma-Aldrich, ≥99%.

2,2-Bis(bromomethyl)-1,3-propanediol: Sigma-Aldrich, 98%.

3-Allyloxy-1,2-propanediol: Sigma-Aldrich, 99%.

P-Toluenesulfonyl chloride: Sigma-Aldrich, ≥99%.

Aniline: Sigma-Aldrich, ≥99.5%.

Calcium hydride (CaH2): Sigma-Aldrich, ≥97%.

e-Caprolactone (CL): Sigma-Aldrich , 97%, dried with CaH2 overnight, distilled in vaccuo twice, and stored under argon.

2.2 Instruments

Nuclear Magnetic Resonance (NMR): NMR spectra were obtained using a Varian

NMRS 300 MHz spectrometer. Relaxation time of 1H NMR was 10 sec and scan numbers were 128. Relaxation time of 13C NMR was 1 sec and scan numbers were 10000. Data analysis was done by ACD/NMR Processor Software. Chemical shifts were reported in parts per million (δ) and referenced to the chemical shifts of the residual solvent (1H NMR, 75

13 CDCl3 7.27 ppm, DMSO-d6 2.50 ppm, D2O 4.80 ppm; C, CDCl3 77.00 ppm, DMSO-d6

39.50 ppm).

Fourier Transform Infrared (FT-IR): FT-IR spectra were recorded on a Shimadzu

MIRacle 10 ATR-FTIR spectrometer from 600 to 4000 cm-1. Resolution was 4 cm-1 and scan numbers were 128. Data analysis was done by Win-IP Pro Software and the data were plotted in Origin 8 Software. Polymer films were prepared either by solvent casting or compression molding. Hydrogel sponges were ultra-dried before testing.

Thin-layered Chromatographic Analyses (TLC): TLC of organic compounds were performed by spotting and developing samples on flexible silica gel plates (Selecto

Scientific, Silica Gel 60, F-254 with fluorescent indicator) using customized solvents as eluents.

Size Exclusion Chromatography (SEC): SEC analyses were performed using a

TOSOH HLC-8320 gel permeation chromatograph (GPC) equipped with refractive index

(RI) detector. SEC used two separation columns of SuperAW3000 and SuperAW-H (TSK-

GEL, Tosoh). N,N-Dimethylformamide (DMF) (with 0.01 M LiBr) was used as the eluent with a flow rate of 0.8 mL/min at 50 °C. The molecular mass was calculated from universal calibration based on polystyrene standards.

Mass Spectra: Matrix-assisted laser desorption/ionization-time of flight (MALDI-

TOF) mass spectra were carried out on a Bruker Ultraflex-III TOF/TOF mass spectrometer

(Bruker Daltonics, Inc., Billerica, MA) equipped with a Nd:YAG laser (355 nm). All spectra were measured in positive reflection mode. The instrument was calibrated using external polystyrene or PMMA standards at the molecular mass under consideration. Trans-

2-[3-(4-tert-butylphenyl)-2-methyl-2-propenylidene]-malononitrile (DCTB, Santa Cruz 76

Biotechnology, Inc., >99%) which was used as matrix was prepared in CHCl3 at a concentration of 20 mg/mL. Sodium trifluoroacetate (NaTFA) which was used as cationizing salt was prepared in MeOH/CHCl3 (v/v = 1/3) at a concentration of 10 mg/mL.

All the samples were dissolved in CHCl3. Matrix and cationizing salt were mixed in a ratio of 10/1 (v/v). The sample preparation involved depositing 0.5 μL of matrix and salt mixture on the wells of a 384-well ground-steel plate, after the spots dried, depositing 0.5

μL of sample on the matrix spot, adding another 0.5 μL of matrix and salt mixture on the dry sample spot. After evaporation of solvent, the target plate was inserted into the

MALDI source. The laser was adjusted and attenuated to minimize the undesired polymer fragmentation and to optimize the peak intensity. Electrospray ionization (ESI) was performed using an HCT Ultra II quadrupole ion trap mass spectrometer (Bruker Daltonics,

Billerica, MA) equipped with an electrospray ionization source.

UV-Visible Spectroscopy (UV-Vis): UV-Vis analyses were performed on a Synergy

Mx spectrophotometer (Biotek Inc.) using quartz cuvette. 2 mL solution was tested for each sample. Spectrum was performed from 230-400 nm with each data point recorded for every

1 nm. Reader control was via BioTeck’s Gen5TM Data Analysis Software.

Quartz Crystal Microbalance (QCM): A Q-sense E4 operator from Biolin Scientific

AB was used to study the peptide uptake of polymer thin films. The SiO2-coated crystal sensor X301(5 MHz resonant frequency) was chosen as substrate for polymer spin-coating.

Sensors were cleaned as follows: 1) washed with customized solvents to remove polymer layers and then dried in nitrogen flow; 2) exposed to UV-ozone environments 15 min; 3) soaked in 2 wt% sodium dodecyl sulfate (SDS) aqueous solution for 30 min, washed with deionized water, and dried in nitrogen flow; 4) exposed to UV-ozone environments 15 min. 77

After UV-ozone cleaning, sensors were used for spin-coating immediately. Polymer solutions were prepared by dissolving polymers (30 mg) in chloroform (2 mL) and filtered with a 0.4 μm PTFE filter. The films were spun at 2000 rpm for 1 min, and the acceleration time was 10 s. All coatings were annealed overnight at 25 oC. All QCM testing was performed at 20 ± 0.1 °C using a flow rate of 0.1 mL/min. In a typical experiment, polymer- coated QCM sensors were exposed to the respective buffer solution overnight to reach a hydrated equilibrium. The baseline was established over a 10 min interval. The reaction formulation was introduced using a 10 min continuous flow to make sure that the entire chamber was saturated with the reaction formulation. After 1 h, the sensors were rinsed thoroughly with buffer solution until a plateau was reached. All presented data corresponded to the normalized frequency of the seventh overtone. On the basis of the

Sauerbrey model, the following equation:

△ f 푚 = C n is often utilized to convert the frequency shift into the mass change per area:(1) where m is the mass change per area (mg/m2), C is the sensitivity constant, −0.177 (mg/(m2·Hz)), Δf is the change in resonant frequency (Hz), and n is the overtone number.

Thermal Analysis: All samples were vacuum dried over night before testing. The degradation temperatures (Td) of polymers were determined by TGA (TA instruments,

Q500 TGA) across a temperature range of 30-1000 °C at a scanning rate of 20 °C min−1 under nitrogen. Sample mass was around 10 mg for each testing. The thermal transitions of polymers were characterized by DSC (TA Instruments, Q2000 DSC) at a scanning rate of 10 °C min−1 under nitrogen. Temperature range was from -50 to 250 oC. Equilibrium

78 time at highest/lowest temperature was 5 min. Aluminum pan was used. Mass difference between the sample pan and the control pan was controlled less than 0.1 mg. Sample mass was around 5 mg for each testing. All data were reported from the second heating cycle.

Values for the thermal transitions were determined from three individual measurements.

Dynamic mechanical analysis (DMA): DMA analysis was performed using a TA

Q800 DMA instrument with stress-strain tensile testing mode at 25 oC. Sample dimensions were 40 mm × 2.0 mm × 0.2 mm. Clamp force was 2 N and preload force was 0.001 N.

The stain rate was 1.5% s-1. Using small strains (<0.15%) the Young’s moduli were determined using the slope of the tangent line in the linear regime. Stress–strain data were reported using the TA Universal Analysis software. The data were plotted in Origin 8 and

Young’s modulus values were calculated using regression analysis in the linear regime.

Values for Young’s moduli and standard deviations were determined from four individual measurements.

Destructive Tensile Testing: The elastic modulus and tensile properties of polymers were measured using an Instron 3365 universal materials testing machine. The preload force was 0.01 N. The gauge length was 20 mm and the crosshead speed was set at 3 mm min−1. Polymer film specimens were 40 mm × 2.0 mm × 0.2 mm. Polymer fiber specimens were 40 mm × 2.0 mm × 0.01 mm. Stress–strain data were reported using the Instron

Bluehill software. The data were plotted in Origin 8 and elastic modulus values were calculated using regression analysis in the linear regime prior to the yield point. Results presented are average values for six individual measurements. Elastic modulus was calculated using the slope of the tangent line of the data curve prior to the yield point.

Mechanical Analysis of Hydrogels: Mechanical strength and gelation kinetics of hydrogels were tested using ARES G2 Rheometer (TA Instrument) with 8 mm parallel plate geometry at room temperature. Normal force was 10 N and frequency ramped from 100 to

0.1 rad s–1 with 1% strain amplitude. Gels were soaked in customized buffer during the whole mechanical strength testing. In gelation kinetics, frequency was 10 rad/s and strain amplitude was 10%. The top of gel formulation was covered with silicon oil to avoid dehydration.

Fluorescence Microscopy Testing: Florescence Images were recorded on an

OLYMPUS IX 81 fluorescence microscope and were unaltered. Data analysis was performed using MetaMorph software.

Scanning Electron Microscopy (SEM): SEM testing was performed using JSM-

7401F (JEOL, Peabody, MA). The acceleration voltage for SEM imaging was 5.00 kV.

Samples were vaccuo dried overnight before testing.

Static Contact Angles: The testing was performed using an advanced goniometer

(Ramé-Hart Instrument Co., model 500) at 25 °C using ultrapure water (1 μL) (18 MΩ cm–

1) as the probe fluid and analyzed by a drop shape analysis method.

CHAPTER III

RESORBABLE, AMINO ACID-BASED POLY (ESTER UREA)S CROSSLINKED

WITH OGP PEPTIDE WITH ENHANCED MEAHCNICAL PROPERTIES AND

BIOACTIVITY

Portions of this work have been published previously as

Kimberly Sloan Stakleff, Fei Lin, Laura A. Smith Callahan, Mary Beth Wade,

Andrew Esterle, James Miller, Matthew Graham, and Matthew L. Becker

Acta Biomaterialia, 2013, 9, 5132–5142

3.1 Outline

Materials currently used for the treatment of bone defects include ceramics, polymeric scaffolds and composites, which are often impregnated with recombinant growth factors and other bioactive substances. While these materials have seen instances of success, each has inherent shortcomings including prohibitive expense, poor protein stability, poorly defined growth factor release and less than desirable mechanical properties.

We have developed a novel class of amino acid-based poly(ester urea)s (PEU)s materials which are biodegradable in vivo and possess mechanical properties superior to conventionally used polyesters (<3.5 GPa) available currently to clinicians and medical providers. We report the use of a short peptide derived from osteogenic growth peptide

(OGP) as a covalent crosslinker for the PEU materials. In addition to imparting specific

81 bioactive signaling, our crosslinking studies show that the mechanical properties increase proportionally when 0.5% and 1.0% concentrations of the OGP crosslinker are added. Our results in vitro and in an in vivo subcutaneous rat model show the OGP-based crosslinkers, which are small fragments of growth factors that are normally soluble, exhibit enhanced proliferative activity, accelerated degradation properties and concentration dependent bioactivity when immobilized.

3.2 Introduction

Synthetic, degradable polymers have been used in a myriad of ways for regenerative medicine and orthopedic applications. However, they generally lack the mechanical properties necessary for load-bearing surgical interventions.18 Numerous examples are found in the literature where degradable polymers have been used successfully in orthopedic applications, including poly(l-lactic acid) (PLLA)347-350, polycaprolactone and poly(propylene fumarate)92,103,351-353; however, the maximum reported mechanical properties for these polymers hover near 3.0–3.5 GPa (Young’s modulus).354-356 For comparison, the elastic modulus of cortical bone within the mid- diaphysis of a long bone, along the axis of the bone, is approximately 18 GPa.357,358 Most of the clinical community has acknowledged that PLLA has insufficient mechanical properties to sustain load-bearing applications. Resorbable biomaterials that possess high moduli are needed for numerous regenerative medicine and orthopedic applications.4,359

Ideally the mechanical properties of the scaffold must be appropriate to regenerate bone in load-bearing sites.360 It is unlikely that stand-alone polymers will attain those numbers. Composite and blending approaches have increased the mechanical properties of 82 degradable materials, yet producing engineering polymeric materials with sufficient mechanical properties that retain the ability to degrade fully has remains a challenge.

Traditional methods to mechanically reinforce the polymers, including covalent crosslinking, generally limit or prevent the biodegradation. One strategy has been to use naturally occurring amino acids as building blocks for monomer precursors.113,175 However, conventional poly(α-amino acids), despite their biological origin, possess distinct physical, chemical and biodegradation properties which limit their synthetic utility.361 However, the poly(ester urea) materials described herein are a significant step in the right direction in that they are both strong, yet degradable.

Significant limitations in bringing new materials to the clinic include the findings that fully synthetic materials lack cell specific receptors and have poorly defined serum adsorption properties, which can vary widely depending on the amount and nature of the adsorbed layer.362 Recent advances in synthetic polymer chemistry have enabled the synthesis of polymeric materials designed to elicit specific cellular functions and to direct cell–cell interactions.5,113,363,364 Biomimetic approaches based on polymers derivatized with adhesive receptor-binding peptides, glycoproteins and tethered growth factors have been reported to enhance interactions at the biological–synthetic interface.365-368 Potential solutions have included doping with proteins or peptides or decorating the polymer with covalently tethered peptides that mimic the extracellular matrix or growth factors have been employed over the years.369,370 One example is osteogenic growth peptide (OGP).

OGP is a naturally occurring 14-mer peptide growth factor found in serum at micromolar concentrations.371,372 As a soluble peptide, OGP regulates proliferation, differentiation and matrix mineralization in osteoblast lineage cells.372,373 The active portion of OGP, the 83

OGP(10–14) region, is cleaved from the peptide and binds to the OGP receptor which activates multiple signaling pathways including the MAP kinase, Src and RhoA pathways.374-376 When administered intravenously to animals, OGP and OGP(10–14) promote increased bone density and stimulate healing, suggesting a potential use in bone- tissue-engineering applications.377 We, and others, have shown that OGP retains the ability to initiate potent osteoinductive activity when tethered to surfaces and scaffolds.378,379

Third-generation tissue-engineering materials are designed to stimulate specific cellular responses at the molecular level, mimic the dimensionality of the native tissue, and degrade at the rate in which the tissue is repaired with by-products that are benign and resorbable.2 Further advances in both synthetic methodology and scaffold fabrication are needed to drive these efforts forward. While bioactive approaches have been demonstrated to aid biochemical signaling and integration into host tissues, they generally reduce the mechanical properties of the material. This report describes our efforts to develop a new class of crosslinked, mechanically robust polymeric material for orthopedic applications.

Our methods include enhanced mechanical properties in addition to imparting specific osteogenic signaling motifs. To mechanically reinforce our polymers and stimulate specific biological activity we have incorporated OGP-based crosslinkers. Peptide-crosslinked phenylalanine and leucine-based poly(ester urea) (PEU) homopolymers were synthesized and tethered with 0.5% and 1.0% OGP(10–14). In addition, the semicrystalline nature of poly(ester urea)s afford non-chemical methods in which the mechanical properties, chemical stability and biodegradation rates can be tailored. This report describes in detail the chemical, mechanical, in vitro and in vivo data which demonstrate enhanced moduli, biocompatibility and resorption of the PEU materials. Furthermore, the data herein 84 highlights the many opportunities that the clinical community will find for these materials in regenerative medicine applications.

3.3 Experimental Section

3.3.1 Materials and Methods

Unless listed otherwise, all solvents and reagents were purchased from Sigma-

Aldrich and used as received. Fluorenylmethyloxycarbonyl (FMOC)-protected amino acids and preloaded Wang-resins were purchased from CEM Corp. Alpha minimum essential medium (α-MEM) and ultraculture media were purchased from Lonza. All other cell culture reagents were purchased from Invitrogen Corp. All reagents were used as received.

3.3.2 Synthesis of di-p-toluenesulfonic acid salt monomers

Di-p-toluenesulfonic acid salts of bis-l-phenylalanine and bis-l-leucine esters were prepared using procedures published previously,162 as shown in Figure 3.1. Briefly, l- leucine (1.31 g, 10 mmol), 1,6-hexanediol (0.48 g, 4 mmol), p-toluenesulfonic acid (1.92 g,

10 mmol) and toluene (20 ml) were mixed in a 250 ml three-neck flask equipped with Dean

Stark trap and a magnetic stirrer bar. The system was purged with nitrogen for 30 min after which the reaction mixture was heated at 135 °C under nitrogen for 20 h. The reaction mixture was allowed to cool to ambient temperature and the crude product was isolated by vacuum filtration. The organic residue was recrystallized four times using 25 ml water to yield 2.26 g (82%) of l-leucine based-monomer, the di-p-toluenesulfonic acid salt of bis-l-

85 leucine ester, as a white power. The product was characterized with 1H and 13C nuclear magnetic resonance (NMR), and melting point measurements.

Figure 3.1. The two-step general synthetic route of amino acid-based poly(ester urea)s

(PEU)s. First 1,6-hexane diol was condensed with 2 equivalents of α-amino acids catalyzed and protonated with p-toluenesulfonic acid. Following the diamine salt synthesis, interfacial polycondensation with triphosgene yielded the PEU homopolymer. To both enhance the mechanical properties and impact specific biological signaling to the polymer, highly reactive vinyl groups were coupled to a short fragment of OGP peptide and the construct was crosslinked photochemically. Small mole fractions of crosslinker and relatively high molecular masses enabled increased mechanical properties despite inducing radical-based chain scission.

Di-p-toluenesulfonic acid salt of bis-l-leucine hexane-1,6-diester(1-LEU-6): mp:

186–188 oC; 1H-NMR (300 MHz, DMSO): 0.90 (d, 12H) 1.34 (s, 4H) 1.45–1.80 (m, 8H) 86

2.29 (s, 6H) 3.99 (t, 2H) 4.15 (d, 4H) 7.13 (d, 4H) 7.49 (d, 4H) 8.31 (s, urea H); 13C-NMR

(75 MHz, DMSO): 169.91, 145.34, 137.35, 129.10, 125.48, 65.52, 50.62, 27.76, 24.75,

23.79, 23.13, 21.92, 20.79.

Di-p-toluenesulfonic acid salt of bis-l-phenylalanine hexane-1,6-diester (1-PHE-6): mp: 215–217 oC; 1H-NMR (300 MHz, DMSO): 0.90–1.15 (m, 4H) 1.38 (s, 4H) 1.25–1.50

(m, 4H) 2.23 (s, 6H) 2.91–3.09 (m, 2H) 3.10–3.21 (m, 2H) 4.01 (t, 4H) 4.30 (t, 2H) 7.11

(d, 4 H) 7.19–7.40 (m, 10H) 7.49 (d, 4H) 8.43 (s, active H); 13C-NMR (75 MHz, DMSO):

169.06, 145.00, 138.12, 134.70, 129.32, 128.55, 128.21, 127.54, 125.53, 65.45, 53.34,

36.20, 27.63, 24.70, 20.82.

3.3.3. Interfacial polycondensation of 1-LEU-6 and 1-PHE-6

A general scheme for PEU synthesis is given in Figure 3.1. Monomer 1-LEU-6

(6.89 g, 10 mmol), sodium carbonate (3.18 g, 30 mmol) and water (150 mL) were mixed in 500 ml three-neck flask equipped with an overhead mechanical stirrer and a thermometer.

The mixture was heated with a warm water bath at 40 °C for 30 min. The water bath was removed and replaced with an ice-salt bath. When the inside temperature decreased to about 0 °C, pre-prepared triphosgene solution (1.035 g, 3.30 mmol in 30 ml chloroform) was added to the reaction system quickly (5 s) with fast mechanical stirring. The reaction was allowed proceed for 30 min and then additional aliquots of triphosgene (0.108 g,

0.330 mmol, total) were dissolved in chloroform (5 ml) and 1 ml aliquots were added into the reaction system every 10 min. After the addition of the triphosgene, the organic phase was precipitated into hot water, filtered and dried in vacuum to yield a white solid (3.2 g,

74.5% yield). The product was characterized by 1H-NMR, 13C-NMR, Fourier transform 87 infrared (FT-IR) spectroscopy, size exclusion chromatography (SEC), thermogravimetric analysis (TGA) and differential scanning calorimetry (DSC). The molecular weights and thermal properties of the polymers are listed in Table 3.1.

Table 3.1. Characterization data summary for the amino acid-based poly(ester ureas).

Samples Mw Mw/Mn Tg Tm Td G’(GPa)

Poly(1-LEU-6) 76,800 2.12 57 126 275 4.4 ± 0.9

Poly(1-PHE-6) 84,000 2.42 77 153 335 6.1± 1.1

Poly(1-LEU-6): FT-IR (cm−1) 1740 [–C(CO)–O–], 1648, 1542 [–NH–C(O)–NH–],

3283 [–NH–C(O)–NH–]; 1H-NMR (300 MHz, DMSO): 0.91 (d, 12H) 1.20–2.00 (m, 14H)

4.21 (t, 4H) 4.45 (d, 2H) 5.35–5.80 (m, active H); 13C-NMR (75 MHz, DMSO): 174.64,

157.08, 65.00, 51.60, 65.00, 51.00, 42.31, 28.20, 25.29, 24.74, 22.61, 22.12.

Poly(1-PHE-6): FT-IR (cm−1): 1736 [–C(CO)–O–], 1649, 1553 [–NH–C(O)–NH–],

3384 [–NH–C(O)–NH–]; 1H-NMR (300 MHz, DMSO): 0.91 (d, 12H) 1.20–2.00 (m, 14H)

4.21 (t, 4H) 4.45 (d, 2H) 5.35–5.80 (m, active H); 13C-NMR (75 MHz, DMSO): 174.64,

157.08, 65.00, 51.60, 65.00, 51.00, 42.31, 28.20, 25.29, 24.74, 22.61, 22.12.

3.3.4. Peptide crosslinker

A symmetric vinyl functionalized OGP(10–14) was synthesized using the

(Aloc)KYGFGGK(Aloc) sequence by solid-phase FMOC chemistry. Chemical structure of this peptide is shown in Figure 3.2. Peptides were cleaved from resin using standard conditions (45 min, 95% trifluoroacetic acid, 2.5% triisopropylsilane, and 2.5% water (by 88 volume)) and precipitated in cold diethyl ether. Following two trituration cycles, the peptides were dialyzed in deionized water (molecular weight (MW) cut-off 100 g mol−1, cellulose membrane, Pierce), and molecular weight was verified with matrix-assisted laser desorption ionization-time of flight spectroscopy (FW (+H) 924.50 g mol−1, required

924.44 g mol−1).

Figure 3.2. The mitogenic OGP(10–14) peptide, YGFGG, is symmetrically functionalized with lysine (K) residues at both the N- and C-termini (KYGFGGK) with a highly reactive vinyl groups coupled on the side chain of each lysine residue.

3.3.5 Molecular mass characterization

Number-average (Mn) and weight-average (Mw) molecular weights and molecular weight distribution (Mw/Mn) were determined by SEC. The instrument was equipped a guard column and set of 50, 100, 104 Å, and linear (50–104 Å) Styragel 5 μm columns, a

Waters 486 tunable UV–Vis detector, and a Waters 410 differential refractometer. All the analyses were carried out with a flow rate of 1 ml min−1 using an RI detector.

Dimethylformamide (DMF) was used as the eluent at 50 °C. The respective molecular weights and molecular weight distributions of each polymer were determined using a

89 universal calibration curve, which was obtained by plotting ln-([ɳ]/Mn) as a function of elution volume, after calibration with polystyrene standards (Polymer Laboratories).

3.3.6 Thermal characterization

The degradation temperatures (Td) of poly(1-LEU-6) and poly(1-PHE-6) materials were determined by TGA (TA instruments, Q500 TGA) across a temperature range of 30–

500 °C at a scanning rate of 20 °C min−1 under nitrogen. The thermal transitions of the poly(1-LEU-6) and poly(1-PHE-6) materials were characterized by DSC (TA Instruments,

Q2000 DSC) at a scanning rate of 10 °C min−1. Values for the thermal transitions were determined from three individual measurements.

3.3.7 Crosslinked scaffold fabrication

Polymer plugs were prepared by a compression molding fabrication process.

Polymer materials ∼1 g, and a corresponding amount of peptide crosslinker and photoinitiator Irgacure 2959 were dissolved in 20 ml hexafluoride-2-propanol. The clear solution was photo-irradiated with 365 nm UV light for 45 min. The solvent was evaporated in a fume hood at room temperature for 24 h followed by vacuum drying at

80 °C for 24 h. The composite was crushed into small pieces and melt pressed into sheets using a Carver Hydraulic unit model 3912 with pressure 1800 psi at the designed temperature (130 °C for poly(1-LEU-6) and 180 °C for poly(1-PHE-6)). The polymeric block was cooled to room temperature, and annealed in vacuum at proper temperature for

24 h (80 °C for poly(1-LEU-1) and 130 °C for poly(1-PHE-6)). The final polymeric block

90 was cut into small circular plugs 0.5 cm in diameter for testing. All plugs were stored in a nitrogen atmosphere at −20 °C for future use.

3.3.8 Mechanical characterization

3.3.8.1 Dynamic mechanical analysis (DMA)

The Young’s moduli of the poly(1-PHE-6), 0.5% OGP poly(1-PHE-6) and 1.0%

OGP poly(1-PHE-6) data were determined using a DMA ( TA instruments, Q800 DMA) with sample dimensions 40 mm × 2.0 mm × 0.2 mm at ambient temperature. The strain rate was 1.5% s−1. Using small strains (<0.15%) the Young’s moduli were determined using the slope of the tangent line in the linear regime. Stress–strain data were reported using the TA Universal Analysis software. The data were plotted in Origin 8 and Young’s modulus values were calculated using regression analysis in the linear regime. Values for

Young’s moduli and standard deviations were determined from four individual measurements.

3.3.8.2. Destructive tensile testing

The elastic modulus and tensile properties of the poly(1-LEU-6) and the poly(1-

PHE-6) were measured using an Instron 3365 universal materials testing machine. The gauge length was 20 mm and the crosshead speed was set at 30 mm min−1. The specimens were 40 mm long, 4 mm wide and 0.2 mm thick. Stress–strain data were reported using the

Instron Bluehill software. The data were plotted in Origin 8 and elastic modulus values were calculated using regression analysis in the linear regime prior to the yield point.

Results presented are average values for six individual measurements. The elastic modulus was calculated using the slope of the tangent line of the data curve prior to the yield point.

3.3.9 In vitro cell culture and characterization

Primary human foreskin fibroblasts were obtained from stock cultures isolated from infant male circumcision tissue specimens maintained at the Calhoun Research Laboratory,

Akron General Medical Center. MC3T3E1 osteoblasts were obtained from Riken.

Fibroblasts were maintained in DMEM with high glucose (Gibco, 11965) and MC3T3-E1 osteoblasts in α-MEM (GibcoA10490; Invitrogen, Carlsbad, CA). Each media was supplemented with 10% fetal bovine serum (F6178, Sigma, St. Louis, MO) and 1% penicillin–streptomycin–fungizone (10,000 U/10,000 μg/25 μg) (Lonza Bio Whittaker,

BW17-745E; Fisher Scientific, Waltham, MA). Cells were maintained in an incubator at

37 °C and 5% carbon dioxide/95% air. In preparation for morphological and proliferation experiments, cells were released from the flasks using 2.5% trypsin–EDTA (Invitrogen,

15090-046) and counted on a hemacytometer for viability and concentration determination using 0.4% Trypan Blue (Gibco, 15250). A preliminary morphological assessment was performed with plugs of the poly(1-PHE-6) and poly(1-LEU-6) polymers tethered with 1.0%

OGP in the presence of fibroblast cells at a concentration of 105 cells ml−1 in six-well plates.

After 48 h, cells were viewed with an Olympus CKX41 inverted microscope (Center

Valley, PA) and digital images were captured using QImaging software and a

Micropublisher Real Time Viewing (RTV) 5.0 charge-coupled device (CCD) color cooled camera (QImaging, Princeton, NJ). Thin plugs of polymers were sterilized with ethylene oxide and stored prior to use. 92

Cellular proliferation was measured using a WST-1 viability assay (Dojindo

Molecular Technologies, W201-10; Rockville, MD). Briefly, plugs of poly(lactic acid)

(PLA), poly(1-PHE-6) or poly(1-LEU-6) base polymer and tethered with 0.5% or 1.0%

OGP were sterilized by rinsing in isopropyl alcohol and applied to six-well plates containing 4 × 105 fibroblast or MC3T3 cells. Following incubation for 48 h, the polymers were transferred to a 96-well plate, with care taken not to disrupt cells on polymer surfaces, and rinsed in Tyrode’s Hepes buffer. The Wst-1assay solution was added, incubated with the polymers for 2 h, and the resultant reaction solution was transferred to clean wells where the absorbance was read at 450 nm. The reported absorbance values and standard deviations were determined from four individual measurements.

3.3.10 Animal surgeries

All animal protocols regarding the handling, care, maintenance and surgical procedures were reviewed and approved by the Institutional Animal Care and Use

Committee of Akron General Medical Center. A total of 16 male Sprague–Dawley rats

(Harlan Laboratories, Indianapolis, IN) weighing >250 g were divided into groups containing eight animals each per base polymer type, poly(1-PHE-6) or poly(1-LEU-6).

All animals were pre-anesthetized by a subcutaneous injection with 10–12 mg kg−1 but orphanol mixed with 0.04 mg kg−1 atropine. Following inhalant anesthesia induction at 3% isofluorane in 100% oxygen, the animal was maintained at 1–1.5% isofluorane in 100% for the duration of the surgical procedures.

On the dorsum, four incisions were created with a sterile scalpel blade, two in the left and right lateral direction, approximately 1 cm from the spine and approximately 2 cm 93 apart. A subcutaneous pocket was tunneled using hemostats in the posteroanterior direction.

Thin plugs of polymers sterilized by ethylene oxide were inserted into each pocket and skin incisions were closed with Michel clips. Within the four subcutaneous spaces, each animal received: (1) PLA; (2) either poly(1-LEU-6) or poly(1-PHE-6); and, poly(1-LEU-6) or poly(1-PHE-6) tethered with either (3) 0.5% or (4) 1.0% OGP. Positions 1–4 were randomized for each animal, while retaining diagonal distribution of control and test materials, to account for variability in positioning on the back. The sample size was four animals per polymer construct per time point. After 4 or 12 weeks post-surgical insertion, four animals from each group were euthanized and tissues containing the polymers were collected (2 cm × 2 cm), preserved in formaldehyde and prepared for histological evaluation.

3.3.11 Histology and histomorphometric image analysis

Tissue sections (5 μm) were cut (Leica RM2235 micrometer) and stained by hematoxylin and eosin (Ventana ST5020 Automated Stainer, Hematoxylin 7211 and Eosin

71204) to show normal tissue architecture, Mallory’s trichrome (Ventana NEXES Special

Stains, Trichrome II Staining Kit 860-013) for collagen deposition and cellular infiltration, and Alizarin Red to detect mineralization of calcium as evidence of bone cell activity. For

Alizarin Red S staining, histological sections stained with 40 mM Alizarin Red S solution, pH 4.2, at ambient temperature for 10 min, rinsed five times in distilled water and washed for 15 min in 1 × PBS. Histological sections were counterstained with hematoxylin, dehydrated with ethanol and rinsed in xylene before mounting with Permamount. Eight sections were cut per sample from the middle of the disk and three were selected randomly 94 from each animal (12 totals) for analysis. The sections were made through the middle of the polymer disk and represent the center of the construct. The data were tabulated in a blinded fashion. All slides were examined with an Olympus BX51 light microscope to identify the location of the polymer within the subcutaneous region and digital images were captured using the QImaging camera and software. Histomorphometric features were analyzed on the digital images of the Trichrome sections at 40× using BioQuant Nova

Primev.6.75.10 image analysis software (Nashville, TN). The total area (μm2) of connective tissue surrounding and including the polymer was delineated as the region of interest (ROI). A separate area measurement (μm2) was defined for the regions of tissue extending from the connective tissue capsule into the region containing the polymer, indicating polymer degradation. Within the ROI, the threshold was selected to identify the regions of red stain and a video count array measured the total pixel area (μm2) of collagen deposition/cellular infiltration. The percentages of degradation and cellular infiltration were calculated in relation to the respective total area ROI. The width of the connective tissue capsule was measured at five random locations surrounding the polymer and averaged. The numbers of giant cells and blood vessels were counted within and adjacent to the ROI.

3.3.12 Statistics

Histomorpometric data for comparisons of the polymer constructs were statistically evaluated using one-way analysis of variance (ANOVA) and linear discriminate correlation. Mean values and standard errors are reported, unless otherwise noted.

Student’s t-tests were performed to identify individual comparative differences. Statistical 95 analysis for all other measurements was performed using one-way ANOVA and a comparison-wise Tukey test at 95% confidence. Mean values and standard deviations are reported, unless otherwise noted. The standard deviations of the mean were used as an estimate for the standard uncertainty associated with each measurement technique.

3.4 Results

3.4.1 Synthesis and characterization

In 1997, Katsarava et al. published a synthesis of homo-PEUs, without using diisocyanates, via active polycondensation.380 In this procedure, active carbonates (e.g. di- p-nitrophenyl carbonate) interact with di-p-toluenesulfonic acid salts of bis(α-amino acid)-

α,ω-alkylenediesters. The molecular weight can be tailored by altering the molecular weights of the constituent monomers and the degree of polymerization. We have used a modified version of the process for the synthesis of our base PEU materials which are described as x-amino acid-y where x and y are the number of C atoms in the chain as shown in Figure 3.1. The LEU and PHE amino acid-based PEUs utilized 1,6-hexane diol, and are denoted 1-LEU-6 and 1-PHE-6. The molecular weight, molecular weight distribution and thermal properties of the poly(1-LEU-6) and poly(1-PHE-6) were measured (Table 3.1).

At ambient temperature, poly(1-PHE-6) is not soluble in conventional organic solvents, but is soluble in hexafluoroisopropanol, Dimethylformamide, dimethyl sulfoxide and 3:1 mixtures of tetrachloroethane/phenol. When the poly(1-LEU-6) was melt processed, the Tg remained the same but no melting peak is observed, indicating that no crystallinity was present. This indicates that polymer crystallinity can be suppressed using the appropriate processing method. The degradation temperatures (Td) of the poly(1-LEU-6) and poly(1- 96

PHE-6) materials were over 100 °C higher than the melting temperature, indicating that both materials can be melt processed with limited impact of thermal degradation. These characteristics allow processing techniques such as molding and melt processing to be used to fabricate our scaffolds.

3.4.2 Mechanical properties

The mechanical properties of the PEU plugs where characterized by Instron and

DMA methods reported in Table 3.2, which characteristically report the Young’s and elastic modulus, respectively. The Instron data show clearly that values for both the poly(1-

LEU-6) and poly(1-PHE-6) (4.4 and 6.1 GPa, respectively) exceed the published values for PLLA (2.9 GPa)381 and poly(ε-caprolactone) (280 MPa)382. To maximize accuracy and provide additional information, Instron testing was used to measure the yield strength (YS) and tensile strength (TS). Poly(1-LEU-6) and poly(1-PHE-6) had TS values of ∼470% and

510%, respectively (Figure 3.3). The absolute value of TS does not depend on the size of the test specimen which enables our results to be extrapolated to larger constructs. However, it is influenced by other factors, including sample preparation, defects and temperature. TS does not correlate well with compressive strength as the loading properties and the resulting values can be quite different. DMA data yielded a value of 3.05 ± 0.24 GPa for the elastic modulus of poly(1-PHE-6) and showed that the elastic modulus increased proportionally with increased levels of OGP crosslinking (Table 3.2)). The linear regime of the stress– strain curve (Figure 3.3) was used to calculate the Young’s modulus of the poly(1-PHE-6) homopolymer as well as the 0.5% and 1.0% OGP crosslinked materials. While the materials are unoptimized, these data are significantly stronger than degradable polymers 97 currently available clinically including tyrosine derived polycarbonates (∼1–2 GPa),

PLLA (∼3–3.5 GPa), and poly(propylene fumarate) (2.2 GPa)362.

Figure 3.3. The elastic modulus and tensile properties of the poly(1-LEU-6) and the poly(1-

PHE-6) were measured using an Instron 3365 universal materials testing machine. The

Young’s moduli of the poly(1-PHE-6) (black), 0.5% OGP poly(1-PHE-6) (red), and 1.0%

OGP poly(1-PHE-6) (blue) data were determined using a TA Q800DMA instrument. Using small strains (<0.15%), the Young’s moduli were determined using the slope of the tangent line in the linear regime. Values for Young’s moduli and standard deviations were determined from four individual measurements.

Table 3.2. Summary of the mechanical properties of peptide-crosslinked PEU

Mechanical testing (Instron) DMA Young’ Samples Modulus Elastic Modulus Tensile Tensile stress (GPa) (GPa) strain (%) (MPa) Poly(1-LEU-6) 4.4 ± 0.9 470 ± 50 43±9 - Poly(1-PHE-6) 6.1 ± 1.1 510 ± 30 45±3 3.05 ± 0.24 Poly(1-PHE-6) - - - 3.41 ± 0.06 0.5% OGP Poly(1-PHE-6) - - - 4.18±0.14 1.0% OGP

3.4.3 In vitro proliferation and biocompatibility

Initial in vitro screening for biocompatibility and biodegradation was performed with primary human foreskin fibroblast cells and MC3T3-E1 osteoblasts. The fibroblast morphology showed an adherent state (image not shown) following cell seeding in the presence of poly(1-LEU-6) and poly(1-PHE-6) tethered with 1.0% OGP(10–14). The cells did not display features of cytotoxicity or apoptosis. WST-1 proliferation assays were performed to confirm cellular viability and proliferative rates with poly(1-LEU-6) or poly(1-PHE-6) and polymers each tethered with 0.5% and 1.0% OGP, respectively (Figure

3.4). Comparisons by two-way ANOVA of PLA, poly(1-LEU-6) and poly(1-PHE-6) controls showed no significant differences in fibroblast proliferation. Osteoblast proliferation was increased significantly for poly(1-LEU-6) with 0.5% OGP and for poly(1-PHE-6) with 1.0% OGP compared to polymer controls (∗P < 0.05). Thus, it appears that in addition to providing the mechanical reinforcement, the crosslinked OGP peptide is bioavailable to the cell surface receptors and active for signal initiation.

Figure 3.4. WST-1 proliferation assay of MC3T3-E1osteoblast and primary murine fibroblast cells. Consistent with the biphasic, concentration-dependent proliferative effect as previously reported371, the unfunctionalized homopolymers (PLLA, poly(1-LEU-6) or poly(1-PHE-6)) did not exhibit cell-type-dependent proliferative activity. However, the increasing proliferative trends in both the 0.5% and 1.0% OGP functionalized materials for the osteoblasts relative to the corresponding decreasing trends in the fibroblasts showed that the peptide was bioavailable to the receptor and that we were in a bioactive concentration regime. ∗P < 0.05 compared to polymer controls.

3.4.4 In vivo biocompatibility and degradation by histological image analysis

After demonstrating no appreciable effects on cellular viability, a subcutaneous implant plug model was used to assess biodegradation, cellular infiltration, capsule thickness, inflammatory response and the number of blood vessels. During necropsy for tissue collection, no evidence of fibrosis, granulomatous reactions, necrosis or bacterial

100 infection was observed. After 4 weeks implantation, little evidence is seen in the histological sections for the degradation of either the poly(1-LEU-6) and poly(1-PHE-6) homopolymers (Figure 3.5). The degradation rates for both of the homopolymers are not statistically different from what was measured in the PLLA control (Figure 3.6A).

However, when crosslinked with OGP, both polymer sets show significant levels of degradation which propagate further at 12 weeks. We attribute this increase in degradation to additional free volume in the polymer plugs imparted by the OGP crosslinker and increased water uptake into the bulk material due to the peptides.

The percentage of cellular infiltration was quantitated as a pixel count of the red trichrome stain within the ROI containing the implant in the tissue section. The red stain included the nuclei of cells (predominantly lymphocytes, macrophages and fibroblasts) and the presence of collagen. The amounts of cellular infiltration in the respective data sets showed little significance other than a suggestive trend at 12 weeks in the 1% OGP crosslinked poly(1-LEU-6) relative to PLLA (Figure 3.6B). The relatively low percentages of cellular migration and collagen production within the regions indicate that there were minimal inflammatory and fibrotic responses to the implanted materials.

Promising trends were seen with regard to capsule formation in both data sets

(Figure 3.6C). While the overall thickness data at 4 weeks were similar for PLLA, poly(1-

LEU-6) and poly(1-PHE-6) homopolymers, the capsule thickness of the OGP crosslinked poly(1-PHE-6) materials did not increase appreciably, while the poly(1-PHE-6) containing

0.5% and 1.0% OGP showed less capsule formation at 12 weeks relative to PLLA. The generally small capsule thicknesses (all <500 μm) confirm the lack of any significant fibrotic or granulomatous responses. 101

The numbers of giant cells were counted as an indicator of foreign body tissue response.383 Giant cell counts were similar for PLLA and poly(1-LEU-6); however, the 1.0%

OGP crosslinked poly(1-PHE-6) materials did induce increased numbers of giant cells relative to controls (Figure 3.6D). This may suggest 1% OGP is too high a peptide concentration, yet even though the giant cell numbers were greater, the overall numbers were all comparatively small (all <20). We did not expect to find indications of calcium mineralization via the Alazirin Red stain (Figure 3.5, row C) due to the subcutaneous implantation site, absence of porosity and stem cell source, and follow-up time frame.

Biodegradation was complemented by a higher number of blood vessels in the poly(1-PHE-6) groups (Figure 3.6E) compared to other test groups. An effect on vascularity has not been reported previously in the OGP literature, and we are investigating this phenomenon further. Since the increased blood vessel formation was not accompanied by a significant increase in inflammatory cell infiltration, it appears that recruitment and integration of blood vessels may facilitate biodegradation of the polymers.

The in vivo biocompatibility studies confirmed the in vitro results by demonstrating no adverse response from the integral tissues in contact with the polymeric materials.

Capsule formation around the polymers was similar to what is normally observed for PLA.

In comparison to poly(1-LEU-6), the poly(1-PHE-6) polymers with 0.5% and 1.0% OGP exhibited a significantly favorable interaction with the amount of biodegradation and incorporation of tissue into the polymer materials compared to other test groups.

102

Figure 3.5. Digital images of histology stained slides stained with Masson’s Trichrome at

100×. Tissues were removed at 4 weeks (row A) and 12 weeks (row B) post-implantation.

Row C, serial sections of 12 week histology slides shown in row B stained with Alizarin

Red. 103

Figure 3.6. Quantitative histological analysis of the respective measurements as collected from Masson’s Trichrome analysis at 4 weeks (black) and 12 weeks (grey). Analysis representative of biodegradation (A) by measuring area of tissue migration within the polymer space, cellular infiltration (B), capsule thickness (C), immune response (D) by number of giant cells present and vascularization (E) by counting the number of associated blood vessels. Statistical significance (P < 0.05) was indicated by 1 = compared to PLA;

2 = compared to base polymer; 3 = 4 week vs. 12 week results; 4 = 0.5% OGP vs. 1.0%

OGP; 5 = poly(1-PHE-6) vs. poly(1-LEU-6). 104

3.5 Discussion

Resorbable polymers that are stable at physiological temperatures (Tg > 40 °C) and possess high elastic moduli (>3.5 GPa) are needed for many medical and pharmaceutical applications.368,384-388 To ease the regulatory approval process, naturally occurring amino acids are often used as building blocks of monomer precursors. Polymers created with monomeric units that are comprised of biomimetic and simple structures are generally associated with minimal risks for biocompatibility and do not cause serious in vivo immunostimulatory side effects.6,173,175,177,178,389-391 However, conventional poly(α-amino acids), despite their biological origin, possess distinct physical, chemical and biodegradation properties which limit their synthetic utility. The polymeric synthesis of poly(α-amino acids) is generally difficult and expensive and the resultant materials have a limited property range, especially mechanically. As a result, considerable efforts have been made to replace the amide linkage in the poly(α-amino acids) with a variety of non-amide bonds, yielding novel α-amino acid-based polymeric systems.

One family of α-amino acid-based polymers are PEUs, which are prepared from bis(α-aminoacyl)-α,ω-diol-diester monomers.162 The degradation of PEUs can be tailored to vary over a 6 week to 2 year time-frame depending on factors such as implant size, ester linkages formed by the various amino acid residues, hydrophilicity of the PEU polymer and extent of crystallinity.362 To increase the bioactivity of the PEU materials, the polymer chains were photochemically crosslinked with OGP.373 As a soluble peptide, OGP regulates proliferation, differentiation and matrix mineralization in osteoblast lineage cells.373,374 The soluble form of OGP increases cell proliferation in a biphasic-dependent manner, which is indicative of an autocrine/paracrine mode of regulation.372 The active 105 portion of OGP, the OGP(10–14) region (YGFGG), is cleaved from the peptide and binds to the OGP receptor which activates the MAP kinase, Src and RhoA pathways.374 In animal studies, OGP and OGP(10–14) promoted increased bone density and stimulated fracture healing, suggesting a potential use in bone-tissue-engineering applications.374,378,392

In previous and our own studies, the presence of suboptimal and optimal levels of exogenous soluble OGP triggers a 2- to 5-fold increase in endogenous OGP expressed from osteoblast lineage cells.374 The MAPK-based signaling initiates and sustains significant soluble growth factor by-product and the secondary paracrine effect drives the osteoinductive signaling over an extended period. Interestingly, the tethered form of

OGP(10–14), which could not be enzymatically cleaved from the surface, retained potent signaling capacity long after the cleaved fragments had been released. Figure 3.4 clearly shows that the OGP retains the ability to accelerate MC3T3-E1 osteoblast cell proliferation when utilized as a crosslinker and that the 0.5–1.0% is within the relevant concentration regime first identified by Greenberg et al.372,393 If the OGP crosslinked materials perform well in higher-order animal models, which are being explored currently, the approach will be transformative to future designs of high-strength polymeric materials.

We acknowledge that during the peptide crosslinking process, radical formation in situ is undoubtedly leading to chain scission in the polymer backbone. While one would generally expect rapid loss of mechanical properties, this does not happen in this instance due to the low mole fraction of crosslinker and the high initial molecular mass. If the peptide crosslinker approaches 5 mol.% or the initial molecular mass is less than 50 kDa, one does not see any increase in mechanical strength upon crosslinking. The OGP-based crosslinking is far from an optimized process. However, our mechanical and biological 106 data show this to be a viable process. Our next-generation designs include methods to crosslink individual chains while avoiding chain scission through the radical processes.

For our in vitro and in vivo experiments we chose to fabricate solid plug materials.

The solid plugs yield an implant for mechanical testing and provide information as to the rate and mechanism of in vivo degradation. In addition, the subcutaneous experiments provide an early indication that there is minimal potential for inflammatory processes that may be initiated in vivo. Future application-specific models for tissue engineering will include porous networks to allow cell infiltration and tissue in growth. Preliminary in vitro and in vivo studies on the biological reaction to poly(ester urea) polymers based on 1-LEU-

6 and 1-PHE-6 tethered with OGP showed that the materials were biocompatible according to the simple assays which we have described. The observed lack of cytotoxicity and enhanced proliferation activity showed OGP to be a bioactive crosslinking agent.

Interestingly, the lack of inflammation, despite ester-based hydrolysis, suggests that the polymers are self-buffering. The urea linkage is in close proximity to the ester linkage and the relative pKa suggest that the cleaved bond would not lead to an acceleration of degradation locally due to the acid neutralization effect. The in vivo data suggest the potential of these materials for developing tissue engineering scaffolds with controlled degradation and replacement by bone.

Based on the success of the crosslinked PEU material in vivo, we are actively pursuing crosslinking strategies that do not require chain scission. The peptide crosslinked

PEU materials allow us to address the three basic tenets of bone regrowth for treatment of bone defects: incorporation of necessary growth factors for osteoinduction (e.g. bone morphogenetic proteins), suitable scaffolding for generating osteoconduction and 107 stimulating/directing progenitor stem cells towards osteogenesis.394 In addition, full-length recombinant proteins generally denature when exposed to synthetic polymers and the long- term impact of this phenomenon is currently being debated.395

3.6 Conclusion

Our data show that the homopolymer poly(1-PHE-6) PEU materials have mechanical properties (elastic modulus = 6.1 GPa) that are nearly twice that of PLA

(elastic modulus = 2.9 GPa). PLLA is generally acknowledged to be insufficient for stand- alone, load-bearing materials. Yet, to the best of our knowledge, these are the first degradable polymeric constructs with moduli in the range of 6.0 GPa, which is substantially higher than other, commercially available and widely used polyesters. In addition, these data on our innovative process for developing PEUs including peptide crosslinking show that the mechanical properties increase proportionally with 0.5% and

1.0% concentrations of OGP. Further refinements offered by copolymerization of mixed monomers, and the extent of crosslinking therein, will provide significant advantages over existing materials with regard to optimizing the degradation profiles, especially if these materials are carried forward for more advanced surgical procedures in bone defect repair.

Our in vivo data shows the general biocompatibility and resorption of the peptide- crosslinked PEU polymers in an in vivo rat subcutaneous experiment. While subcutaneous implants are not a clinical orthopedic model, it is an important step to demonstrate the utility of this new class of biomaterials. The peptide-based crosslinkers have been shown to promote integration between the polymer construct and host. The initial results will guide further experiments to optimize both the mole fraction and ratio of peptides for 108 mechanical strength and stiffness and optimize the porosity (pore size and volume fraction) for osteoinduction. When optimized, translation into higher-order animal models, including segmental defects in dogs or sheep, and eventually in humans, should be relatively straightforward due to the biocompatibility of the polymer, related to many of the peptides already used clinically, and the limitations of current treatment methods.

3.7 Acknowledgements

The authors thank Professor Robert A. Weiss from the Department of Polymer

Engineering for assistance with the rheology analysis, John Elias, PhD for consultation on project goals for orthopedic applications and experimental design parameters, and Rochak

Vig, MS for statistical analysis of the histomorphometric data. The authors gratefully acknowledge financial support from the Austen Bioinnovation Institute in Akron, the

University of Akron Research Foundation and Akron General Medical Center summer fellowship program.

109

CHAPTER IV

POST-ELECTROSPINNING “CLICK” MODIFICATION OF DEGRADABLE AMINO

ACID-BASED POLY(ESTER UREA) NANOFIBERS

Portions of this work have been published previously as

Fei Lin, Jiayi Yu, Wen Tang, Jukuan Zheng, Sibai Xie, and Matthew L. Becker

Macromolecules, 2013, 46 (24), 9515–9525

4.1. Outline

Amino acid-based poly(ester urea)s (PEU) are emerging as a new class of degradable polymers that have shown promise in regenerative medicine applications.

Herein, we report the synthesis of PEUs carrying pendent “clickable” groups on modified tyrosine amino acids. The pendent species include alkyne, azide, alkene, tyrosine-phenol, and ketone groups. PEUs with Mw exceeding to 100k Da were obtained via interfacial polycondensation methods and the concentration of pendent groups was varied using a copolymerization strategy. The incorporation of derivatizable functionalities is demonstrated using 1H NMR and UV-Vis spectroscopy methods. Electrospinning was used to fabricate PEU nanofibers with a diameters ranging from 350 nm to 500 nm. The nanofiber matricies possess mechanical strengths suitable for tissue engineering (Young’s modulus: 300±45 MPa; tensile stress: 8.5±1.2 MPa). A series of bioactive peptides and 110 fluorescent molecules were conjugated to the surface of the nanofibers following electrospinning using bio-orthogonal reactions in aqueous media. The ability to derivatize

PEUs with biological molecules using translationally relevant chemical methods will significantly expand their use in vitro and in vivo.

4.2 Introduction

Over the last few decades, biodegradable polymers have been applied to a number of applications in drug delivery and regenerative medicine.5,364,396,397 While naturally derived degradable polymers have distinct bioactivity and cell binding properties, they are difficult to isolate, derivatize and purify, and also have the potential for immunogenic responses.398,399 Synthetic degradable polymers have a number of advantages over natural materials, especially the chemical diversity of monomers that can be utilized to tailor the chemical, mechanical and degradation properties.400-403 There are a number of biodegradable polymers including poly(ε-caprolactone) (PCL), poly(lactic acid) (PLA), poly(glycolide) (PGA), and copolymers thereof that used clinically and while their properties in vitro and in vivo are largely understood, their range of physical and chemical properties is somewhat limited.404,405 Efforts have been made to diversify the pool of synthetic polymers to meet design criteria for more advanced applications. Currently a wide range of polymers including polyurethanes406,407, polycarbonates173,178 and poly(α- amino acids)408,409 have been utilized in biomedical and regenerative applications.

We have previously demonstrated the synthesis of L-phenylalanine-based poly(ester urea)s (PEU).410 We have also shown that when PEUs are crosslinked with osteogenic growth peptide (OGP)372,373 the mechanical properties increase significantly 111 while the OGP retains its bioactivity.410 The use of specific amino acids influences the physical and chemical properties of the resulting polymers and also provides significant variability in chemical structures. Amino acid-based PEUs are semi-crystalline and thermal or solution-based processing methods offer non-chemical routes to tune the mechanical properties, chemical stability and degradation rates. To further enhance the biological interactions, we and others are focused on attaching bioactive molecules to the polymers.216,241,411,412 However, attaching bioactive molecules, peptides or proteins prior to processing is generally difficult and the biological activity is often lost due to denaturation or degradation.

The incorporation of reactive sites into biodegradable polymers provides a platform for the conjugation of biological cues.216,261,413,414 To meet the challenge of regiospecific biomolecular derivation, it is particularly attractive to employ orthogonal “click” chemistry methods. The “click” concept was introduced by Sharpless,189 and currently it represents a number of reactions, which are robust, selective, efficient, and high yielding.189,190,415 The catalog of “click” reactions includes copper (I) catalyzed azide-alkyne cycloaddition

(CuAAC)189,416, thiol-ene radical addition250,251, oxime ligation305,315, Michael- addition241,252, etc. They are widely utilized for protein and DNA conjugation, cell modification, surface functionalization, and in vivo signaling.242,417-419 Other reactions, such as thiol-maleimide and NHS-ester coupling, are also widely used in the field of material and life science.420,421 Recently, “click” chemistry has also been used for protein and peptide conjugation to tyrosine-based phenol residues using both Mannich-type addition422,423 and “ene-type” addition424-426 reactions. Compared to the large abundance of lysine residues typically found in proteins, the tyrosine content is much lower. In 112 addition, unlike the disulfide linkages and bridges enabled by cysteine residues in close proximity, tyrosine is available for chemical modification without additional protection/deprotection steps.

Polymeric nanofiber matrices are routinely used in biomedical applications, due to their morphological and structural similarities to the natural extra-cellular matrix

(ECM).427,428 Nanofibers are obtained via electrospinning of polymer solutions or from the melt. The physical and dimensional properties of fiber matrices can be tuned precisely.

Fiber diameter, alignment, surface-to-volume ratio, and porosity can be controlled in the electrospinning process.429-431 Recently we have shown that nanofiber matrices surface functionalized with YIGSR peptides derived from laminin significantly enhance the neural differentiation of embryonic stem cells.53 This approach, which was shown to be effective in vitro, may enhance the interactions in an in vivo environment.432-434 The combination of

“click” reactions and electrospinning provides a versatile platform to prepare ECM-like materials with biological functionalities via post-fabrication modification of nanofiber surfaces.242,435,436 In order to explore the feasibility of the possible chemical functionalities,

L-phenylalanine and L-tyrosine-based PEU copolymers bearing various “clickable” pendent groups, including alkyne, azide, alkene, tyrosine-phenol, and ketone were synthesized. Model reactions with peptides or fluorescent molecules demonstrate the presence of “clickable” sites on the surfaces of electrospun nanofibers and that they are available for post-fabrication functionalization.

113

4.3 Experimental Section

4.3.1 Materials and Methods

All chemicals and reagents were purchased from Sigma or Fisher Scientific, and used as received unless noted otherwise. Chromeo 488 azide and Alexa Fluor 568 hydrazide were purchased from Life Technologies (Carlsbad, CA). Rhodamine (TRITC)- conjugated streptavidin was purchased from Fisher Scientific (Pittsburgh, PA). Chloroform was distilled after drying overnight over CaH2.

NMR spectra were obtained using a Varian NMRS 300. Chemical shifts are reported in ppm (δ), and referenced to the chemical shifts of the residual solvent resonances

1 13 TM (DMSO-d6 H 2.50 ppm, C 39.50 ppm). UV-Vis spectra were recorded using a Synergy

Mx spectrophotometer (Biotek Inc.). Fluorescence microscopy images were recorded on an OLYMPUS IX 81 fluorescence microscope and are unaltered. The morphology of the nanofiber was characterized using field-emission scanning electron microscopy (SEM)

(JSM-7401F, JEOL, Peabody, MA). The acceleration voltage for SEM imaging was 5.00 kV. Mechanical properties were recorded using an Instron 3365 under universal tensile testing conditions. Size exclusion chromatographic analyses (SEC) were performed using a TOSOH HLC-8320 SEC. Dimethylformamide (DMF) with 0.01M LiBr was used as the eluent with a flow rate of 0.8 mL/min at 50 oC. The molecular mass and mass distributions were calculated from polystyrene standards. Electrospray ionization (ESI) was performed using a HCT Ultra II quadripole ion trap mass spectrometer (Bruker Daltonics, Billerica,

MA) equipped with an electrospray ionization source.

114

4.3.2 General Procedures of Carbodiimide Coupling Esterification

The reagent diol (1 eq.), acid (1.2 eq. per hydroxyl unit), and 4-(N,N- dimethylamino)pyridinium-4-toluenesulfonate (DPTS, 0.2 eq. per hydroxyl unit) were dissolved using a minimum amount of DMF. After set in ice bath for 10 min, 1, 3- diisopropyl cabodiimide (DIPC, 1.5 eq. per hydroxyl unit) was added via syringe. The white precipitate was observed in minutes. The reaction was continuously stirred for 24 h.

After filtration to remove the solid, the collected solution was concentrated for further purification.

4.3.3 General Procedures for tert-Butyloxycarbonyl (Boc) Deprotection

Boc-protected precursors were dissolved in HCl/dioxane (4M) solutions under nitrogen atmosphere. The reaction was continuously stirred for 24 h under nitrogen environment, followed by lyophilization to remove the organic solvent. The solid residue was further washed with diethyl ether twice and dried in vacuum, affording the desired product.

4.3.4 Synthesis of Monomer 4.1 (M1) for PEU-1

Scheme 4.1. The synthetic route of monomer 4.1 (M1) for PEU-1. 115

The synthesis of 4.1 (M1) was described in Scheme 4.1, as reported elsewhere.410

1 + H NMR (300 MHz, DMSO-d6): δ= 8.45 (br, 6H, NH3-), 7.00-7.53 (m, 18H, aromatic),

+ 4.30 (t, 2H, NH3CHCOO-), 4.00 (t, 4H, -COOCH2CH2-), 2.90-3.35 (m, 4H, -CHCH2-Ar),

2.29 (s, 6H, CH3Ar-), 1.25-1.50 (br, 4H, -COOCH2CH2CH2-), 0.95-1.15 (br, 4H, -

13 COOCH2CH2CH2-); C NMR (75 MHz, DMSO-d6): δ= 169.1, 145.0, 138.2, 134.7, 129.4,

128.6, 128.3, 127.3, 125.6, 65.5, 53.4, 36.2, 27.7, 24.7, 20.8. (Appendix Figure 1 and 2)

4.3.5 Synthesis of Monomer 4.2 (M2) for PEU-2

Scheme 4.2. Chemical structure of monomer 4.2 (M2) for PEU-2.

The chemical structure of M2 was described in Scheme 4.2. The synthesis of M2 was similar to the synthesis process of M1, except using o-propargyltyrosine437 instead of

L-phenylalanine. M2 was purified by recrystallization with a mixture of water and ethanol

(5:1, v/v) four times. The product came out as a white powder (yield 75%). 1H NMR (300

+ MHz, DMSO-d6): δ= 8.41 (s, 6H, NH3-), 7.50 (d, 4H, aromatic), 7.10-7.20 (m, 8H,

+ aromatic), 6.95 (d, 4H, aromatic), 4.76 (d, 4H, -OCH2C≡CH), 4.23 (t, 2H, NH3CHCOO-),

4.04 (t, 4H, -COOCH2CH2-), 3.55 (t, 2H, -OCH2C≡CH), 2.90-3.25 (m, 4H, -CHCH2-Ar), 116

2.29 (s, 6H, CH3Ar-), 1.25-1.50 (br, 4H, -COOCH2CH2CH2-), 0.95-1.15 (br, 4H, -

13 COOCH2CH2CH2-); C NMR (75 MHz, DMSO-d6): δ= 169.1, 156.5, 145.2, 138.0, 130.5,

128.2, 127.1, 125.5, 114.9, 79.3, 78.3, 65.5, 55.4, 53.3, 35.3, 27.7, 24.8, 20.8. (Appendix

Figure 3 and 4)

4.3.6 Synthesis of Monomer 4.6(M4) for PEU-4

The synthesis route of monomer 4.6 (M4) was described in Scheme 4.3.

Intermediate 4.3 and 4.4 were prepared following the previous reference.438 Intermediate

4.5 was synthesized through the general esterification process. Monomer 4.6 (M4) was obtained by the general procedure of Boc deprotection.

1 Intermediate 4.3: H NMR (300 MHz, CDCl3): δ= 6.95-7.15 (m, 2H), 6.75-6.92 (m,

2H), 5.75-5.95 (m, 1H), 4.85-5.15 (m, 3H), 4.45-4.65 (m, 1H), 3.95 (t, 2H), 3.72 (s, 3H),

2.85-3.15 (m, 2H), 2.18-2.30 (m, 2H), 1.80-1.95 (m, 2H), 1.42 (s, 9H). (Appendix Figure

1 Intermediate 4.4: H NMR (300 MHz, CDCl3): δ= 8.95-10.10 (br, 1H), δ6.95-7.15

(m, 2H), 6.75-6.92 (m, 2H), 5.75-5.95 (m, 1H), 4.85-5.15 (m, 3H), 4.45-4.65 (m, 1H), 3.95

(t, 2H), 2.85-3.15 (m, 2H), 2.18-2.30 (m, 2H), 1.80-1.95 (m, 2H), 1.42 (s, 9H). (Appendix

Figure 6)

1 Intermediate 4.5: H NMR (300 MHz, DMSO-d6): δ= 8.31 (s, 2H), 7.15-7.25 (m,

2H), 7.05-7.15 (m, 4H), 6.75-6.90 (m, 4H), 5.55-5.95 (m, 2H), 4.90-5.10 (m, 4H), 3.80-

4.15 (m, 10H), 2.65-2.95 (m, 4H), 2.10-2.23 (m, 4H), 1.68-1.85 (m, 4H), 1.10-1.60 (m, 26

H). (Appendix Figure 7)

117

1 + PEU-4 monomer 4.6(M4): H NMR (300 MHz, DMSO-d6): δ= 8.73 (s, 6H, NH3-),

7.12 (d, 4H, aromatic), 6.84 (d, 4H, aromatic), 5.55-5.95 (m, 2H, -CH2CH=CH2), 4.90-

+ 5.10 (m, 4H, -CH2CH=CH2), 3.80-4.15 (m, 10H, NH3CHCOO-, -COOCH2CH2-, -

ArOCH2CH2-), 2.85-3.20 (m, 4H, -CHCH2-Ar), 2.10-2.23 (m, 4H, -

OCH2CH2CH2CH=CH2), 1.68-1.85 (m, 4H, -OCH2CH2CH2CH=CH2), 1.25-1.50 (br, 4H,

13 -COOCH2CH2CH2-), 0.95-1.15 (br, 4H, -COOCH2CH2CH2-); C NMR (75 MHz,

DMSO-d6): δ= 169.5, 158.3, 138.4, 130.9, 126.9, 115.7, 114.9, 67.2, 65.8, 53.8, 35.7, 30.1,

28.3, 28.1, 25.2. (Appendix Figure 8 and 9)

Scheme 4.3. The synthetic route of PEU-4 monomer 4.6 (M4).

118

4.3.7 Synthesis of Monomer 4.10 (M3) for PEU-3

The synthesis of monomer for PEU-3 was similar to that of 4.6 (M4), except using

3-azidopropyl 4-methylbenzenesulfonate439 instead of 5-Bromo-1-pentene. It was described in Scheme 4.4.

Scheme 4.4. The synthetic route of PEU-3 monomer 3.10 (M3).

1 Intermediate 4.7: H NMR (300 MHz, CDCl3) : δ= 6.95-7.15 (m, 2H), 6.75-6.90

(m, 2H), 4.96 (d, 1H), 4.45-4.65 (m, 1H), 4.03 (t, 2H), 3.52 (t, 2H), 2.95-3.12 (m, 2H),

1.97-2.10 (2H), 1.43 (s, 9H). (Appendix Figure 10)

1 Intermediate 4.8: H NMR (300 MHz, CDCl3) : δ= 8.50-10.00 (b, 1H), δ= 7.02-

7.15 (m, 2H), 6.75-6.90 (m, 2H), 4.96 (d, 1H), 4.45-4.65 (m, 1H), 4.03 (t, 2H), 3.52 (t, 2H),

2.95-3.25 (m, 2H), 1.97-2.10 (2H), 1.43 (s, 9H). (Appendix Figure 11)

119

1 Intermediate 4.9: H NMR (300 MHz, CDCl3): δ= 6.95-7.15 (m, 4H), 6.75-6.90 (m,

4H), 4.96 (d, 2H), 4.45-4.65 (m, 2H), 3.85-4.23 (m, 8H), 3.52 (t, 4H), 2.85-3.15 (m, 4H),

1.97-2.10 (4H), 1.20-1.75 (m, 26H). (Appendix Figure 12)

1 PEU-3 monomer 4.10 (M3): H NMR (300 MHz, DMSO-d6): δ= 8.73 (s, 6H,

+ + NH3-), 7.12 (d, 4H, aromatic), 6.84 (d, 4H, aromatic), , 4.17 (t, 2H, NH3CHCOO-), 3.80-

4.10 (m, 8H, -CH2CH2N3, -COOCH2CH2-), 3.50 (t, 4H, -CH2CH2N3), 2.85-3.20 (m, 4H, -

CHCH2-Ar), 1.90-2.05 (m, 4H, -OCH2CH2CH2N3), 1.25-1.50 (br, 4H, -

13 COOCH2CH2CH2-), 0.95-1.15 (br, 4H, -COOCH2CH2CH2-); C NMR (75 MHz, DMSO- d6): δ= 169.0, 157.6, 130.4, 126.5, 114.5, 65.3, 64.5, 53.3, 47.6, 35.2, 28.1, 27.6, 24.7.

(Appendix Figure 13 and 14)

4.3.8 Synthesis of Monomer 4.12 (M5) for PEU-5

Scheme 4.5. The synthetic route of PEU-5 monomer 4.12 (M5).

The monomer of PEU-5 (M5) was synthesized through the general esterification between Boc-O-benzyl-L-tyrosine and 1, 6-hexanediol, followed by Boc deprotection

(Scheme 4.5).

120

1 Intermediate 4.11: H NMR (300 MHz, CDCl3): δ= 7.25-7.50 (m, 10H), 7.00-7.10

(m, 4H), 6.85-6.95 (m, 4H), 4.85-5.15 (m, 6H), 4.4-0-4.60 (m, 2H), 3.95-4.20 (m, 4H),

2.85-3.15 (m, 4H), 1.15-1.70 (m, 26H). (Appendix Figure 15)

1 PEU-5 monomer 4.12 (M5): H NMR (300 MHz, DMSO-d6): δ= 8.71 (br, 6H,

+ NH3-), 7.25-7.50 (m, 10H, benzyl unit aromatic), 7.10-7.20 (m, 4H, tyrosine unit aromatic), 6.85-7.00 (m, 4H, tyrosine unit aromatic), 5.06 (s, 4H, -Ar-OCH2-Ar), 4.07-4.20

+ (m, 2H, NH3CHCOO-), 4.01 (t, 4H, -COOCH2CH2-), 2.90-3.25 (m, 4H, -CHCH2-Ar),

13 1.30-1.55 (m, 4H, -COOCH2CH2CH2-), 1.10-1.25 (m, 4H, -COOCH2CH2CH2-); C NMR

(75 MHz, DMSO-d6): δ= 169.6, 158.0, 137.5, 131.0, 128.9, 128.0, 127.2, 115.2, 69.6, 66.8,

53.8, 35.6, 29.0, 28.3, 25.6. (Appendix Figure 16 and 17)

4.3.9 Synthesis of O-(prop-2-yn-1-yl)hydroxylamine (4.14)

Scheme 4.6: The synthetic route of compound 4.14.

The synthesis of O-(prop-2-yn-1-yl)hydroxylamine (compound 4.14) was described in Scheme 4.6 as reported elsewhere.241

1 Intermediate 4.13: H NMR (300M Hz, CDCl3) 7.73-7.95 (m, 4H, aromatic), 4.89

(d, 2H, -OCH2C≡CH), δ= 2.60 (t, 1H, -OCH2C≡CH). (Appendix Figure 18)

121

1 O-(prop-2-yn-1-yl)hydroxylamine (4.14): H NMR (300M Hz, DMSO-d6): δ=

+ 13 11.10 (br, 3H, NH3O-), 4.75 (m, 2H, -OCH2C≡CH), 3.87(m, 1H, -OCH2C≡CH); C

NMR (75 MHz, DMSO-d6): δ= 81.1, 76.5, 61.7. (Appendix Figure 19 and 20)

4.3.10 O-(pent-4-en-1-yl)hydroxylamine (4.16)

Scheme 4.7. The synthetic route of compound 4.16.

The synthetic process of compound 4.16 was similar to 4.14, except using 5-

Bromo-1-pentene instead of propargyl bromide, as shown in Scheme 4.7.

1 Intermediate 4.15: H NMR (300 MHz, CDCl3): δ=7.70-7.90 (m, 4H, aromatic H),

5.75-6.60 (m, 1H, CH2=CHCH2-), 4.95-5.35 (m, 2H, CH2=CHCH2-), 4.23 (t, 2H, -

CH2CH2O-), 2.25-2.35 (m, 2H, CH2=CHCH2CH2CH2O-), 1.83-1.97 (m, 2H,

CH2=CHCH2CH2CH2O-). (Appendix Figure 21)

1 O-(pent-4-en-1-yl)hydroxylamine (4.16): H NMR (300 MHz, DMSO-d6): δ=11.11

+ (br, 3H, NH3O-), 5.75-6.60 (m, 1H, CH2=CHCH2-), 4.95-5.35 (m, 2H, CH2=CHCH2-),

4.02 (t, 2H, -CH2CH2O-), 1.95-2.15 (m, 2H, CH2=CHCH2CH2CH2O-), 1.55-1.75 (m, 2H,

13 CH2=CHCH2CH2CH2O-); C NMR (75 MHz, DMSO-d6): δ=136.5, 115.4, 73.4, 29.2,

26.4. (Appendix Figure 22 and 23)

122

4.3.11 Synthesis of Alkyne Derived Cyclic Diazodicarboxamide (4.21)424,425

Preparation of cyclic diazodicarboxamide (compound 4.21) was involved multi- step organic synthesis, as shown in Scheme 4.8.

Intermediate 4.17: N-Boc-4-hydroxyaniline (2.0g, 10 mmol, 1 eq.) and K2CO3

(2.0g, 15 mmol, 1.5 eq.) were added into 50 mL DMF, followed by the addition of propargyl bromide (2.3g, 15 mmol, 1.5 eq.) drop by drop. The resulted suspension was allowed to stir at room temperature for 24h. The solid residue was removed by filtration.

The collected organic solution was concentrated for further column chromatography purification on silica gel. The product was afforded as a whit solid. (2.1 g, yield 85%). 1H

NMR (300M Hz, DMSO-d6): δ= 9.15 (s, 1H, -NH-), 7.25-7.45 (m, 2H, aromatic), 6.80-

6.92 (m, 2H, aromatic), 4.70 (d, 2H, -OCH2C≡CH), 3.52 (t, 1H, -OCH2C≡CH), 1.44 (s,

9H, (CH3)3CCO-). (Appendix Figure 24)

Scheme 4.8. The synthetic route of alkyne derived cyclic diazodicarboxamide (4.21)

123

Intermediate 4.18: This compound was prepared from 4.17 after the deprotection

1 + of Boc groups. H NMR (300M Hz, DMSO-d6): δ= 10.36 (s, 3H, NH3-), 7.30-7.40 (m,

2H, aromatic), 7.00-7.13 (m, 2H, aromatic), 4.81 (d, 2H, -OCH2C≡CH), 3.60 (t, 1H, -

OCH2C≡CH). (Appendix Figure 25)

Intermediate 4.19: In a 100 mL flask, ethyl hydrazinecarboxylate (0.5g, 5 mmol, 1. eq.) and 1,1'-carbonyldiimidazole (0.89g, 5.5 mol, 1.1 eq.) were dissolved in 30 mL anhydrous CHCl3 and stirred for 2h at room temperature. Intermediate 4.18 (0.9g, 5 mmol,

1 eq.) and triethylamine (1.4 ml, 10 mmol, 2eq.) were added into the reaction mixture and stirred overnight. After diluted with 50 mL CHCl3, the organic solution was washed with

1 M HCl solution three times followed by water washing twice. The organic layer was collected and dried with MgSO4. After removal of all the organic solvent, the product came out as a light brown solid. It was used directly for the next step without further purification

1 (1.1g, yield 78%). H NMR (300M Hz, DMSO-d6): δ= 8.86 (s, 1H, -ArNHCONHNHCO-),

8.54 (s, 1H, -ArNHCONHNHCO-), 7.90 (s, 1H, -ArNHCONHNHCO-), 7.30-7.40 (m, 2H, aromatic), 6.83-6.95 (m, 2H, aromatic), 4.72 (d, 2H, -OCH2C≡CH), 4.00-4.12 (m, 2H, -

NHCOOCH2CH3), 3.51 (t, 1H, -OCH2C≡CH), 1.19 (t, 3H, -NHCOOCH2CH3). (Appendix

Figure 26)

Intermediate 4.20: Compound 4.19 (1.0g, 3.6 mmol, 1 eq.) was dissolved in 30 mL methanol, followed by the addition of K2CO3 (1.0g, 7.2 mmol, eq.) This suspension was allowed to reflux overnight. The reaction mixture was acidified to the pH of 2 with 12 N

HCl. Solid residue was filtered off. After the removal of solvent, the solid residue was purified using column chromatography on silica gel with ethyl acetate as elute. The product

1 came out as white solid. (0.38 g, yield 45%). H NMR (300M Hz, DMSO-d6): δ= 10.36 (s, 124

2H, -NHNH-), 7.30-7.43 (m, 2H, aromatic), 7.03-7.15 (m, 2H, aromatic), 4.83 (d, 2H, -

13 OCH2C≡CH), 3.57 (t, 1H, -OCH2C≡CH); C NMR (75 MHz, DMSO-d6): δ= 156.8, 154.1,

128.0, 125.6, 115.5, 79.5, 78.8, 56.1. (Appendix Figure 27 and 28)

Alkyne derived cyclic diazodicarboxamide (4.21): In a 20 mL vial, compound 4.20

(23 mg, 10 mmol, 1eq.) and pyridine (8 uL, 10 mmol, 1 eq.) was dissolve in 3 mL CH3CN, followed by the addition of N-bromosuccinimide (17.8 mg, 10 mmol, 1 eq.). The reaction mixture was allowed to stir at room temperature for 5 min, and directly used for the next step without further purification. In all experiments, compound 4.21 was freshly made for use.

4.3.12 General Procedures of Interfacial Polymerization

The synthesis route of all polymers was shown in Scheme 4.9. PEU-1, PEU-2,

PEU-3, PEU-4 and PEU-5 were synthesized using the following procedure. Monomer (1 eq.) and sodium carbonate (3.5 eq.) were dissolved in 500 mL of water (0.1 M for monomer) in a 4-neck 2 L round bottom flask equipped with mechanical stirring. The cloudy solution was placed in a 35 oC water bath. After the temperature of the reaction solution reached 35 oC, it was allowed to stir for an additional 30 min. The water bath was then removed and replaced with a brine-ice bath. When the reaction formulation temperature dropped to 0 °C, a solution of triphosgene (0.6 M in chloroform, 1.1 eq.) was added within one minute with vigorous stirring. After 30 min of stirring, the cooling bath was removed. An additional aliquot of triphosgene solution (0.6 M, 0.1 eq.) was added drop-wise during over an additional 30 min period. The two phases were then separated. The organic phase containing the polymers was collected and washed with water twice. The collected organic 125 phase was precipitated slowly into 3 L of hot water using mechanical stirring. After cooling to room temperature, pure polymer was obtained after filtration and drying.

PEU-1 (L-phenylalanine based PEU without functional groups): 1H NMR (300

MHz, DMSO-d6): δ= 7.00-7.40 (m, 10H, aromatic H), 6.40-6.55 (d, 2H, -NH-), 4.25-4.45

(m, 2H, -NHCHCOO-), 3.94 (t, 4H, -CHCOOCH2CH2-), 2.75-3.05 (m, 4H, -CHCH2Ar),

1.30-1.60 (m, 4H, -COOCH2CH2CH2-), 1.00-1.30 (m, 4H, -COOCH2CH2CH2-).

Scheme 4.9. The synthetic route for PEUs using interfacial polycondensation methods.

PEU (1) no chemical functionalities; PEUs possessing functional alkyne (2), azide (3), alkene (4), benzyl-protected tyrosine (5), unprotected tyrosine (6), and ketone (7) can be synthesized using a copolymerization approach that is highly versatile.

As to PEU-2, PEU-3, PEU-4, PEU-5, PEU-6, and PEU-7, NMR shifts of the L- phenylalanine units are identical as those in PEU-1. The chemical shifts of functionalized tyrosine units are described below.

126

1 PEU-2 (Alkyne-tyrosine units 2.5%): H NMR (300 MHz, DMSO-d6): δ= 7.00-

7.10 (m, 4H, aromatic), 6.82-6.90 (m, 4H, aromatic), 4.71 (d, 4H, -OCH2C≡CH), 3.52 (t,

2H, -OCH2C≡CH).

1 PEU-3 (Azide-tyrosine units 5%): H NMR (300 MHz, DMSO-d6): δ= 7.00-7.10

(m, 4H, aromatic), 6.82-6.90 (m, 4H, aromatic), 4.12 (t, 4H, -OCH2CH2CH2N3), 3.44 (t,

-1 4H, -OCH2CH2CH2N3), 1.93 (m, 4H, -OCH2CH2CH2N3). FT-IR (cm ): N3 stretch 2099 cm-1.

1 PEU-4 (Alkene-tyrosine units 5%): H NMR (300 MHz, DMSO-d6): δ= 7.00-7.10

(m, 4H, aromatic), 6.82-6.90 (m, 4H, aromatic), 5.75-5.95 (m, 2H, -CH2CH=CH2), 4.90-

5.10 (m, 4H, -CH2CH=CH2), 2.05-2.25 (m, 4H, -OCH2CH2CH2CH=CH2), 1.65-1.85 (m,

4H, -OCH2CH2CH2CH=CH2).

1 PEU-5 (Benzyl protected tyrosine units 5%): H NMR (300 MHz, DMSO-d6): δ=

7.30-7.50 (shoulder, Bzl unit aromatic H), 7.00-7.10 (m, 4H, tyrosine unit aromatic), 6.82-

6.90 (m, 4H, tyrosine unit aromatic), 5.03 (s, 4H, -ArOCH2-Ar).

4.3.13 Synthesis of Tyrosine-Phenol Derived Polymer (PEU-6)

PEU-5 (2.00 g, tyrosine units 5 mol%) was dissolved in DMF (20 mL), followed by the addition of palladium/carbon (0.20 g, 10 wt% of Pd). The suspension was stirred under hydrogen (60 PSI) at 50 oC for 24 h. The carbon was removed by filtration through

Celite 545. The collected light brown solution was concentrated and precipitated into water.

Pure polymer was obtained as a white solid after filtration and drying (1.75 g, yield 87%).

1 Tyrosine unit H NMR (300 MHz, DMSO-d6): δ= 9.20 (s, 2H, -Ar-OH phenol), 6.85-6.98

(m, 4H, aromatic H), 6.60-6.70 (m, 4H, aromatic H). 127

4.3.14 Synthesis of ketone derived polymer (PEU-7)

PEU-6 (2.00 g) was dissolved in DMF (20 mL), followed by the addition of levulinic acid (1.2 eq. per phenol unit) and 4-(N,N-dimethylamino) pyridinium-4- toluenesulfonate (DPTS, 0.2 eq. per phenol unit). After cooling to 0 oC, 1,3-diisopropyl cabodiimide (DIPC, 1.5 eq. per phenol unit) was added via syringe. The reaction mixture was allowed to warm up to ambient temperature and continuously stirred 24 h. Following precipitation into methanol, filtration and drying, the product PEU-7 was obtained as a

1 white solid (1.70 g, yield 85%). Modified tyrosine unit H NMR (300 MHz, DMSO-d6):

δ= 6.93-7.02 (m, 4H, aromatic), 2.75-2.85 (shoulder, -OOCCH2CH2COCH3), 2.60-2.75 (m,

8H, -OOCCH2CH2COCH3), 2.12 (s, 6H, -OOCCH2CH2COCH3).

4.3.15 Azide-alkyne Huisgen Cycloaddition in Organic Solution

Alkyne-PEU (PEU-2) was used to demonstrate the general nature of the methods.

PEU-2 (2.00 g), 3-azidopropan-1-ol439 (2 eq. per alkyne unit) and PMDETA (20 uL) were dissolved in 20 mL DMF in a 100 mL schlenk flask under a argon atmosphere, followed by the addition of CuIBr (2 mg). Following three cycles of degassing, the reaction mixture

o was allowed to stir at 50 C for 24 h. The copper salt was removed using neutral Al2O3 column chromatography. The collected light brown solution was concentrated and precipitated into water, yielding the desired triazole functionalized polymer as a white solid

(1.68 g, yield 84%).

Modified tyrosine unit of PEU-2 after “click” reaction with 3-azidopropan-1-ol: 1H

NMR (300 MHz, DMSO-d6): δ= 8.16 (s, 2H, triazole), 7.00-7.10 (m, 4H, aromatic), 6.82-

128

6.90 (m, 4H, aromatic), 5.07 (s, 4H, -OCH2-triazole), 4.64 (t, 4H, triazole-

CH2CH2CH2OH), 1.95 (m, 4H, triazole-CH2CH2CH2OH).

Modified tyrosine unit of PEU-3 after “click” reaction with propargyl alcohol: 1H

NMR (300 MHz, DMSO-d6): δ= 7.96 (s, 2H, triazole), 5.12 (t, triazole-CH2OH), 4.48 (m,

-1 8H, -OCH2CH2CH2-triaziol), 2.10-2.35 (m, 4H, -OCH2CH2CH2-triaziol). FT-IR (cm ): N3 stretch is not visible (Appendix Figure 29).

4.3.16 Functionalization of Alkene Polymer (PEU-4) via Thiol-ene Addition

PEU-4 (2.00 g), mercaptopropionic acid (5 eq. per alkene unit) and I-2959 photo initiator (0.05 eq. per alkene unit) were dissolved in 20 mL DMF in a 100 mL beaker. The reaction mixture was irradiated using a hand-held UV lamp (365 nm, intensity 10W/cm2) for 30 min. The polymer solution was precipitated into methanol, yielding the desired thiol- functionalized polymer as a white solid (1.72 g, yield 86%). The modified tyrosine unit of

PEU-4 after thiol-ene addition with mercaptopropionic acid 1H NMR (300 MHz, DMSO- d6): δ= 7.00-7.10 (m, 4H, aromatic), 6.82-6.90 (m, 4H, aromatic), 2.65 (t, 4H, -

CH2CH2SCH2COOH), 1.50-1.75 (shoulder, -OCH2CH2CH2CH2CH2S-).

4.3.17 Functionalization of Ketone Polymer (PEU-7) via Oxime Ligation

PEU-7 (2.00 g) was dissolved in DMF (20 mL). O-(pent-4-en-1-yl)hydroxylamine

(compound 4.16, 1.2 eq. to ketone unit), triethylamine (1.2 eq. to ketone unit) and p- toluenesulfonic acid (5 mg) were added into the polymer solution. After overnight stirring at room temperature, polymer solution was precipitated into methanol, yielding the desired oxime product polymer as a white solid (1.72 g, yield 86%). In the 1H NMR spectra, there 129 were no changes in the L-phenylalanine monomer units. Special details of the L-tyrosine units are listed below. Modified tyrosine unit of PEU-7 after oxime-ligation with O-(pent-

1 4-en-1-yl)hydroxylamine (compound 4.16): H NMR (300 MHz, DMSO-d6): δ= 6.93-7.02

(m, 4H, aromatic), 5.75-5.95 (m, 2H, -CH2CH=CH2), 4.90-5.10 (m, 4H, -CH2CH=CH2), -

OOCCH2CH2COCH3), 1.97-2.12 (m, 4H, -OCH2CH2CH2CH=CH2), 1.79 (s, 6H, -

OOCCH2CH2C(=N-)CH3), 1.55-1.70 (m, 4H, -OCH2CH2CH2CH=CH2).

4.3.18 Peptide Synthesis

Three peptides bearing “clickable” groups were synthesized by standard solid phase

FMOC methodology using a CEM automatic microwave synthesis instrument. Peptides were cleaved from the resin using standard conditions (45 min, 95% trifluoroacetic acid

(TFA), 2.5% triisopropylsilane (TIPS), 2.5% water (by volume)) and precipitated in cold diethyl ether. The crude solid product was isolated by centrifugation, washed twice with diethyl ether and dialyzed in deionized water (molecular weight cut off (MWCO) 500 g/mol, cellulose, Pierce), followed by lyophilization. For alkyne-GRGDSK(Biotin)-

COOH), 5-hexynoic acid was coupled to N-terminus using standard coupling conditions prior to cleavage. For fluorescein-5(6)-isothiocyanate (FITC)-RGD-thiol (sequence FITC-

GRGDSCS), FITC (Sigma) was coupled to the N-terminus in DMF overnight prior to cleavage.440,441 For Fmoc-RGD-thiol (sequence Fmoc-CGRGDS), after the coupling of a cysteine residue at the N-terminus, the peptide was directly cleaved without the deprotection of Fmoc groups. The chemical structures and purity of the peptides and conjugates were confirmed by mass spectroscopy.

130

ESI: Alkyne-RGD-Biotin calculated 939.1, measured 939.5; MALDI: FITC-RGD- thiol calculated 1070.1, measured 1070.5; Fmoc-RGD-thiol calculated 815.8, measured

816.4. FITC-RGD-thiol exhibited an excitation peak at 492 nm, and an emission peak at

512 nm in PBS buffer. The absorption maxima peak of Fmoc-RGD-thiol was 300 nm in

PBS buffer. (Appendix Figure 35 ESI spectrum of alkyne-RGD-biotin; Appendix Figure

36 MALDI spectrum of FITC-RGD-thiol; Appendix Figure 37 MALDI spectrum of Fmoc-

RGD-thiol)

4.3.19 Nanofiber Fabrication via Electrospinning

PEU was dissolved in hexafluoroisopropanol (HFIP) (10 wt%). The jet was fabricated by pulling a torch-heated PTFE container terminal equipped with stainless metal needle of 25 gauge. The flow rate was controlled at 1 mL/h via a gas pump. A voltage of

12 kV was used and the aluminum foil collector was grounded. The distance between needle and collector was controlled at 25 cm. (Appendix Figure 30 and 31, optical microscope image of PEU nanofibers)

4.3.20 Mechanical Testing of Nanofiber Matrix

The Young’s modulus and tensile properties of a PEU nanofiber matrix were measured using universal tensile testing on an Instron 3365. The gauge length was 30 mm and the crosshead speed was set at 3 mm/min. The specimens were 50 mm long, 4 mm wide and 0.01 mm thick. The Young’s modulus was calculated using the slope of tangent line of the stress-strain curve in small strain region (1%). The results presented are average values for three individual measurements. 131

4.3.21 Surface Modification of Nanofibers via Wet Chemistry

A series of fluorescent probes were used to visualize the nanofiber surface functionalization. Nanofibers were deposited on glass slides for surface wet chemistry in all experiments. A summary of the individual reaction formulations for both the experimental groups and control groups are listed in Table 4.1.

PEU-2: Nanofibers were incubated at ambient temperature in phosphate buffered saline (PBS, x10) solution containing following reagents: CuSO4 (0.1 mg/mL), sodium

washed thoroughly with PBS followed by deionized water. The fibers were dried under the flow of nitrogen and imaged with fluorescence microscope.

PEU-2: In the control groups, there was no copper catalyst in the buffer solution.

Fluorescent reagent (Chromeo 488 azide) would not be attached to the fibers compared to experimental groups.

PEU-3: Following surface conjugation with alkyne-RGD-biotin is similar to the method described in PEU-2. After the surface derivation with alkyne-RGD-biotin, the nanofibers were incubated in PBS (x1) solution containing a rhodamine (TRITC)-

thoroughly with PBS followed by deionized water. The fibers were dried under the flow of nitrogen and imaged with fluorescence microscope.

PEU-3: In the control groups, there was no copper catalyst in the buffer solution.

Alkyne-RGD-biotin would not be attached to the fibers compared to experimental groups.

In the following step, there would be no specific absorption of (TRITC)-conjugated streptavidin to fibers. 132

Table 4.1 Reaction formulation summary of experiment groups and control groups for nanofiber surface modification.

Polymer Reaction Experiment Group Control Group

PEU-2 CuAAC Chromeo 488 azide, CuSO4, Chromeo 488 azide, sodium ascorbate sodium ascorbate PEU-3 CuAAC Two steps Two Steps 1. alkyne-RGD-biotin, 1. alkyne-RGD-biotin,

CuSO4, sodium ascorbate sodium ascorbate 2. (TRITC)-conjugated 2. (TRITC)-conjugated streptavidin streptavidin PEU-4 Thiol-ene FITC-RGD-thiol, I-2959 FITC-RGD-thiol

PEU-6 Ene-type Two steps Two steps 1. Compound 4.21 1. Compound 4.21 2. Chromeo 488 azide, 2. Chromeo 488 azide, CuSO4, sodium ascorbate sodium ascorbate PEU-7 Hydrazine Alexa Fluor 488 hydrazide PEU-1 as fiber substrates, Alexa Fluor 488 hydrazide PEU-7 Aminooxy Two steps Two steps 1. Compound 4.14 1. Compound 4.14 2. Chromeo 488 azide, 2. Chromeo 488 azide,

CuSO4, sodium ascorbate sodium ascorbate

PEU-4: Nanofibers were covered with 0.2 mL of PBS solution containing FITC-

RGD-thiol (0.1 mg/mL) and I-2959 (0.01 mg/mL). The fibers were exposed to UV light for 2 min (365 nm, intensity 10 mW/cm2), and then washed thoroughly with PBS followed

133 by deionized water. The fibers were dried under the flow of nitrogen and imaged with fluorescence microscope. Fmoc-RGD-thiol was coupled to the nanofiber surface also using thiol-ene radical addition. Nanofibers mats were soaked in PBS solution containing Fmoc-

RGD-thiol (0.5 mg/mL) and I-2959 (0.01 mg/mL). Those fibers were exposed to UV light for 2 min (365 nm, intensity 10W/cm2), and then washed thoroughly with PBS followed by deionized water. The fibers were dried under the flow of nitrogen. UV-Vis absorption curve was recorded after dissolving the fibers in HFIP.

PEU-4: In the control groups, there was no photo initiator I-2959, and then no thiol- ene radical addition reaction. FITC-RGD-thiol would not be attached to the fibers through a chemical pathway.

PEU-6: Nanofibers were incubated in freshly made alkyne derived cyclic diazodicarboxamide (Compound 4.21) solution (1 mg/mL, PBS buffer: CH3CN=10:1 by volume) for 30 min. After being thoroughly washed with H2O, the nanofibers were treated with Chromeo 488 azide using CuAAC method described in PEU-2.

PEU-6: In the control groups, there was no copper catalyst in the second step.

Fluorescent reagent (Chromeo 488 azide) would not be attached to the fibers compared to experimental groups.

PEU-7: Nanofibers were incubated in acetic buffer solution (pH 4.5) containing

Alexa Fluor 488 hydrazide for 10 min, followed by thorough wash with H2O. The fibers were dried under the flow of nitrogen and imaged with fluorescence microscope. In two- step derivation, O-(prop-2-yn-1-yl)hydroxylamine (compound 4.14) was bonded to fiber surface via CuAAC, followed by the conjugation with Chromeo 488 azide.

134

PEU-7: There were two control experiments. As to the hydrazine reaction, in the control group the PEU-1 nanofibers without tyrosine units were used as the templates.

Fluorescent reagent (Alexa Fluor 488 hydrazide) would not be attached. As to the oxime condensation, in the control groups, there was no copper catalyst in the second step.

Fluorescent reagent (Chromeo 488 azide) would not be attached to the fibers compared to experimental groups.

4.4 Results and Discussion

4.4.1 Monomer and Polymer Synthesis

Five monomers representing the various derivatization chemistries were synthesized and characterized with 1H NMR and 13C NMR.

Figure 4.1(a) shows the spectra of PEU-1 monomer without functional groups. As shown in Figure 4.1(b), the alkyne-derived monomer was used for the synthesis of PEU-2.

The alkyne CH peak is observed at 3.55 ppm, and the corresponding methylene is located at 4.76 ppm. The aromatic hydrogens of tyrosine have chemical shift 7.06 ppm and 6.86 ppm, which are different from those of phenylalanine located between 7.00 ppm and 7.53 ppm.

In Figure 4.1(c), the PEU-3 monomer containing azide, the peak at 3.50 ppm is assigned to the methylene adjacent to the azide.

In Figure 4.1(d), the PEU-4 monomer shows the characteristic alkene peaks at 5.85 ppm and 5.00 ppm.

135

For the benzyl protected tyrosine monomer in Figure 4.1(e), the benzyl group has chemical shifts in the range of 7.25 ppm to 7.50 ppm, and the singlet at 5.07 ppm is assigned to the benzyl methylene.

Within the limit of 1H NMR detection, there are no mono-amine salts in all monomers, which is very important if high molecular mass polymers are to be obtained in the AA-BB-type step-growth polymerization. All of the monomers are stored as the quantitative amine salt. The free amine is generated in situ using sodium carbonate at the beginning of the polymerization.

136

1 Figure 4.1. H NMR of monomers (DMSO-d6). (a) L-phenylalanine based monomer for

PEU-1 (M1); (b) Alkyne-monomer for PEU-2 (M2): alkyne signals 4.76 and 3.55 ppm; (c)

Azide-monomer for PEU-3 (M3): azide signal 3.50 ppm; (d) Alkene-monomer for PEU-4

(M4): alkene signals 5.85 and 5.00 ppm; (e) Benzyl-protected tyrosine monomer for PEU-

5 (M5): Bzl signal 5.07 ppm. Solvent residues are marked with asterisks.

137

Figure 4.2. Size-exclusion chromatography (SEC) elution curves of the individual PEUs

(RI signal, eluent DMF with 0.01M LiBr, flow rate 0.8 mL/min, temperature 50 oC, polystyrene standards).

The PEUs were obtained via interfacial polycondensation between the respective monomers and triphosgene (Scheme 4.9). The resulting polymers have a weight average molecular mass (Mw) up near 100k Da. The size exclusion chromatography (SEC) elution curves are shown in Figure 4.2 and the corresponding mass data are summarized in Table

4.2. Interfacial polymerization facilitates fast reaction rates and is nearly quantitative yield at low temperature. In addition, precise stoichiometric control between monomers and triphosgene is less critical as interfacial polymerization is kinetically controlled.442 All of the polymers are soluble in polar organic solvents, including N,N-dimethylformamide

(DMF), dimethyl sulfoxide (DMSO), N-methyl-2-pyrrolidone (NMP), and hexafluoroisopropanol (HFIP). The solubility properties provide access to a number of

138 approaches for solution processing methods for PEU, including solution casting, spinning coating and electrospinning.

Table 4.2. Molecular mass data of PEUs. aMass average molecular weight (g/mol); b Distribution of molecular mass (Dm, Mw/Mn) of polymers after precipitation in water, which significantly narrows the molecular mass distribution from what is expected in a step growth polymerization.

PEU-1 PEU-2 PEU-3 PEU-4 PEU-5 PEU-6 PEU-7

a Mw 132,100 117,600 118,900 110,200 96,300 89,400 93,200

b Dm 1.21 1.28 1.21 1.37 1.37 1.50 1.39

Random copolymerization of functionalized tyrosine-based monomers and phenylalanine-based monomers yielded PEU with “clickable” pendent groups (Scheme

4.9). The content of the functional units was controlled by the feed ratio of the respective monomers. Five different “clickable” groups are incorporated separately into the polymers.

PEU-2 (alkyne-PEU) and PEU-3 (azide-PEU) are designed for copper(I) catalyzed azide- alkyne cycloaddition reactions (CuAAC). PEU-4 (alkene-PEU) can be used for thiol-ene radical addition derivation, and PEU-7 (keto-PEU) is able to react with aminooxy and hydrazine groups through a condensation process. PEU-6 with tyrosine phenol residue was synthesized to explore the polymer modification with cyclic diazodicarboxamides via ene- type addition.424,425 In addition, the phenol group of tyrosine is esterification active, which offers an approach for polymer modification using carboxylic acids through carbodiimide

139 coupling. 1H NMR spectroscopy was used to confirm the successful incorporation of the

“clickable” groups. All of the peak assignments are detailed in Figure 4.3. For example,

Figure 4.3(a) describes chemical composition of phenylalanine-based PEU-1 polymer without functional groups. In Figure 4.3(b), the alkyne units show characteristic peaks at

4.71 ppm and 3.52 ppm, and the chemical shifts of aromatic benzyl protecting group on the tyrosine are found at 6.77 ppm and 7.00 ppm. In Figure 4.3(c), the chemical shift of the methylene adjacent to the azide groups is located at 3.46 ppm. The corresponding azide stretch peak is also observed at 2099 cm-1 in the FT-IR spectra of PEU-3 (Appendix Figure

29). In Figure 4.3(d), the existence of alkene groups is verified by multi-split peaks at 5.80 and 5.00 ppm. In Figure 4.3(f) of PEU-6 spectra, the hydroxyl group of tyrosine has a singlet at 9.20 ppm after the deprotection of the benzyl (Bzl) group in PEU-5, and the methylene of Bzl is no longer present. PEU-7 (Keto-PEU) was prepared by the carbodiimide coupling between PEU-6 and levulinic acid. In Figure 4.3(g), the single peak at 4.12 ppm is assigned to the methyl end group adjacent to the ketone, and the peak of phenolic hydroxyl group is not visible after esterification reaction.

The existence of tyrosine units was verified by UV-Vis absorption. The absorption curve of PEU-1 is flat in the 278 nm region, while others have a shoulder, which is the characteristic absorption of tyrosine424 (Figure 4.4). PEU-2 has a smaller absorption at 278 nm than the other functionalized PEUs, since the content of tyrosine units in PEU-2 is 2.5%, and in the other PEUs, it is 5%. The peak at 257 nm is assigned to the π-π* absorption of phenylalanine.424

140

1 Figure 4.3. H NMR spectra of PEU (DMSO-d6): (a) PEU-1, without chemical functionalities; (b) Alkyne PEU-2; (c) Azide PEU-3; (d) Alkene PEU-4; (e) Benzyl protecting PEU-5; (f) Tyrosine PEU-6; (g) Ketone PEU-7. The content of alkyne functionalized units in PEU-2 is 2.5%; the number is 5% in other functionalized PEUs.

Solvent residues are marked with asterisks. 141

Figure 4.4. Normalized UV-Vis Spectra of PEUs. The peak at 257 nm is assigned to phenylalanine absorption; the shoulder at 278 nm is assigned to tyrosine unit absorption.

The content of alkyne derived tyrosine units in PEU-2 is 2.5%; in other PEUs, that number is 5%.

4.4.2 Derivation of “Clickable” Group Functionalized PEU

A series of small molecules were chosen to verify the reactivities of PEU “clickable” groups for post-polymerization modification. 3-azidopropan-1-ol439 was utilized for the derivation of PEU-2 through the CuAAC method. In Figure 4.5(a), the triazole proton signal is observed at 8.16 ppm and the methylene adjacent to the alkyne shifts from 4.72 ppm to 5.07 ppm after the alkyne-azide cycloaddition reaction. Peaks at 4.64 ppm and 1.93 ppm are assigned to the 3-azidopropan-1-ol protons. Propargyl alcohol was coupled to

PEU-3 also through cycloaddition between alkyne and azide groups. In Figure 4.5(b), the

142 triazole proton is found at 7.96 ppm and the methylene protons adjacent to the alkyne are located at 5.11 ppm. The methylene group protons adjacent to the azide shift from 3.44 ppm to 4.48 ppm. The success of propargyl alcohol coupling was furthered demonstrated via FT-IR spectroscopy. The azide stretch peak is not visible after the reaction (Appendix

Figure 29 red line), indicating the complete conversion of the azide groups within the limits of FTIR detection.

Thiol-ene radical additions are also a well-established click process for both polymer and materials synthesis as there is no need for the metal catalyst and it is able to proceed at room temperature.251,443 In PEU-4, the alkene is incorporated as a pendent group for molecular grafting via thiol-ene reaction. Mercaptopropionic acid was utilized to modify PEU-4 via photochemical thiol-ene addition in the presence of Irgacure-2959 photoinitiator. In Figure 4.5(c), after half an hour irradiation with 365 nm UV at room temperature, the signal corresponding to the alkene groups is not present. There is a new peak at 2.65 ppm, which is assigned to the methylene group adjacent to sulfur atom after the reaction.

PEU-6 with unprotected tyrosine units were synthesized for functionalization of polymers at the phenolic side chain with cyclic diazodicarboxamides, through an “ene”- type addition. This reaction emerges as a new type of “click” like reactions for the modification of small molecules, proteins, and peptides in aqueous environment.424,425

Compared to other kinds of reactive groups such as azide and alkyne, phenolic side chains of tyrosine are natural. However, in our system, there was no reaction observed between

PEU-6 and alkyne derived cyclic diazodicarboxamide (compound 4.21) in organic solvent, when triethylamine (TEA) or N, N-diisopropylethylamine (DIPEA) was used as the base 143 in the environment of DMF or DMSO. According to the discussion in previous literatures424, a strong base such as sodium hydride was used to activate the phenol groups for ene-type addition in organic media. We argue that the basicity of TEA (pKa 10.8) and

DIPEA (pKa 11.4) is not sufficient for phenol activation in DMF or DMSO. We didn’t try sodium hydride, since in that kind of harsh condition, PEU or biomolecules would be destroyed. Due to that kind of nature, the application of the ene-type addition would be limited, especially when material modification would be done in organic solvent. However, in the surface modification to PEU-6 nanofibers, the reaction actually happened in the mixture of phosphate buffered saline (PBS buffer, pH 7.4) and acetonitrile (10:1 by volume). This result agrees with the previous findings.424,425 This kind of reaction would have potential application in the polymer or scaffold modification in aqueous environment.

On the other hand, the hydroxyl group of tyrosine units is still available for esterification, which is verified in the synthesis PEU-7. It allows us to functionalize polymers using carbodiimide coupling.

The oxime ligation, a condensation reaction between aldehyde or ketone groups and aminooxy groups, is a particularly attractive, facile and versatile for polymer modification. It occurs readily without major side reactions at room temperature, and the oxime bond is stable in physiological conditions. It is widely used for protein conjugation, polymeric hydrogel fabrication and surface functionalization.241,315,322 O-(pent-4-en-1-yl) hydroxylamine (compound 4.16) was synthesized for derivation of ketone groups in PEU-

7. In Figure 4.5(d), the protons of the methyl groups adjacent to the ketone shift from 2.12 ppm to 1.79 ppm following the imine formation and the characteristic peaks of alkene are observed at 5.00 ppm and 5.80 ppm. 144

Figure 4.5. 1H NMR spectra of PEUs following derivation with small molecule probes

(DMSO-d6). (a) CuAAC between PEU-2 and 3-azidopropan-1-ol; (b) CuAAC product between PEU-3 and propargyl alcohol; (c) Thiol-ene addition product between PEU-4 and mercaptopropionic acid; (d) Oxime ligation product between PEU-7 and o-(pent-4-en-1- yl) hydroxylamine.

145

Those results demonstrate that alkyne, azide, alkene, phenol, and ketone groups in

PEU are available for the post-polymerization functionalization of nanofibers. Collectively these methods provide a toolbox of various functional groups to tune PEU properties via highly efficient bio-orthogonal reactions, which occur without the use of extreme reaction conditions, solvents and reagents. Biomolecules such as peptides and proteins are readily attached to the PEU polymers utilizing CuAAC, thiol-ene addition or oxime ligation techniques. The conjugation to polymers can improve the biomolecules’ stability in vitro and in vivo, and thus enhance their biological effects.444

4.4.3 Nanofiber Processing and Mechanical Testing

Nanofibers of all PEUs were prepared via electrospinning. The fabrication process was stable and continuous, and diameters of the resulting nanofibers have a narrow distribution, varying from 350 nm to 500 nm (Figure 4.6(a) and 4.6(b)). No differences were observed between the PEU homopolymer and the functionalized PEUs in the electrospinning process. The survival of the respective functional groups after electrospinning was confirmed by 1H NMR spectra of the fiber mesh (Appendix Figure 38-

42). The mechanical properties of the PEU fiber meshes were characterized using universal tensile testing. In Figure 4.6(c), the Young’s modulus of fiber matrix is 300 ± 45 MPa in small strain region; the tensile strength is 8.5 ± 1.2 MPa; and the tensile strain was 65 ± 8%

(n=3). In tissue engineering scaffolds, the mechanical strength of the nanofiber meshes are critical to the performance in vivo. 445 As shown in Table 4.3, the mechanical strength of

PEU nanofibers is comparable to that of PLGA when the diameter of fibers is around 500

146 nm446. PEU nanofiber meshes have sufficient mechanical properties, which meet the requirements for some trabecular bones447, cartilage448 and skin449 repairs.

Figure 4.6. SEM micrograph of PEU nanofibers (a) scale bar 1 um and (b) scale bar 100 nm; Nanofibers with a narrow diameter distribution (350 nm to 500 nm) are obtained. (c)

Stress-strain curve of PEU nanofiber matrices, where the Young’s modulus is 300 ± 45

MPa, ultimate tensile stress is 8.5 ± 1.2 MPa, and the ultimate tensile strain is 65 ± 8 %

(n=3). PEU nanofiber matrices have sufficient mechanical properties, which meet the requirements for some trabecular bones447, cartilage448 and skin449 defect repairs.

147

Table 4.3. Comparison of tensile properties of nanofiber matrices: PEU, PCL, PLGA, gelatin, cartilage and skin. Nanofibers of PEU, PCL, PLGA, and gelatin are randomly oriented. PEU fibers’ diameter: 350-500 nm; PCL and gelatin fibers’ diameter: 10-1000 nm;

PLGA fibers’ diameter: 500-800 nm.

PEU PCL450 PLGA446 Gelatin450 Cartilage448 Skin449 Young’s modulus 300 4.98 323 105 130 15-150 (MPa) Ultimate tensile 8.5 2.7 23 2.5 19 5-30 stress (MPa) Ultimate tensile 65 126 96 64 20-120 35-115 strain (%)

4.4.4 Surface Modification of PEU Nanofibers

The availability of “clickable” groups on the nanofiber surfaces was demonstrated using a series of wet chemistry method with fluorescence probes with complimentary reactive groups. In Figure 4.7(a), Chromeo 488 azide was used for PEU-2 nanofiber labeling via CuAAC in PBS buffer solution. The green color present on the nanofibers after reaction indicates the success of surface coupling. In Figure 4.7(b), alkyne-RGD-biotin was first coupled to PEU-3 fiber surface, followed by the incubation in aqueous environment containing rhodamine (TRITC)-conjugated streptavidin451. It yielded the red fluorescent fiber matrices of PEU-3. In Figure 4.7(c), fluorescein-5-isothiocyanate (FITC)- conjugated RGD-thiol was tethered to PEU-4 fibers via thiol-ene radical addition reaction, yielding a green fluorescence. In addition, Fmoc-RGD-thiol was also used for PEU-4 nanofiber modification. Successful conjugation was verified by UV-Vis. Nanofibers after modification exhibit the Fmoc absorption peak at 300 nm452 in HFIP (Appendix Figure 32),

148 indicating the attachment of Fmoc-RGD-thiol to the fiber surface. For the surface functionalization of PEU-6 nanofibers in aqueous media, alkyne derived cyclic diazodicarboxamide (compound 4.21) was first covalently tethered to the fiber surface via ene-type reaction424,425 at the phenolic side chains of tyrosine, and then Chromeo 488 azide was conjugated to the nanofiber surface via CuAAC methods at surface available alkyne sites. The green fluorescence indicates the successful modification of PEU-6 fiber surface via ene-type addition between tyrosine-phenol and cyclic diazodicarboxamide. Figure

4.4(d) shows the green fluorescence of the resulting fibers. As to the PEU-7 fibers, the reactivities of the surface ketone groups with hydrazine and aminooxy were both explored.

Alexa fluor 568 hydrazide fluorescence was directly coupled to fiber surface via hydrazone bond formation and provided a red color (Figure 4.7(e)). For oxime ligation, o-(prop-2-yn-

1-yl)hydroxylamine (compound 4.14) was utilized to conjugate to the fiber surface first, followed by the attachment of Chromeo 488 azide. A green fluorescence is yielded after the two-step derivation (Appendix Figure 33). In all surface modification, controls were conducted to exclude the physical absorption influence. Compared to the experimental groups, the fluorescence in all the controls was much weaker (Figure 4.8(a), 4.8(b) and

Appendix Figure 34), which indicates that those fluorescent probes were actually attached to nanofiber surfaces via chemical reactions. As shown in Figure 4.8(a), the fluorescence picture was generated from the PEU-1 nanofibers after 1h incubation in the buffer solution of Alexa Fluor 488 hydrazide. In Figure 4.8(b), PEU-6 nanofibers were directly immersed into the solution of Chromeo 488 azide without the coupling step with cyclic diazodicarboxamide (compound 4.21). In both figures, the fluorescence was very weak,

149 since fluorescent probes were attached via physical absorption, and the attached amount was much smaller than that due to chemical reactions.

Results above demonstrate that “clickable” groups on the nanofiber surface of

PEUs are chemically available for the bio-orthogonal conjugation of biologically active molecules. In that area of regenerative medicine, one of the major challenges is to design and fabricate degradable scaffolds with highly specific surface functional groups, which are able to promote cell attachment, proliferation and differentiation.453,454 In conventional methods for surface modification, plasma treatment is not efficient for modification of degradable polymers especially within a 3D porous structure455, and most surface hydrolytic methods require extreme conditions456. In this case, a number of bio-orthogonal

“click” reactions are utilized in the surface functionalization of nanofibers that facilitates the covalent attachment of biomolecules. The methods provide a versatile and simple platform to precisely tune the surface properties of ECM like scaffolds in biological environment.

150

Figure 4.7. Fluorescent images of nanofibers that were modified post-electrospinning (a-e) and the corresponding chemical reactions (a1-e1). (a) PEU-2 labeling with Chromeo 488 azide via CuAAC click methods; (a1) Chemical reaction site between PEU-2 and Chromeo

488 azide via CuAAC; (b) PEU-3 modified with alkyne-RGD-biotin via CuAAC methods, followed by labeling with TRITC-conjugated streptavidin; (b1) Chemical reaction site between PEU-3 and alkyne-RGD-biotin via CuAAC; (c) PEU-4 labeling with FITC-RGD- thiol via thiol-ene reaction; (c1) Chemical reaction site between PEU-4 and FITC-RGD- thiol via thiol-ene reaction; (d) PEU-6 modified with alkyne-derivatized diazodicarboxamide (compound 4.21) via ene-type addition, followed by labeling with

Chromeo 488 azide; (d1) Chemical reaction scheme between PEU-6 and alkyne derived diazodicarboxamide; (e) PEU-7 labeling with Alexa fluor 568 hydrazide fluorescence via hydrazone formation; (e1) Chemical reaction site between PEU-7 and Alexa fluor 568 hydrazide fluorescence via hydrazone formation. Fluorescent images of labeled PEU nanofibers (x20, scale bar 50 m).

151

Figure 4.8. Fluorescent images of nanofibers after physical absorption as control groups.

(a) PEU-1 nanofibers were soaked in PBS buffer solution containing Alexa Fluor 488 hydrazide; (b) PEU-6 nanofibers were directly incubated in PBS buffer solution containing

Chromeo 488 azide without the step of “ene-type” modification. (x20, scale bar 50 m)

4.5 Conclusion

The synthesis of highly functional tyrosine-based di-amine monomers and poly(ester urea)s were described. Those monomers were used to prepare high molecular mass poly (ester urea) using interfacial polycondensation methods. “Clickable” pendent groups, including alkyne, azide, alkene, tyrosine-phenol, and ketone, were incorporated to polymers. Post-polymerization functionalization of PEU was accomplished via robust and efficient “click” type reactions. PEU nanofiber matrices were fabricated using electrospinning methods. The measured mechanical properties are in regimes useful for tissue engineering applications and the ability to chemically derivatize the surface of the nanofiber surfaces with peptides and fluorescent probes using bio-orthogonal reaction strategies will enable a number of translationally relevant constructs to be used in biomedical applications.

152

4. 6 Acknowledgement

The authors gratefully acknowledge financial support from The Akron Functional

Materials Center and the National Science Foundation (DMR-BMAT 1105329). The authors also gratefully acknowledge mass spectrometry confirmation of our peptide precursors from Kai Guo and Professor Chrys Wesdemiotis from the Department of

Chemistry at The University of Akron.

153

CHAPTER V

PEPTIDE-FUNCTIONALIZED OXIME HYDROGELS WITH TUNABLE

MECHANICAL PROPERTIES AND GELATION BEHAVIOR

Portions of this work have been published previously as

Fei Lin, Jiayi Yu, Wen Tang, Jukuan Zheng, Adrian Defante, Kai Guo,

Chrys Wesdemiotis, and Matthew L. Becker

Biomacromolecules, 2013, 14 (10), 3749–3758.

5.1 Outline

We demonstrate the formation of polyethylene glycol (PEG) based hydrogels via oxime ligation and the photo-initiated thiol-ene 3D patterning of peptides within the hydrogel matrix post-gelation. The gelation process and final mechanical strength of hydrogels can be tuned using pH and the aniline catalyst concentration. The time scale to reach the gel point and complete gelation can be shortened from hours to seconds using both pH and aniline catalyst, which facilitates the tuning of the storage modulus from 0.3 kPa to over 15 kPa. Azide and alkene functionalized hydrogels were also synthesized and we have shown the post gelation “click” type Husigen alkyne-azide cycloaddition and thiol-ene photo-initiated radical reactions for spatially defined peptide incorporation. These materials are the initial demonstration for translationally relevant hydrogel materials that

154 possess tunable mechanical regimes attractive to soft tissue engineering and possess atom neutral chemistries attractive for post gelation patterning in the presence or absence of cells.

5.2 Introduction

Hydrogels represent a diverse class of polymeric networks that absorb significant amount of water without dissolution. Hydrogels are utilized widely in tissue engineering, drug and protein delivery, cell culture, and sensors due to their unique and highly tunable chemical, mechanical and transport properties.457-461 In biomedical applications, cells and tissues sense and respond to the rigidity of their microenvironments and are able to alter their migration, signaling, differentiation and proliferation responses.462-465 It is essential that hydrogels targeted to tissue engineering scaffolds match the mechanical properties of target tissues.397,466 Therefore, the biomaterials community continually searches for new hydrogel materials where the mechanical properties and network formation times are highly tunable.467-471

Polyethylene glycol (PEG) based hydrogels are utilized frequently due to the lack of cytotoxicity, unique hydration properties and ease of end group modification for crosslinking or network formation.472-474 Although much success can be attributed to the fabrication of PEG hydrogels via conventional photo-initiated free-radical chain polymerization,475-477 trends have been moving toward building hydrogels through more robust, selective, efficient, and tunable chemical reactions.292,478 Various types of “click” chemistry strategies have been employed to prepare hydrogels, including Michael- addition479-481, thiol-ene radical addition482,483, tetrazine–norbornene addition293,484, tetrazine-alkene photo click reaction485, copper(I)-catalyzed alkyne-azide cycloaddition 155

(CuAAC)486-489 or metal-free strain-promoted alkyne-azide cycloadditions

(SPAAC)244,246,490, etc. While these strategies are used as crosslinking methodologies, there have been increasing interests to combine several “click” reactions into one system to create more complex and biologically-patterned microenvironments. Anseth and coworkers demonstrated that photo-initiated thiol-ene chemistry can be employed to spatiotemporally incorporate peptides and proteins in PEGs hydrogels via tetrazine- norbornene addition293, CuAAC489 or SPAAC.246,483 Hawker et al also utilized thiol-ene reactions to immobilize bioactive and diagnostic molecules to hydrogel microarray surfaces294. Dynamic 3D patterning of biological cues were developed via CuAAC where

Cu(I) was produced by photo initiated Cu(II) reduction.487 Shoichet and co-workers further exploited thiol-maleimide Michael-addition for spatial patterning of agarose hydrogels.491,492

Recently, the condensation between aldehyde or ketone groups and aminooxy functional groups, which is also known as an oxime ligation, has grown in popularity as an emerging “click” reaction (Figure 5.2(c)). It occurs readily in aqueous solutions without major side reactions, 295,298 and is used widely for the conjugation and labeling of biological molecules, cell surface modification, scaffolds preparation, and injectable hydrogels315,321,322,493. Though oxime bond is more stable than hydrazone295, it is still found to show degradation behaviors in some biologically relevant environment494. Oxime ligation is particularly interesting due to its pH sensitive reaction kinetics, dependence on catalyst concentration, and varying degrees of reversibility.298,495 For fast reaction speeds, it is necessary to maintain a slightly acidic reaction environment where the proton serves as a catalyst. In neutral or slightly basic conditions, nucleophilic catalysts such as 156 aniline298,301 are necessary to obtain the desired reaction rates and extent of conversion.

Currently, significant efforts have been focused on the development of non-toxic catalysts, like 4-aminophenylalanine302. These unique features inspire us to create hydrogels with tunable properties such as mechanical strength and gelation time by modulating the reaction conditions.

Our experimental results show using oxime ligation to fabrication hydrogels with tunable mechanical and gelation properties, as well as chemical functionalities for post- gelation biological cues incorporation is an attractive and highly versatile strategy. We have designed and synthesized two precursors: bi-aldehyde functionalized PEG (PEG-bCHO) and a 4-arm aminooxy crosslinker. Using identical precursor compositions, we found that the hydrogel properties were influenced greatly by both pH and the presence or absence of an aniline catalyst. The gelation process was completed in minutes at a pH of 4.5 and the resulting storage modulus was 14 kPa, while at pH of 7.4 the gelation process takes hours to form a much softer gel with a modulus of 0.3 kPa. The aniline catalyst helped to improve the gel mechanical properties and shorten the gelation process. In addition, azide functionalized oxime hydrogels were also prepared for post-gelation incorporation of RGD peptides via CuAAC and SPAAC. 3D photo patterning of peptides via thiol-ene chemistry was also explored in alkene derived oxime hydrogels.

157

5.3 Experimental Section

5.3.1 Materials and Methods

All commercial reagents and solvents were purchased from Aldrich or Fisher

Scientific and used without further purification unless noted otherwise. All reactions were performed under a blanket of nitrogen unless noted otherwise.

NMR spectra were obtained by Varian NMRS 300 MHz spectrometer. All chemical shifts are reported in ppm (δ), and referenced to the chemical shifts of residual solvent

1 13 resonances ( H NMR CDCl3 7.27 ppm, D2O 4.80 ppm, DMSO-d6 2.50 ppm); C CDCl3

77.00 ppm, DMSO-d6 39.50 ppm). The following abbreviations were used to explain the multiplicities: s = singlet, d = doublet, t = triplet, br = broad singlet, m = multiplet. FT-IR spectra were recorded on a SHIMADZU MIRacle 10 ATR-FTIR, and Raman spectra were recorded on a LabRAM HR 800 Spectrophotometer. Fluorescence microscopy images were recorded on OLYMPUS IX 81 Microscope and are unaltered. Size exclusion chromatographic analyses (SEC) were performed using a Waters 150-C Plus instrument equipped with three HR-Styragel columns [100 Å, mixed bed (50/500/103/104 Å), mixed bed (103, 104, 106 Å)], and a differential refractometer (Waters 410) detector. THF was used as eluent with a flow rate of 1.0 mL/min at 30 °C. The molecular mass and mass distribution were calculated from polystyrene standards. Electrospray ionization (ESI) was performed using a HCT Ultra II quadrupole ion trap mass spectrometer (Bruker Daltonics,

Billerica, MA) equipped with an electrospray ionization source. MALDI-TOF mass spectra were carried out on a Bruker Ultraflex-III TOF/TOF mass spectrometer (Bruker Daltonics,

Inc., Billerica, MA) equipped with a Nd:YAG laser (355 nm). All spectra were measured in positive reflection mode. Trans-2-[3-(4-tert-butylphenyl)-2-methyl-2-propenylidene]- 158 malononitrile (DCTB, Aldrich, >98%) and sodium trifluoroacetate served as the matrix and cationizing salt, respectively.

5.3.2 Synthesis of Bi-aldehyde Functionalized PEG Precursor (PEG-bCHO)

The synthesis route is described as Scheme 5.1(a). Both PEG (4.6k) and 4- formylbenzoic acid were vacuum dried overnight prior to use. In a 250 mL round flask,

PEG (4600 g/mol, 9.2 g, 2 mmol), 4-(N,N-dimethylamino) pyridinium-4-toluenesulfonate

(DPTS, 0.5 g, 1.5 mmol) and 4-formylbenzoic acid (1.2 g, 8 mmol) were dissolved in 50 mL of anhydrous DMF. The reaction mixture was placed in an ice bath for 10 min, followed by the addition of 1,3-diisopropyl cabodiimide (DIPC, 1.5 mL, 10 mmol) via syringe. The white precipitation was observed in minutes. The reaction was allowed to warm up to room temperature and stir for 24 h. After centrifugation to remove solids, the top layer solution was diluted with 1 L of methanol. The resulting solution was kept at -20 oC overnight. The resulting white solid was isolated by centrifugation and washed with cold methanol three times followed by ethyl ether twice. After vacuum drying, a white solid product (8.2 g, yield 82%) was stored -20 oC for future use.

1 H NMR (300 MHz, CDCl3): δ=10.09 (s, 1H, -PhCHO), 8.21 (m, 2H, aromatic H),

7.95 (m, 2H, aromatic H), 4.50 (t, 2H, -COOCH2CH2O-), 3.84 (t, 2H, -COOCH2CH2O-),

-1 3.63 (s, ~220 H, -CH2CH2O-, PEG main chain). FT-IR (cm ): 2881, 1719, 1701, 1466,

1359, 1341, 1279, 1240, 1202, 1145, 1095, 1059, 961, 841, 760, 681.

159

Scheme 5.1. (a) Synthesis route of bi-aldehyde functionalized PEG Precursor (PEG-bCHO, n=104). (b) Synthesis route of 4-arm aminooxy crosslinker. (c) Chemical structure of azide- aminooxy extender (azide-ONH2). (d) Chemical structure of alkene-aminooxy extender

(alkene-ONH2). (e) Model reaction of oxime ligation between PEG-bCHO and o-(pent-4- en-1-yl) hydroxylamine (Intermediate 5.2).

5.3.3 Synthesis of 4-arm Aminooxy Crosslinker

The synthesis route is described as Scheme 5.1(b). Intermediate 5.1 was first prepared via conventional esterification. Briefly, in a 100 mL round bottom flask, pentaerythritol (1 g, 7.5 mmol), DPTS (0.25 g, 0.75 mmol) and (Boc-aminooxy)acetic acid

(5.8 g, 30 mmol) were dissolved in 30 mL of anhydrous DMF. The reaction mixture was placed in an ice bath for 10 min, followed by the addition of DIPC (5.5 mL, 36 mmol) via 160 syringe. White precipitation was observed within minutes. The reaction formulation was allowed to warm to room temperature and was stirred for 24 h. After removal of the solid by filtration, the collected solution was concentrated for chromatography purification on silica gel. (Ethyl acetate: hexane=3:1(v/v)). The resulting product was a white solid (7.9 g,

1 yield 65%). H NMR (300 MHz, CDCl3): δ=7.93 (s, 4H, -NHCO-), 4.47 (s, 8H, -

13 NHOCH2CO-), 4.26 (s, 8H, -COOCH2C-), 1.49 (s, 36H, (CH3)3CO). C NMR (300 MHz,

CDCl3): δ=169.06, 159.27, 82.41, 72.38, 61.91, 42.76, 28.15.

4-arm aminooxy crosslinker was received by the deprotection of tert-

Butyloxycarbonyl (BOC) group from Intermediate 5.1. Briefly, in a 100 mL round bottom flask, Intermediate 5.1 (2.00 g, 2.4 mmol) was dissolved in HCl/dioxane solution (20 mL,

4M). The solution was allowed to stir at room temperature for 24 h. The white precipitate was isolated by filtration and washed with dioxane three times followed by ethyl ether twice. After drying in vacuum, product came out as a white solid (1.35 g, quantitative). 1H

+ NMR (300 MHz, CDCl3): δ=4.85 (s, 2H, NH3OCH2CO-), 4.45 (s, 2H, -COOCH2C-)).

FT-IR (cm-1): 3710-2159 broad, 1743, 1679, 1513, 1409, 1368, 1205, 1115, 1060, 1015,

887, 873, 717. ESI: (+H) theory 429.5, experimental 429.5 (Appendix Figure 43).

5.3.3 Synthesis of Azide-aminooxy Extender (azide-ONH2)

The synthesis route of azide-ONH2 is shown in Scheme 5.2. It was synthesized in a similar process as 4-arm aminooxy crosslinker.

Intermediate 5.3 was synthesized as reported previously.496

Intermediate 5.4 was synthesized via general esterification catalyzed by

1 DPTS/DIPC. H NMR (300 MHz, CDCl3): δ= 7.87 (s, 2H, -NH-), 4.47 (s, 4H, - 161

NHOCH2CO), 4.15 (s, 4H, -COOCH2C-), 3.46 (s, 4H, -CCH2N3), 1.49 (s, 18H,

13 (CH3)3CCO-). C NMR (300 MHz, CDCl3): δ= 169.0, 156.2, 82.4, 72.4, 63.1, 51.2, 43.0,

28.2.

Azide-aminooxy extender was obtained by the Boc deprotection of Intermediate

1 + 5.4. H NMR (300 MHz, D2O): δ= 4.67-4.73 (m, 4H, NH3OCH2COO-), 4.15-4.25 (m, 4H,

-CH2COOCH2C-), 3.42-3.52 (m, 4H, -CCH2N3).

Scheme 5.2. Synthesis route of azide-aminooxy extender (azide-ONH2).

5.3.4 Synthesis of Alkene-aminooxy Extender (alkene-ONH2)

The synthesis route of alkene-ONH2 is shown in Scheme 5.3. It was synthesized in a similar process as 4-arm aminooxy crosslinker.

Intermediate 5.5 was synthesized via general esterification catalyzed by

1 DPTS/DIPC. H NMR (300 MHz, CDCl3): δ= 7.87 (s, 2H, -NH-), 5.75-6.00 (m, 1H, -

CH=CH2), 5.15-5.40 (m, 3H, -COOCH2CHOOC-, -CH=CH2), 4.47-4.55 (m, 1H,-

COOCH2CHOOC-), 4.45 (d, 4H, -NHOCH2COO-), 4.28-4.38 (m, 1H, - 162

COOCH2CHOOC-), 4.00 (d, 2H, -OCH2CH=CH2), 3.60 (d, 2H, -CHCH2OCH2CH=CH2),

1.48 (s, 18H, (CH3)3CCO-).

Alkene-aminooxy extender was obtained by the Boc deprotection of Intermediate

1 5.5. H NMR (300 MHz, D2O): 5.75-6.05 (m, 1H, -CH=CH2), 5.10-5.50 (m, 3H, -

+ COOCH2CHOOC-, -CH=CH2), 4.55-4.80 (m, 4H, NH3OCH2COO-), 4.30-4.50 (m, 2H,

-COOCH2CHOOC-), 3.95-4.15 (m, 2H, -OCH2CH=CH2), 3.78(d, 2H, -

CHCH2OCH2CH=CH2).

Scheme 5.3. Synthesis route of alkene-aminooxy extender (alkene-ONH2).

5.3.5 Model Reaction of Oxime Ligation (PEG-b-alkene)

This model reaction is described in Scheme 5.4. Intermediate 5.2 was first synthesized and then tethered to the chain ends of PEG-bCHO.

Intermediate 5.6: In a 250 mL round bottom flask, N-Hydroxyphthalimide (8.2 g,

50 mmol), 5-bromo-1-pentene (11.3 g, 75 mmol), and NaHCO3 (8.0 g, 75 mmol) were mixed in 80 mL of THF. The red suspension was allowed to reflux for 24 h. After filtration to remove the solid, the collected solution was concentrated and re-diluted with 100 mL

CH2Cl2. The new solution was washed with saturated NaHCO3 solution until the aqueous layer was colorless. After drying with MgSO4, the organic layer was collected and

163 concentrated for chromatography purification on silica gel with CH2Cl2 as elute fluid. The

1 resulting product was a white solid (9.2 g, yield 80%). H NMR (300 MHz, CDCl3):

δ=7.70-7.90 (m, 4H, aromatic H), 5.75-6.60 (m, 1H, CH2=CHCH2-), 4.95-5.35 (m, 2H,

CH2=CHCH2-), 4.23 (t, 2H, -CH2CH2O-), 2.25-2.35 (m, 2H, CH2=CHCH2CH2CH2O-),

1.83-1.97 (m, 2H, CH2=CHCH2CH2CH2O-).

O-(pent-4-en-1-yl) hydroxylamine (Intermediate 5.2): In a 250 mL round bottom flask, Intermediate 5.6 (2.3 g, 10 mmol) was dissolved in 80 mL ether, followed by the addition of hydrazine monohydrate. White solid was observed within minutes. This reaction mixture was allowed to stir at room temperature 24 h. The white solid byproduct was removed via filtration. Dry HCl gas was purged into the collected organic solution for

30 min. The white solid was collected, dried in vacuum, and stored at room temperature

1 for further use (1.1 g, yield 78%). H NMR (300 MHz, DMSO-d6): δ=11.11 (s, 3H,

+ NH3O-), 5.75-6.60 (m, 1H, CH2=CHCH2-), 4.95-5.35 (m, 2H, CH2=CHCH2-), 4.02 (t,

2H, -CH2CH2O-), 1.95-2.15 (m, 2H, CH2=CHCH2CH2CH2O-), 1.55-1.75 (m, 2H,

13 CH2=CHCH2CH2CH2O-). C NMR (300 MHz, DMSO-d6): δ=136.5, 115.4, 73.4, 29.2,

26.4.

Bi-alkene derived PEG (PEG-b-alkene): In a 4 mL glass vial, PEG-bCHO (100 mg,

0.02 mmol) and Intermediate 5.2 (8.4mg, 0.06 mmol) was dissolved in acetic buffer (pH

4.5) and stirred at room temperature for 10 min. The reaction mixture was subjected for

MALDI-TOF characterization without purification. Pure sample was obtained by dialyzation in deionized water (molecular weight (MW) cut off 3000 g/mol, cellulose

1 membrane, Pierce), followed by lyophilization. H NMR (300 MHz, CDCl3): δ=8.05-

8.10 (m, 2H, aromatic H), 7.58-7.68 (m, 2H aromatic H), 5.75-5.95 (m, 1H, CH2=CHCH2-), 164

4.93-5.13 (m, 2H, CH2=CHCH2-), 4.40-4.55(t, 2H, -COOCH2CH2O-), 4.15-4.25 (t, 2H, -

CH2CH2ON=C-), 3.80-3.90(m, 2H, -COOCH2CH2O-), 3.55-3.75 (s, ~220H, -CH2CH2O-,

PEG main chain), 2.10-2.25 (m, 2H, CH2=CHCH2CH2CH2O-), 1.75-1.90 (m, 2H,

CH2=CHCH2CH2CH2O-).

Scheme 5.4. Model reaction of oxime ligation between PEG-bCHO and Intermediate 5.2.

5.3.6 Synthesis of Functionalized Peptides

Two kinds of peptides bearing “clickable” groups were synthesized via standard solid phase FMOC methodology. Peptides were cleaved from resin using standard conditions (45 min, 95% trifluoroacetic acid (TFA), 2.5% Triisopropylsilane (TIPS), 2.5% water (by volume)) and precipitated in cold diethyl ether. The crude solid product was isolated by centrifuge, washed twice with diethyl ether and dialyzed in deionized water

(molecular weight (MW) cut off 500 g/mol, cellulose membrane, Pierce), followed by lyophilization. As to alkyne-RGD-biotin (sequence Alkyne-GRGDSK(Biotin)-COOH), 5- hexynoic acid was coupled to N-terminus using standard amino acid coupling condition before cleavage. For fluorescein-5(6)-isothiocyanate (FITC)-RGD-thiol (sequence FTIC- 165

GRGDSCS), FITC (Sigma) was coupled to N-terminus in DMF overnight before cleavage.440,441 Alkyne-RGD-Biotin: ESI calculated 939.1, found 939.5 (Appendix Figure

44). FTIC-RGD-thiol: MALDI calculated 1070.1, found 1070.5 (Appendix Figure 45).

Excitation 492 nm, emission 512 nm (PBS buffer, pH 7.4).

5.3.7 Hydrogel Fabrication and Characterization

Citrate-phosphate buffer (10 mM) was used with different pH (2.5, 4.5, 6.0, 6.6,

6.8, 7.0, 7.2, 7.4, and 7.6) for hydrogel formation. Hydrochloric acid-potassium chloride buffer (pH 1.5) was used as the highest acid concentration condition. Concentration of the aniline catalyst was controlled as (0.1 wt %) when used. Hydrogels were prepared with the stoichiometric balanced formulations (1:1 total aminooxy: aldehyde), and total precursor concentration was 10 wt %. Two precursors were pre-dissolved separately in those different buffer solutions to control the pH. Mixing of the two solutions generated oxime hydrogels.

For oscillatory shear measurement, mixture of PEG-bCHO solution and 4-arm aminooxy crosslinker solution was placed in silicon mold for 12 h to make sure the gelation reactions are complete. Hydrogels were then soaked in phosphate buffered saline (PBS, 10 mM, pH

7.0) overnight to reach swell equilibrium. The equilibrium modulus measurements were recorded by ARES G2 Rheometer (TA Instrument) using 8 mm parallel plate geometry at room temperature, and gel thickness was around 1 mm. Hydrogel were immersed in PBS during the whole testing process. Frequency ramped from 100 rad s-1 to 0.1 rad s-1 with 10% strain. To evaluate the gel formation kinetics, storage modulus (G’) and loss modulus (G”) were observed at the constant frequency 10 rad s-1 as a function of time with a strain 10%.

The top of gel formulation was covered with silicon oil to avoid dehydration. 166

5.3.8 Azide Derived Hydrogels Fabrication and Further Functionalization

The citrate-phosphate buffer (10 mM, pH 4.5) was used as the gelation medium.

The ratio between the total aminooxy and aldehyde stoichiometry was controlled 1:1. The concentration of PEGs was 10 wt%. First the PEGs and bi-aminooxy extender (0.1 eq. to aldehyde) were mixed in buffer solution and stirred for 10 min, followed by the addition of the 4-arm aminooxy crosslinker (0.9 eq. to aldehyde) solution. After 12 h of gelation time, the hydrogels were immersed in PBS (pH 7.0) overnight to reach swollen equilibrium for further peptide functionalization.

Azide hydrogels were incubated overnight in alkyne-RGD-biotin solution (1 mg/mL) in PBS, together with copper sulfate (0.1 mg/mL) and sodium ascorbate (1 mg/mL). Hydrogels were then placed in fresh media for 2 d to remove copper salt and any unbonded peptides. For FT-IR testing, solvent exchange was further carried out in deionized water for 1 d before lyophilization. This dry gel sponge was then subjecting for

FT-IR.

For protein binding assay, after copper salt removal, peptide functionalized hydrogels were incubated in tetramethylrhodamine-5(6)-isothiocyanate (TRITC) labeled streptavidin PBS solution (10 ug/mL) for 2 h, followed by incubation in fresh media overnight to remove unbound protein. The resulting hydrogels were imaged by fluorescence microscopy. Control experiment was conducted with plain azide gel without peptide conjugation.

167

5.3.9 Alkene Derived Hydrogels Fabrication and 3D Biochemical Patterning

Alkene hydrogels were fabricated similarly to azide hydrogels except using alkene aminooxy extender to replace the azide one. The thickness of the hydrogel was 1mm or less. Alkene hydrogels were incubated in PBS containing Irgacure 2959 (0.1 mg/mL) and the patterning agent FTIC-RGD-thiol (0.01 mg/mL) for 30 min. Using conventional photolithographic techniques, gels were exposed to ultraviolet light (365nm wavelength) through a 300 mesh TEM grid mask for 2 min. After patterning was complete, hydrogels were washed overnight with fresh media to remove any unbound peptides. The final patterned hydrogels were imaged by fluorescence microscopy.

5.4 Results and Discussion

5.4.1 Hydrogel Precursor Synthesis

Bis-aldehyde functionalized PEG (PEG-bCHO) was synthesized via carbodiimide coupling with 4-formylbenzoic acid, in the presence of DPTS and DIPC. The aldehyde group of 4-formylbenzoic acid is highly reactive due to electron deficiency. DPTS/DIPC was employed as catalyst system to obtain high reaction efficiency and yields under mild conditions497. The complete conversion of hydroxyl to aldehyde was verified by matrix- assisted desorption ionization time-of-flight mass spectrometry (MALDI-TOF-MS, Figure

5.1(a)), which shows a single distribution. The monoisotopic mass of each peak matches well with that expected for PEG-bCHO (e.g., for [103mer+Na]+, found 4839.7 Da vs calculated 4839.6 Da). The difference between neighboring peaks is equal to the mass of a

-CH2CH2O- repeating unit (44.0 Da). Analysis by size-exclusion chromatography (SEC) with polystyrene standard indicated a number average molecular weight (Mn) of 5200 Da 168

(polydispersity index (PDI) = 1.08) (Figure 5.2, Table 5.1). It is higher than the Mn of PEG raw materials, which is found 4860 Da.

Figure 5.1. MALDI-TOF mass spectra of (a) Bi-aldehyde functionalized PEG (PEG- bCHO); (b) Model reaction product between PEG-bCHO and o-(pent-4-en-1-yl) hydroxylamine (Intermediate 5.2) show monomodal distributions and complete reaction conversion.

Prior to hydrogel fabrication, a model reaction was conducted to verify the reaction efficiency and reaction rate of oxime ligation. Intermediate 5.2 was used as a model substrate and coupled to PEG-bCHO in pH 4.5 buffer solution. The raw formulation was submitted to MALDI-TOF-MS without purification after mixing the reagents for 10 min.

The single mono-modal distribution of the mass spectra (Figure 5.1(b)) demonstrates the 169 complete conversion of aldehyde to oxime. The monoisotopic mass peak at 5005.5 Da in

Figure 5.1(b) corresponds to the peak at 4839.7 Da in Figure 5.1(a) after end capping with o-(pent-4-en-1-yl) hydroxylamine. In SEC analysis, the Mn of PEG-b-alkene is 5390 Da

(PDI=1.12) (Figure 5.2, Table 5.1). The exhibited fast reaction rate and high reaction efficiency of the oxime ligation outlined the promising strategy to fabricate the hydrogels.

Table 5.1. Molecular mass data of polymers. Mn: number average molecular weight; PDI: polydispersity index. Mn and PDI were calculated from polystyrene standards.

PEG PEG-bCHO PEG-b-alkene

Mn (Da) 4860 5200 5390

PDI 1.10 1.08 1.12

Figure 5.2. SEC elution curves of PEG raw materials (black), PEG-bCHO (red), and PEG- b-alkene (blue). THF was used as eluent with a flow rate of 1.0 mL/min at 30 °C.

5.4.2 Hydrogel fabrication and chemical characterization

170

Figure 5.3. (a) The chemical structure of bi-aldehyde-functionalized PEG (PEG-bCHO);

(b) The chemical structure of 4-arm aminooxy crosslinker; (c) The general reaction scheme for oxime ligations. The reaction is pH and nucleophile catalyst dependent; (d) A schematic of hydrogel network formation via oxime bond ligation using a 4-arm aminooxy crosslinker; (e) Photos of hydrogels formed at pH 7.4 and hydrogels formed at pH 4.5 (f) show clearly the mechanical differences in the hydrogels with the fast crosslinking kinetics.

Gels obtained at pH 4.5 are stronger than those at pH 7.4 at similar time intervals. (g) The

FT-IR spectra of PEG-bCHO (black) and hydrogels (red) clearly show the extent of oxime formation in the hydrogels; (h) Raman spectra of phenyl and C=N transitions for PEG- bCHO (black) and hydrogels (red) compliment the FT-IR data showing the extent of reaction during the gel formation.

171

Hydrogels were formed using a 10% (mass) precursor solution containing a 1:1 ratio of aldehyde to aminooxy functionalities (Figure 5.3(d)). FT-IR was used to examine the chemical structure of resulted hydrogels. In Figure 5.3(g), both the aldehyde C=O stretch peak at 1701 cm-1 and the C-C stretch peak at 1202 cm-1 (black line) of PEG-bCHO are much weaker in the spectra of crosslinked hydrogels (red line), indicating that the high conversion of the aldehyde groups was achieved in the network formation process.. A new peak at 1762 cm-1 within the hydrogels is assigned to the ester C=O stretch of 4-arm aminooxy crosslinker. Raman spectroscopy was also used to further verify this crosslinking process (Figure 5.3(h)). The characteristic stretch of the oxime C=N bond is found as a sharp peak at 1612 cm-1 (red line). Meanwhile, the small peak at 1612 cm-1 in black line is assigned to the phenyl stretch of PEG-bCHO, which is shifted to 1565 cm-1 in red line after oxime ligation due to the conjugation of π electrons498.

5.4.3 pH Control Over Hydrogel Mechanical Properties and Gelation Kinetics

It is very important to control the intrinsic mechanics of hydrogels, which would influence the interaction between the scaffolds and biological environment, including cell behaviors, gene expression, and tissue regeneration.499-502 Recently we have shown that mechanical properties of polyethylene glycol dimethacrylate (PEGDM) hydrogels have a significant impact in modulating human osteoarthritic chondrocyte behaviors and tissue formation.465 Study of Ifkovitsa and coworkers found that injectable hyaluronic acid hydrogels with high mechanical strength can improve the post-myocardial infarction remodeling process.503

172

Figure 5.4. The extent of crosslinking and mechanical properties are highly dependent on pH. (a) Modulus vs. Frequency behavior of hydrogels formed at pH 4.5 and 7.4 show that the gels formed at low pH are much stronger at comparable time intervals; (b) The storage modulus (G’)-pH plot (frequency 10 rad/s, strain 10%) show that the mechanical properties can be tuned via pH. The data are displayed as the mean and the error bars represent the standard deviation of three independent measurements for each condition.

To quantitate the influence of pH on hydrogel formation, a series of buffer solutions with different pH values, from 1.5 to 7.6, was used as the gel formation media. The gel precursor composition was kept identical (10%, mass), as well as the 1:1 ratio of aldehyde

173 to aminooxy functional groups. Figure 5.4(a) shows the modulus-frequency curve of two hydrogels formed in buffer at two different pH values. At a pH of 4.5, the resulting gel has a storage modulus 14.1±1.2 kPa. It is nearly 50 times stronger than the gel formed at pH of

7.4, which resulted in a storage modulus of only 0.3±0.1 kPa. Figure 5.3(e) and 5.3(f) demonstrate the strength difference by appearance. The plot in Figure 5.4(b) shows the pH control over the resulting storage modulus. In the first region from pH 1.5 to 4.5, the storage modulus increases with increasing pH. In the pH range between 4.5 and 7.4, the storage modulus decreases with an increase in pH, especially in slightly basic environments. The storage modulus of gels at pH of 7.0 is 6.2±0.9 kPa, while at pH of 7.4 it drops to 0.3±0.1 kPa. There is no coherent hydrogel formation at pH 7.6 which results in a very viscous fluid.

The mechanical strength of hydrogels is directly related to the crosslink density, which is determined by functional group conversion. The theoretical gel-point conversion for step-growth polymerization can be calculated by the Flory-Stockmayer equation504,505

1 ρ = (5.1) √푟(1−푓푎푙푑푒ℎ푦푑푒)(1−푓푎푚𝑖푛표표푥푦)

where ρ is the gel-point conversion, r is the stoichiometric ratio, and faldehyde and faminooxy are the degree of functionality for PEG-bCHO and 4-arm linker, respectively. In the current system, we have r=1, faldehyde=2, and faminooxy=4. The critical conversion is approximately

0.58 to achieve gelation. As to oxime ligation, the reaction conversion and equilibrium is largely influenced by pH.298,495 When pH decreases from 4.5 to 1.5, the reverse reactions are favored due to the protonation of imines (pKb ~ 10), which is the rate-limiting step for

174 oxime hydrolysis. The equilibrium of the oxime ligation shifts to the left and the functional group conversion and crosslink density are reduced. As a result, the obtained hydrogels are softer and have a smaller storage modulus. Meanwhile, at pH values of 1.5 and 2.5, the gel was formed immediately after mixing the two precursors (see gelation kinetics data, Figure

5.5(b)). The fast gelation largely restricts the mobility of both aldehyde and aminooxy, and further crosslink density is reduced. On the other hand, fast gelation results in microscopic inhomogeneity, and then leads to weaker hydrogels.285 In the pH range from 4.5 to 7.6, the reaction conversion is reduced due to the increase of pH, and the critical conversion is never reached at pH 7.6. The optimal environment for oxime ligation is mildly acidic.

Under these conditions, the attenuated basicity of the aminooxy groups leaves it unprotonated for further attack on the aldehyde, and the imine is also unprotonated due to the α-effect296, which suspends the hydrolysis process.

In the time dependent gelation experiment (Figure 5.5(a)), the crossover of storage modulus (G’) and loss modulus (G’’) represents the gel point, which means the transformation from a viscous liquid to a viscoelastic solid. The storage modulus plateau is treated as the complete of gelation. In the entire pH range we tested, from strongly acidic to slightly basic, a decrease in acid concentration lead to the faster gel point and complete gelation time (Figure 5.5(b). At pH 1.5 and 2.5, the gel point is not detected due to the fast gelation behavior; only the complete gelation time is obtained (240 s at pH 1.5 and 400 s at pH 2.5). At pH 7.4, the gel point is achieved after 5 h, and 10 h is required for complete gelation. This process is kinetically controlled. The oxime bond formation is catalyzed by acid, which activates aldehyde by protonation and accelerates the dehydration step. When pH decreases, the reaction rate is increased, and then the critical conversion 0.58 and final 175 conversion is achieved faster. It indicates that less time for gelation point and complete gelation is required.

Figure 5.5. The gelation kinetics are highly dependent on pH. (a) Modulus-time behavior of hydrogels formed at pH 4.5; (b) The plot of pH vs time captures the gel point and the times to reach complete gelation. At low pH the gelation time is too fast to capture. The asterisks (*) in (b) represent time intervals that are too short to detect. The data represent the mean and the error bars represent the standard deviation of three independent samples for each condition.

176

5.4.4 Influence of Aniline Catalyst Concentration on Hydrogel Mechanical Properties and

Gelation Kinetics

Additional studies quantified the influence of the nucleophilic catalyst, aniline, on hydrogel mechanical properties and the gelation behavior with a pH ranging from 6.6 to

7.6. Aniline is a widely used catalyst for oxime gelation and has been shown to improve the reaction rate and efficiency.298,506 As shown in Figure 5.6(a), aniline greatly influences the mechanical strength of hydrogels. The catalyst leads to a large increase in the storage modulus, especially at neutral and slightly basic conditions. At pH 7.4, the storage modulus of hydrogels with aniline is 4.7±0.3 kPa, while the storage modulus without aniline is only

0.3±0.1 kPa. Particularly at pH 7.6, there was no gel formation in the absence of catalyst.

Following the addition of catalyst, gels with the storage modulus of 2.4±0.2 kPa were obtained. The gelation kinetics are also catalyst dependent. The time scale of both the gel point (Figure 5.6(b)) and time to complete gelation (Figure 5.6(c)) is reduced from hours to minutes with the addition of aniline. Adding catalyst results in stronger hydrogels by increasing the functional group conversion and crosslink density, and enables the critical conversion to be reached at pH 7.6. At the same time, aniline also accelerates oxime ligation, and reduces time to achieve critical conversion.298,301

177

Figure 5.6. Influence of aniline catalyst (0.1 wt%) on hydrogels storage modulus (a), time to reach the gel point (b) and the complete gelation time (c) as a function of pH. The data represent the mean and the error bars represent the standard deviation of three independent samples for each condition.

178

5.4.5 Hydrogels with Chemical Functionalities and its Further Derivation with Peptides

Azide and alkene functionalized hydrogels were prepared for further conjugation of biological molecules via a cascade approach as shown in Figure 5.7(a). The 2-arm aminooxy precursor with azide (azide-ONH2, Scheme 5.1(c)) or alkene (alkene-ONH2,

Scheme 5.1(d)) handles was used as the chain extender. It reacted with PEG-bCHO, resulting in a new precursor formulation in situ. Aldehyde groups are present on the chain end, while the other two arms contain azide or alkene groups. The content of the PEG with

“clickable” groups in aldehyde mixtures was controlled via feed ratio between the extender and PEG-bCHO. The 4-arm crosslinker solution was added into aldehyde mixtures to form crosslinked hydrogels.

Azide incorporation into oxime hydrogels was confirmed via FT-IR spectroscopy.

The characteristic azide stretch peak is present at 2103 cm-1 (Figure 5.7(b1), black line).

Azide groups provide a platform to functionalize hydrogels using azide-alkyne Huisgen cycloaddition, which is widely used in biomedical engineering for bioconjugation and tissue labeling applications193,235,240,507. Alkyne-RGD-biotin was used as a model peptide to verify the availability of azide groups for post-gelation reactions. The disappearance of the azide stretch peak after the “click” cyclization reaction indicates the peptide incorporation into hydrogel matrix was successful (Figure 5.7(b1), red line). An additional protein binding experiment was conducted with tetramethylrhodamine-5(6)-isothiocyanate

(TRITC) labeled streptavidin.451,508 The strong red fluorescence of labeled streptavadin protein confirms the existence and bioavailability of biotin in the hydrogels (Figure 5.7(b2).

The dark region is the glass slide background. The reaction and binding sites are described in Figure 5.7(b3). 179

Figure 5.7. (a) Fabrication procedure of oxime hydrogels with “clickable” groups. PEG- bCHO was initially mixed together with Alkene/Azide-ONH2 for minutes, followed by the addition of 4-arm crosslinker. (b1) FT-IR spectra of azide functionalized hydrogels before

(black) and after (red) “click” conjugation with alkyne-RGD-biotin peptides. (b2)

Fluorescence microscopy image of azide hydrogels after bonding with TRITC labeled streptavidin. Red areas are from labeled hydrogels and black areas are the glass side background. (b3) Schematic of the “click” cycloaddition reaction and the binding sites for peptide or protein conjugation. (c1) Raman spectra of alkene hydrogels before (black) and after (red) the thiol-ene addition of 2-mercaptoethanol. (c2) Fluorescence microscopy image of alkene functionalized hydrogels after 3D patterning with FTIC-RGD-thiol. (c3)

Schematic of thiol-ene addition sites. (Scale bar 100 um) 180

Oxime hydrogels bearing alkene groups were also utilized for peptide conjugation via photo-initiated thiol-ene reactions. In the Raman spectra of the alkene functionalized gels in Figure 5.7(c1) (black line), the stretch of the alkene is not observed due to the overlap with the imine peak between 1600 cm-1 and 1700 cm-1. Following the photo- initiated addition of 2-mercaptoethanol with Irgacure 2959 as the radical formation initiator, the C-S stretch peak at 630 cm-1 is observed (Figure 5.7(c1), red line), which indicates the occurrence of the thiol-ene radical addition. This reaction is further used to create spatial patterns of peptides in oxime hydrogels using a simple photomask. Fluorescein-5- isothiocyanate (FTIC) conjugated RGD with pendant thiol groups (FTIC-RGD-SH) was employed as the model bioactive signal to explore the potential for photochemical patterning. Photocoupling was achieved using a UV light source (365 nm) and the biocompatible initiator Irgacure 2959. A square pattern of fluorescent-labeled peptide was generated with a 300 mesh TEM copper grid functioning as a mask (Figure 5.7(c2)). The contrast of the exposed square areas and covered lines is obvious, although background fluorescence was observed due to light scattering in the photo pattering process. The tethered sites are described in Figure 5.7(c3).

5.5 Conclusion

In conclusion, we have demonstrated the fabrication of covalently crosslinked

PEG-based hydrogels via oxime ligation. The crosslinking process is both pH and catalyst dependent. The tunable mechanical properties are related to the condensation equilibrium, reaction rate and the precursors mobility509. The results demonstrate that oxime ligation is an efficient method for the fabrication of hydrogels with variable mechanical and gelation 181 kinetics using identical precursors. At 10 wt% concentrations, the storage modulus can be controllably varied from 0.3 kPa to 15 kPa and the gel formation times can be tuned from seconds to hours. Oxime hydrogels with controlled amounts of residual azide and alkene groups were also obtained. The hydrogels were successfully used for pattern formation and peptide incorporation via robust and highly efficient reaction conditions. These materials are the initial demonstration for translationally relevant hydrogel materials that possess tunable mechanical regimes attractive to soft tissue engineering and possess atom neutral chemistries attractive for post gelation patterning in the presence or absence of cells.

5.6 Acknowledgements

The authors gratefully acknowledge financial support from the Akron Functional

Materials Center and RESBIO: Integrated Technologies for Polymeric Biomaterials (NIH-

NIBIB P41 EB001046). The authors also acknowledge partial support for AD from the

NSF (DMR- 1105370).

182

CHAPTER VI CASCADING “TRI-CLICK”FUNCTIONALIZATION OF POLYCAPROLACTONE THIN FILMS QUANTIFIED VIA QCM

Portions of this work have been published previously as

Fei Lin, Jukuan Zheng, Jiayi Yu, Jinjun Zhou, and Matthew L. Becker

Biomacromolecules, 2013, 14 (8), 2857–2865.

6.1 Outline

A series of mono- and multi-functionalized degradable polyesters bearing various

“clickable” groups, including ketone, alkyne, azide, and methyl acrylate (MA) are reported.

Using this approach we demonstrate a cascade approach to immobilize and quantitate three separate bioactive groups onto poly(caprolactone) thin films. The materials are based on tunable copolymer compositions of ε-caprolactone and 2-oxepane-1,5-dione. Quartz crystal microbalance (QCM) was used to quantify the rate and extent of surface conjugation between RGD peptide and polymer thin films using “click” chemistry methods. The results show that alkyne functionalized polymer has a highest conversion efficiency, followed by

MA and azide polymers, while polymer films possessing keto groups are less amenable to surface functionalization. The successful conjugation was further confirmed by static contact angle measurements, with a smaller contact angle correlating directly with lower levels of surface peptide conjugation. Quartz crystal microbalance (QCM) results

183 quantify the sequential immobilization of peptides on the PCL thin films, and indicate that

Michel-Addition must occur first, followed by azide-alkyne Huisgen cycloadditions.

6.2 Introduction

Synthetic aliphatic polyesters, such as poly (ε-caprolactone) (PCL), poly (lactides)

(PLA), and poly (glycolide) (PGA) are used widely in biomedical applications such as scaffolds for tissue engineering, drug delivery devices, coatings and sutures.73,81,510,511

However, the expanded use of degradable polyesters into applications requiring functional species to enhance the biointerfacial interactions are limited by the availability of reactive sites for the covalent immobilization of bioactive drugs, peptides, and proteins. In response to this limitation significant efforts have focused on introducing chemically reactive pendent groups into the monomer species, which afford the placement of functional groups by postpolymerization modification.217,261,263,304,308,402,413,512,513

Copolymerizing commercial cyclic monomers like CL, LA, or GA with analogs possessing protected functional groups is a common strategy to introduce reactive handles. High reaction efficiency and reproducibility are critical when attaching bioactive molecules as they are expensive relative to the polymeric components and showing efficacy in translational applications is dependent on the ability to correlate structure to function, especially on functional surfaces.342,514-517

Click chemistry approaches have provided multiple avenues to address these challenges but optimizing the reaction conditions and quantifying the reaction efficiency remain difficult. The concept of “click chemistry” was introduced by Sharpless and coworkers189 and the term now represents a wide range of reactions, including copper(I)- 184 catalyzed azide-alkyne Huisgen cycloaddition (CuAAC), thiol-ene radical additions,

Michael-additions, Diels-Alder reactions, oxime ligation, etc.190 “Click” reactions are highly selective, highly efficient, and are orthogonal to other chemical groups and work in mild reaction conditions.190,250,252,298 They are widely used for the post-polymerization functionalization of polymers and the surface modification of films and nanofibers.242,261,518,519 In biological applications metal-free strain promoted azide-alkyne cycloadditions (SPAAC) are attractive in which alkynes are activated using ring-strain, or incorporating electron-withdrawing groups, or both.233,239,240,520 Strained alkynes have been shown to react with azide groups directly at room temperature without metal catalyst.242,490,521 SPAAC has been used to immobilize functional molecules on polymer surfaces using “click” reactions, though it is still challenging to attach biomolecules on the same surface with different chemical reactions.451,518,522

The combination of ring opening polymerization (ROP) with highly efficient

(“click”) reactions provides a platform for the construction of degradable polyester materials possessing multiple functional groups.216,217 Emrick and co-workers synthesized alkyne and azide derived PCL and further functionalized polymer with bioactive RGD peptide 216. Alkene sulfone functionalized polyester was prepared in the

Zhong and Dove group for biomolecules attachment via Michael-addition261,413. PCL with pendent ketone groups has been synthesized by multiple research groups307,523,524. The ketone derived copolymer: poly (ε-caprolactone-co-2-oxepane-1,5-dione) (P(CL-co-OPD)) has a higher glass transition and melting temperature, degrades faster and has stronger mechanical properties than PCL400,403,524. The electropilic ketones in the copolymer backbone can undergo oxime ligation reactions with hydrazine, semicarbazide, and 185 aminooxy functionalized molecules. These reactions have been used for the post- polymerization modification of P(CL-co-OPD), with peptides, drugs, and poly(ethylene oxide).304,308,319,322

Inspired by the “clickable” characteristics of the oxime ligation chemistry, we demonstrate a highly versatile method (Scheme 6.4) to synthesize PCL possessing mono- and multi- “clickable” reactive groups. We show using a cascade approach that only one monomer is required for the preparation of polymers with different reactive sites.

Incorporation of additional “clickable” groups was accomplished via postpolymerization modification through oxime ligation with reactive amino-oxy groups. Quartz crystal microbalance (QCM) and static contact angle measurements were used to quantify the conjugation efficiency of the bioactive peptides to a poly(caprolactone) film. 1H NMR and

FT-IR also helped to confirm the availability of surface functional groups for chemical reactions.

6.3 Experimental Section

6.3.1 Materials and Methods

All commercial reagents and solvents were purchased from Aldrich or Fisher

Scientific and used without additional purification unless specifically noted. ε- caprolactone (CL) was dried with CaH2 overnight and distilled in vaccuo twice and stored under argon. All reactions were performed under a blanket of nitrogen unless noted otherwise.

NMR spectra were obtained using a Varian NMRS 300 MHz spectrometer.

Chemical shifts are reported in ppm (δ), and referenced to the chemical shifts of the residual 186

1 13 solvent ( H NMR CDCl3 7.27 ppm, DMSO-d6 2.50 ppm, D2O 4.80 ppm; C CDCl3 77.00 ppm, DMSO-d6 39.50ppm). The following abbreviations were used to explain the multiplicities: s = singlet, d = doublet, t = triplet, bs = broad singlet, m = multiplet. FT-

IR spectra were recorded on a SHIMADZU MIRacle 10 ATR-FTIR. Size exclusion chromatographic analyses (SEC) were performed using TOSOH HLC-8320 GPC. N,N-

Dimethylformamide (DMF) (with 0.01M LiBr) was used as eluent with a flow rate of 0.8 mL/min at 50 oC. The molecular mass was calculated from universal calibration based on polystyrene standards. Electrospray ionization (ESI) was performed using a HCT Ultra II quadrupole ion trap mass spectrometer (Bruker Daltonics, Billerica, MA) equipped with an electrospray ionization source. MALDI-TOF mass spectra were carried out on a Bruker

Ultraflex-III TOF/TOF mass spectrometer (Bruker Daltonics, Inc., Billerica, MA) equipped with a Nd:YAG laser (355 nm). All spectra were measured in positive reflection mode. Sodium trifluoroacetate was used as cationizing salt. The thermal properties of the materials were recorded using a TA Instruments DSC Q2000. The temperature ramping rate was 10 oC/min. Data were collected in the second heating cycle. The static contact angles were measured using an Advanced Goniometer (Ramé-Hart Instrument Co., Model

500) at 25 °C using ultrapure water (1 μL) (18 MΩ cm-1) as the probe fluid and analyzed by a drop shape analysis method.

6.3.2 Quartz Crystal Microbalance (QCM)

Q-sense E4 operator from Biolin Scientific AB was used to study the peptide uptake of polymer thin film. SiO2 coated crystal sensor X301(5 MHz resonant frequency) was chosen for spinning coating substrate. All presented data corresponded to the normalized 187 frequency of the seventh overtone. Based on Sauerbrey model525, equation (5.1) is often utilized to convert frequency shift into mass change per area.

△푓 m = C (6.1) n where m is the mass change per area (mg/m2), C is the sensitivity constant, -0.177 (mg/

(m2 •Hz)), Δf is the change in resonant frequency (Hz), and n is the overtone number.

6.3.3 Synthesis of Intermediates R-ONH2

6.3.3.1 Synthesis of Intermediate 6.1: 2-(2-(aminooxy)acetoxy)ethyl acrylate

Scheme 6.1: Synthesis route for 2-(2-(aminooxy)acetoxy)ethyl acrylate (6.1).

The synthesis route is described in Scheme 6.1. There are two steps.

Compound 6.1(a): 2-Hydroxyethyl acrylate (1.1g, 10 mmol, 1 eq.), (Boc-aminooxy) acetic acid (2.5g, 13 mmol, 1.3 eq.), and 4-Dimethylaminopyridine (DMAP) (0.3g, 3mmol,

0.4 eq.) were dissolved in 20 mL anhydrous CH2Cl2 at ice-water bath. N, N’ -

Diisopropylcarbodiimide (DIPC) (1.8ml, 15 mmol, 1.5 eq.) was added into the solution above via syringe. The reaction was allowed to warm up to room temperature and proceed overnight. The result suspension was diluted with 100 mL CH2Cl2 and washed with HCl solution (1M, 50 mL x 3), followed drying over MgSO4. After evaporation of solvent under reduced pressure, the crude product was purified by flash column chromatography on silica 188 gel with CH3OH/CHCl3 (v/v =3/100) as eluent to afford the product as a colorless oil (1.8g,

1 90%). H NMR (300M Hz, CDCl3): δ=1.48 (m, 9H, -C(CH3)3), 4.38-4.43 (m, 4H, -

OCH2CH2O-), 4.47 (s, 2H, -OCCH2O-), 5.83-5.93 (m, 1H, CH2=CH-), 6.08-6.23 (m, 1H,

13 CH2=CH-), 6.35-6.50 (m, 1H, CH2=CH-), 7.80 (s, 1H, -ONH-); C NMR (300 Hz,

CDCl3): 28.1, 61.9, 62.8, 72.5, 82.1, 127.9131.6, 156.2, 165.8, 169.4

Intermediate 6.1: Compound 6.1(a) (1.5g) was dissolved with 8 mL mixture of

CH2Cl2 and trifluoroacetic acid (v/v=3:1) at ice bath. The reaction was allowed to warm up to room temperature and proceed for 8h. Reaction solution was diluted with 50 mL

CH2Cl2, and then washed with saturated NaHCO3 aqueous solution three times. Collected organic phase was dried with MgSO4 and evaporated under reduced pressure to yield intermediate 6.1 as colorless oil (0.84g, yield 85%). It was directly used for the next step

1 without further purification. H NMR (300M Hz, CDCl3): δ= 4.27 (s, 2H, NH2OCH2-),

4.36-4.47 (m, 4H, -OCH2CH2O-), 5.80-5.95 (m, 3H, CH2=CH- and NH2O-), 6.08-6.23 (m,

13 1H, CH2=CH-), 6.35-6.50 (m, 1H, CH2=CH-); C NMR (300 Hz, CDCl3): 62.0, 62.4,

72.3, 127.9, 131.4, 165.8, 170.5; ESI(+H): 191.2 (191.2). (Appendix Figure 46)

6.3.3.2 Synthesis of Intermediate 6.2: O-(6-azidohexyl)hydroxylammonium chloride

The synthesis route is described in Scheme 6.2. There are three steps.

Compound 6.2(a): 1,6-dibromohexane (5g, 20.6 mmol, 1 eq.), N- hydroxyphthalimide(10g, 62 mmol, 3eq) and triethylamine (TEA) (11ml, 62 mmol, 5eq.) were suspended in 500 mL THF. The deep red reaction system was refluxed at 80oC for

36h with magnetic stirring. Solvent was removed by reduced pressure, and solid residue was redissolved in 200 mL CHCl3. This solution was washed with saturated sodium 189 carbonate solution until the aqueous phase was colorless. Organic phase was collected and dried with MgSO4 and then evaporated under reduced pressure. The crude product was purified by flash column chromatography on silica gel with CH3OH/CHCl3 (v/v = 1/100)

1 as eluent to afford the product as a white solid (4.8g, 75%). H NMR (300M Hz, CDCl3):

δ= 1.40-1.65 (m, 4H, -OCH2CH2CH2), 1.70-2.00 (m, 4H, -OCH2CH2CH2), 3.44 (t, J=6.9,

13 2H, Br-CH2CH2), 4.22 (t, J=6.4, 2H, -O-CH2CH2), 7.70-7.90 (m, 4H, aromatic); C NMR

(300 Hz, CDCl3): 24.5, 25.3, 27.0, 32.5, 33.6, 77.7, 123.4, 128.9, 134.4, 163.6.

Scheme 6.2: Synthesis route for O-(6-azidohexyl)hydroxylammonium (6.2)

Compound 6.2(b): Compound 6.2(a) (3g, 9.2 mmol, 1eq.) was dissolved in 100 mL

CH2Cl2 resulting a clear solution. Hydrazine monohydrate (0.75 mL, 13.8 mmol, 1.5 eq.) was added to reaction system, which was allowed to stir at room temperature overnight.

After filtration to remove white solid, the organic phase was washed with water three times.

After dried with MgSO4 overnight, the collected organic phase was evaporated under reduced pressure. The colorless viscous liquid was redissolved in 100 mL diethyl ether, which was acidified with dry HCl gas subsequently. The precipitation was collected and dried in vacuum oven to afford the product as a white solid. (1.5g, yield 82%). 1H NMR

(300M Hz, D2O): δ= 1.30-1.66 (m, 4H, -OCH2CH2CH2), 1.60-2.00 (m, 4H, -

13 OCH2CH2CH2), 3.53 (t, J=6.7, 2H, Br-CH2CH2), 4.10 (t, J=6.6, 2H, -O-CH2CH2); C

NMR (300 Hz, D2O): 23.8, 26.7, 26.8, 31.8, 34.9, 75.4

190

Intermediate 6.2: Compound 6.2(b) (1.5g, 6.4 mmol, 1 eq.), Na2CO3 (0.75g, 7.1 mmol, 1.1 eq.), and NaN3 (1.3g, 23 mmol, 3 eq.) were dissolved in 50 mL H2O. The reaction system was stirred at 80oC overnight. The aqueous cloudy solution was extracted with

CH2Cl2 (50 mL x 3). The combined organic phase was dried over with MgSO4 and evaporated under reduced pressure. The colorless oil was redissolved in 100 mL diethyl ether, which was acidified with dry HCl gas subsequently. The precipitation was collected and dried in vacuum oven to afford the product as a white solid. (1.15g, yield 92%). 1H

NMR (300M Hz, D2O) δ= 1.30-1.55 (m, 4H, -OCH2CH2CH2), 1.55-1.80 (m, 4H, -

13 OCH2CH2CH2), 3.35 (t, J=6.7, 2H, Br-CH2CH2), 4.10 (t, J=6.6, 2H, -O-CH2CH2); C

NMR (300 Hz, D2O) 24.2, 25.4, 26.7, 27.7, 51.0, 75.4. (Appendix Figure 47)

6.3.3.3 Synthesis of Intermediate 6.3: Propargyl hydroxylammonium chloride

Schemes 6.3: Synthesis route for propargyl hydroxylammonium chloride (6.3)

The synthesis route is described in Scheme 6.3. There are two steps.

Compound 6.3(a): N-hydroxyphthalimide (6g, 36.4 mmol, 1.3eq.) and NaHCO3

(4g, 37.7 mmol, 1.4 eq.) were suspended in 100 mL DMF. Propargyl bromide (3 mL 80% in toluene, 27.5 mmol, 1eq.) was added to the above suspension dropwise. The reaction system was heated to 80oC and stirred at that temperature for 20h. DMF was removed under reduced pressure, and the residue was redissolved into 200 mL CH2Cl2. This solution was

191 washed with saturated sodium carbonate solution until the aqueous phase was colorless.

Organic phase was collected and dried with MgSO4 overnight, and evaporated under reduced pressure. The crude product was purified by flash column chromatography on silica gel with CH3OH/CHCl3 (v/v = 2/100) as eluent to afford the product as a white solid

1 (5.0g, 92.6%). H NMR (300M Hz, CDCl3): δ= 2.60 (t, 1H, CH≡C-), 1.55-1.80 (d, 2H, -

13 CH2O-), 7.73-7.95 (m, 4H, aromatic); C NMR (300 Hz, CDCl3): 65.2, 76.7, 78.3, 123.9,

134.8, 163.5.

Intermediate 6.3: Compound 6.3(a) (5g, 25 mmol, 1eq.) and hydrazine monohydrate (1.5 mL, 30 mmol, 1.2 eq.) were added into 200 mL diethyl ether. The suspension was stirred at room temperature for 4h, and then filtered to remove solid byproduct. The collected organic phase was dried over MgSO4. After acidification with dry

HCl gas, the product was afforded as a white solid. (2.3g, yield 92%). 1H NMR (300M Hz,

13 DMSO-d6) δ= 3.87(m, 1H, CH≡C-), 4.75 (m, 2H, -CH2O-), 11.10 (bs, 2H, NH2O-); C

NMR (75M Hz, DMSO-d6) 61.7, 76.5, 81.1. (Appendix Figure 48)

6.3.4 Synthesis of Mono-functionalized PCL

The synthesis route is described in Scheme 6.4. Random copolymers of P (CL-co-

OPD) were prepared by ROP of CL and 1,4,8-trioxaspiro[4.6]-9-undecanone (TOSUO), followed by the deprotection of the ketone groups using triphenylcarbenium tetrafluoroborate.400,523,524 Oxime ligation between keto-PCL and the respective intermediates resulted in three mono-functionalized PCL: PCL-Alkyne, PCL-Azide, and

PCL-MA. In a typical experiment, a 20 mL glass vial equipped with magnetic stir bar 192 were charged keto-PCL copolymer (0.500 g, ketone group as 0.4 mmol), intermediate

6.1 (0.085 g, 0.45 mmol, 1.1 eq.), p-toluenesulfonic acid (catalyst amount, 2.0 mg), and

THF (5.0 mL). The reaction was allowed to proceed for 4 h at room temperature. The resulting polymer was isolated by precipitation in clod methanol, filtration and drying in vacuum, giving the product as a white solid in a quantitative yield.

Scheme 6.4. The synthetic route for the ring opening polymerization of keto-derivatized poly(caprolactone). The use of multiple functionalized amino-oxy reactive groups enables the post polymerization modification of multiple “clickable” groups on PCL-based copolymer films.

1 P(CL-co-TOSUO): H NMR (300M Hz, CDCl3): δ= 1.25-1.48 (m, 2 H of the CL units, - CH2CH2CH2C(O)O-), 1.50-1.75 (m, 4H of the CL units, -

CH2CH2CH2CH2CH2C(O)O-), 1.90-2.10 (m, 4 H of TOSUO units, -

193

CH2C(OCH2CH2O)CH2-), 2.31 (t, J=7.5, 2H of the CL units, -CH2CH2C(O)O-), 2.37 (t, J

= 7.2 Hz, 2H of TOSUO units,- CH2C(O)OCH2-) , 3.94 (s, 4H of the TOSUO units, -

CH2C(OCH2CH2O)CH2-), 4.07 (t, J=6.6, 2H of CL units, -C(O)OCH2CH2-), 4.17 (t, J=,

-1 2H of TOSUO units, -C(O)OCH2CH2-); FT-IR (cm ): 2945, 2867, 1727, 1470, 1464, 1396,

o o 1366, 1295, 1243, 1190, 1108, 1048, 962, 733.; DSC: Tg= -56.2 C, Tm= 51.0 C.; SEC: Mn

=39115, Mw=82869, PDI=2.1.

1 P(CL-co-OPD) (keto-PCL): H NMR (300M Hz, CDCl3): δ= 1.25-1.48 (m, 2 H of the CL units, - CH2CH2CH2C(O)O-), 1.50-1.75 (m, 4H of the CL units, -

CH2CH2CH2CH2CH2C(O)O-), 2.31 (t, J=7.5, 2H of the CL units, -CH2CH2C(O)O-), 2.60

(t, J=6.1, 2H of OPD units, -CH2C(O)CH2CH2C(O)O-), 2.68-2.85 (m, 4H of OPD units, -

C(O)CH2CH2C(O)O-), 4.07 (t, J=6.6, 2H of CL units, -C(O)OCH2CH2-), 4.35 (t, J=6.1,

-1 2H of OPD units, -C(O)OCH2CH2C(CO)-).; FT-IR (cm ): 2944, 2866, 1725, 1471, 1439,

o o 1366, 1294, 1243, 1190, 1107, 1065, 1047, 933, 732; DSC: Tg= -56.1 C, Tm= 57.8 C; SEC:

Mn =41119, Mw=71752, PDI=1.7.

1 PCL-azide: H NMR (300M Hz, CDCl3): δ= 1.25-1.48 (m, 2 H of the CL units, -

CH2CH2CH2C(O)O-), 1.50-1.75 (m, 4H of the CL units, -CH2CH2CH2CH2CH2C(O)O-),

2.31 (t, J=7.5, 2H of the CL units, -CH2CH2C(O)O-), 2.50-2.70 (m, 6H of azide units, -

CH2C(N)CH2CH2C(O)O-), 3.28 (t, J=6.9, 2H of azide units, -CH2N3), 3.98 (t, J=6.1, 2H of azide units, -C(N)OCH2-), 4.07 (t, J=6.6, 2H of CL units, -C(O)OCH2CH2-), 4.24 (t,

-1 J=6.5, 2H of azide units, -C(O)OCH2CH2C(N)-); FT-IR (cm ): 2944, 2865, 2096, 1725,

o o 1420, 1395, 1366, 1295, 1244, 1191, 1107, 1046, 962, 732; DSC: Tg= -62.9 C, Tm1= 40.1 C,

o Tm2= 46.0 C; SEC: Mn =39115, Mw=82869, PDI=2.1.

194

1 PCL-alkyne: H NMR (300M Hz, CDCl3): δ= 1.25-1.48 (m, 2 H of the CL units, -

CH2CH2CH2C(O)O-), 1.50-1.75 (m, 4H of the CL units, -CH2CH2CH2CH2CH2C(O)O-),

2.31 (t, J=7.5, 2H of the CL units, -CH2CH2C(O)O-), 2.44 (t, 1H of alkyne units, -

CH2C≡CH), 2.50-2.70 (m, 6H of alkyne units, -CH2C(N)CH2CH2C(O)O-), 4.07 (t, J=6.6,

2H of CL units, -C(O)OCH2CH2-), 4.24 (t, J=6.5, 2H of alkyne units, -

-1 C(O)OCH2CH2C(N)-), 4.55-4.70 (m, 2H of alkyne units, -OCH2C≡CH); FT-IR (cm ):

3264, 2944, 2865, 1730, 1462, 1410, 1392, 1364, 1295, 1243, 1164, 1105, 1045, 1007, 962,

o o o 732; DSC: Tg= -52.5 C, Tm1= 40.4 C, Tm2= 46.7 C; SEC: Mn =41100, Mw=74044,

PDI=1.8.

PCL-MA: Molecular weight of PCL-MA was not characterized, due to the possibility of double bond crosslink in the SEC condition, which would clog the SEC column. The situation is the same for other polymers containing MA units. 1H NMR (300M

Hz, CDCl3): δ= 1.25-1.48 (m, 2 H of the CL units, - CH2CH2CH2C(O)O-), 1.50-1.75 (m,

4H of the CL units, -CH2CH2CH2CH2CH2C(O)O-), 2.31 (t, J=7.5, 2H of the CL units, -

CH2CH2C(O)O-), 2.50-2.70 (m, 6H of CL units, -CH2C(N)CH2CH2C(O)O-), 4.07 (t,

J=6.6, 2H of CL units, -C(O)OCH2CH2-), 4.38(s, 4H of MA units,

=NOCH2COOCH2CH2O-), 4.60 (d, 2H of MA units, =NOCH2COO-), 5.83-5.93 (m, 1H of MA units, CH2=CHCOO-), 6.08-6.21 (m, 1H of MA units, CH2CHCOO-), 6.38-6.50

-1 (m, 1H of MA units, CH2CHCOO-). FT-IR (cm ): 3076, 2944, 2865, 1641, 1462, 1419,

o o 1392, 1365, 1295, 1242, 1190, 1106, 1065, 962, 916, 732; DSC: Tg= -47.9 C, Tm1= 39.6 C,

o Tm2= 46.1 C.

195

6.3.5 Synthesis of Multi-functionalized PCL.

Three types of multi-functionalized PCL were synthesized: PCL-MA-Alkyne,

PCL-MA-Azide, and PCL-MA-Alkyne-Azide. The synthetic process was similar to that of mono-functionalized PCL, except a mixture of intermediates was used in one pot to replace the single functional species. In the initial study, the content of each “clickable” group was controlled using a defined feed ratio.

1 PCL-MA-azide: H NMR (300M Hz, CDCl3): δ= 1.25-1.48 (m, 2 H of the CL units,

- CH2CH2CH2C(O)O-), 1.50-1.75 (m, 4H of the CL units, -CH2CH2CH2CH2CH2C(O)O-),

2.31 (t, J=7.5, 2H of the CL units, -CH2CH2C(O)O-), 2.45-2.75 (m, 12H of MA-azide units,

-CH2C(N)CH2CH2C(O)O-), 3.28 (t, J=6.9, 2H of azide units, -CH2N3), 3.98 (t, J=6.1, 2H of azide units, -C(N)OCH2-), 4.07 (t, J=6.6, 2H of CL units, -C(O)OCH2CH2-), 4.15-4.35

(m, 4H of MA-azide units, -C(O)OCH2CH2C(N)-)), 4.38(s, 4H of MA units,

=NOCH2COOCH2CH2O-), 4.60 (d, 2H of MA units, =NOCH2COO-), 5.83-5.93 (m, 1H of MA units, CH2=CHCOO-), 6.08-6.21 (m, 1H of MA units, CH2CHCOO-), 6.38-6.50

-1 (m, 1H of MA units, CH2CHCOO-); FT-IR (cm ) 2943, 2865, 2096, 1725, 1635, 1465,

o o 1417, 1394, 1366, 1295, 1242, 1190, 1106, 1045, 962, 732; DSC: Tg= -55.1 C, Tm1= 40.5 C,

o Tm2= 46.1 C.

1 PCL-MA-alkyne: H NMR (300M Hz, CDCl3): δ= 1.25-1.48 (m, 2 H of the CL units, - CH2CH2CH2C(O)O-), 1.50-1.75 (m, 4H of the CL units, -

CH2CH2CH2CH2CH2C(O)O-), 2.31 (t, J=7.5, 2H of the CL units, -CH2CH2C(O)O-), 2.44

(t, 1H of alkyne units, -CH2C≡CH), 2.45-2.75 (m, 12H of MA-alkyne units, -

CH2C(N)CH2CH2C(O)O-), 4.07 (t, J=6.6, 2H of CL units, -C(O)OCH2CH2-), 4.15-4.35

(m, 4H of MA-azide units, -C(O)OCH2CH2C(N)-)), 4.38(s, 4H of MA units, 196

=NOCH2COOCH2CH2O-), 4.55-4.65 (m, 4H of MA-alkyne units, =NOCH2COO- and -

OCH2C≡CH), 5.83-5.93 (m, 1H of MA units, CH2=CHCOO-), 6.08-6.21 (m, 1H of MA

-1 units, CH2CHCOO-), 6.38-6.50 (m, 1H of MA units, CH2CHCOO-).; FT-IR (cm ) 3264,

2944, 2865, 1725, 1471, 1419, 1396, 1367, 1295, 1244, 1191, 1105, 1066, 1046, 962, 732;

o o o DSC: Tg= -48.5 C, Tm1= 40.8 C, Tm2= 46.9 C.

1 PCL-MA-azide-alkyne: H NMR (300M Hz, CDCl3): δ= 1.25-1.48 (m, 2 H of the

CL units, - CH2CH2CH2C(O)O-), 1.50-1.75 (m, 4H of the CL units, -

CH2CH2CH2CH2CH2C(O)O-), 2.31 (t, J=7.5, 2H of the CL units, -CH2CH2C(O)O-), 2.44

(t, 1H of alkyne units, -CH2C≡CH), 2.45-2.75 (m, 18H of MA-azide-alkyne units, -

CH2C(N)CH2CH2C(O)O-), 3.28 (t, J=6.9, 2H of azide units, -CH2N3), 3.98 (t, J=6.1, 2H of azide units, -C(N)OCH2CH2-),4.07 (t, J=6.6, 2H of CL units, -C(O)OCH2CH2-), 4.15-

4.35 (m, 6H of MA-azide-alkyne units, -C(O)OCH2CH2C(N)-)), 4.38(s, 4H of MA units,

=NOCH2COOCH2CH2O-), 4.55-4.65 (m, 4H of MA-alkyne units, =NOCH2COO- and -

OCH2C≡CH), 5.83-5.93 (m, 1H of MA units, CH2=CHCOO-), 6.08-6.21 (m, 1H of MA

-1 units, CH2CHCOO-), 6.38-6.50 (m, 1H of MA units, CH2CHCOO-); FT-IR (cm ) 3293,

2944, 2865, 2096, 1725, 1635, 1470, 1419, 1396, 1367, 1295, 1244, 1191, 1105, 1046, 962,

o o o 732; DSC: Tg= -53.2 C, Tm1= 40.1 C, Tm2= 47.1 C.

6.3.6 Synthesis of RGD Peptides

Standard solid phase FMOC methodology was utilized to synthesize RGD possessing different reactive end groups: RGD-hydroxylamine, RGD-thiol, RGD-alkyne,

RGD-azide, and RGD-cyclooctyne. The reactive chemical handles were attached at the

N-terminus during the synthesis. Peptides were cleaved from resin using standard 197 conditions (45 min, 95% trifluoroacetic acid (TFA), 2.5% Triisopropylsilane (TIPS), 2.5% water (by volume)) and precipitated in cold diethyl ether. The crude solid product was isolated by centrifuge, triturated, washed twice with diethyl ether and dialyzed in deionized water (molecular weight (MW) cut off 500 g/mol, cellulose membrane, Pierce), followed by lyophilization. Products were isolated as a white powder. The molecular masses were verified by ESI.

RGD-alkyne: After the sequence GRGDS, 5-hexynoic acid was used to couple to the N-terminus before cleavage. (Calculated MW(+H+) 585.2 g/mol, ESI 585.0 g/mol).

RGD-azide: Bromohexanoic acid was coupled to GRGDS at the N-terminus. After cleavage, Br terminated RGD was further derived with NaN3 (3 eq.) in aqueous environment. After dialysis and lyophilization, RGD-azide was collected as white solid.

(Calculated MW(+H+) 630.3 g/mol, ESI 630.1 g/mol).

RGD-thiol: The sequence of this peptide was CGRGDS. Thiol of cysteine provided as the functional group. (Calculated MW(+H+) 594.2 g/mol, ESI 594.2 g/mol).

Scheme 6.5: The synthesis route of RGD-cyclooctyne.

198

RGD-cyclooctyne: The Synthesis is described in Scheme 6.5. GRGDS with free amine at N-terminus (50 mg, 0.1 mmol, 1eq.), compound 6.4242 (75 mg, 0.2 mmol, 2 eq.) and TEA (55 uL, 0.4 mmol, 4 eq.) were dissolved in 4 mL DMSO. This solution was stirred at room temperature for 48h. After solvent removal by reduced pressure, solid residue was redissolved in 20 mL PBS buffer (x1, Ph 7.4). After centrifuge, upped layer clear solution was collected, followed by dialysis and lyophilization, yielding product as a white powder

(68mg, 91%) (Calculated MW(+H+) 737.3 g/mol, ESI 737.3 g/mol, Appendix Figure 49).

The downer solid layer in the centrifuge process was recovered as compound 6.4 and ready to use.

6.3.7 Polymer Coated QCM Sensors

SiO2 coated crystal sensor X301(5 MHz resonant frequency) were cleaned thoroughly using standard methods prior to spin coating. Polymer solutions were prepared by dissolving the functional polyester precursors (30 mg) in chloroform (2 mL) and filtered with 0.4 um PETF filter. The films were spun at 2000 rpm for 1 min and the acceleration time was 10 sec. The thickness of the thin films was measured using spectroscopic ellipsometry as described in Table 6.1.

Table 6.1: Thickness of Spinning coated polyester films

MA- MA- Polymer Ketone Alkyne Azide MA MA-Alkyne Alkyne- Azide Azide Thickness 129.8 120.2 122.5 127.1 122.1 119.2 123.5 (nm) . 199

6.3.8 Surface-RGD Conjugation Studies via QCM

All experiments were performed at 20 ±0.1 °C using flow rates of 0.1 mL/min. The concentration of the RGD peptide in buffer solutions was 1 mM. Acetate buffer of

298 CH3OONa/CH3OOH (100 mM, pH 4.5) containing 10 mM aniline catalyst was prepared for oxime ligation of keto-PCL, and phosphate buffered saline (PBS) solution (100 mM, pH 7.4) was used in the Michael-addition and SPAAC reactions. In the copper catalyzed “click” reaction, sodium ascorbate (2mM) and CuSO4 · 5H2O (0.5 mM) were added into the PBS solution as the Cu (I) catalyst supply. In a typical experiment, the polymer-coated QCM sensors were exposed to the respective buffer solution overnight to reach a hydrated equilibrium. The baseline was established over a 10 min interval. The reaction formulation (buffer solution plus peptide) was introduced using a 10 min continuous flow to make sure that the entire chamber was saturated with the reaction formulation. After 1 h, sensors were rinsed thoroughly with buffer solution until a plateau was reached.

6.3.9 Surface Chemistry Studies via 1H NMR

The reaction buffer solutions were identical to the QCM experiments, and the concentrations of the small molecules were fixed at 10 mM. Polymer films were prepared via solution casting of polymer dichloromethane solution (5 mg/mL) onto aluminum pan, and following drying, were immersed in the reaction formulation. After 1h, films were washed extensively with a buffer solution and pure water in sequence, and dried in vacuum for further characterization. The o-(prop-2-yn-1-yl)hydroxylamine (intermediates 6.3),

2-hydroxy-1-ethanethiol, 3-azidopropan-1-ol and propargyl alcohol were chosen as small 200 molecules to investigate the surface reactions of keto-PCL, PCL-MA, PCL-alkyne, and

PCL-azide respectively.

6.4 Results and discussion

6.4.1 Synthesis of Functionalized PCL

Scheme 6.4 shows the synthetic strategy employed in this study. Three bi-functional intermediates were synthesized, which includes an aminooxy group, and an additional

“clickable” functional group: alkyne, azide, or methyl acrylate (MA). Using an oxime ligation, a number of functional intermediates were grafted to keto-PCL, resulting in polyesters containing different “clickable” groups on the pendant side chains. The conversion reaction was completed in several hours using a stoichiometric feed ratio between the ketone and aminooxy groups at room temperature. Using the keto-PCL precursor, four mono-functionalized polyesters were prepared: keto-PCL for oxime ligation, PCL-azide/alkyne for Huisgen 1, 3-dipolar cycloaddition, and PCL-MA for

Michael-addition. Their chemical structures were first characterized with 1H NMR as shown in Figure 6.1. The spectra of keto-PCL is shown in Figure 6.1(a) and is identical to spectra reported previously304,400. The reactions were confirmed using the chemical shifts of H at “a” position 4.35 ppm in Figure 6.1(a) to peak “i” 4.25 ppm in Figure 6.1(b), and peaks “c” and “d” shifted to “j”. In Figure 6.1(b), the resonances at 4.62 ppm (h) and 2.44 ppm (k) are assigned to the alkyne bond. The chemical shift at 3.20 ppm (m) in Figure

6.1(c) was assigned to the methylene bonded to the azide. In Figure 6.1(d) three resonance peaks from 5.85 ppm to 6.45 ppm (o, p, and q) come from alkene group in PCL-

MA. MALDI-TOF was utilized to further confirm the chemical composition of PCL-MA. 201

From the mass spectra (Appendix Figure 52), there are only two repeat units: one is 114.1 from the CL unit, and the other is 299.1 exactly from OPD-MA unit. FT-IR also confirmed the existence of “clickable” groups in PCL copolymers, as described in Figure 6.2. In the

FT-IR spectra of PCL-alkyne in Figure 6.2(a), the peak at 3264 cm-1 corresponds to the C-

H stretch of triple bond in PCL-alkyne. The N3 stretch in PCL-azide is reported at 2096 in

Figure 6.2(b). Characteristic absorption peaks of the alkene groups in PCL-MA are present at 3077 cm-1 and 1642 cm-1 in Figure 6.2(c). Compared to keto-PCL, the glass transition temperature of PCL-alkyne, PCL-azide and PCL-MA does not change

o significantly. However, the melting temperature (Tm) was around 40 C, much lower than

o that of keto-PCL, which was 60 C. The suppression of Tm was attributed to the decrease in chain packing due to existence of pendant side chains.

202

1 Figure 6.1. H NMR spectrum of mono-functionalized PCL in CDCl3: (a) PCL-Ketone, (b)

PCL-Alkyne, (c) PCL-Azide, (d) PCL-MA. The oxime ligation is quantitative and proceeds under reaction conditions that minimize the degradation of the polymer. Peaks of CDCl3 and water are marked with asterisk.

Three multi-functionalized PCL were also prepared via a “clickable group conversion” method for evaluation of the reaction efficiency: PCL-MA-alkyne, PCL-MA- azide, and PCL-MA-alkyne-azide. The resulting chemical structures are confirmed with

1H NMR as shown in Figure 6.3. Each of the three multi-functionalized polymers contained a MA unit. The characteristic alkene peaks of MA are marked inside of the red dashed lines between 5.85 ppm and 6.45 ppm. In Figure 6.3(a) and 6.3(b), PCL-MA-alkyne and

PCL-MA-alkyne-azide both possess alkyne group signatures as shown by the singlet at 203

2.45 ppm marked inside of the green dashed line. In Figure 6.3(b) and 6.3(c), both polymers,

PCL-MA-alkyne-azide and PCL-MA-azide, show azide groups with a chemical shift at

3.20 ppm assigned to the methylene group adjacent to the azide.

Figure 6.2. FT-IR spectra of mono-functionalized polymers indicate the presence of the

-1 functional groups. (a) PCL-Alkyne: -C≡C-H stretch 3264 cm ; (b) PCL-Azide: -N3 stretch

2096 cm-1; (c) PCL-MA: C=C-H stretch 3077 cm-1 and C=C stretch 1642 cm-1.

The synthesis of mono- and multi-functionalized PCL required one synthesized polymer. A distinct advantage to this method is the oxime ligation was conducted postpolymerization in one pot under mild reaction conditions. Furthermore the content of each functional groups in the final polymer can be tuned using the hydroxyl amine feed ratio. This “clickable group conversion” method avoids the challenges associated with new monomer design and synthesis, which involves the optimization of polymerization conditions, including catalyst, initiator, temperature, solvent conditions, etc. Using a 204

“clickable group conversion” strategy, a diverse library of functionalized PCL could be synthesized using a combination of aminooxy intermediates.

1 Figure 6.3. H NMR spectra of functionalized PCL polymers in CDCl3. Peaks between 5.75 ppm and 6.50 ppm are assigned to alkene group of MA marked within red dash line. (a)

PCL-MA-alkyne: characteristic peak of alkyne 2.45 ppm marked within green dash line;

(b) PCL-MA-alkyne-azide: characteristic peak of methyl group bonded to azide 3.25 ppm marked within blue dash line. (c) PCL-MA-azide.

6.4.2 Conjugation of RGD to Thin Film of Mono-functionalized PCL

A RGD peptide was chosen as a model peptide model to investigate the reactivity of biomolecules immobilization to polyester surfaces. RGD mediates integrin based cell adhesion526 and it is still bioavailable when tethered to surfaces120,527 and polymers216,522,528.

Using a quartz crystal microbalance (QCM) monitoring experiment, the reaction efficiency of five independent reactive peptide-polymer pairs was investigated. Control experiments 205 were conducted to assess any physical absorption of RGD in each pairing. In Figure 6.4 the frequency shifts of the experimental pairings (black line) and controls (colored lines) are shown. A decrease in frequency during the experiment indicates an increase in mass on the surface of the QCM sensor. The increase in mass correlates directly with peptide functionalization and/or protein absorption.529-531

Table 6.2. Summary of (a) QCM frequency shift (△f) after peptide conjugation to mono- functionalized polymers surface, (b) Amount of covalently bound RGD calculated from △f based on Sauerbrey model, (c) reaction efficiency was calculated by the ratio between reacted functional groups and the whole amount in entire film; (d) change of contact angle

(△θ) after peptide conjugation.

PCL- PCL- PCL-Azide PCL-MA Ketone Alkyne CuAAC SPAAC △fa (-Hz) 10.0±1.1 29.8±3.5 47.8±1.9 24.5±1.9 42.7±5.7 RGDb (nmol) 0.5±0.1 1.4±0.2 2.4±0.1 1.2±0.1 1.6±0.2 Efficiency c (%) 3.5±0.4 9.5±1.1 16.4±0.6 7.9±0.6 11.1±1.5 △θd (-o) 4.1±1.0 10.9±2.9 15.8±2.1 9.5±0.9 10.2±1.8

206

Figure 6.4. QCM frequency shift: (a) RGD-aminooxy and PCL-Ketone; (b) RGD-thiol and

PCL-MA; (c) RGD-azide and PCL-Alkyne, (d) RGD-alkyne and PCL-Azide via CuAAC;

(e) RGD-cyclooctyne and PCL-Azide via SPAAC. 207

After the introducing of functionalized RGD buffer solution to the functionalized

PCL-coated QCM modules, the experimental frequency shifts decreased rapidly indicating the chemical reactions happened almost immediately. For all five experimental polymer pairings, the extent of the reaction reached ~ 80% during the first 20 min. The reactions reached their maximum conversion in about 1h. This was determined by the establishment of a frequency plateau that corresponds to the complete surface conjugation of the available functional groups. The frequency increased slightly during the buffer rinse after the conjugation plateau and indicated that a small amount of residual physically absorbed peptide was removed. In the control experiments (colored lines), the small frequency drop (2~5 Hz) due to physical absorption is much smaller than measured in the conjugation reactions. Figure 6.4(a) shows the conjugation reaction between keto-PCL and RGD-hydroxylamine. The control experiment uses a sensor coated with plain PCL instead of keto-PCL. Figure 6.4(b) shows the reaction of PCL-MA and RGD-thiol system and two control experiments: unfunctionalized PCL was used to quantify the non-specific absorption of RGD-thiol peptide; and a RGD peptide with free amine at the N-terminus was used to exclude any reaction between the amine and MA alkene group under the QCM incubation conditions. The frequency drops in each control is much smaller than the corresponding reaction experiment, indicating the covalent attachment of the peptide.

Figure 6.4(c) and 6.4(d) described the CuAAC conjugation between PCL-alkyne and RGD- azide, and PCL-azide and RGD-alkyne. In control experiments, there was no copper sulfate in the control buffer formulation. The frequency shift was very small without Cu (I) catalyst.

Figure 6.4(e) shows the conjugation behavior between PCL-azide and RGD-cyclooctyne.

In another control experiment unfunctionalized PCL was used instead of alkyne 208 functionalized PCL. The small frequency decrease of the control experiment was attributed to physical absorption, and the sharp drop in the experimental pair was due to a covalent reaction between the azide and the strained cyclooctyne via SPAAC.

Based on equation (6.1) from the Sauerbrey Model, changes in frequency are converted into an absorbed mass and correspond to a molar amount, as listed in Table 6.2.

While QCM provides a measure of how much peptide is covalently tethered to the surface, the exact amount of residual surface available functional groups is unknown and therefore the reaction efficiency of surface immobilization cannot be determined. The reported reaction efficiency was calculated using the ratio between the converted functional groups and the total content in the whole film. According to the QCM results, the amounts of surface conjugated peptides and the respective reaction efficiencies are different for each functionalized polymer and peptide system, though the bulk contents of functional groups are the same in each polymer. Many factors influence the rate and extent of reaction on a surface especially steric hindrance, reactivity differences and diffusion limitations of some reagents. In the five pairings we described, keto-PCL reacted with the smallest amount of

RGD and the reaction efficiency, 3.5%, is also lowest. Steric hindrance can account for this as the side chain of keto-PCL is the smallest, and the opportunity for the peptide to be in the correct orientation with the ketone group is reduced. The PCL-azide system reacted with the peptide derived with dibenzyl ring fused cyclooctyne (DIBO) using SPACC resulted in higher peptide conversion than traditional CuAAC, even though the N-terminus of RGD-cyclooctyne is more bulky than RGD-alkyne. The copper mediated cycloaddition was recently shown to involve two copper atoms532 and the resulting CuAAC intermediate is a bulky complex between copper (I) and the alkyne , which results in a 209 smaller amount of peptide reaction. Second, the DIBO group is very lipophilic and as a result may be driven to the surface of the polymer film, resulting in more opportunities for the chemical reaction to occur.533 The PCL-MA derivatized film reacted with more peptide through the Michael-addition than PCL-azide via CuAAC due primarily to steric hindrance from Cu (I)-alkyne complex. Interestingly, the PCL-alkyne resulted in the highest RGD immobilization and was double the amount reported in the reverse of PCL- azide pairing even though CuAAC intermediates were necessary in both systems. It is possible that more alkyne groups are present on the surface of PCL-alkyne film but very difficult to confirm this. Another possible explanation is related to the steric hindrance as described above and associated diffusion.

Figure 6.5. Static contact angle of PCL-alkyne thin film before (a) and after (b) RGD-azide peptide conjugation. Contact angle drops after reaction indicates the increase of surface hydrophylicity due to water-soluble RGD attachment to film.

Together with QCM data, contact angle (CA) measurements were used to confirm the changes in hydrophobicity of polymer surfaces. The PCL film surfaces are hydrophobic with a contact angel around 78o. Following the peptide conjugation, the polymer surfaces are more hydrophilic due to the peptide attachment, which is confirmed with a smaller contact angle (Table 6.2, Figure 6.5). For example, the contact angle of PCL- 210 alkyne film dropped from 78o to 63o following the peptide reaction. The magnitude of contact angle decrease correlates directly with the amount of peptides tethered to the surface, and agrees with the data acquired by QCM.

6.4.3 Conjugation of Small Molecule Probes to PCL Films

To further confirm the availability of functional groups on polymer surface for chemical reactions, a series of small molecules were chosen to react with the polymer films as model reactions. The 1H NMR spectra are shown in Figure 6.6, where the red line denoted the labeled chemical shifts following the reactions. O-(prop-2-yn-1- yl)hydroxylamine (intermediate 6.3) was used to react with keto-PCL. A close comparison of the two spectra in Figure 6.6(a) shows a new resonance at 4.62 ppm that is assigned to alkyne bond, and a small wide shoulder between 4.2 ppm and 4.3 ppm that proves the occurrence of the oxime ligation.

In Figure 6.6(b) that shows the PCL-alkyne precursor, the evidence for a surface chemical reaction is obvious. In addition to peaks labeled b, c and d as noted in Figure

6.6(b), small peak “a” at 7.8 ppm is assigned to the H of the triazole ring following the

CuAAC cycloaddition. The decrease of the stretch associated with the alkyne was also observed via FT-IR (Appendix Figure 51).

Similarly in Figure 6.6(c) of PCL-Azide, the small peak at 7.8 ppm following the reaction corresponds to the triazole H. Peak “c” at 4.45 ppm is assigned to the methyl groups after reaction, which shifts downfield from 3.2 ppm in original PCL-azide polymer.

Meanwhile, the corresponding decrease of azide stretch in the FT-IR (Appendix Figure 50) confirms the CuAAC reaction. 211

1 Figure 6.6. H NMR spectra of solution cast films in CDCl3 before (red) and after (black) surface reactions: (a) Keto-PCL with o-(prop-2-yn-1-yl)hydroxylamine (intermediate 6.3);

(b) PCL-alkyne with 3-azidopropan-1-ol; (c) PCL-azide with propargyl alcohol; (4) PCL-

MA with 2-hydroxyl-1-ethanethiol.

In PCL-MA system of Figure 6.6(d), there is a new wide multi-peak resonance between 2.75 ppm to 3.75 ppm after the reaction, which is assigned to the product of

Michael-addition between 2-hydroxy-1-ethanethiol and MA. The peak at 3.5 ppm is assigned to methylene group adjacent to the hydroxyl.

212

The surface conjugation process of small molecules to polymer film happened in a similar way to that of peptide-polymer film conjugation. Both of them occurred via “click” reactions. In the QCM experiment, only peptide with corresponding reactive sites can bond the surface resulting the big decrease of frequency; and in the study of small molecular probes, the NMR signal change before and after conjugation demonstrated the occurrence of “click” reactions.

6.3.4 Cascade Conjugation of RGD to Multi-functionalized PCL Thin Films

QCM was used to monitor the peptide conjugation via orthogonal “click” chemistry to thin film surface of multi-functionalized polymer in a series of cascading sequential reactions. In Figure 6.7(a) the PCL-MA-Alkyne system, a Michael-addition was designed to perform the initial reaction followed by CuAAC reaction. After introducing the RGD- thiol, the frequency decreased quickly by 15Hz due to peptide conjugation to the MA units.

Compared to the frequency shift of the mono-functionalized PCL-MA described above this shift is smaller because the MA content in PCL-MA-alkyne is only 5%, which is half of that in the PCL-MA polymer. Following the Michael-addition the sensors were rinsed thoroughly with buffer solution until equilibrium, and then reaction formulation of RGD- azide for the second conjugation was loaded into the QCM chamber. Following the addition of azide RGD solution, the frequency again dropped quickly, indicating the covalent

CuAAC addition of RGD to the polymer surface. Figure 6.7(b) shows the frequency behavior in PCL-MA-azide system, which is very similar to that in PCL-MA-alkyne system. The initial Michael-addition again showed a sharp frequency drop followed by the

213

SPAAC reaction. In the two systems, both the copper catalyzed and the strain promoted

“click” reactions are available for reaction following the Michael-addition.

Figure 6.7. The shift in QCM frequency for the cascading conjugation between peptide and surface of multi-functionalized polymer thin film. (a) PCL-MA-alkyne: Michael addition between RGD-thiol and MA units first, CuAAC between RGD-azide and alkyne units second; (b) PCL-MA-azide: Michael addition between RGD-thiol and MA units first,

SPAAC between RGD-cyclooctyne and azide units second; (c) PCL-MA-alkyne-azide:

Michael addition between RGD-thiol and MA units first, SPAAC between RGD- cyclooctyne and azide units second, CuAAC between RGD-azide and alkyne units third.

In the tri-functionalized polymer PCL-MA-Alkyne-Azide, the reactions follow a predetermined sequence: Michael-Addition, SPAAC, and CuAAC respectively. Figure

6.7(c) shows the obvious irreversible frequency shift at each step following the peptide addition indicating the successful conjugation of RGD to the film surface. In all three multi- functionalized PCL systems, the Michal-Addition between MA double bond and thiol group was active only when it was conducted at the first step. If CuAAC or SPAAC was performed first, there was no frequency shift attributed to the Michael addition in subsequent reactions. Figure 6.8 shows a small frequency shift in the second step after

214

RGD-thiol, following the initial SPAAC reaction. This distinction could be due to a number of factors such as a change in surface micro-environment, since MA-thiol conjugation is pH influenced534,535; or the interaction between triazole and MA may deactivate MA due to electron-density increase of MA double bond due to triazole electron donation536. The precise reason remains unknown at this time.

Figure 6.8. The shift in QCM frequency in the PCL-MA-azide system. In the first step

(black line) RGD-cyclooctyne for SPAAC was introduced. A significant decrease of frequency indicates the extent of a chemical reaction. In the second step (red line) RGD- thiol for Michael Addition was introduced. A slight drop of frequency indicates that little to no chemical reaction occurred.

6.5 Conclusion

In summary, a series of mono and multi-functionalized polyesters were synthesized by grafting “clickable” groups to PCL-co-OPD (keto-PCL) using an oxime ligation at room temperature. The materials demonstrated the ability to immobilize bioactive peptides to the 215 polymer surface using several highly efficient (“click”) reactions. The extent of the individual reactions was confirmed by Quartz Crystal Microbalance (QCM) and contact angle experiments. 1H NMR and FT-IR were used to prove the availability of “clickable” groups on the surface of the polymer film. Using the reactivity differences of the pendent side chain groups, a cascade of orthogonal reactions was also demonstrated. Three multifunctional polymers precursors were prepared for the sequential immobilization of peptides on the polymer surface. We further determined that the Michel-addition must occur first in the cascading approach. Changing the order of the reaction sequence was not possible in using these reactions. In general, the described methodology provides a tool to synthesize functional PCL bearing different “clickable” groups. This new system is likely to find applications in the biomolecular conjugation of peptides and drugs.451,518

6.6 Acknowledgements

The authors gratefully acknowledge financial support from the National Science

Foundation (DMR-1105329) and the Akron Functional Materials Center. The authors also acknowledge the use of a Quartz Crystal Microbalance which was acquired as part of an NSF-MRI award (DMR-1126544). The authors gratefully acknowledge mass spectrometry confirmation of our peptide precursors from Ms Kai Guo and Professor Chrys

Wesdemiotis at The University of Akron.

216

REFERENCES

(1) The Williams Dictionary of Biomaterials; Liverpool University Press, 1999.

(2) Hench, L. L.; Polak, J. M. Science 2002, 295, 1014.

(3) Hench, L. Science 1980, 208, 826.

(4) Navarro, M.; Michiardi, A.; Castaño, O.; Planell, J. A. Journal of The Royal Society Interface 2008, 5, 1137.

(5) Nair, L. S.; Laurencin, C. T. Progress in Polymer Science 2007, 32, 762.

(6) Kohn, J. Nature Materials 2004, 3, 745.

(7) Zhang, S. Nat Biotech 2003, 21, 1171.

(8) Karageorgiou, V.; Kaplan, D. Biomaterials 2005, 26, 5474.

(9) Cen, L.; Liu, W.; Cui, L.; Zhang, W.; Cao, Y. Pediatr Res 2008, 63, 492.

(10) Mitragotri, S.; Lahann, J. Nat Mater 2009, 8, 15.

(11) Albertsson, A.-C.; Varma, I. In Degradable Aliphatic Polyesters; Springer Berlin Heidelberg: 2002; Vol. 157, p 1.

(12) Okada, M. Progress in Polymer Science 2002, 27, 87.

(13) Edlund, U.; Albertsson, A. C. Advanced Drug Delivery Reviews 2003, 55, 585.

(14) Löfgren, A.; Albertsson, A.-C.; Dubois, P.; Jérôme, R. Journal of Macromolecular Science, Part C 1995, 35, 379.

(15) Labet, M.; Thielemans, W. Chemical Society Reviews 2009, 38, 3484.

(16) Vert, M. Biomacromolecules 2004, 6, 538.

(17) Göpferich, A. Macromolecules 1997, 30, 2598.

217

(18) Middleton, J. C.; Tipton, A. J. Biomaterials 2000, 21, 2335.

(19) Farng, E.; Sherman, O. Arthroscopy: The Journal of Arthroscopic & Related Surgery 2004, 20, 273.

(20) Chujo, K.; Kobayashi, H.; Suzuki, J.; Tokuhara, S.; Tanabe, M. Die Makromolekulare Chemie 1967, 100, 262.

(21) Chujo, K.; Kobayashi, H.; Suzuki, J.; Tokuhara, S. Die Makromolekulare Chemie 1967, 100, 267.

(22) Mikos, A. G.; Temenoff, J. S. Electronic Journal of Biotechnology 2000, 3, 23.

(23) Maurus, P. B.; Kaeding, C. C. Operative Techniques in Sports Medicine 2004, 12, 158.

(24) Ashammakhi, N.; Rokkanen, P. Biomaterials 1997, 18, 3.

(25) Chu, C. C. Journal of Biomedical Materials Research 1981, 15, 19.

(26) Chu, C. C. Journal of Applied Polymer Science 1981, 26, 1727.

(27) Chu, C. C. Journal of Biomedical Materials Research 1981, 15, 795.

(28) P. A. Gunatillake, R. A. Europe Cell and Materials 2003, 5, 1.

(29) Törmälä, P.; Vasenius, J.; Vainionpää, S.; Laiho, J.; Pohjonen, T.; Rokkanen, P. Journal of Biomedical Materials Research 1991, 25, 1.

(30) Böstman, O. M.; Päivärinta, U.; Partio, E.; Manninen, M.; Vasenius, J.; Majola, A.; Rokkanen, P. Clinical Orthopaedics and Related Research 1992, 285, 263.

(31) Päivärinta, U.; Böstman, O.; Majola, A.; Toivonen, T.; Törmälä, P.; Rokkanen, P. Arch Orthop Trauma Surg 1993, 112, 71.

(32) Gilding, D. K.; Reed, A. M. Polymer 1979, 20, 1459.

(33) Rokkanen, P. U.; Böstman, O.; Hirvensalo, E.; Mäkelä, E. A.; Partio, E. K.; Pätiälä, H.; Vainionpää, S.; Kimmo, V.; Törmälä, P. Biomaterials 2000, 21, 2607.

(34) Hiraoka., Y.; Kimura., Y.; Ueda., H.; Tabata., Y. Tissue Engineering 2003, 9, 1101.

(35) Boland, E. D.; Wnek, G. E.; Simpson, D. G.; Pawlowski, K. J.; Bowlin, G. L. Journal of Macromolecular Science, Part A 2001, 38, 1231.

218

(36) Day, R. M.; Boccaccini, A. R.; Shurey, S.; Roether, J. A.; Forbes, A.; Hench, L. L.; Gabe, S. M. Biomaterials 2004, 25, 5857.

(37) Mooney, D. J.; Mazzoni, C. L.; Breuer, C.; McNamara, K.; Hern, D.; Vacanti, J. P.; Langer, R. Biomaterials 1996, 17, 115.

(38) Mehta, R.; Kumar, V.; Bhunia, H.; Upadhyay, S. N. Journal of Macromolecular Science, Part C 2005, 45, 325.

(39) Garlotta, D. Journal of Polymers and the Environment 2001, 9, 63.

(40) Lim, L. T.; Auras, R.; Rubino, M. Progress in Polymer Science 2008, 33, 820.

(41) Miller, R. A.; Brady, J. M.; Cutright, D. E. Journal of Biomedical Materials Research 1977, 11, 711.

(42) Walton, M.; Cotton, N. J. Journal of Biomaterials Applications 2007, 21, 395.

(43) Pihlajamäki, H.; Böstman, O.; Tynninen, O.; Laitinen, O. Bone 2006, 39, 932.

(44) Luckachan, G. E.; Pillai, C. K. S. Carbohydrate Polymers 2006, 64, 254.

(45) Kim, K.; Yu, M.; Zong, X.; Chiu, J.; Fang, D.; Seo, Y.-S.; Hsiao, B. S.; Chu, B.; Hadjiargyrou, M. Biomaterials 2003, 24, 4977.

(46) Böstman, O.; Pihlajamäki, H. Biomaterials 2000, 21, 2615.

(47) Bergsma, J. E.; de Bruijn, W. C.; Rozema, F. R.; Bos, R. R. M.; Boering, G. Biomaterials 1995, 16, 25.

(48) Zilberman, M.; Nelson, K. D.; Eberhart, R. C. Journal of Biomedical Materials Research Part B: Applied Biomaterials 2005, 74B, 792.

(49) Cooper, J. A.; Lu, H. H.; Ko, F. K.; Freeman, J. W.; Laurencin, C. T. Biomaterials 2005, 26, 1523.

(50) Yang, F.; Murugan, R.; Ramakrishna, S.; Wang, X.; Ma, Y. X.; Wang, S. Biomaterials 2004, 25, 1891.

(51) Hu, J.; Sun, X.; Ma, H.; Xie, C.; Chen, Y. E.; Ma, P. X. Biomaterials 2010, 31, 7971.

(52) Sabir, M.; Xu, X.; Li, L. J Mater Sci 2009, 44, 5713.

219

(53) Smith Callahan, L. A.; Xie, S.; Barker, I. A.; Zheng, J.; Reneker, D. H.; Dove, A. P.; Becker, M. L. Biomaterials 2013, 34, 9089.

(54) PiŞKiN, E.; KaiTiAn, X.; Denkbaş, E. B.; Küçükyavuz, Z. Journal of Biomaterials Science, Polymer Edition 1996, 7, 359.

(55) Vester, H.; Wildemann, B.; Schmidmaier, G.; Stöckle, U.; Lucke, M. Injury 2010, 41, 1053.

(56) Du, K.; Shi, X.; Gan, Z. Langmuir 2013, 29, 15293.

(57) Passerini, N.; Craig, D. Q. M. Journal of Controlled Release 2001, 73, 111.

(58) Makadia, H. K.; Siegel, S. J. Polymers 2011, 3, 1377.

(59) Klose, D.; Siepmann, F.; Elkharraz, K.; Siepmann, J. International Journal of Pharmaceutics 2008, 354, 95.

(60) Danhier, F.; Ansorena, E.; Silva, J. M.; Coco, R.; Le Breton, A.; Préat, V. Journal of Controlled Release 2012, 161, 505.

(61) Pan, Z.; Ding, J. Interface Focus 2012, 2, 366.

(62) Vozzi, G.; Flaim, C. J.; Bianchi, F.; Ahluwalia, A.; Bhatia, S. Materials Science and Engineering: C 2002, 20, 43.

(63) Shin, H. J.; Lee, C. H.; Cho, I. H.; Kim, Y.-J.; Lee, Y.-J.; Kim, I. A.; Park, K.-D.; Yui, N.; Shin, J.-W. Journal of Biomaterials Science, Polymer Edition 2006, 17, 103.

(64) Munirah, S.; Kim, S.; Ruszymah, B.; Khang, G. Journal of European Cells and Materials 2008, 15, 41.

(65) Engelberg, I.; Kohn, J. Biomaterials 1991, 12, 292.

(66) Woodruff, M. A.; Hutmacher, D. W. Progress in Polymer Science 2010, 35, 1217.

(67) Sun, H.; Mei, L.; Song, C.; Cui, X.; Wang, P. Biomaterials 2006, 27, 1735.

(68) Woodward, S. C.; Brewer, P. S.; Moatamed, F.; Schindler, A.; Pitt, C. G. Journal of Biomedical Materials Research 1985, 19, 437.

(69) Pitt, G. G.; Gratzl, M. M.; Kimmel, G. L.; Surles, J.; Sohindler, A. Biomaterials 1981, 2, 215.

220

(70) Sinha, V. R.; Bansal, K.; Kaushik, R.; Kumria, R.; Trehan, A. International Journal of Pharmaceutics 2004, 278, 1.

(71) Gaucher, G.; Dufresne, M.-H.; Sant, V. P.; Kang, N.; Maysinger, D.; Leroux, J.-C. Journal of Controlled Release 2005, 109, 169.

(72) Gref, R.; Lück, M.; Quellec, P.; Marchand, M.; Dellacherie, E.; Harnisch, S.; Blunk, T.; Müller, R. H. Colloids and Surfaces B: Biointerfaces 2000, 18, 301.

(73) Coombes, A. G. A.; Rizzi, S. C.; Williamson, M.; Barralet, J. E.; Downes, S.; Wallace, W. A. Biomaterials 2004, 25, 315.

(74) Rai, B.; Teoh, S. H.; Hutmacher, D. W.; Cao, T.; Ho, K. H. Biomaterials 2005, 26, 3739.

(75) Frazza, E. J.; Schmitt, E. E. Journal of Biomedical Materials Research 1971, 5, 43.

(76) Ng, K. W.; Achuth, H. N.; Moochhala, S.; Lim, T. C.; Hutmacher, D. W. Journal of Biomaterials Science, Polymer Edition 2007, 18, 925.

(77) Lowry, K. J.; Hamson, K. R.; Bear, L.; Peng, Y. B.; Calaluce, R.; Evans, M. L.; Anglen, J. O.; Allen, W. C. Journal of Biomedical Materials Research 1997, 36, 536.

(78) Lo, H.-Y.; Kuo, H.-T.; Huang, Y.-Y. Artificial Organs 2010, 34, 648.

(79) Alves da Silva, M. L.; Martins, A.; Costa-Pinto, A. R.; Costa, P.; Faria, S.; Gomes, M.; Reis, R. L.; Neves, N. M. Biomacromolecules 2010, 11, 3228.

(80) Kweon, H.; Yoo, M. K.; Park, I. K.; Kim, T. H.; Lee, H. C.; Lee, H.-S.; Oh, J.-S.; Akaike, T.; Cho, C.-S. Biomaterials 2003, 24, 801.

(81) Yoshimoto, H.; Shin, Y. M.; Terai, H.; Vacanti, J. P. Biomaterials 2003, 24, 2077.

(82) Zhou, Y.; Hutmacher, D. W.; Varawan, S.-L.; Lim, T. M. Polymer International 2007, 56, 333.

(83) Lei, Y.; Rai, B.; Ho, K. H.; Teoh, S. H. Materials Science and Engineering: C 2007, 27, 293.

(84) Woodruff, M. A.; Lange, C.; Chen, F.; Fratzl, P.; Hutmacher, D. W. Advanced Healthcare Materials 2013, 2, 546.

(85) Lam, C. X. F.; Hutmacher, D. W.; Schantz, J.-T.; Woodruff, M. A.; Teoh, S. H. Journal of Biomedical Materials Research Part A 2009, 90A, 906. 221

(86) Lam, C. X. F.; Teoh, S. H.; Hutmacher, D. W. Polymer International 2007, 56, 718.

(87) Shung, A. K.; Timmer, M. D.; Jo, S.; Engel, P. S.; Mikos, A. G. Journal of Biomaterials Science, Polymer Edition 2002, 13, 95.

(88) Kasper, F. K.; Tanahashi, K.; Fisher, J. P.; Mikos, A. G. Nat. Protocols 2009, 4, 518.

(89) Kharas, G. B.; Kamenetsky, M.; Simantirakis, J.; Beinlich, K. C.; Rizzo, A.-M. T.; Caywood, G. A.; Watson, K. Journal of Applied Polymer Science 1997, 66, 1123.

(90) Temenoff, J. S.; Mikos, A. G. Biomaterials 2000, 21, 2405.

(91) DiCiccio, A. M.; Coates, G. W. Journal of the American Chemical Society 2011, 133, 10724.

(92) Wang, S.; Lu, L.; Yaszemski, M. J. Biomacromolecules 2006, 7, 1976.

(93) He, S.; Timmer, M. D.; Yaszemski, M. J.; Yasko, A. W.; Engel, P. S.; Mikos, A. G. Polymer 2001, 42, 1251.

(94) Yaszemski, M. J.; Payne, R. G.; Hayes, W. C.; Langer, R.; Mikos, A. G. Biomaterials 1996, 17, 2127.

(95) Domb, A. J.; Manor, N.; Elmalak, O. Biomaterials 1996, 17, 411.

(96) Peter, S. J.; Kim, P.; Yasko, A. W.; Yaszemski, M. J.; Mikos, A. G. Journal of Biomedical Materials Research 1999, 44, 314.

(97) Suggs, L. J.; Mikos, A. G. Cell Transplantation 1999, 8, 345.

(98) Jo, S.; Engel, P. S.; Mikos, A. G. Polymer 2000, 41, 7595.

(99) Lee, K.-W.; Wang, S.; Fox, B. C.; Ritman, E. L.; Yaszemski, M. J.; Lu, L. Biomacromolecules 2007, 8, 1077.

(100) Cooke, M. N.; Fisher, J. P.; Dean, D.; Rimnac, C.; Mikos, A. G. Journal of Biomedical Materials Research Part B: Applied Biomaterials 2003, 64B, 65.

(101) Lan, P.; Lee, J.; Seol, Y.-J.; Cho, D.-W. J Mater Sci: Mater Med 2009, 20, 271.

(102) Timmer, M. D.; Carter, C.; Ambrose, C. G.; Mikos, A. G. Biomaterials 2003, 24, 4707.

222

(103) Fisher, J. P.; Vehof, J. W. M.; Dean, D.; van der Waerden, J. P. C. M.; Holland, T. A.; Mikos, A. G.; Jansen, J. A. Journal of Biomedical Materials Research 2002, 59, 547.

(104) Shi, X.; Hudson, J. L.; Spicer, P. P.; Tour, J. M.; Krishnamoorti, R.; Mikos, A. G. Nanotechnology 2005, 16, S531.

(105) He, S.; J. Yaszemski, M.; Yasko, A. W.; Engel, P. S.; Mikos, A. G. Biomaterials 2000, 21, 2389.

(106) Cai, Z.-Y.; Yang, D.-A.; Zhang, N.; Ji, C.-G.; Zhu, L.; Zhang, T. Acta Biomaterialia 2009, 5, 628.

(107) Peter, S. J.; Lu, L.; Kim, D. J.; Mikos, A. G. Biomaterials 2000, 21, 1207.

(108) Hedberg, E. L.; Tang, A.; Crowther, R. S.; Carney, D. H.; Mikos, A. G. Journal of Controlled Release 2002, 84, 137.

(109) Hacker, M. C.; Haesslein, A.; Ueda, H.; Foster, W. J.; Garcia, C. A.; Ammon, D. M.; Borazjani, R. N.; Kunzler, J. F.; Salamone, J. C.; Mikos, A. G. Journal of Biomedical Materials Research Part A 2009, 88A, 976.

(110) Payne, R. G.; McGonigle, J. S.; Yaszemski, M. J.; Yasko, A. W.; Mikos, A. G. Biomaterials 2002, 23, 4381.

(111) Fisher, J. P.; Holland, T. A.; Dean, D.; Mikos, A. G. Biomacromolecules 2003, 4, 1335.

(112) Timmer, M. D.; Ambrose, C. G.; Mikos, A. G. Biomaterials 2003, 24, 571.

(113) Sun, H.; Meng, F.; Dias, A. A.; Hendriks, M.; Feijen, J.; Zhong, Z. Biomacromolecules 2011, 12, 1937.

(114) Cappello, J.; Crissman, J.; Dorman, M.; Mikolajczak, M.; Textor, G.; Marquet, M.; Ferrari, F. Biotechnology Progress 1990, 6, 198.

(115) Rehm, B. H. A. Nat Rev Micro 2010, 8, 578.

(116) Deming, T. J. Advanced Materials 1997, 9, 299.

(117) Kisiday, J.; Jin, M.; Kurz, B.; Hung, H.; Semino, C.; Zhang, S.; Grodzinsky, A. J. Proceedings of the National Academy of Sciences 2002, 99, 9996.

(118) Merrifield, R. B. Journal of the American Chemical Society 1963, 85, 2149.

223

(119) Ruoslahti, E. Annual Review of Cell and Developmental Biology 1996, 12, 697.

(120) Gallant, N. D.; Lavery, K. A.; Amis, E. J.; Becker, M. L. Advanced Materials 2007, 19, 965.

(121) Gazit, E. Chemical Society Reviews 2007, 36, 1263.

(122) Maude, S.; Tai, L. R.; Davies, R. P. W.; Liu, B.; Harris, S. A.; Kocienski, P. J.; Aggeli, A. In Peptide-Based Materials; Deming, T., Ed.; Springer Berlin Heidelberg: 2012; Vol. 310, p 27.

(123) Deming, T. J. Journal of Polymer Science Part A: Polymer Chemistry 2000, 38, 3011.

(124) Cheng, J.; Deming, T. In Peptide-Based Materials; Deming, T., Ed.; Springer Berlin Heidelberg: 2012; Vol. 310, p 1.

(125) Poché, D. S.; Thibodeaux, S. J.; Rucker, V. C.; Warner, I. M.; Daly, W. H. Macromolecules 1997, 30, 8081.

(126) Huang, J.; Habraken, G.; Audouin, F.; Heise, A. Macromolecules 2010, 43, 6050.

(127) Yu, M.; Nowak, A. P.; Deming, T. J.; Pochan, D. J. Journal of the American Chemical Society 1999, 121, 12210.

(128) Schappacher, M.; Soum, A.; Guillaume, S. M. Biomacromolecules 2006, 7, 1373.

(129) Caillol, S.; Lecommandoux, S.; Mingotaud, A.-F.; Schappacher, M.; Soum, A.; Bryson, N.; Meyrueix, R. Macromolecules 2003, 36, 1118.

(130) Rypáček, F.; Pytela, J.; Kotva, R.; Škarda, V.; Cífková, I. Macromolecular Symposia 1997, 123, 9.

(131) Li, C. Advanced Drug Delivery Reviews 2002, 54, 695.

(132) Hayashi, T.; Tabata, Y.; Nakajima, A. Polym J 1985, 17, 463.

(133) Roweton, S.; Huang, S. J.; Swift, G. J Environ Polym Degr 1997, 5, 175.

(134) Uhrich, K. E.; Cannizzaro, S. M.; Langer, R. S.; Shakesheff, K. M. Chemical Reviews 1999, 99, 3181.

224

(135) Matsumura, Y.; Hamaguchi, T.; Ura, T.; Muro, K.; Yamada, Y.; Shimada, Y.; Shirao, K.; Okusaka, T.; Ueno, H.; Ikeda, M.; Watanabe, N. Br J Cancer 2004, 91, 1775.

(136) Edwards, D. A.; Hanes, J.; Caponetti, G.; Hrkach, J.; Ben-Jebria, A.; Eskew, M. L.; Mintzes, J.; Deaver, D.; Lotan, N.; Langer, R. Science 1997, 276, 1868.

(137) Singer, J. W.; De Vries, P.; Bhatt, R.; Tulinsky, J.; Klein, P.; Li, C.; Milas, L.; Lewis, R. A.; Wallace, S. Annals of the New York Academy of Sciences 2000, 922, 136.

(138) Singer, J. W.; Bhatt, R.; Tulinsky, J.; Buhler, K. R.; Heasley, E.; Klein, P.; de Vries, P. Journal of Controlled Release 2001, 74, 243.

(139) Patil, S.; Rhodes, D.; Burgess, D. AAPS J 2005, 7, E61.

(140) Jeong, J. H.; Park, T. G. Journal of Controlled Release 2002, 82, 159.

(141) Lee, Y.; Nam, H. Y.; Kim, J.; Lee, M.; Yockman, J. W.; Shin, S. K.; Kim, S. W. Mol Ther 2012, 20, 1360.

(142) Zhang, X.; Oulad-Abdelghani, M.; Zelkin, A. N.; Wang, Y.; Haîkel, Y.; Mainard, D.; Voegel, J.-C.; Caruso, F.; Benkirane-Jessel, N. Biomaterials 2010, 31, 1699.

(143) Nicol, F.; Wong, M.; MacLaughlin, F. C.; Perrard, J.; Wilson, E.; Nordstrom, J. L.; Smith, L. C. Gene Therapy 2002, 9, 1351.

(144) Sun, Y.; Tang, Y.; Chu, M.; Song, S.; Xin, Y. International Journal of Nanomedicine 2008, 3, 249.

(145) Cook, A. D.; Hrkach, J. S.; Gao, N. N.; Johnson, I. M.; Pajvani, U. B.; Cannizzaro, S. M.; Langer, R. Journal of Biomedical Materials Research 1997, 35, 513.

(146) Peng, H.; Xiao, Y.; Mao, X.; Chen, L.; Crawford, R.; Whittaker, A. K. Biomacromolecules 2008, 10, 95.

(147) Deming, T. J. Progress in Polymer Science 2007, 32, 858.

(148) Nowak, A. P.; Breedveld, V.; Pakstis, L.; Ozbas, B.; Pine, D. J.; Pochan, D.; Deming, T. J. Nature 2002, 417, 424.

(149) Deming, T. J. Soft Matter 2005, 1, 28.

(150) Bellomo, E. G.; Wyrsta, M. D.; Pakstis, L.; Pochan, D. J.; Deming, T. J. Nat Mater 2004, 3, 244.

225

(151) Quadir, M. A.; Martin, M.; Hammond, P. T. Chemistry of Materials 2013, 26, 461.

(152) Pitarresi, G.; Saiano, F.; Cavallaro, G.; Mandracchia, D.; Palumbo, F. S. International Journal of Pharmaceutics 2007, 335, 130.

(153) Otani, Y.; Tabata, Y.; Ikada, Y. Biomaterials 1998, 19, 2091.

(154) Sekine, T.; Nakamura, T.; Shimizu, Y.; Ueda, H.; Matsumoto, K.; Takimoto, Y.; Kiyotani, T. Journal of Biomedical Materials Research 2001, 54, 305.

(155) Waite, J. H.; Hansen, D.; Little, K. J Comp Physiol B 1989, 159, 517.

(156) WAITE, J. H.; TANZER, M. L. Science 1981, 212, 1038.

(157) Yamamoto, H.; Hayakawa, T. Polymer 1978, 19, 1115.

(158) Yu, M.; Deming, T. J. Macromolecules 1998, 31, 4739.

(159) Yamamoto, H.; Hayakawa, T. Biopolymers 1982, 21, 1137.

(160) Yamamoto, H.; Hayakawa, T. Biopolymers 1979, 18, 3067.

(161) Rodriguez-Galan, A.; Franco, L.; Puiggali, J. Polymers 2010, 3, 65.

(162) Pang, X.; Chu, C.-C. Biomaterials 2010, 31, 3745.

(163) Deng, M.; Wu, J.; Reinhart-King, C. A.; Chu, C.-C. Biomacromolecules 2009, 10, 3037.

(164) Guo, K.; Chu, C. C. Journal of Polymer Science Part A: Polymer Chemistry 2007, 45, 1595.

(165) Asín, L.; Armelin, E.; Montané, J.; Rodríguez-Galán, A.; Puiggalí, J. Journal of Polymer Science Part A: Polymer Chemistry 2001, 39, 4283.

(166) Karimi, P.; Rizkalla, A. S.; Mequanint, K. Materials 2010, 3, 2346.

(167) Guo, K.; Chu, C. C.; Chkhaidze, E.; Katsarava, R. Journal of Polymer Science Part A: Polymer Chemistry 2005, 43, 1463.

(168) Wu, D.-Q.; Wu, J.; Chu, C.-C. Soft Matter 2013, 9, 3965.

(169) Yamanouchi, D.; Wu, J.; Lazar, A. N.; Craig Kent, K.; Chu, C.-C.; Liu, B. Biomaterials 2008, 29, 3269. 226

(170) Horwitz, J. A.; Shum, K. M.; Bodle, J. C.; Deng, M.; Chu, C.-C.; Reinhart- King, C. A. Journal of Biomedical Materials Research Part A 2010, 95A, 371.

(171) Guo, K.; Chu, C.-C. Biomaterials 2007, 28, 3284.

(172) Pulapura, S.; Kohn, J. Biopolymers 1992, 32, 411.

(173) Ertel, S. I.; Kohn, J. Journal of Biomedical Materials Research 1994, 28, 919.

(174) Pulapura, S.; Li, C.; Kohn, J. Biomaterials 1990, 11, 666.

(175) Bourke, S. L.; Kohn, J. Advanced Drug Delivery Reviews 2003, 55, 447.

(176) Asikainen, A.; Pelto, M.; Noponen, J.; Kellomäki, M.; Pihlajamäki, H.; Lindqvist, C.; Suuronen, R. J Mater Sci: Mater Med 2008, 19, 53.

(177) Tangpasuthadol, V.; Pendharkar, S. M.; Kohn, J. Biomaterials 2000, 21, 2371.

(178) Tangpasuthadol, V.; Pendharkar, S. M.; Peterson, R. C.; Kohn, J. Biomaterials 2000, 21, 2379.

(179) Hooper, K. A.; Macon, N. D.; Kohn, J. Journal of Biomedical Materials Research 1998, 41, 443.

(180) Choueka, J.; Charvet, J. L.; Koval, K. J.; Alexander, H.; James, K. S.; Hooper, K. A.; Kohn, J. Journal of Biomedical Materials Research 1996, 31, 35.

(181) Magno, M. H. R.; Kim, J.; Srinivasan, A.; McBride, S.; Bolikal, D.; Darr, A.; Hollinger, J. O.; Kohn, J. Journal of Materials Chemistry 2010, 20, 8885.

(182) Yu, C.; Kohn, J. Biomaterials 1999, 20, 253.

(183) Yu, C.; Mielewczyk, S. S.; Breslauer, K. J.; Kohn, J. Biomaterials 1999, 20, 265.

(184) C, M.; R, D.; P, S.; VP, H.; J, K. Journal of Biomaterials Science Polymer Edition 2006, 17, 1039.

(185) Bourke, S. L.; Kohn, J.; Dunn, M. G. Tissue Engineering 2004, 10, 43.

(186) Sheihet, L.; Dubin, R. A.; Devore, D.; Kohn, J. Biomacromolecules 2005, 6, 2726.

(187) Ertel, S. I.; Kohn, J.; Zimmerman, M. C.; Parsons, J. R. Journal of Biomedical Materials Research 1995, 29, 1337. 227

(188) James, K.; Levene, H.; Russell Parsons, J.; Kohn, J. Biomaterials 1999, 20, 2203.

(189) Kolb, H. C.; Finn, M. G.; Sharpless, K. B. Angewandte Chemie International Edition 2001, 40, 2004.

(190) Iha, R. K.; Wooley, K. L.; Nyström, A. M.; Burke, D. J.; Kade, M. J.; Hawker, C. J. Chemical Reviews 2009, 109, 5620.

(191) Kolb, H. C.; Sharpless, K. B. Drug Discovery Today 2003, 8, 1128.

(192) Huisgen, R. Pure and Applied Chemistry 1989, 61, 613.

(193) Tornøe, C. W.; Christensen, C.; Meldal, M. The Journal of Organic Chemistry 2002, 67, 3057.

(194) Rostovtsev, V. V.; Green, L. G.; Fokin, V. V.; Sharpless, K. B. Angewandte Chemie International Edition 2002, 41, 2596.

(195) Himo, F.; Lovell, T.; Hilgraf, R.; Rostovtsev, V. V.; Noodleman, L.; Sharpless, K. B.; Fokin, V. V. Journal of the American Chemical Society 2004, 127, 210.

(196) Bock, V. D.; Hiemstra, H.; van Maarseveen, J. H. European Journal of Organic Chemistry 2006, 2006, 51.

(197) Meldal, M.; Tornøe, C. W. Chemical Reviews 2008, 108, 2952.

(198) Hein, J. E.; Fokin, V. V. Chemical Society Reviews 2010, 39, 1302.

(199) Chan, T. R.; Hilgraf, R.; Sharpless, K. B.; Fokin, V. V. Organic Letters 2004, 6, 2853.

(200) Wang, Q.; Chan, T. R.; Hilgraf, R.; Fokin, V. V.; Sharpless, K. B.; Finn, M. G. Journal of the American Chemical Society 2003, 125, 3192.

(201) Pérez-Balderas, F.; Ortega-Muñoz, M.; Morales-Sanfrutos, J.; Hernández- Mateo, F.; Calvo-Flores, F. G.; Calvo-Asín, J. A.; Isac-García, J.; Santoyo-González, F. Organic Letters 2003, 5, 1951.

(202) Binder, W. H.; Sachsenhofer, R. Macromolecular Rapid Communications 2007, 28, 15.

(203) Lutz, J.-F. Angewandte Chemie International Edition 2007, 46, 1018.

(204) Moses, J. E.; Moorhouse, A. D. Chemical Society Reviews 2007, 36, 1249. 228

(205) Breinbauer, R.; Köhn, M. ChemBioChem 2003, 4, 1147.

(206) Brik, A.; Wu, C.-Y.; Wong, C.-H. Organic & Biomolecular Chemistry 2006, 4, 1446.

(207) Qin, A.; Lam, J. W. Y.; Tang, B. Z. Macromolecules 2010, 43, 8693.

(208) Qin, A.; Lam, J. W. Y.; Tang, B. Z. Chemical Society Reviews 2010, 39, 2522.

(209) Chen, Y.; Guan, Z. Journal of the American Chemical Society 2010, 132, 4577.

(210) Geng, J.; Mantovani, G.; Tao, L.; Nicolas, J.; Chen, G.; Wallis, R.; Mitchell, D. A.; Johnson, B. R. G.; Evans, S. D.; Haddleton, D. M. Journal of the American Chemical Society 2007, 129, 15156.

(211) Johnson, J. A.; Finn, M. G.; Koberstein, J. T.; Turro, N. J. Macromolecular Rapid Communications 2008, 29, 1052.

(212) Kushwaha, D.; Tiwari, V. K. The Journal of Organic Chemistry 2013, 78, 8184.

(213) Vieyres, A.; Lam, T.; Gillet, R.; Franc, G.; Castonguay, A.; Kakkar, A. Chemical Communications 2010, 46, 1875.

(214) Tempelaar, S.; Mespouille, L.; Coulembier, O.; Dubois, P.; Dove, A. P. Chemical Society Reviews 2013, 42, 1312.

(215) Pounder, R. J.; Dove, A. P. Polymer Chemistry 2010, 1, 260.

(216) Parrish, B.; Breitenkamp, R. B.; Emrick, T. Journal of the American Chemical Society 2005, 127, 7404.

(217) Riva, R.; Schmeits, S.; Stoffelbach, F.; Jerome, C.; Jerome, R.; Lecomte, P. Chemical Communications 2005, 5334.

(218) Wu, W.-X.; Wang, N.; Liu, B.-Y.; Deng, Q.-F.; Yu, X.-Q. Soft Matter 2014, 10, 1199.

(219) Rana, S.; Lee, S.; Cho, J. Polym. Bull. 2010, 64, 401.

(220) Borreguero, A. M.; Sharma, P.; Spiteri, C.; Velencoso, M. M.; Carmona, M. S.; Moses, J. E.; Rodríguez, J. F. Reactive and Functional Polymers 2013, 73, 1207.

(221) Engler, A. C.; Lee, H.-i.; Hammond, P. T. Angewandte Chemie International Edition 2009, 48, 9334. 229

(222) Bonduelle, C.; Lecommandoux, S. Biomacromolecules 2013, 14, 2973.

(223) Tang, H.; Zhang, D. Biomacromolecules 2010, 11, 1585.

(224) Jiang, X.; Lok, M. C.; Hennink, W. E. Bioconjugate Chemistry 2007, 18, 2077.

(225) Li, M.; De, P.; Gondi, S. R.; Sumerlin, B. S. Macromolecular Rapid Communications 2008, 29, 1172.

(226) Crescenzi, V.; Cornelio, L.; Di Meo, C.; Nardecchia, S.; Lamanna, R. Biomacromolecules 2007, 8, 1844.

(227) Okay, O. In Hydrogel Sensors and Actuators; Gerlach, G., Arndt, K.-F., Eds.; Springer Berlin Heidelberg: 2010; Vol. 6, p 1.

(228) Ossipov, D. A.; Hilborn, J. Macromolecules 2006, 39, 1709.

(229) Malkoch, M.; Vestberg, R.; Gupta, N.; Mespouille, L.; Dubois, P.; Mason, A. F.; Hedrick, J. L.; Liao, Q.; Frank, C. W.; Kingsbury, K.; Hawker, C. J. Chemical Communications 2006, 2774.

(230) Johnson, J. A.; Finn, M. G.; Koberstein, J. T.; Turro, N. J. Macromolecules 2007, 40, 3589.

(231) Xu, X.-D.; Chen, C.-S.; Wang, Z.-C.; Wang, G.-R.; Cheng, S.-X.; Zhang, X.- Z.; Zhuo, R.-X. Journal of Polymer Science Part A: Polymer Chemistry 2008, 46, 5263.

(232) Wolbers, F.; ter Braak, P.; Le Gac, S.; Luttge, R.; Andersson, H.; Vermes, I.; van den Berg, A. ELECTROPHORESIS 2006, 27, 5073.

(233) Sletten, E. M.; Bertozzi, C. R. Angewandte Chemie International Edition 2009, 48, 6974.

(234) Jewett, J. C.; Bertozzi, C. R. Chemical Society Reviews 2010, 39, 1272.

(235) Agard, N. J.; Prescher, J. A.; Bertozzi, C. R. Journal of the American Chemical Society 2004, 126, 15046.

(236) Agard, N. J.; Baskin, J. M.; Prescher, J. A.; Lo, A.; Bertozzi, C. R. ACS Chemical Biology 2006, 1, 644.

(237) Ning, X.; Guo, J.; Wolfert, M. A.; Boons, G.-J. Angewandte Chemie International Edition 2008, 47, 2253.

(238) Li, Z.; Seo, T. S.; Ju, J. Tetrahedron Letters 2004, 45, 3143. 230

(239) Inglis, A. J.; Barner-Kowollik, C. Macromolecular Rapid Communications 2010, 31, 1247.

(240) Baskin, J. M.; Prescher, J. A.; Laughlin, S. T.; Agard, N. J.; Chang, P. V.; Miller, I. A.; Lo, A.; Codelli, J. A.; Bertozzi, C. R. Proceedings of the National Academy of Sciences 2007, 104, 16793.

(241) Lin, F.; Zheng, J.; Yu, J.; Zhou, J.; Becker, M. L. Biomacromolecules 2013, 14, 2857.

(242) Zheng, J.; Liu, K.; Reneker, D. H.; Becker, M. L. Journal of the American Chemical Society 2012, 134, 17274.

(243) Mbua, N. E.; Guo, J.; Wolfert, M. A.; Steet, R.; Boons, G.-J. ChemBioChem 2011, 12, 1912.

(244) DeForest, C. A.; Sims, E. A.; Anseth, K. S. Chemistry of Materials 2010, 22, 4783.

(245) DeForest, C. A.; Anseth, K. S. Nat Chem 2011, 3, 925.

(246) DeForest, C. A.; Polizzotti, B. D.; Anseth, K. S. Nat Mater 2009, 8, 659.

(247) Azagarsamy, M. A.; Anseth, K. S. Angewandte Chemie International Edition 2013, 52, 13803.

(248) DeForest, C. A.; Anseth, K. S. Angewandte Chemie International Edition 2012, 51, 1816.

(249) Posner, T. Berichte der deutschen chemischen Gesellschaft 1905, 38, 646.

(250) Hoyle, C. E.; Bowman, C. N. Angewandte Chemie International Edition 2010, 49, 1540.

(251) Lowe, A. B. Polymer Chemistry 2010, 1, 17.

(252) Mather, B. D.; Viswanathan, K.; Miller, K. M.; Long, T. E. Progress in Polymer Science 2006, 31, 487.

(253) Griesbaum, K. Angewandte Chemie International Edition in English 1970, 9, 273.

(254) Hoyle, C. E.; Lee, T. Y.; Roper, T. Journal of Polymer Science Part A: Polymer Chemistry 2004, 42, 5301.

231

(255) Cramer, N. B.; Reddy, S. K.; Cole, M.; Hoyle, C.; Bowman, C. N. Journal of Polymer Science Part A: Polymer Chemistry 2004, 42, 5817.

(256) Northrop, B. H.; Coffey, R. N. Journal of the American Chemical Society 2012, 134, 13804.

(257) Morgan, C. R.; Magnotta, F.; Ketley, A. D. Journal of Polymer Science: Polymer Chemistry Edition 1977, 15, 627.

(258) Stewart, I. C.; Bergman, R. G.; Toste, F. D. Journal of the American Chemical Society 2003, 125, 8696.

(259) Hvilsted, S. Polymer International 2012, 61, 485.

(260) van Dijk, M.; Rijkers, D. T. S.; Liskamp, R. M. J.; van Nostrum, C. F.; Hennink, W. E. Bioconjugate Chemistry 2009, 20, 2001.

(261) Wang, R.; Chen, W.; Meng, F.; Cheng, R.; Deng, C.; Feijen, J.; Zhong, Z. Macromolecules 2011, 44, 6009.

(262) Zhou, J.; Chen, P.; Deng, C.; Meng, F.; Cheng, R.; Zhong, Z. Macromolecules 2013, 46, 6723.

(263) Tempelaar, S.; Mespouille, L.; Dubois, P.; Dove, A. P. Macromolecules 2011, 44, 2084.

(264) Billiet, L.; Gok, O.; Dove, A. P.; Sanyal, A.; Nguyen, L.-T. T.; Du Prez, F. E. Macromolecules 2011, 44, 7874.

(265) Stanford, M. J.; Pflughaupt, R. L.; Dove, A. P. Macromolecules 2010, 43, 6538.

(266) Truong, V. X.; Dove, A. P. Angewandte Chemie International Edition 2013, 52, 4132.

(267) Lowe, A. B.; Hoyle, C. E.; Bowman, C. N. Journal of Materials Chemistry 2010, 20, 4745.

(268) Fairbanks, B. D.; Scott, T. F.; Kloxin, C. J.; Anseth, K. S.; Bowman, C. N. Macromolecules 2008, 42, 211.

(269) Gustafson, T. P.; Lonnecker, A. T.; Heo, G. S.; Zhang, S.; Dove, A. P.; Wooley, K. L. Biomacromolecules 2013, 14, 3346.

232

(270) Zhang, S.; Zou, J.; Zhang, F.; Elsabahy, M.; Felder, S. E.; Zhu, J.; Pochan, D. J.; Wooley, K. L. Journal of the American Chemical Society 2012, 134, 18467. (271) Sun, J.; Schlaad, H. Macromolecules 2010, 43, 4445.

(272) Gress, A.; Völkel, A.; Schlaad, H. Macromolecules 2007, 40, 7928.

(273) Mergy, J.; Fournier, A.; Hachet, E.; Auzély-Velty, R. Journal of Polymer Science Part A: Polymer Chemistry 2012, 50, 4019.

(274) Slavin, S.; Burns, J.; Haddleton, D. M.; Becer, C. R. European Polymer Journal 2011, 47, 435.

(275) Bae, J. W.; Lee, E.; Park, K. M.; Park, K. D. Macromolecules 2009, 42, 3437.

(276) Obermeier, B.; Frey, H. Bioconjugate Chemistry 2011, 22, 436.

(277) Connal, L. A.; Kinnane, C. R.; Zelikin, A. N.; Caruso, F. Chemistry of Materials 2009, 21, 576.

(278) Hayashi, K.; Ono, K.; Suzuki, H.; Sawada, M.; Moriya, M.; Sakamoto, W.; Yogo, T. Chemistry of Materials 2010, 22, 3768.

(279) Hordyjewicz-Baran, Z.; You, L.; Smarsly, B.; Sigel, R.; Schlaad, H. Macromolecules 2007, 40, 3901.

(280) Geng, Y.; Discher, D. E.; Justynska, J.; Schlaad, H. Angewandte Chemie International Edition 2006, 45, 7578.

(281) van der Ende, A. E.; Harrell, J.; Sathiyakumar, V.; Meschievitz, M.; Katz, J.; Adcock, K.; Harth, E. Macromolecules 2010, 43, 5665.

(282) van der Ende, A.; Croce, T.; Hamilton, S.; Sathiyakumar, V.; Harth, E. Soft Matter 2009, 5, 1417.

(283) Kottari, N.; Chabre, Y. M.; Shiao, T. C.; Rej, R.; Roy, R. Chemical Communications 2014, 50, 1983.

(284) Killops, K. L.; Campos, L. M.; Hawker, C. J. Journal of the American Chemical Society 2008, 130, 5062.

(285) Tibbitt, M. W.; Kloxin, A. M.; Sawicki, L. A.; Anseth, K. S. Macromolecules 2013, 46, 2785.

(286) Cui, J.; Lackey, M. A.; Madkour, A. E.; Saffer, E. M.; Griffin, D. M.; Bhatia, S. R.; Crosby, A. J.; Tew, G. N. Biomacromolecules 2012, 13, 584. 233

(287) Cui, J.; Lackey, M. A.; Tew, G. N.; Crosby, A. J. Macromolecules 2012, 45, 6104.

(288) Chawla, K.; Yu, T. B.; Liao, S. W.; Guan, Z. Biomacromolecules 2011, 12, 560.

(289) Peng, K.; Tomatsu, I.; van den Broek, B.; Cui, C.; Korobko, A. V.; van Noort, J.; Meijer, A. H.; Spaink, H. P.; Kros, A. Soft Matter 2011, 7, 4881.

(290) Fu, Y.; Kao, W. J. Journal of Biomedical Materials Research Part A 2011, 98A, 201.

(291) Jin, R.; Moreira Teixeira, L. S.; Krouwels, A.; Dijkstra, P. J.; van Blitterswijk, C. A.; Karperien, M.; Feijen, J. Acta Biomaterialia 2010, 6, 1968.

(292) Azagarsamy, M. A.; Anseth, K. S. ACS Macro Letters 2012, 2, 5.

(293) Alge, D. L.; Azagarsamy, M. A.; Donohue, D. F.; Anseth, K. S. Biomacromolecules 2013, 14, 949.

(294) Gupta, N.; Lin, B. F.; Campos, L. M.; Dimitriou, M. D.; Hikita, S. T.; Treat, N. D.; Tirrell, M. V.; Clegg, D. O.; Kramer, E. J.; Hawker, C. J. Nat Chem 2012, 4, 424.

(295) Kalia, J.; Raines, R. T. Angewandte Chemie International Edition 2008, 47, 7523.

(296) Sander, E. G.; Jencks, W. P. Journal of the American Chemical Society 1968, 90, 6154.

(297) Dawson, P. E.; Kent, S. B. H. Annual Review of Biochemistry 2000, 69, 923.

(298) Dirksen, A.; Hackeng, T. M.; Dawson, P. E. Angewandte Chemie International Edition 2006, 45, 7581.

(299) Rashidian, M.; Mahmoodi, M. M.; Shah, R.; Dozier, J. K.; Wagner, C. R.; Distefano, M. D. Bioconjugate Chemistry 2013, 24, 333.

(300) Wendeler, M.; Grinberg, L.; Wang, X.; Dawson, P. E.; Baca, M. Bioconjugate Chemistry 2013, 25, 93.

(301) Dirksen, A.; Dawson, P. E. Bioconjugate Chemistry 2008, 19, 2543.

(302) Blanden, A. R.; Mukherjee, K.; Dilek, O.; Loew, M.; Bane, S. L. Bioconjugate Chemistry 2011, 22, 1954.

234

(303) Crisalli, P.; Kool, E. T. The Journal of Organic Chemistry 2013, 78, 1184.

(304) Van Horn, B. A.; Wooley, K. L. Soft Matter 2007, 3, 1032.

(305) Van Horn, B. A.; Iha, R. K.; Wooley, K. L. Macromolecules 2008, 41, 1618.

(306) Maynard, H. D.; Broyer, R. M.; Kolodziej, C. M. In Click Chemistry for Biotechnology and Materials Science; John Wiley & Sons, Ltd: 2009, p 53.

(307) Tian, D.; Dubois, P.; Jérôme, R. Macromolecules 1997, 30, 2575.

(308) Taniguchi, I.; Mayes, A. M.; Chan, E. W. L.; Griffith, L. G. Macromolecules 2004, 38, 216.

(309) Iha, R. K.; van Horn, B. A.; Wooley, K. L. Journal of Polymer Science Part A: Polymer Chemistry 2010, 48, 3553.

(310) Grover, G. N.; Lee, J.; Matsumoto, N. M.; Maynard, H. D. Macromolecules 2012, 45, 4958.

(311) Liu, J.; Li, R. C.; Sand, G. J.; Bulmus, V.; Davis, T. P.; Maynard, H. D. Macromolecules 2012, 46, 8.

(312) Kopping, J. T.; Tolstyka, Z. P.; Maynard, H. D. Macromolecules 2007, 40, 8593.

(313) Stukel, J. M.; Li, R. C.; Maynard, H. D.; Caplan, M. R. Biomacromolecules 2009, 11, 160.

(314) Vázquez-Dorbatt, V.; Tolstyka, Z. P.; Maynard, H. D. Macromolecules 2009, 42, 7650.

(315) Heredia, K. L.; Tolstyka, Z. P.; Maynard, H. D. Macromolecules 2007, 40, 4772.

(316) Kochendoerfer, G. G.; Chen, S.-Y.; Mao, F.; Cressman, S.; Traviglia, S.; Shao, H.; Hunter, C. L.; Low, D. W.; Cagle, E. N.; Carnevali, M.; Gueriguian, V.; Keogh, P. J.; Porter, H.; Stratton, S. M.; Wiedeke, M. C.; Wilken, J.; Tang, J.; Levy, J. J.; Miranda, L. P.; Crnogorac, M. M.; Kalbag, S.; Botti, P.; Schindler-Horvat, J.; Savatski, L.; Adamson, J. W.; Kung, A.; Kent, S. B. H.; Bradburne, J. A. Science 2003, 299, 884.

(317) Shao, H.; Crnogorac, M. M.; Kong, T.; Chen, S.-Y.; Williams, J. M.; Tack, J. M.; Gueriguian, V.; Cagle, E. N.; Carnevali, M.; Tumelty, D.; Paliard, X.; Miranda, L. P.; Bradburne, J. A.; Kochendoerfer, G. G. Journal of the American Chemical Society 2005, 127, 1350. 235

(318) Kochendoerfer, G. G. Current Opinion in Chemical Biology 2005, 9, 555.

(319) Ossipov, D. A.; Yang, X.; Varghese, O.; Kootala, S.; Hilborn, J. Chemical Communications 2010, 46, 8368.

(320) Prestwich, G. D.; Marecak, D. M.; Marecek, J. F.; Vercruysse, K. P.; Ziebell, M. R. Journal of Controlled Release 1998, 53, 93.

(321) Grover, G. N.; Braden, R. L.; Christman, K. L. Advanced Materials 2013, 25, 2937.

(322) Grover, G. N.; Lam, J.; Nguyen, T. H.; Segura, T.; Maynard, H. D. Biomacromolecules 2012, 13, 3013.

(323) Deng, G.; Li, F.; Yu, H.; Liu, F.; Liu, C.; Sun, W.; Jiang, H.; Chen, Y. ACS Macro Letters 2012, 1, 275.

(324) Lin, F.; Yu, J.; Tang, W.; Zheng, J.; Defante, A.; Guo, K.; Wesdemiotis, C.; Becker, M. L. Biomacromolecules 2013, 14, 3749.

(325) McKinnon, D. D.; Domaille, D. W.; Cha, J. N.; Anseth, K. S. Advanced Materials 2013, n/a.

(326) Zhou, L.; Cheng, R.; Tao, H.; Ma, S.; Guo, W.; Meng, F.; Liu, H.; Liu, Z.; Zhong, Z. Biomacromolecules 2011, 12, 1460.

(327) Filippov, S. K.; Chytil, P.; Konarev, P. V.; Dyakonova, M.; Papadakis, C.; Zhigunov, A.; Plestil, J.; Stepanek, P.; Etrych, T.; Ulbrich, K.; Svergun, D. I. Biomacromolecules 2012, 13, 2594.

(328) Chytil, P.; Etrych, T.; Koňák, Č.; Šírová, M.; Mrkvan, T.; Bouček, J.; Říhová, B.; Ulbrich, K. Journal of Controlled Release 2008, 127, 121.

(329) Esser-Kahn, A. P.; Francis, M. B. Angewandte Chemie International Edition 2008, 47, 3751.

(330) Etrych, T.; Mrkvan, T.; Chytil, P.; Koňák, Č.; Říhová, B.; Ulbrich, K. Journal of Applied Polymer Science 2008, 109, 3050.

(331) Hrubý, M.; Koňák, Č.; Ulbrich, K. Journal of Controlled Release 2005, 103, 137.

(332) Dutta, D.; Pulsipher, A.; Luo, W.; Mak, H.; Yousaf, M. N. Bioconjugate Chemistry 2011, 22, 2423.

236

(333) Luo, W.; Yousaf, M. N. Journal of the American Chemical Society 2011, 133, 10780.

(334) Dutta, D.; Pulsipher, A.; Luo, W.; Yousaf, M. N. Journal of the American Chemical Society 2011, 133, 8704.

(335) Barrett, D. G.; Merkel, T. J.; Luft, J. C.; Yousaf, M. N. Macromolecules 2010, 43, 9660.

(336) Pulsipher, A.; Westcott, N. P.; Luo, W.; Yousaf, M. N. Journal of the American Chemical Society 2009, 131, 7626.

(337) Barrett, D. G.; Yousaf, M. N. Macromolecules 2008, 41, 6347.

(338) Barrett, D. G.; Lamb, B. M.; Yousaf, M. N. Langmuir 2008, 24, 9861.

(339) Barrett, D. G.; Yousaf, M. N. ChemBioChem 2008, 9, 62.

(340) Christman, K. L.; Broyer, R. M.; Schopf, E.; Kolodziej, C. M.; Chen, Y.; Maynard, H. D. Langmuir 2010, 27, 1415.

(341) Christman, K. L.; Broyer, R. M.; Tolstyka, Z. P.; Maynard, H. D. Journal of Materials Chemistry 2007, 17, 2021.

(342) Christman, K. L.; Maynard, H. D. Langmuir 2005, 21, 8389.

(343) Christman, K. L.; Schopf, E.; Broyer, R. M.; Li, R. C.; Chen, Y.; Maynard, H. D. Journal of the American Chemical Society 2008, 131, 521.

(344) Lamb, B. M.; Park, S.; Yousaf, M. N. Langmuir 2010, 26, 12817.

(345) Dutta, D.; Pulsipher, A.; Yousaf, M. N. Langmuir 2010, 26, 9835.

(346) Park, S.; Yousaf, M. N. Langmuir 2008, 24, 6201.

(347) Athanasiou, K. A.; Agrawal, C. M.; Barber, F. A.; Burkhart, S. S. Arthroscopy: The Journal of Arthroscopic & Related Surgery 1998, 14, 726.

(348) Athanasiou, K. A.; Niederauer, G. G.; Agrawal, C. M. Biomaterials 1996, 17, 93.

(349) Anselme, K.; Flautre, B.; Hardouin, P.; Chanavaz, M.; Ustariz, C.; Vert, M. Biomaterials 1993, 14, 44.

(350) Liao, S. S.; Cui, F. Z.; Zhang, W.; Feng, Q. L. Journal of Biomedical Materials 237

Research Part B: Applied Biomaterials 2004, 69B, 158.

(351) Mistry, A. S.; Cheng, S. H.; Yeh, T.; Christenson, E.; Jansen, J. A.; Mikos, A. G. Journal of Biomedical Materials Research Part A 2009, 89A, 68.

(352) Williams, J. M.; Adewunmi, A.; Schek, R. M.; Flanagan, C. L.; Krebsbach, P. H.; Feinberg, S. E.; Hollister, S. J.; Das, S. Biomaterials 2005, 26, 4817.

(353) Choong, C.; Triffitt, J. T.; Cui, Z. F. Food and Bioproducts Processing 2004, 82, 117.

(354) Ma, P. X.; Choi, J. W. Tissue Engineering 2001, 7, 23.

(355) Tsuji, H.; Ikada, Y. Polymer 1995, 36, 2709.

(356) Wei, G.; Ma, P. X. Biomaterials 2004, 25, 4749.

(357) Reilly, D. T.; Burstein, A. H.; Frankel, V. H. Journal of Biomechanics 1974, 7, 271.

(358) Rho, J. Y.; Ashman, R. B.; Turner, C. H. Journal of Biomechanics 1993, 26, 111.

(359) Daniels, A. U.; Chang, M. K. O.; Andriano, K. P.; Heller, J. Journal of Applied Biomaterials 1990, 1, 57.

(360) Babis, G. C.; Soucacos, P. N. Injury 2005, 36, S38.

(361) Katchalski-Katzir, E. Cellular and Molecular Life Sciences 1997, 53, 780.

(362) Rydholm, A. E.; Bowman, C. N.; Anseth, K. S. Biomaterials 2005, 26, 4495.

(363) Lutolf, M. P.; Hubbell, J. A. Nat Biotech 2005, 23, 47.

(364) Laurencin, L. S. N. C. T. Adv Biochem Engin/Biotechnol 2006, 102, 47.

(365) Holzwarth, J. M.; Ma, P. X. Biomaterials 2011, 32, 9622.

(366) Zhong, C.; Chu, C. C. Journal of Materials Chemistry 2012, 22, 6080.

(367) Kim, I. L.; Mauck, R. L.; Burdick, J. A. Biomaterials 2011, 32, 8771.

(368) Shin, H.; Jo, S.; Mikos, A. G. Biomaterials 2003, 24, 4353.

238

(369) Liberelle, B. t.; Boucher, C.; Chen, J.; Jolicoeur, M.; Durocher, Y.; De Crescenzo, G. Bioconjugate Chemistry 2010, 21, 2257.

(370) Ito, Y.; Inoue, M.; Liu, S. Q.; Imanishi, Y. Journal of Biomedical Materials Research 1993, 27, 901.

(371) Bab, I.; Gazit, D.; Chorev, M.; Muhlrad, A.; Shteyer, A.; Greenberg, Z.; Namdar, M.; Kahn, A. EMBO Journal 1992, 11, 1867.

(372) Greenberg, Z.; Chorev, M.; Muhlrad, A.; Shteyer, A.; Namdar-Attar, M.; Casap, N.; Tartakovsky, A.; Vidson, M.; Bab, I. Journal of Clinical Endocrinology & Metabolism 1995, 80, 2330.

(373) Chen, Z.-x.; Chang, M.; Peng, Y.-l.; Zhao, L.; Zhan, Y.-r.; Wang, L.-j.; Wang, R. Regulatory Peptides 2007, 142, 16.

(374) Gabarin, N.; Gavish, H.; Muhlrad, A.; Chen, Y.; Namdar-Attar, M.; Nissenson, R. A.; Chorev, M.; Bab, I. Journal of Cellular Biochemistry 2001, 81, 594.

(375) Mattii, L.; Battolla, B.; Moscato, S.; Fazzi, R.; Galimberti, S.; Bernardini, N.; Dolfi, A.; Petrini, M. Medical Science Monitor 2008, 14, BR103.

(376) Mattii, L.; Fazzi, R.; Moscato, S.; Segnani, C.; Pacini, S.; Galimberti, S.; D'Alessandro, D.; Bernardini, N.; Petrini, M. Journal of Cellular Biochemistry 2004, 93, 1231.

(377) Gabet, Y.; Müller, R.; Regev, E.; Sela, J.; Shteyer, A.; Salisbury, K.; Chorev, M.; Bab, I. Bone 2004, 35, 65.

(378) Moore, N. M.; Lin, N. J.; Gallant, N. D.; Becker, M. L. Biomaterials 2010, 31, 1604.

(379) Shuqiang, M.; Kunzheng, W.; Xiaoqiang, D.; Wei, W.; Mingyu, Z.; Daocheng, W. Journal of Plastic, Reconstructive & Aesthetic Surgery 2008, 61, 1558.

(380) Katsarava, R.; Beridze, V.; Arabuli, N.; Kharadze, D.; Chu, C. C.; Won, C. Y. Journal of Polymer Science Part A: Polymer Chemistry 1999, 37, 391.

(381) Kasuga, T.; Ota, Y.; Nogami, M.; Abe, Y. Biomaterials 2001, 22, 19.

(382) Avella, M.; Errico, M. E.; Laurienzo, P.; Martuscelli, E.; Raimo, M.; Rimedio, R. Polymer 2000, 41, 3875.

(383) Anderson, J. M.; Rodriguez, A.; Chang, D. T. Seminars in Immunology 2008, 20, 86. 239

(384) Crane, G. M.; Ishaug, S. L.; Mikos, A. G. Nature Medicine 1995, 1, 1322.

(385) Hubbell, J. A. Bio/Technology 1995, 13, 565.

(386) Kohn, J.; Langer, R. L. In An Introduction to Materials in Medicine, Academic Press; Ratner, B. D., Hoffman, A. S., Schoen, F. J., Lemon, J. E., Eds.; Academic Press: San Diego 1997, p 65.

(387) Langer, R.; Tirrell, D. A. Nature 2004, 428, 487.

(388) Yaszemski, M. J.; Payne, R. G.; Hayes, W. C.; Langer, R.; Mikos, A. G. Biomaterials 1996, 17, 175.

(389) Aamer, K. A.; Genson, K. L.; Kohn, J.; Becker, M. L. Biomacromolecules 2009, 10, 2418.

(390) Bailey, L. O.; Becker, M. L.; Stephens, J. S.; Gallant, N. D.; Mahoney, C. M.; Washburn, N. R.; Rege, A.; Kohn, J.; Amis, E. J. Journal of Biomedical Materials Research - Part A 2006, 76, 491.

(391) Muggli, D. S.; Burkoth, A. K.; Keyser, S. A.; Lee, H. R.; Anseth, K. S. Macromolecules 1998, 31, 4120.

(392) Greenberg, Z.; Chorev, M.; Muhlrad, A.; Shteyer, A.; Namdar-Attar, M.; Casap, N.; Tartakovsky, A.; Vidson, M.; Bab, I. Journal of Clinical Endocrinology and Metabolism 1995, 80, 2330.

(393) Greenberg, Z.; Gavish, H.; Muhlrad, A.; Chorev, M.; Shteyer, A.; Attar- Namdar, M.; Tartakovsky, A.; Bab, I. Journal of Cellular Biochemistry 1997, 65, 359.

(394) M, S.; R., G. Orthopedics 2002, 25, 601.

(395) Raghuvanshi, R. S.; Goyal, S.; Singh, O.; Panda, A. K. Pharmaceutical Development and Technology 1998, 3, 269.

(396) Kohane, D. S.; Langer, R. Pediatr Res 2008, 63, 487.

(397) Seal, B. L.; Otero, T. C.; Panitch, A. Materials Science and Engineering: R: Reports 2001, 34, 147.

(398) Varghese, S.; Elisseeff, J. In Polymers for Regenerative Medicine; Werner, C., Ed.; Springer Berlin Heidelberg: 2006; Vol. 203, p 95.

(399) Puppi, D.; Chiellini, F.; Piras, A. M.; Chiellini, E. Progress in Polymer Science 2010, 35, 403. 240

(400) Latere, J.-P.; Lecomte, P.; Dubois, P.; Jérôme, R. Macromolecules 2002, 35, 7857.

(401) Gurusamy-Thangavelu, S. A.; Emond, S. J.; Kulshrestha, A.; Hillmyer, M. A.; Macosko, C. W.; Tolman, W. B.; Hoye, T. R. Polymer Chemistry 2012, 3, 2941.

(402) Zhang, S.; Li, A.; Zou, J.; Lin, L. Y.; Wooley, K. L. ACS Macro Letters 2012, 1, 328.

(403) Dwan'Isa, J.-P. L.; Lecomte, P.; Dubois, P.; Jérôme, R. Macromolecular Chemistry and Physics 2003, 204, 1191.

(404) Angela, L. S.; Chia-Chih, C.; Bryan, P.; Todd, E. In Degradable Polymers and Materials: Principles and Practice (2nd Edition); American Chemical Society: 2012; Vol. 1114, p 237.

(405) Madhavan Nampoothiri, K.; Nair, N. R.; John, R. P. Bioresource Technology 2010, 101, 8493.

(406) Grad, S.; Kupcsik, L.; Gorna, K.; Gogolewski, S.; Alini, M. Biomaterials 2003, 24, 5163.

(407) Grenier, S.; Sandig, M.; Mequanint, K. Journal of Biomedical Materials Research Part A 2007, 82A, 802.

(408) Anderson, J. M.; Gibbons, D. F.; Martin, R. L.; Hiltner, A.; Woods, R. Journal of Biomedical Materials Research 1974, 8, 197.

(409) Chow, D.; Nunalee, M. L.; Lim, D. W.; Simnick, A. J.; Chilkoti, A. Materials Science and Engineering: R: Reports 2008, 62, 125.

(410) Stakleff, K. S.; Lin, F.; Smith Callahan, L. A.; Wade, M. B.; Esterle, A.; Miller, J.; Graham, M.; Becker, M. L. Acta Biomaterialia.

(411) Lele, B. S.; Murata, H.; Matyjaszewski, K.; Russell, A. J. Biomacromolecules 2005, 6, 3380.

(412) Heredia, K. L.; Maynard, H. D. Organic & Biomolecular Chemistry 2007, 5, 45.

(413) Onbulak, S.; Tempelaar, S.; Pounder, R. J.; Gok, O.; Sanyal, R.; Dove, A. P.; Sanyal, A. Macromolecules 2012, 45, 1715.

(414) Danial, M.; Root, M. J.; Klok, H.-A. Biomacromolecules 2012, 13, 1438.

241

(415) Becer, C. R.; Hoogenboom, R.; Schubert, U. S. Angewandte Chemie International Edition 2009, 48, 4900.

(416) Su, H.; Zheng, J.; Wang, Z.; Lin, F.; Feng, X.; Dong, X.-H.; Becker, M. L.; Cheng, S. Z. D.; Zhang, W.-B.; Li, Y. ACS Macro Letters 2013, 2, 645.

(417) Gogoi, K.; Mane, M. V.; Kunte, S. S.; Kumar, V. A. Nucleic Acids Research 2007, 35, e139.

(418) Wu, P.; Shui, W.; Carlson, B. L.; Hu, N.; Rabuka, D.; Lee, J.; Bertozzi, C. R. Proceedings of the National Academy of Sciences 2009.

(419) Devaraj, N. K.; Hilderbrand, S.; Upadhyay, R.; Mazitschek, R.; Weissleder, R. Angewandte Chemie International Edition 2010, 49, 2869.

(420) Shen, B.-Q.; Xu, K.; Liu, L.; Raab, H.; Bhakta, S.; Kenrick, M.; Parsons- Reponte, K. L.; Tien, J.; Yu, S.-F.; Mai, E.; Li, D.; Tibbitts, J.; Baudys, J.; Saad, O. M.; Scales, S. J.; McDonald, P. J.; Hass, P. E.; Eigenbrot, C.; Nguyen, T.; Solis, W. A.; Fuji, R. N.; Flagella, K. M.; Patel, D.; Spencer, S. D.; Khawli, L. A.; Ebens, A.; Wong, W. L.; Vandlen, R.; Kaur, S.; Sliwkowski, M. X.; Scheller, R. H.; Polakis, P.; Junutula, J. R. Nat Biotech 2012, 30, 184.

(421) Luo, Y.; Prestwich, G. D. Bioconjugate Chemistry 1999, 10, 755.

(422) Joshi, N. S.; Whitaker, L. R.; Francis, M. B. Journal of the American Chemical Society 2004, 126, 15942.

(423) McFarland, J. M.; Joshi, N. S.; Francis, M. B. Journal of the American Chemical Society 2008, 130, 7639.

(424) Ban, H.; Gavrilyuk, J.; Barbas, C. F. Journal of the American Chemical Society 2010, 132, 1523.

(425) Ban, H.; Nagano, M.; Gavrilyuk, J.; Hakamata, W.; Inokuma, T.; Barbas, C. F. Bioconjugate Chemistry 2013, 24, 520.

(426) Jones, M. W.; Mantovani, G.; Blindauer, C. A.; Ryan, S. M.; Wang, X.; Brayden, D. J.; Haddleton, D. M. Journal of the American Chemical Society 2012, 134, 7406.

(427) Liang, D.; Hsiao, B. S.; Chu, B. Advanced Drug Delivery Reviews 2007, 59, 1392.

(428) Agarwal, S.; Wendorff, J. H.; Greiner, A. Polymer 2008, 49, 5603.

242

(429) Frenot, A.; Chronakis, I. S. Current Opinion in Colloid & Interface Science 2003, 8, 64.

(430) Zong, X.; Kim, K.; Fang, D.; Ran, S.; Hsiao, B. S.; Chu, B. Polymer 2002, 43, 4403.

(431) Reneker, D. H.; Yarin, A. L. Polymer 2008, 49, 2387.

(432) Kim, T. G.; Park, T. G. Biotechnology Progress 2006, 22, 1108.

(433) Sahoo, S.; Ang, L.-T.; Cho-Hong Goh, J.; Toh, S.-L. Differentiation 2010, 79, 102.

(434) Patel, S.; Kurpinski, K.; Quigley, R.; Gao, H.; Hsiao, B. S.; Poo, M.-M.; Li, S. Nano Letters 2007, 7, 2122.

(435) Yoo, H. S.; Kim, T. G.; Park, T. G. Advanced Drug Delivery Reviews 2009, 61, 1033.

(436) Lancuški, A.; Fort, S.; Bossard, F. ACS Applied Materials & Interfaces 2012, 4, 6499.

(437) Deiters, A.; Cropp, T. A.; Mukherji, M.; Chin, J. W.; Anderson, J. C.; Schultz, P. G. Journal of the American Chemical Society 2003, 125, 11782.

(438) Wang, Y.-S.; Fang, X.; Wallace, A. L.; Wu, B.; Liu, W. R. Journal of the American Chemical Society 2012, 134, 2950.

(439) Mantovani, G.; Ladmiral, V.; Tao, L.; Haddleton, D. M. Chemical Communications 2005, 2089.

(440) Bryson, D. I.; Zhang, W.; Ray, W. K.; Santos, W. L. Molecular BioSystems 2009, 5, 1070.

(441) Bahta, M.; Liu, F.; Kim, S.-E.; Stephen, A. G.; Fisher, R. J.; Burke, T. R. Nat. Protocols 2012, 7, 686.

(442) Karode, S. K.; Kulkarni, S. S.; Suresh, A. K.; Mashelkar, R. A. Chemical Engineering Science 1998, 53, 2649.

(443) Kade, M. J.; Burke, D. J.; Hawker, C. J. Journal of Polymer Science Part A: Polymer Chemistry 2010, 48, 743.

(444) Caliceti, P.; Veronese, F. M. Advanced Drug Delivery Reviews 2003, 55, 1261.

243

(445) Hutmacher, D. W. Biomaterials 2000, 21, 2529.

(446) Li, W.-J.; Laurencin, C. T.; Caterson, E. J.; Tuan, R. S.; Ko, F. K. Journal of Biomedical Materials Research 2002, 60, 613.

(447) Goldstein, S. A. Journal of Biomechanics 1987, 20, 1055.

(448) Kempson, G. E.; Muir, H.; Pollard, C.; Tuke, M. Biochimica et Biophysica Acta (BBA) - General Subjects 1973, 297, 456.

(449) Edwards, C.; Marks, R. Clinics in Dermatology 1995, 13, 375.

(450) Zhang, Y.; Ouyang, H.; Lim, C. T.; Ramakrishna, S.; Huang, Z.-M. Journal of Biomedical Materials Research Part B: Applied Biomaterials 2005, 72B, 156.

(451) Deng, X.; Friedmann, C.; Lahann, J. Angewandte Chemie International Edition 2011, 50, 6522.

(452) Saha, A.; De, S.; Stuparu, M. C.; Khan, A. Journal of the American Chemical Society 2012, 134, 17291.

(453) Cao, Y.; Croll, T.; Cooper-White, J.; Connor, A.; Stevens, G. In Biopolymer Methods in Tissue Engineering; Hollander, A., Hatton, P., Eds.; Humana Press: 2004; Vol. 238, p 87.

(454) Ma, Z.; Gao, C.; Gong, Y.; Shen, J. Biomaterials 2005, 26, 1253.

(455) Chu, P. K.; Chen, J. Y.; Wang, L. P.; Huang, N. Materials Science and Engineering: R: Reports 2002, 36, 143.

(456) Yoon Sung, N.; Joon Jin, Y.; Jae Gwan, L.; Tae Gwan, P. Journal of Biomaterials Science, Polymer Edition 1999, 10, 1145.

(457) Qiu, Y.; Park, K. Advanced Drug Delivery Reviews 2001, 53, 321.

(458) Slaughter, B. V.; Khurshid, S. S.; Fisher, O. Z.; Khademhosseini, A.; Peppas, N. A. Advanced Materials 2009, 21, 3307.

(459) Lee, K. Y.; Mooney, D. J. Chemical Reviews 2001, 101, 1869.

(460) Dong, L.; Agarwal, A. K.; Beebe, D. J.; Jiang, H. Nature 2006, 442, 551.

(461) Hoare, T. R.; Kohane, D. S. Polymer 2008, 49, 1993.

(462) Nemir, S.; West, J. Ann Biomed Eng 2010, 38, 2. 244

(463) Nemir, S.; Hayenga, H. N.; West, J. L. Biotechnology and Bioengineering 2010, 105, 636.

(464) Carr, L. R.; Krause, J. E.; Ella-Menye, J.-R.; Jiang, S. Biomaterials 2011, 32, 8456.

(465) Smith Callahan, L. A.; Ganios, A. M.; Childers, E. P.; Weiner, S. D.; Becker, M. L. Acta Biomaterialia 2013, 9, 6095.

(466) Cloyd, J.; Malhotra, N.; Weng, L.; Chen, W.; Mauck, R.; Elliott, D. Eur Spine J 2007, 16, 1892.

(467) Jeon, O.; Bouhadir, K. H.; Mansour, J. M.; Alsberg, E. Biomaterials 2009, 30, 2724.

(468) Forget, A.; Christensen, J.; Lüdeke, S.; Kohler, E.; Tobias, S.; Matloubi, M.; Thomann, R.; Shastri, V. P. Proceedings of the National Academy of Sciences 2013, 110, 12887.

(469) Herrick, W. G.; Nguyen, T. V.; Sleiman, M.; McRae, S.; Emrick, T. S.; Peyton, S. R. Biomacromolecules 2013, 14, 2294.

(470) Song, F.; Zhang, L.-M. The Journal of Physical Chemistry B 2008, 112, 13749.

(471) Nguyen, M. K.; Lee, D. S. Chemical Communications 2010, 46, 3583.

(472) Mano, J. F.; Sousa, R. A.; Boesel, L. F.; Neves, N. M.; Reis, R. L. Composites Science and Technology 2004, 64, 789 . (473) Zhu, J. Biomaterials 2010, 31, 4639.

(474) Browning, M. B.; Russell, B.; Rivera, J.; Höök, M.; Cosgriff-Hernandez, E. M. Biomacromolecules 2013, 14, 2225.

(475) Sawhney, A. S.; Pathak, C. P.; Hubbell, J. A. Macromolecules 1993, 26, 581.

(476) Elisseeff, J.; McIntosh, W.; Anseth, K.; Riley, S.; Ragan, P.; Langer, R. Journal of Biomedical Materials Research 2000, 51, 164.

(477) Callahan, L. A. S.; Ganios, A. M.; McBurney, D. L.; Dilisio, M. F.; Weiner, S. D.; Horton, W. E.; Becker, M. L. Biomacromolecules 2012, 13, 1625.

(478) Nimmo, C. M.; Shoichet, M. S. Bioconjugate Chemistry 2011, 22, 2199.

245

(479) Elbert, D. L.; Pratt, A. B.; Lutolf, M. P.; Halstenberg, S.; Hubbell, J. A. Journal of Controlled Release 2001, 76, 11.

(480) Pritchard, C. D.; O’Shea, T. M.; Siegwart, D. J.; Calo, E.; Anderson, D. G.; Reynolds, F. M.; Thomas, J. A.; Slotkin, J. R.; Woodard, E. J.; Langer, R. Biomaterials 2011, 32, 587.

(481) Lutolf, M. P.; Lauer-Fields, J. L.; Schmoekel, H. G.; Metters, A. T.; Weber, F. E.; Fields, G. B.; Hubbell, J. A. Proceedings of the National Academy of Sciences 2003, 100, 5413.

(482) Lin, C.-C.; Raza, A.; Shih, H. Biomaterials 2011, 32, 9685.

(483) Fairbanks, B. D.; Schwartz, M. P.; Halevi, A. E.; Nuttelman, C. R.; Bowman, C. N.; Anseth, K. S. Advanced Materials 2009, 21, 5005.

(484) Zhou, H.; Woo, J.; Cok, A. M.; Wang, M.; Olsen, B. D.; Johnson, J. A. Proceedings of the National Academy of Sciences 2012, 109, 19119.

(485) Fan, Y.; Deng, C.; Cheng, R.; Meng, F.; Zhong, Z. Biomacromolecules 2013, 14, 2814.

(486) Truong, V.; Blakey, I.; Whittaker, A. K. Biomacromolecules 2012, 13, 4012.

(487) Adzima, B. J.; Tao, Y.; Kloxin, C. J.; DeForest, C. A.; Anseth, K. S.; Bowman, C. N. Nat Chem 2011, 3, 256.

(488) Hu, X.; Li, D.; Zhou, F.; Gao, C. Acta Biomaterialia 2011, 7, 1618.

(489) Polizzotti, B. D.; Fairbanks, B. D.; Anseth, K. S. Biomacromolecules 2008, 9, 1084.

(490) Zheng, J.; Smith Callahan, L. A.; Hao, J.; Guo, K.; Wesdemiotis, C.; Weiss, R. A.; Becker, M. L. ACS Macro Letters 2012, 1, 1071.

(491) Wosnick, J. H.; Shoichet, M. S. Chemistry of Materials 2007, 20, 55.

(492) Wylie, R. G.; Shoichet, M. S. Biomacromolecules 2011, 12, 3789.

(493) Zeng, Y.; Ramya, T. N. C.; Dirksen, A.; Dawson, P. E.; Paulson, J. C. Nat Meth 2009, 6, 207.

(494) Orbán, E.; Mező, G.; Schlage, P.; Csík, G.; Kulić, Ž.; Ansorge, P.; Fellinger, E.; Möller, H.; Manea, M. Amino Acids 2011, 41, 469.

246

(495) Jencks, W. P. Journal of the American Chemical Society 1959, 81, 475.

(496) Grandjean, C.; Boutonnier, A.; Guerreiro, C.; Fournier, J.-M.; Mulard, L. A. The Journal of Organic Chemistry 2005, 70, 7123.

(497) Moore, J. S.; Stupp, S. I. Macromolecules 1990, 23, 65.

(498) Lee, M.; Lee, J.-P.; Rhee, H.; Choo, J.; Gyu Chai, Y.; Kyu Lee, E. Journal of Raman Spectroscopy 2003, 34, 737.

(499) Drury, J. L.; Mooney, D. J. Biomaterials 2003, 24, 4337.

(500) Gunn, J. W.; Turner, S. D.; Mann, B. K. Journal of Biomedical Materials Research Part A 2005, 72A, 91.

(501) Marklein, R. A.; Soranno, D. E.; Burdick, J. A. Soft Matter 2012, 8, 8113.

(502) Banerjee, A.; Arha, M.; Choudhary, S.; Ashton, R. S.; Bhatia, S. R.; Schaffer, D. V.; Kane, R. S. Biomaterials 2009, 30, 4695.

(503) Ifkovits, J. L.; Tous, E.; Minakawa, M.; Morita, M.; Robb, J. D.; Koomalsingh, K. J.; Gorman, J. H.; Gorman, R. C.; Burdick, J. A. Proceedings of the National Academy of Sciences 2010, 107, 11507.

(504) Flory, P. J. Journal of the American Chemical Society 1941, 63, 3091.

(505) Stockmayer, W. H. The Journal of chemical physics 1943, 11, 45.

(506) Byeon, J.-Y.; Limpoco, F. T.; Bailey, R. C. Langmuir 2010, 26, 15430.

(507) Gierlich, J.; Burley, G. A.; Gramlich, P. M. E.; Hammond, D. M.; Carell, T. Organic Letters 2006, 8, 3639.

(508) Weber, P.; Ohlendorf, D.; Wendoloski, J.; Salemme, F. Science 1989, 243, 85.

(509) Boekhoven, J.; Poolman, J. M.; Maity, C.; Li, F.; van der Mee, L.; Minkenberg, C. B.; Mendes, E.; van EschJan, H.; Eelkema, R. Nat Chem 2013, 5, 433.

(510) Rapoport, N. Progress in Polymer Science 2007, 32, 962.

(511) Lou, C.-W.; Yao, C.-H.; Chen, Y.-S.; Hsieh, T.-C.; Lin, J.-H.; Hsing, W.-H. Textile Research Journal 2008, 78, 958.

(512) Lecomte, P.; Riva, R.; Schmeits, S.; Rieger, J.; Van Butsele, K.; Jérôme, C.; Jérôme, R. Macromolecular Symposia 2006, 240, 157. 247

(513) Ji, S.; Bruchmann, B.; Klok, H.-A. Macromolecules 2011, 44, 5218.

(514) Tseng, Y. Y.; Li, W.-H. Proceedings of the National Academy of Sciences 2012, 109, 1170.

(515) Lempens, E. M.; Helms, B.; Merkx, M. In Bioconjugation Protocols; Mark, S. S., Ed.; Humana Press: 2011; Vol. 751, p 401.

(516) Costa, F.; Carvalho, I. F.; Montelaro, R. C.; Gomes, P.; Martins, M. C. L. Acta Biomaterialia 2011, 7, 1431.

(517) Gevrek, T. N.; Ozdeslik, R. N.; Sahin, G. S.; Yesilbag, G.; Mutlu, S.; Sanyal, A. Macromolecular Chemistry and Physics 2012, 213, 166.

(518) Spruell, J. M.; Wolffs, M.; Leibfarth, F. A.; Stahl, B. C.; Heo, J.; Connal, L. A.; Hu, J.; Hawker, C. J. Journal of the American Chemical Society 2011, 133, 16698.

(519) Leibfarth, F. A.; Kang, M.; Ham, M.; Kim, J.; Campos, L. M.; Gupta, N.; Moon, B.; Hawker, C. J. Nat Chem 2010, 2, 207.

(520) Debets, M. F.; van der Doelen, C. W. J.; Rutjes, F. P. J. T.; van Delft, F. L. ChemBioChem 2010, 11, 1168.

(521) Ma, Y.; Zheng, J.; Amond, E. F.; Stafford, C. M.; Becker, M. L. Biomacromolecules 2013.

(522) Prime, E. L.; Hamid, Z. A. A.; Cooper-White, J. J.; Qiao, G. G. Biomacromolecules 2007, 8, 2416.

(523) Tian, D.; Dubois, P.; Grandfils, C.; Jérôme, R. Macromolecules 1997, 30, 406.

(524) Latere Dwan'Isa, J.-P.; Lecomte, P.; Dubois, P.; Jérôme, R. Macromolecules 2003, 36, 2609.

(525) Sauerbrey, G. Z. Physik 1959, 155, 206.

(526) Pierschbacher, M. D.; Ruoslahti, E. Nature 1984, 309, 30.

(527) Acharya, A. P.; Dolgova, N. V.; Moore, N. M.; Xia, C.-Q.; Clare-Salzler, M. J.; Becker, M. L.; Gallant, N. D.; Keselowsky, B. G. Biomaterials 2010, 31, 7444.

(528) Shu, J. Y.; Panganiban, B.; Xu, T. Annual Review of Physical Chemistry 2013, 64, 631.

248

(529) Orelma, H.; Teerinen, T.; Johansson, L.-S.; Holappa, S.; Laine, J. Biomacromolecules 2012, 13, 1051.

(530) Montañez, M. I.; Hed, Y.; Utsel, S.; Ropponen, J.; Malmström, E.; Wågberg, L.; Hult, A.; Malkoch, M. Biomacromolecules 2011, 12, 2114.

(531) Zhang, Y.; Luo, S.; Tang, Y.; Yu, L.; Hou, K.-Y.; Cheng, J.-P.; Zeng, X.; Wang, P. G. Analytical Chemistry 2006, 78, 2001.

(532) Worrell, B. T.; Malik, J. A.; Fokin, V. V. Science 2013, 340, 457.

(533) Debets, M. F.; van Berkel, S. S.; Dommerholt, J.; Dirks, A. J.; Rutjes, F. P. J. T.; van Delft, F. L. Accounts of Chemical Research 2011, 44, 805.

(534) Elbert, D. L.; Hubbell, J. A. Biomacromolecules 2001, 2, 430.

(535) Lutolf, M. P.; Tirelli, N.; Cerritelli, S.; Cavalli, L.; Hubbell, J. A. Bioconjugate Chemistry 2001, 12, 1051.

(536) Ermakova, T.; Kuznetsova, N.; Tatarova, L.; Lopyrev, V. Russian Journal of General Chemistry 2007, 77, 132.

249

APPENDIX

1 Figure A. 1. H NMR (DMSO-d6) spectrum of PEU-1 monomer 4.1 (M1)

250

13 Figure A. 2. C NMR (DMSO-d6) spectrum of PEU-1 monomer 4.2(M1)

1 Figure A. 3. H NMR (DMSO-d6) spectrum of PEU-2 monomer 4.2(M2)

251

13 Figure A. 4. C NMR (DMSO-d6) spectrum of PEU-2 monomer 4.2(M2)

1 Figure A. 5. H NMR (CDCl3) spectrum of intermediate 4.3.

252

1 Figure A. 6. H NMR (CDCl3) spectrum of intermediate 4.4.

1 Figure A. 7. H NMR (CDCl3) spectrum of intermediate 4.5.

253

1 Figure A. 8. H NMR (DMSO-d6) spectrum of PEU-4 monomer 4.6 (M4).

13 Figure A. 9. C NMR (DMSO-d6) spectrum of PEU-4 monomer 4.6 (M4).

254

1 Figure A. 10. H NMR (CDCl3) spectrum of intermediate 4.7.

1 Figure A. 11. H NMR (CDCl3) spectrum of intermediate 4.8. 255

1 Figure A. 12. H NMR (CDCl3) spectrum of intermediate 4.9.

1 Figure A. 13. H NMR (DMSO-d6) spectrum of PEU-3 monomer 4.10 (M3).

256

13 Figure A. 14. C NMR (DMSO-d6) spectrum of PEU-3 monomer 4.10 (M3).

1 Figure A. 15. H NMR (CDCl3) spectrum of intermediate 4.11.

257

1 Figure A. 16. H NMR (DMSO-d6) spectrum of PEU-5 monomer 4.12 (M5).

258

13 Figure A. 17. C NMR (DMSO-d6) spectrum of PEU-5 monomer 4.12 (M5).

1 Figure A. 18. H NMR (CDCl3) spectrum of intermediate 4.13. 259

1 Figure A. 19. H NMR (DMSO-d6) spectrum of intermediate 4.14.

13 Figure A. 20. C NMR (DMSO-d6) spectrum of intermediate 4.14.

260

1 Figure A. 21. H NMR (CDCl3) spectrum of intermediate 4.15.

1 Figure A. 22. H NMR (DMSO-d6) spectrum of intermediate 4.16.

261

13 Figure A. 23. C NMR (DMSO-d6) spectrum of intermediate 4.16.

1 Figure A. 24. H NMR (DMSO-d6) spectrum of intermediate 4.17.

262

1 Figure A. 25. H NMR (DMSO-d6) spectrum of intermediate 4.18.

1 Figure A. 26. H NMR (DMSO-d6) spectrum of intermediate 4.19.

263

1 Figure A. 27. H NMR (DMSO-d6) spectrum of intermediate 4.20.

13 Figure A. 28. C NMR (DMSO-d6) spectrum of intermediate 4.20.

264

Figure A. 29. FT-IR spectrum of PEU-3 (azide-PEU) before (black) and after (red)

CuAAC.

265

Figure A. 30. Optical microscope image of PEU nanofibers (x20).

(Scale bar 50 um)

Figure A. 31. Optical microscope image of PEU nanofibers (x100).

(Scale bar 10 um)

266

Figure A. 32. UV-Vis absorption of PEU-4 nanofibers after thiol-ene reaction with Fmoc-

RGD-thiol.

Figure A. 33. Fluorescent microscope image of alkyne derived PEU-7 (keto-PEU)

nanofibers labeled with Chroemo 488 azide (x20). Scale bar 50 um.

267

Figure A. 34. Fluorescent images of nanofibers in controls (x20, 50 um).

(a) PEU-2, without copper catalyst.

(b) PEU-3, without copper catalyst.

(d) PEU-6, directly coupled with Chromeo 488 azide without intermediate 4.21

(e) PEU-7, using PEU-1 nanofibers as control.

(f) Alkyne derived PEU-7 nanofibers, directly coupled with Chromeo 488 azide without

intermediate 4.14.

268

Figure A. 35. ESI spectrum of alkyne-RGD-biotin.

269

Figure A. 36. MALDI spectrum of FITC-RGD-thiol.

270

Figure A. 37. MALDI spectrum of Fmoc-RGD-thiol.

Figure A. Figure 38. 1H NMR of PEU-2 nanofibers.

271

Figure A. Figure 39. 1H NMR of PEU-3 nanofibers.

Figure A. 40. 1H NMR of PEU-4 nanofibers.

272

Figure A. 41. 1H NMR of PEU-6 nanofibers.

Figure A. 42. 1H NMR of PEU-7 nanofibers.

273

Figure A. 43. ESI spectrum of 4-arm aminooxy crosslinker.

274

Figure A. 44. ESI spectrum of Alkyne-RGD-Biotin.

275

Figure A. 45. MALDI-TOF spectrum of FTIC-RGD-thiol.

276

Figure A. 46. 1H NMR spectrum of 2-(2-(aminooxy)acetoxy)ethyl acrylate (intermediate

5.1)

Figure A. 47. 1H NMR spectrum of O-(6-azidohexyl) hydroxylammonium chloride

(intermediate 5.2)

277

Figure A. 48. 1H NMR spectrum of Propargyl hydroxylammonium chloride (intermediate

5.3)

Figure A. 49. ESI spectrum of RGD-cyclooctyne

278

Figure A. 50. FT-IR spectra of PCL-azide solution casting film before (red) and after

(black) reaction with propargyl alcohol.

279

Figure A. 51. FT-IR spectra of PCL-alkyne solution casting film before (red) and after

(black) reaction with 3-azidopropan-1-ol.

280

Figure A. 52: MS Spectrum (Na+ as ion source) of PCL-MA with propargyl alcohol as

initiator.

CL unit: 114.1

MA unit: 299.1

Initiator: 56.06

Each peak can be assigned to the combination of CL, MA, and initiator.

281