<<

© 2018 Vrushali Bhagat ALL RIGHTS RESERVED

POLY(ESTER ) BASED BIOMIMETIC BONE AND SOFT

ADHESIVES

A Dissertation

Presented to

The Graduate Faculty of The University of Akron

In Partial Fulfillment

of the Requirements for the Degree

Doctor of Philosophy

Vrushali Bhagat

May, 2018

POLY(ESTER UREA) BASED BIOMIMETIC BONE AND SOFT TISSUE

ADHESIVES

Vrushali Bhagat

Dissertation

Approved: Accepted:

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

Committee Chair Dean of the College Dr. Ali Dhinojwala Dr. Eric J. Amis

Committee Member Dean of the Graduate School Dr. Chrys Wesdemiotis Dr. Chand Midha

Committee Member Date Dr. Mesfin Tsige

Committee Member Dr. Bryan Vogt ii

ABSTRACT

Sutures and staples are an integral part of and also the gold standard for surgical closure techniques. Additionally, in the case of orthopedic surgeries, metallic implants like plates, pins or screws are usually inserted as bone grafts. However, these techniques are quite invasive in nature, may cause wound dehiscence, secondary tissue damage, microbial infection, poor cosmetic outcome and more importantly discomfort to the patient. In some cases, the patient has to undergo another for removal or replacement of sutures or bone grafts. These shortcomings have led to an urgent need to develop alternative, less invasive surgical closure techniques. Use of tissue adhesives for wound closure is an attractive alternative over the invasive methods owing to ease of preparation and application, strong adhesion and gradual degradation with tissue reconstruction. However, their toxic degradation products and potentially harmful cross- linking strategies have limited their medical applications.

Recently, biomimetic adhesives have become a popular choice for application as tissue adhesives. Biomimetic adhesives are inspired from examples of adhesion in nature like , sandcastle worms, , caddisflies, geckos and spiders to name a few. The adhesion characteristics and properties of mussels are of particular interest due to their strong wet and reversible adhesion, steadfast hold under variation in temperature, pH and currents. The adhesive is rich in polyphenolic - 3,4-

dihydroxyphenylalanine (DOPA). The catechol functionality in DOPA is assumed to be iii the source of strong underwater adhesion in mussels. In this work, we have synthesized a

catechol functionalized soluble poly(ester urea) (poly(CA-Ser-co-Leu-co-PPG) copolymer for soft tissue application. Incorporation of 20 mol% PPG units in the backbone facilitates ethanol solubility making these adhesives clinically relevant. The polymer was characterized using NMR, IR and UV-VIS spectroscopy to confirm its structure and catechol content. The physical properties of the polymer like Tg, Td, Mn, Mw

and ÐM were characterized by DSC, TGA and SEC respectively. The lap shear adhesion

strength of this polymer on aluminum adherends was ~ 3.2 ± 0.8 MPa after cross-linking

with tetrabutylammonium periodate. On wet porcine skin, the adhesion strength was ~

10.6 ± 2.1 kPa after 4 h of curing for a catechol:crosslinker = 10:1 with minimal to no

toxicity. Moreover, the adhesion strength on wet porcine skin was much stronger than

Tisseel – commercial fibrin glue.

Caddisflies are aquatic organisms that spin their own adhesive underwater to form protective casing, capture prey and for locomotion. Caddisfly adhesive silk is comprised

of H-fibroin and L-fibroin in a 1:1 ratio and has abundant divalent cations like Ca2+. The

H-fibroin silk is heavily phosphorylated with repeating pSXn units where pS is

phosphoserine and X is a more hydrophobic amino acid like valine or isoleucine. The

phosphate groups play a role in underwater adhesion in addition to interacting with the

divalent cations to impart stiffness and tensile strength to the fiber. Phosphate groups also

have a strong electrostatic interaction with the Ca2+ on the bone surface which aids in

promoting bone adhesion. To test this hypothesis, we synthesized phosphate

functionalized poly(ester urea) – poly(pSer-co-Val) with 2% and 5% phosphate functionality. As confirmed via 1H, 13C, 31P NMR and ATR-IR spectroscopy. The

iv physical properties of the were characterized using DSC, TGA and SEC. The maximum lap shear adhesion strength on an aluminum adherend was 1.17 ± 0.19 MPa and 439 ± 203 kPa on wet bovine bone after cross-linking with Ca2+. The adhesion strength on bovine bone was comparable to the commercially available bone cement.

These polymeric mimics are degradable, non-toxic and soluble in a clinically relevant which sets them apart from the commercial options.

v

DEDICATION

I would like to dedicate this work to my family – mom, dad, sister and brother-in-law for their patience and unending support.

vi

ACKNOWLEDGEMENT

I would like to take this opportunity to express my gratitude towards people who have supported and helped me throughout this wonderful journey. I would first like to thank my advisor Prof. Matthew L. Becker for giving me excellent opportunities, his cooperation and valuable guidance during my graduate research. His constant support, creative ideas and profound knowledge have instilled me with a passion for my research.

I would also like to extend my gratitude towards my committee members: Prof. Ali

Dhinojwala, Prof. Mesfin Tsige, Prof. Chrys Wesdemiotis, Prof. Alamgir Karim and

Prof. Bryan Vogt for their time, valuable suggestions, meaningful discussions and constructive ideas. I want to thank Dr. Jinjun Zhou for helping me kick-start my research and for being an incredible mentor during my graduate journey.

I am very thankful to the Becker Research Group for their friendship and support which made this journey a smooth ride. I would also like to take this opportunity to thank my friends in Akron who have been like my extended family. With their love and encouragement I was able to stay committed to my goal. I would also like to thank my niece for cheering me up with her beautiful smiles and inspiring me every day to achieve my dream. Finally, I would like to convey my special thanks to my parents, sister and brother-in-law for constantly believing in me. Without their love, prayers and unconditional support I would not have achieved this feat.

vii

TABLE OF CONTENTS

Page

LIST OF TABLES ...... x

LIST OF FIGURES ...... xi

LIST OF SCHEMES ...... xxii

CHAPTER

I. INTRODUCTION ...... 1

1.1. FIBRIN GLUE ...... 4

1.2. -RESORCINOL-FORMALDEHYDE/GLUTARALDEHYDE GLUE (GRFG) ...... 7

1.3. CYANOACRYLATE GLUE ...... 10

1.4. , POLYPEPTIDE OR POLYMERIC ADHESIVES ...... 13

1.5. POLY(ETHYLENE GLYCOL) (PEG) BASED HYDROGEL ADHESIVES ...... 32

1.6. BIOMIMETIC TISSUE ADHESIVES ...... 40

II. MATERIALS AND INSTRUMENTS ...... 73

2.1. MATERIALS ...... 73

2.2. INSTRUMENTS...... 74

III. POLY(ESTER UREA) BASED ADHESIVES: IMPROVED DEPLOYMENT AND ADHESION BY INCORPORATION OF POLY(PROPYLENE GLYCOL) SEGMENTS ...... 78

3.1. OUTLINE ...... 78

viii 3.2. INTRODUCTION ...... 79

3.3. EXPERIMENTAL SECTION ...... 81

3.4. RESULTS AND DISCUSSION ...... 87

3.5. CONCLUSION ...... 98

IV. CADDISFLY INSPIRED PHOSPHORYLATED POLY(ESTER UREA)- BASED DEGRADABLE BONE ADHESIVES ...... 100

4.1. OUTLINE ...... 100

4.2. INTRODUCTION ...... 101

4.3. EXPERIMENTAL SECTION ...... 103

4.4. RESULTS AND DISCUSSION ...... 114

4.5. CONCLUSION ...... 128

V. SUMMARY AND FUTURE SCOPE ...... 130

APPENDIX ...... 161

APPENDIX 5.1 ...... 162

ix

LIST OF TABLES

Table Page

Table 1.1. Composition of commercial fibrin glue...... 7

Table 1.2. Surgical and medical applications of cyanoacrylate glue...... 12

Table 1.3. Polymeric mussel inspired adhesives with their crosslinking conditions and

maximum reported adhesion strengths...... 63

Table 3.1. Physical properties of PPG containing poly(ester-urea)s...... 92

Table 3.2. Normalized % viabilities of fibroblast cells on PEU substrates ...... 97

Table 4.1. Physical properties of the poly(ester urea)s ...... 119

Table 4.2. Polymer Surface Energies ...... 121

x

LIST OF FIGURES

Figure Page

Figure 1.1. Mechanism of clot formation in fibrin glue resembling physiological

coagulation...... 5

Figure 1.2. Crosslinking reaction of resorcin and formaldehyde under basic conditions... 9

Figure 1.3. Mechanism of cyanoacrylate polymerization...... 11

Figure 1.4. Gelatin-poly(L-glutamic acid) (PLGA) based hydrogels. (a) Hydrogels based

on N-hydroxysuccinimide (NHS) activated PLGA. (b) Gelatin-PLGA based

hydrogels: physically crosslinked by hydrogen bonding at low temperature and

chemically crosslinked in the presence of water soluble carbodiimides (WSC).

...... 14

Figure 1.5. N-ethyl-N-(3-dimethylaminopropyl) carbodiimide (EDC) and N-

hydroxysuccinimide (NHS) activated crosslinking of gelatin and alginate based

tissue adhesives ( – Gelatin, Al – Alginate)...... 16

Figure 1.6. Bonding mechanism of 70% CholGltn/Gltn-DST composition on porcine

arterial media. Improved tissue penetration and hydrophobic interactions

between the cholesteryl groups in addition to covalent crosslinking by DST

contribute to increased adhesion strength...... 17

xi Figure 1.7. Synthesis of gelatin-hydroxyphenyl propionic acid (GH) and gelatin-

hydroxyphenyl propionic acid - tyramine (GHT) conjugates. Enzymatic or

peroxide crosslinking yield with 2-3 times higher adhesion strength

compared to fibrin glue……...... 22

Figure 1.8. Mechanism of periodate based crosslinking between caffeic acid

functionalized gelatin and thiol functionalized keratin...... 23

Figure 1.9. Thiolated (CSS) and maleimide functionalized ε-polylysine (EPLM)

undergo rapid crosslinking by Michael addition reaction...... 24

Figure 1.10. Gallic acid conjugated chitin nanofiber precursors crosslink by two methods:

Periodate (NaIO4) – induced covalent hydrogel (color change from white to

brown) and FeCl3 – induced non-covalent, -complex hydrogel (color

change from white to red via purple) ...... 25

Figure 1.11. Thermoresponsive hyperbranched poly(amino acid)s functionalized with

DOPA, arginine, cysteine and/or lysine-acrylamide, demonstrate strong wet

adhesion. . 27

Figure 1.12. Hexamethylene diisocyanate modified trimethylene carbonate (TMC) and

trimethylolpropane ethoxylate (TMPE) hydrogels. a) PEG-(TMCm-HDI)2; b)

TMPE- (TMCm-HDI)3, (m = 1 or 2) ...... 29

Figure 1.13. Synthetic protocol of s-triazine based hyperbranched epoxy resin. L – linear

unit, D – dendritic unit, T – terminal unit...... 32

xii Figure 1.14. Aldehyde terminated PEG-PLA block copolymeric micelles crosslink with

polyamine and adheres to tissue surface by Schiff base reaction...... 35

Figure 1.15. Crosslinking reaction between aldehyde functionalized dextran and amine

functionalized PEG by Schiff base formation to form an adhesive hydrogel. . 36

Figure 1.16. Thiolated chitosan (CSS) and catechol-maleimide functionalized polylysine

crosslink within 10 s via Michael addition. Interactions like hydrogen bonding,

electrostatic interaction, π-π interaction and π-cation interaction contribute to

the strong bulk cohesive force ...... 38

Figure 1.17. Laponite incorporated, modified, 4-arm PEG showed improved

adhesion and cohesion via: (A) polymerization and di-DOPA crosslink

formation by periodate oxidation, (B) interfacial crosslinking with amine

groups on the tissue surface, (C) reversible bonding with Laponite ...... 39

Figure 1.18. Double crosslinked tissue adhesive derived from dopamine functionalized

gelatin macromer crosslinked by Fe3+ (rapid crosslinker) and genipin (GP)

(long term crosslinker) ...... 43

Figure 1.19. Chemical structure of a series of PEG-DOPA adhesives synthesized by Lee

et al...... 45

Figure 1.20. Illustration of the photopolymerization of Y-PPM into an adhesive hydrogel.

Y-PPM is the DOPA functionalized copolymer of methacrylated PEG-PLA block

copolymers...... 47

Figure 1.21. Periodate mediated oxidative cross-linking yields rapid gelation and tissue

adhesion of enzymatically degradable cAAPEG macromonomer. In the

xiii presence of enzyme neutrophil elastase, the Ala-Ala dipeptide linker (blue) is

cleaved to provide degradation sites. Black arrowheads indicate continuation

of the cross-linked hydrogel matrix………………………………………….48

Figure 1.22. (a-c) – Genetically engineered molecular hybrids of mussel adhesive

(Mfps) and amyloid based adhesive proteins in E. coli (CsgA), (d) –

Adhesive molecular hybrids self-assemble with β-sheet amyloid protein

forming the core and the Mfps flanking the exterior...... 51

Figure 1.23. Design strategy of DOPA modified recombinant elastin-like polypeptide

(ELP) synthesized by Liu and coworkers...... 52

Figure 1.24. NHS-thiol condensation based cross-linking of poly(AA-co-AANHS-co-

MDOPA) and thiol terminated 3-armed poly(ethylene glycol); AA – acrylic

acid, AANHS – acrylic acid N-hydroxysuccinimide ester and MDOPA - N-

methacryloyl-3,4- dihydroxy-L-phenylalanine...... 54

Figure 1.25. a) iCMBA pre-polymer synthesis by polycondensation reaction between

citric acid, PEG and dopamine. b) iCMBAs further modified with azide and

alkyne functionalities to facilitate dual crosslinking by catechol oxidation and

click reaction. .. 56

Figure 1.26. Nitrodopamine modified, catechol functionalized 4-arm PEG adheres by

oxidative and metal coordination crosslinking. Nitrodopamine groups are

responsible for debonding on photoirradiation...... 57

Figure 1.27. Chemical structure of catechol modified poly(ester-urea). a) Catechol

modified copolymer based on and leucine amino acids. b) Catechol

xiv modified terpolymer based on serine and leucine amino acid with

poly(propylene glycol) (PPG) groups incorporated in the backbone for ethanol

solubility ...... 59

Figure 1.28. Hydrogel formation or crosslinking reaction between 4-arm PEG-DA

(dopamine functionalized 4-arm PEG) and 4-arm-PEG-PBA (phenyl boronic

acid functionalized 4-arm PEG) ...... 60

Figure 1.29. Structures of PPDAC (Pluronic L-31-poly[(DOPA)-co-(Arg-co-Cys)]) and

PPDAL (Pluronic L-31-poly[(DOPA)-co-(Arg-co-Ac-Lys)]) adhesives.

Different functional groups result in different interactions: (a), (b) Covalent

interaction of catechol groups with tissue surface; (c) di-DOPA crosslinking

and DOPA polymerization; (d) electrostatic interaction between guadinium

(Gu+) and oxoanions on the tissue; (e) disulfide crosslink formation; (f)

covalent interaction between catechol and sulfhydryl groups; (g) thiol-ene click

reaction...... 62

Figure 1.30. Gecko inspired poly(glycerol-co-sebacate acrylate) (PGSA) tissue adhesive.

PGSA prepolymer is nanomolded by UV irradiation, followed by spin

with dextran aldehyde (DXTA). SEM image showed excellent pattern transfer

fidelity ………………………………………………………………………..68

Figure 1.31. Synthetic adhesive mimic of P. californica glue. 1: Structure of Pc3 analog;

2: Structure of Pc1 analog. Model of pH dependent coacervation and adhesion.

(a) At acidic pH 4, polyphosphates (black) and polyamines (grey) form

colloidal polyelectrolyte complexes (PECs) with a net positive charge, (b) At

slightly basic pH of 8.2, the extended polyphosphates form network with xv polyamines and divalent cations with a net negative charge, (c) On oxidation

3,4-dihydroxyphenol (D) initiates covalent crosslinking and surface interaction

via quinones (Q) ...... 70

Figure 1.32. Caddisfly adhesive mimic of phosphate functionalized poly(ester urea)

copolymer based on serine and valine amino acids...... 72

Figure 3.1. Structure of N-(3,4-dihydroxyphenethyl)methacrylamide ...... 85

Figure 3.2. 1H NMR spectra of (A) Poly(bzlSer-co-Leu-co-PPG) (B) Poly(Ser-co-Leu-co-

PPG) (C) Poly(CA-AN-Ser-co-Leu-co-PPG) (D) Poly(CA-Ser-co-Leu-co-PPG)

...... 89

Figure 3.3. 13C NMR of (A) Poly(Ser-co-Leu-co-PPG) (B) Poly(CA-AN-Ser-co-Leu-co-

PPG) and (C) Poly(CA-Ser-co-Leu-co-PPG) ...... 90

Figure 3.4. FT-IR spectra of (A) Poly(Ser-co-Leu-co-PPG) (b) Poly(CA-AN-Ser-co-Leu- co-PPG) (c) Poly(CA-Ser-co-Leu-co-PPG) ...... 91

Figure 3.5. UV-vis spectrum of Poly(CA-Ser-co-Leu-co-PPG) showing a peak around

283 nm characteristic of the π-π* absorption of the catechol group. Inset shows

the calibration curve constructed from N-(3,4-

dihydroxyphenethyl)methacrylamide and used to determine the concentration

of catechol groups in the terpolymer...... 93

Figure 3.6. (A) Illustration of the lap-shear adhesion test272 (B) Comparison of adhesion

strength on aluminum substrates; Ser-PEU: Poly(Ser-co-Leu-co-PPG) –

- Control; CA- PEU: Poly(CA-Ser-co-Leu-co-PPG); CA-PEU+IO4 : Poly(CA-

Ser-co-Leu-co-PPG) crosslinked with Bu4N(IO4). * represents p < 0.05 for

xvi samples compared with the control (n = 10)...... 94

Figure 3.7. Lap shear adhesion strengths on porcine skin substrates. Ser-PEU: Poly(Ser-

co-Leu-co-PPG); CA-PEU: Poly(CA-Ser-co-Leu-co-PPG); CA:IO4-:

Poly(CA-Ser-co- Leu-co-PPG) crosslinked with Bu4N(IO4). * indicates p <

0.05 for 30 min and 4 h curing of each sample. ** indicates p < 0.05 for 30 min

samples and *** indicates p < 0.05 for 4 h samples. (n ≥ 5) ...... 96

Figure 3.8. NIH 3T3 fibroblast cell viability on PEU films. Cells stained green are live

cells and the cells stained red are dead. (A) Blank glass substrate - Control, (B)

Poly(Ser- co-Leu-co-PPG) (Ser-PEU) – Control, (C) Poly(CA-Ser-co-Leu-co-

PPG) (CA-PEU) and (D) Poly(CA-Ser-co-Leu-co-PPG) crosslinked with

Bu4N(IO4) (xlinked CA-PEU). Scale bar: 1mm. (E) Normalized %Cell

Viability of NIH 3T3 cells on PEU films. Cell viability was calculated from a

total of 10 images and 3 replicates of each sample were used. No significant

difference in cell viability was observed among PEU samples (p > 0.05) and

the glass control. ……………………………………………………………98

Figure 4.1. 1H NMR spectra of phosphoserine monomer synthesis. (a) bis-N-boc-O-

benzyl(L-serine)-1,8-octanyl diester (M2), (b) bis-N-boc(L-serine)-1,8-octanyl

diester (M3), (c) bis-N-boc-O-diphenylphosphate(L-serine)-1,8-octanyl diester

(M4), (d) dihydrochloride salt of bis-O-diphenylphosphate(L-serine)-1,8-

octanyl diester (M5) ... ……………………………………………………..115

Figure 4.2. 1H NMR spectra of (a) 5% Poly(SerDPP-co-Val), (b) 5% Poly(pSer-co-Val);

31P NMR spectra of (c) 5% Poly(SerDPP-co-Val), (d) 5% Poly(pSer-co-Val),

referenced to 85% H3PO4 as external standard. Inset shows magnification from xvi

δ ~ 7.1-7.3 ppm and δ ~ 4.0 – 4.5 ppm; characteristic peaks of the diphenyl

protecting groups disappear after deprotection. A triplet at δ ~ 4.37 ppm is

characteristic of the proton environment on the methylene group attached to the

deprotected phosphate group, which is not prominent before deprotection…117

Figure 4.3. ATR-IR spectra of Poly(1-Val-8), Poly(SerDPP-co-Val) and Poly(pSer-co-

Val). The peaks corresponding to the P=O (1040 cm-1) and P-O stretching (710

- 675 cm- 1) are highlighted and labelled...... 118

Figure 4.4. (a) DSC curves to determine the glass transition temperatures (Tg). Tg of the

phosphate functionalized polymers do not show significant change after

deprotection because of the low functionality on the polymer backbone chains.

(b) Thermogravimetric analysis curves to determine the decomposition

temperature (Td) of the polymers. A slight decrease in the decomposition

temperature is observed after deprotection of the diphenyl groups...... 120

Figure 4.5. (a) Lap shear adhesion on aluminum adherends at room temperature.

Adhesion strengths were calculated from average of 10 replicates (n=10) and

reported with standard errors. Aluminum adherends show cohesive failure for

all samples, (b) poly(1-Val-8), (c) PMMA bone cement, (d) 2% poly(pSer-co-

Val), (e) 5% poly(pSer-co- Val), (f) 2% poly(pSer-co-Val) crosslinked with 0.3

eq. Ca2+, (g) 5% poly(pSer-co-Val) crosslinked with 0.3 eq. Ca2+ ...... 122

Figure 4.6. Phosphate groups ( ) on Poly(pSer-co-Val) interact with positive charges on

the bone surface promoting adhesion to the bone. Ca2+ from calcium iodide

(cross-linking agent) interact with phosphate groups in the bulk of the polymer

giving rise to cohesive forces...... 124

xvi

Figure 4.7. (a) End-to-end adhesion on bovine bone at room temperature. Adhesion

strengths were calculated from average of 3 replicates (n=3) and reported with

standard errors, (b) Schematic of end-to-end adhesion on bovine bone sample

(Left) and end-to- end adhesion test on texture analyzer (Right), (c) Image

showing cohesive failure of 5% poly(pSer-co-Val) crosslinked with 0.3 eq. of

Ca2+ ...... 126

Figure 4.8. Cell viability ((a)-(f)) and spreading analysis of MC3T3 cells ((g)-(l)) on day

1 and day 3 respectively. (a) and (g) Glass substrate, (b) and (h) poly(1-Val-8),

(c) and (i) 2% poly(pSer-co-Val), (d) and (j) 5% poly(pSer-co-Val), (e) and (k)

2% poly(pSer-co- Val) crosslinked with 0.3 eq. Ca2+, (f) and (l) 5% poly(pSer-

co-Val) crosslinked with 0.3 eq. Ca2+. Cell viability on different samples was

studied by live-dead assay from atleast 40 images per sample (4x

magnification, scale bar ~ 200 μm). Live cells were stained green by calcein

AM and dead cells were stained red by ethidium homodimer. For spreading

studies cells were stained with rhodamine phalloidin – actin filaments (red),

alexa fluor 488 secondary antibody – focal adhesion points (green) and dapi –

nuclei (blue). Images were taken at 20x magnification (Scale bar ~ 50 μm) and

aspect ratios were calculated from 30 cells per sample...... 127

Figure 4.9. (a) Normalized cell viability of MC3T3 cells on polymers, (b) Comparison of

aspect ratios on different polymers. MC3T3 cells show similar spreading

behavior on all samples...... 128

Figure 5.1. 1H NMR spectrum of 4-(dimethylamino)pyridinium 4-toluenesulfonate

(DPTS) ...... 163

xix Figure 5.2. 1H NMR spectrum of acetonide protected 3,4-dihydroxyhydrocinnamic acid

...... 164

Figure 5.3. 13C NMR spectrum of acetonide protected 3,4-dihydroxyhydrocinnamic acid

...... 164

Figure 5.4. 1H NMR spectrum of di-p-toluenesulfonic acid salt of bis(L-leucine)-1,8-

octanyl diester (M1) ...... 165

Figure 5.5. 13C NMR spectrum of di-p-toluenesulfonic acid salt of bis(L-leucine)-1,8-

octanyl diester (M1) ...... 165

Figure 5.6. 1H NMR spectrum of bis-N-Boc-O-benzyl(L-serine)-1,8-octanyl diester ... 166

Figure 5.7. 13C NMR spectrum of bis-N-Boc-O-benzyl(L-serine)-1,8-octanyl

diester ……………………………………………………………………….166

Figure 5.8. 1H NMR spectrum of di-hydrochloric acid salt of bis-O-benzyl(L-serine)-1,8-

octanyl diester (M2) ...... 167

Figure 5.9. 13C NMR spectrum of di-hydrochloric acid salt of bis-O-benzyl(L-serine)-1,8-

octanyl diester (M2) ...... 167

Figure 5.10. 1H NMR spectra of PPG-PEU Polymer (a) Poly(bzlSer-co-Leu-co-PPG), (b)

Poly(Ser-co-Leu-co-PPG), (c) Poly(CA-AN-Ser-co-Leu-co-PPG), (d)

Poly(CA-Ser-co- Leu-co-PPG) ...... 168

Figure 5.11. 1H NMR spectra of di-p-toluenesulfonic acid salt of bis(L-valine)-1,8-

octanyl diester (M1) ...... 168

Figure 5.12. 1H NMR spectra of Poly(1-Val-8) ...... 169

xx Figure 5.13. 13C NMR spectra of Poly(1-Val-8) ...... 169

Figure 5.14. 1H NMR spectra of dihydrochloride salt of bis-O-diphenylphosphate(L-

serine)-1,8-octanyl diester (M5) ...... 170

Figure 5.15. 1H NMR spectra of 2% Poly(SerDPP-co-Val). R group denotes diphenyl

protected phosphate groups. Inset shows aromatic peaks from the diphenyl

protecting groups ...... 171

Figure 5.16. 1H NMR spectra of 2% Poly(pSer-co-Val). R’ denotes deprotected

phosphate groups. Inset (a) shows disappearance of the aromatic peaks

between 7.0 –7.25 ppm confirming deprotection of the diphenyl groups; (b)

Appearance of a triplet around ~ 4.37 ppm corresponds to the proton

environment on methylene group attached to the deprotected phosphate group

……………………………………………………………………………….171

Figure 5.17. 13C NMR spectra of 2% Poly(SerDPP-co-Val) ...... 172

Figure 5.18. 1H NMR spectra of: (a) 5% Poly(SerDPP-co-Val), (b) 5% Poly(pSer-co-

Val) ...... 172

Figure 5.19. 13C NMR spectra of: (a) 5% Poly(SerDPP-co-Val), (b) 5% Poly(pSer-co-

Val) ...... 173

xxi

LIST OF SCHEMES

Scheme Page

3.1. Synthesis of di-p-toluenesulfonic acid salt of bis(L-leucine)-1,8-octanyl diester (M1).

...... 82

3.2. Synthesis of di-hydrochloric acid salt of bis-O-benzyl(L-serine)-1,8-octanyl diester

(M2)...... 83

3.3. Synthesis route of Poly(CA-Ser0.2-co-Leu0.6-co-PPG0.2)...... 88

4.1. Synthesis of di-p-toluenesulfonic acid salt of bis(L-valine)-1,8-octanyl diester (M1).

...... 104

4.2. Synthesis of 1,8-octanediol-L-valine poly(ester urea) (Poly(1-Val-8))...... 105

4.3. Synthesis of dihydrochloride salt of bis-O-diphenylphosphate(L-serine)-1,8- octanyl

diester from N-boc-O-benzyl(L-serine) and 1,8-octanediol ...... 107

4.4. Synthesis of phosphate functionalized PEU: (a) Copolymerization of di-p-

toluenesulfonic acid salt of bis(L-valine)-1,8-octanyl-diester (M1) and

diphenylphosphoserine monomer (M5) in the presence of triphosgene to obtain

Poly(SerDPP-co-Val), (b) Deprotection of phenyl protecting groups by

hydrogenolysis to obtain Poly(pSer-co-Val) ...... 116

5.1. Synthesis of acetonide-protected 3,4-dihydroxyhydrocinnamic acid...... 163

xxii

CHAPTER I

INTRODUCTION

V Bhagat, ML Becker, Degradable adhesives for surgery and tissue engineering.

Biomacromolecules, 2017, 18(10), 3009-3039

Many surgical procedures are performed worldwide and the number continues to grow every year. In a recent study, over 300 million surgeries were performed in 2012; an increase of 33.6% over the last 8 years.1, 2 Current surgical closure techniques involve use of invasive techniques like sutures, staples or clips which often result in secondary tissue damage, microbial infection, fluid or air leakage and poor cosmetic outcome.3 Despite being the current standard of care, the use of sutures is tricky during intricate and sensitive surgeries such as vascular anastomosis, nerve repair or ocular surgeries because of the high risk factor involved in these surgeries. The use of metallic grafts like plates, pins or screws, is common practice in orthopedic surgeries for assistance in osseointegration. However, the modulus mismatch between the stiff graft material and the bone tissue can result in localized stress at the point of fixation and bone atrophy. The metallic grafts act as support medium that lacks chemical interaction with the bone, often suffer from aseptic loosening resulting in poor bone healing.4 In spite of these shortcomings, sutures, staples and metallic grafts still remain the gold standard3 for tissue reconstruction owing to a lack of non-invasive techniques capable of outperforming

1

them. An appealing option to alleviate the use of these invasive techniques is the use of

tissue adhesives. An adhesive spreads over the entire contact area which eliminates stress

localization facilitating load transfer between the fractured surfaces.4, 5 Additionally,

adhesives are easy to apply, join dissimilar materials, increase design flexibility, improve

cost effectiveness, can act as sealants or hemostatic agents to prevent fluid leakage

through the anastomosed site and cause minimal or no tissue damage at the application

site.6 These attractive advantages of using adhesives over traditional closure techniques have led to a steep rise in adhesive research and development and it is estimated that adhesives currently constitute a market share of ~ $38 billion.7

Bone and tissue adhesives have been around for centuries with one of the most ancient

materials used as a tissue adhesive being ‘Plaster of Paris’. Since then, a number of

naturally derived, semi-synthetic or synthetic adhesives have been developed.

Donkerwolcke and Muster have discussed the evolution of adhesives from as early as the

1700s to the 90s.8 The journey from sutures or metallic implants (metallic plates, pins and

screws) to tissue adhesives,8, 9 the variety of tissue adhesives currently available,3, 9-11

their clinical applications 12-14 and future prospects 11, 15 have been extensively covered in

the previous review articles.16-20

Based on their function tissue glues can be divided as hemostats, sealants and adhesives. Although, they are often addressed interchangeably; they are quite different

from each other. A hemostat is responsible for clotting and fails to function in the

absence of blood. A sealant develops a barrier layer to prevent leakage of fluid or

while an adhesive functions to bind two surfaces firmly and hold them together. While a

hemostat performs effectively in the presence of blood; sealants and adhesives often fail

2

to perform under wet conditions.14, 15 A tissue adhesive should have strong wet adhesion,

stability under physiological conditions, rapid curing/crosslinking without excessive heat

generation, non-toxicity, cytocompatibility, minimum swelling, modulus comparable to

the underlying tissue, biodegradability and bioresorbability. Adhesion between two

substrates is a result of a combination of adhesive and cohesive strengths. For strong

adhesive performance, it is imperative to attain an optimum balance between the adhesive

and cohesive strengths of the material.8 Adhesive and cohesive interactions involve

mechanical interlocking, intermolecular bonding, electrostatic bonding, chain

entanglement or cross-link formation.21, 22

Initial attempts to develop tissue adhesives involved use of epoxy resin, polyurethane foam, poly(methyl methacrylate) and calcium/magnesium phosphate based bone cement,

lactide-methacrylate based systems, zinc polycarboxylate and glass ionomer based

cements. The commercial or clinical application of these materials was limited due to

their lack of degradability, low bonding strengths, high infection rates, foreign body

reactions or leaching of toxic metallic ions into the body.9, 11 Adhesives like fibrin glue,

cyanoacrylate glue and gelatin-resorcin formaldehyde/glutaraldehyde glues, which do have certain limitations like low bonding under wet conditions and poor cytocompatibility, have been approved for clinical use and discussed in the following sections.12, 14, 23 Recently, there has been considerable interest in the development of biomimetic tissue adhesives inspired from natural examples of adhesion like mussels,

sandcastle worms, caddisflies and geckos. Though the exact mechanism of adhesion in

3 these organisms is still under debate, several attempts to mimic these adhesives have been

made by different research groups and will be discussed further in this article.24, 25

The following sections review the structure, composition and functioning mechanism of semi-synthetic and synthetic tissue adhesives, hemostats and sealants. Since, biodegradability and tissue compatibility are essential requirements for tissue adhesives,

only biodegradable tissue adhesives have been discussed in this article. In addition, their

adhesion/bonding strengths, suggested clinical applications and limitations are also

reviewed.

1.1. FIBRIN GLUE

A primitive, unprocessed form of fibrin glue containing fibrinogen and thrombin was

first introduced in the 1940s.26 Alving and coworkers in 1995 summarized different fibrin

compositions, their applications, adverse reactions or complications arising from their

use, possible new applications and the need for controlled trials to determine their clinical

efficiency. The fibrin glues synthesized in Europe were a step ahead compared to USA in

which their compositions involved use of antifibrinolytic agents like aprotinin and

epsilon amino caproic acid; though the efficacy of using such antifibrinolytic agents was

not proven.27 As described by Martinowitz and Saltz, clot formation in fibrin glue

resembles the final step in physiological coagulation.28 Briefly, fibrin sealants consist of

two major components: fibrinogen with factor XIII and thrombin with Ca2+. Thrombin

cleaves off fibrinopeptide A and B from α and β chains (respectively) in fibrinogen to

form fibrin monomer. The as formed monomer physically crosslinks via hydrogen

4 bonding to form an unstable clot. Factor XIII is a fibrin stabilizing factor activated by

thrombin using Ca2+ as a co-factor to form factor XIIIa. Factor XIIIa then acts upon the fibrin monomer or the unstable clot to form crosslinks in the form of amide bonds between glutamine and lysine residues resulting in an insoluble clot resistant to proteolytic cleavage (Figure 1.1).

Figure 1.1. Mechanism of clot formation in fibrin glue resembling physiological

coagulation.

The crosslinking reaction also involves attachment of plasmin inhibitors like α2 plasmin inhibitor (α2-PI), α2-macroglobulin and plasminogen activator inhibitor 2 (PAI-2) to the α chain of fibrin which further strengthens the clot and prevents fibrinolysis. Factor XIII also acts on other adhesive like fibronectin, thrombospondin, vitronectin and von Willebrand factor. A number of crosslinking steps are involved in clot formation, for example, fibrin primarily crosslinks with both collagen and adhesive glycoproteins at the wound site. Simultaneously, crosslinking also occurs between the adhesive glycoproteins with collagen and other tissue proteins. The cumulative result of all the

5 crosslinks at the wound site and the presence of plasmin inhibitors is the formation of a strong, adhesive, insoluble clot; resistant to fibrinolysis.28

Both autologous and homologous fibrin sealants have been developed based on whether the plasma is obtained from the same patient or another person respectively.

Fibrin glue is biocompatible, resorbable and does not cause tissue necrosis, fibrosis or inflammation. The degradation time of fibrin glue varies from a few days to months depending on the composition.14 Despite the use of fibrin glue as a hemostatic agent for a range of surgeries, the risk of virus transmission still prevails. The fibrin glue components are subjected to virus screening and virus inactivation or reduction treatments like pasteurization, two step vapor heat treatment, solvent-detergent cleansing, dry heat treatment, nanofiltration, precipitation, pH treatment and some chromatographic steps. However, a particular treatment is not effective against all the viruses and a combination of these treatments is generally required for medical application.29 An extensive review on the composition and fairly recent (year 2013-2014) medical applications of fibrin glue as a hemostat, sealant or adhesive has been compiled by

Spotnitz.14

The composition of commercially available fibrin glues from different parts of the world compiled by Jackson is presented in Table 1.1.29 A comparison between the adhesives emphasizes that they differ in concentration of their main components – fibrinogen and thrombin, the source of thrombin as well as the method used for virus inactivation. The mechanical strength of the fibrin clot is a function of the fibrinogen concentration and often used as the measure of the glue quality, while thrombin 6 concentration directly relates to the speed of clotting. Therefore, an optimum

concentration of both components is necessary to achieve rapid hemostasis as well as

satisfactory adhesion and mechanical properties.27, 30 Adhesion strength of the fibrin glue

depends on the substrate, composition of the glue, method of fibrinogen preparation,

presence of water, fat or collagen and setting time making it redundant to deduce an

universal value for adhesion strength of fibrin glue.31

Table 1.1. Composition of commercial fibrin glue.

Reprinted from Jackson, M. R. Fibrin sealants in surgical practice: An overview, 1S-7S, Am. J. Surg., 182, Copyright 2001, with permission from Elsevier.

1.2. GELATIN-RESORCINOL-FORMALDEHYDE/GLUTARALDEHYDE GLUE

(GRFG)

Gelatin-resorcin-formaldehyde/glutaraldehyde glue was reported as early as 1966. This

adhesive is a combination of all the individual components. The resorcin-formaldehyde

7 forms a crosslinked polymer in basic conditions as shown in Figure 1.2.32 Adhesives based on formaldehyde have strong initial bonding while glutaraldehyde (GA) based adhesives show higher stability in vivo. Therefore, the adhesive formulations sometimes incorporate both formaldehyde and glutaraldehyde to improve adhesion as well as in vivo stability. Gelatin incorporation in the adhesive imparts elasticity comparable to underlying tissue as well as degradability.32, 33 Bonding strength of GRFG on dry substrate is comparable to cyanoacrylate glue and significantly stronger than fibrin glue but deteriorates under wet conditions.33 Histology studies on rat femoral vessels after

GRFG application indicated that the long term adhesive nature of the glue is not intrinsic but arises from extracellular remodeling around the blood vessels.34 GRFG has been used as a hemostatic agent, adhesive for vascular surgeries, gastrointestinal surgery, thoracoscopic operations and lung surgeries.34-39 GRFG glue with a gelatin- resorcin:formaldehyde/glutaraldehyde ratio of 2:1 has shown maximum adhesion strength of 170.5 ± 41.5 kPa in dry conditions and 47.8 ± 17.6 kPa under wet conditions.

Although, GRFG glue has shown impressive hemostatic properties and satisfactory adhesive properties, the possible carcinogenicity associated with the use of aldehydes limits the clinical usage of this glue.33, 35 Results of the histological studies on tissues treated with GRFG glues from different research groups have been conflicting.34 Short term effects were limited to minimal tissue necrosis and local tissue inflammation in most of the studies; however, more studies are required to evaluate the long term or cytotoxic effects of the glue.

8

Reproduced from Lin C.; Ritter, J. A. Effect of synthesis pH on the structure of carbon xerogels, Carbon, 35, 1271- 1278, Copyright 1997, with permission from Elsevier.

Figure 1.2. Crosslinking reaction of resorcin and formaldehyde under basic conditions.

A few other studies have incorporated GA as a cross-linking agent to promote tissue adhesion. Matsuda and coworkers demonstrated the benefit of adding GA to gelatin films in order to promote tissue adhesion. The dual role of GA crosslinking with the amine groups on gelatin as well as interactions with the amine groups on tissue leads to the impressive adhesion strength of these GA crosslinked gelatin films. The aldehyde content in the system was dependent on temperature, pH, treatment time of gelation with GA and

GA concentration; consequently, the bonding strength increased with increasing aldehyde concentration. The bonding strength significantly decreased after aldehyde reduction confirming the role of aldehydes in promoting adhesion. The maximum bonding strength

9 was 250 gf/cm2 (~ 24.5 kPa) on dry porcine skin and deteriorated to negligible bonding

on wet skin.40 Addition of proteinoids to the gelatin-glutaraldehyde system resulted in

improved adhesion strength along with lower toxicity. Proteinoids are synthetic

copolymers of amino acids, particularly RGDKANE which increase the functionality in

the adhesive leading to improved cross-linking and bonding strength.41 Gelatin-resorcin based adhesives crosslinked with epoxy compound (GRE), water soluble carbodiimide

(GRC) or genipin (GG) instead of formaldehyde or glutaraldehyde showed longer crosslinking time and lower bonding strength. While GRC and GG glue had greater cytocompatibility, GRE glue was not deemed suitable for clinical applications.42 Jebrail

and coworkers used GA as a cross-linker for bovine serum albumin (BSA) adhesive. The bonding strengths were studied on samples and impressive strength of ~ 6.74 MPa was obtained on hydrated wood. However, the mechanism of adhesion was not clearly explained. The bonding strengths on wood samples are not translational to clinical or biological adhesion making it difficult to infer performance in physiologically relevant models.43

1.3. CYANOACRYLATE GLUE

Cyanoacrylate glue is a class of synthetic adhesives made from alkyl α-cyanoacrylates

for tissue adhesive application since the 1980s. Cyanoacrylates polymerize rapidly in the

presence of weak basic conditions, for example, water or blood (Figure 1.3). The amine

groups in proteins present on the tissue surface are also speculated to initiate

cyanoacrylate polymerization, resulting in covalent bonding between tissues and the

10 resulting adhesive layer likely responsible for the impressive adhesive strength of

cyanoacrylates. Cyanoacrylate glue is superior to the rival adhesives in terms of strong

wet adhesion, low cost, rapid curing, inherent bactericidal properties and good cosmetic

outcome.25, 44-47 These properties make cyanoacrylates an attractive choice for wound closure or as hemostatic agents to be used in adjunct with the traditional closure techniques. However, the rapid polymerization of cyanoacrylate monomers is associated with significant heat dissipation at the application site resulting in formation of a hard and brittle film.

Reproduced from ref. 25 with permission of The Royal Society of Chemistry. http://dx.doi.org/10.1039/c2bm00121g.

Figure 1.3. Mechanism of cyanoacrylate polymerization.

The properties of cyanoacrylate glue can be tuned by changing the chain length of the

ester side chain. For example, methyl α-cyanoacrylate elicits a severe inflammatory

response in tissues while octyl α-cyanoacrylate exhibits minimal inflammation.44, 48 N-

butyl-2-cyanoacrylate based commercial glues, Histoacryl (Germany) and Glubran

(Italy), are commercially used in Europe as tissue adhesives during endoscopic surgeries.

2-octyl cyanoacrylate based glue Dermabond (USA) is FDA approved for medical use in

topical applications only. Mizrahi and coworkers replaced the alkyl side chains with

11 alkoxy chains (ether linkages) to improve the elastic and mechanical properties of the glue. 2-octyl cyanoacrylate (Dermabond) however, demonstrated higher adhesion compared to the alkoxy analog.49 Bond strengths of modified cyanoacrylates: alkyl 2- cyanoacryloyl glycolate and 1,2-isopropylidene glyceryl 2-cyanoacrylate even though comparable to alkyl 2-cyanoacrylates, exhibited rapid degradation.50-52 Cyanoacrylates were also modified by copolymerizing with 1,1,2-trichlorobutadiene-1,3-methyl methacrylate (MMA), incorporation of elastomeric polymers and addition of fillers or modifiers like polymeric oxalates to improve their bonding strength, flexibility and biocompatibility.53-58 Table 1.2 is an extensive list of surgical and medical applications of cyanoacrylates compiled by Leggat and coworkers in addition to a few other references

59-62:

Table 1.2. Surgical and medical applications of cyanoacrylate glue.

Specialty Past, present and potential applications

Surgical wound repair

General surgery Control of hemorrhage

Emergency medicine and general practice Traumatic wound repair

Control of variceal bleeding and obliteration of Endoscopy oesophagogastric varices

Ophthalmology Temporary repair of corneal perforations

Arterial surgery Arterial anastomoses

Thoracic surgery Closure of pulmonary leaks Repair of peripheral nerves

Neurosurgery Microanastomosis of sciatic nerve

Otological surgery Ossicular chain reconstruction

12 Embolotherapy of various vascular Interventional radiology and cardiology abnormalities including aneurysms Pediatrics Wound closure in children

Pediatric endoscopic surgery Tissue approximation and hemostasis

Pharmacotherapeutics Drug carriers

Reproduced from ref 59 with permission from John Wiley and Sons. Copyright 2007 Royal Australasian College of Surgeons

Despite the attractive properties and impressive wet adhesion, use of cyanoacrylate

glue is limited to topical applications owing to the toxic nature of the degradation

products. The shorter alkyl chain cyanoacrylates degrade faster resulting in a stronger

tissue inflammatory response while the longer alkyl chain cyanoacrylates degrade rather

slowly preventing the build-up of toxic products and facilitating their removal from the

body resulting in mild inflammation. Nevertheless, cyanoacrylates have caused chronic

inflammation, tissue necrosis in vivo, intimal thickening, arterial ocular lesion,

occupational asthma, dermatitis, in vitro cytotoxicity for cells in direct contact as well as

in leached to list a few.59, 63, 64 Full potential of cyanoacrylate glue cannot be

realized in vivo or commercially unless the problem of toxicity is resolved.

1.4. POLYSACCHARIDE, POLYPEPTIDE OR POLYMERIC ADHESIVES

Polysaccharides, polypeptides or proteins are rich in amine, hydroxyl or carboxylic

acid functionalities. Adhesives developed from these building blocks adhere to amine

groups on tissue surface via covalent interaction by N-hydroxysuccinimide activation or

Schiff base formation, Michael addition reaction, biaryl formation, imine formation or π-

π interaction. The following section is focused on gelatin, chitosan, dextran and alginate

13 based adhesives. In addition, adhesives developed using isocyanate and acrylate functionalized polymers and hyperbranched polymers are also discussed in this section.

Gelatin is a thermally denatured collagen composed of polypeptides and proteins, rich in amines and carboxylic acid groups. The degradability and biocompatibility of gelatin makes it one of the most popular choices for application in tissue adhesives. In the late

90s Ikada and coworkers performed extensive research to develop gelatin based bioresorbable, adhesive hydrogels. Gelatin-poly(L-glutamic acid) (PLGA) based adhesive hydrogels were developed using water soluble carbodiimide as a cross-linker and/or N- hydroxysuccinimide activated PLGA. These hydrogels provided strong adhesion within a short gelation time compared to the commercial fibrin glue on mouse soft tissue (Figure

1.4).65-69

(a) Reprinted from Iwata, H.; Matsuda, S.; Mitsuhashi, K.; Itoh, E.; Ikada, Y., A novel surgical glue composed of gelatin and N-hydroxysuccinimide activated poly(L-glutamic acid) and its gelation with gelatin. Biomaterials, 19, 1869- 1876, Copyright 1998, with permission from Elsevier. (b) Reprinted from Otani, Y.; Tabata, Y.; Ikada, Y., Effect of

additives on gelation and tissue adhesion of gelatin-poly(L-glutamic acid) mixture. Biomaterials, 19, 2167-2173, Copyright 1998, with permission from Elsevier.

Figure 1.4. Gelatin-poly(L-glutamic acid) (PLGA) based hydrogels. (a) Hydrogels based on N-hydroxysuccinimide (NHS) activated PLGA. (b) Gelatin-PLGA based hydrogels:

14 physically crosslinked by hydrogen bonding at low temperature and chemically crosslinked in the presence of water soluble carbodiimides (WSC).

In a similar work, Zilberman and coworkers replaced PLGA with an anionic polysaccharide –alginate, to develop gelatin based tissue adhesives crosslinked with carbodiimides and/or NHS activated route. They also loaded these adhesives with anesthetic, analgesic, antibiotic drugs, hemostatic agents or bioactive (Figure

1.5). Excessive swelling of the adhesive, cytotoxicity of the cross-linker, burst release of the drugs and their unpredictable effect on bonding strength limits the application of these adhesives. Addition of bioactive ceramics indeed improved the adhesion strength on soft tissues from 8.4 ± 2.3 kPa to 18.1 ± 4.0 kPa, while on hard tissue the adhesion strengths were ~ 71.4 ± 28.2 kPa.70-75

Reproduced with permission from ref 73. Copyright 2014, John Wiley and Sons, Ltd.

15 Figure 1.5. N-ethyl-N-(3-dimethylaminopropyl) carbodiimide (EDC) and N- hydroxysuccinimide (NHS) activated crosslinking of gelatin and alginate based tissue adhesives (Gel – Gelatin, Al – Alginate).

Another example of NHS based adhesive was given by Taguchi and coworkers. They used N-hydroxysuccinimide derivative of citric acid to crosslink a collagen based matrix.

This adhesive demonstrated adhesion strength of 19.9 ± 1.9 kPa on porcine soft tissue. In the following studies collagen was replaced with human serum albumin (HSA) owing to the spherical structure and negative charges, which were assumed to promote bonding.

The maximum bonding strength obtained on collagen casings (model soft tissue substrate) was 760 g/cm2 (74.56 kPa); comparable to cyanoacrylate glue and ~ 9 times stronger than fibrin glue. The adhesive showed complete wound closure with gradual degradation and mild inflammatory response on mouse skin.

In a subsequent study, an adhesive based on HSA crosslinked with organic acid based crosslinker - disuccinimidyl tartarate (DST) reached maximum adhesion strength of

489.14 ± 93.06 kPa (on collagen casing) within 5 min of mixing however; it resulted in mild tissue inflammation. When HSA was replaced with cholesteryl functionalized gelatin (Figure 1.6), improved bonding and peeling strength were obtained due to improved tissue penetration by anchoring to the lipid bilayer of cell membranes. The hydrophobic interactions between cholesteryl groups in addition to covalent crosslinking by DST increased the cohesive strength of this adhesive. In the following study, alkaline treated gelatin was modified with hexanoyl, decanoyl and stearoyl chloride to introduce hydrophobic groups in the polymers. The bonding strength on porcine intestinal tissue 16 showed an improvement for shorter alkyl chain modification while the burst strength improved with the degree of functionalization of the hydrophobic moiety.76-83

Reprinted from Matsuda, M.; Ueno, M.; Endo, Y.; Inoue, M.; Sasaki, M.; Taguchi, T., Enhanced tissue penetration- induced high bonding strength of a novel tissue adhesive composed of cholesteryl group-modified gelatin and disuccinimidyl tartarate, Colloids Surf., B, 91, 48-56, Copyright 2012, with permission from Elsevier.

Figure 1.6. Bonding mechanism of 70% CholGltn/Gltn-DST composition on porcine arterial media. Improved tissue penetration and hydrophobic interactions between the cholesteryl groups in addition to covalent crosslinking by DST contribute to increased adhesion strength.

In another work, Nagatomi and coworkers developed acrylate and N- hydroxysuccinimide (NHS) bifunctional Tetronic hydrogels for soft tissue wound closure. The adhesive showed maximum adhesion strength of 74 kPa with low swelling but poor burst pressure compared to the native tissue.84 The crosslinker and reaction by- products (urea) used in these studies were cytotoxic and very likely to cause local tissue inflammation or tissue necrosis. To overcome the cytotoxicity concern of urea by- products, gelatin-polysaccharide based hydrogels by Schiff base formation were

17 subsequently developed by other groups. Schiff base reaction occurs under mild

conditions between an amine and an aldehyde group. An aldehyde functionalized moiety

reacts with the amine groups on the tissue surface to promote adhesion and

simultaneously crosslinks with another amine functionalized moiety to promote

crosslinking as well as cohesion. This chemistry eliminates the need to use crosslinking

agents which is beneficial in terms of biocompatibility of the reagents used.

Based on this concept, Mo and coworkers developed an adhesive hydrogel by

crosslinking aldehyde functionalized alginate with amine functionalized gelatin via Schiff

base reaction. The maximum adhesion strength on porcine skin was around 11.51 ± 1.3

kPa, slightly lower than the commercial fibrin glue (13.54 ± 3.07 kPa). The degradation

products and an increase in the aldehyde content of the hydrogel lowered the cell viability

as observed from the MTT assay.85 A dextran based hydrogel comprising of aminodextran and oxidized dextran (dextran aldehyde) was capable of sealing an incision in the swine uterine horn with maximum burst pressure of 64.6 ± 9.3 mmHg in addition to being non-cytotoxic, low swelling and degradable.86 Ikada and coworkers compared

the bonding and sealing ability of three different biodegradable adhesives comprising of:

modified gelatin+aldehyde dextran (gel-dext), modified gelatin+oxidized (aldehyde)

hydroxyethyl (gel-HES) and chitosan+modified dextran (chit-dext). Chit-dext gels

owing to high stiffness showed lower bonding strength (130 gf/cm2 (12.75 kPa)) and

sealing ability compared to gel-dext and gel-HES (bonding strengths: 210 gf/cm2 (20.60 kPa) and 227 gf/cm2 (22.27 kPa) respectively).87 An acidic of non-crosslinked collagen or gelatin modified by oxidative cleavage forms a bioresorbable tissue adhesive

18 when neutralized to a pH between 6-10. The collagen/gelatin reacts with the proteins in the tissue and undergoes rapid cross-linking when neutralized forming adhesive bonds comparable to fibrin glue.88 In another study, a combination of periodate oxidized dextran

(dextran dialdehyde, DDA) and chitosan hydrochloride (amine functionalized chitosan)

resulted in rapid in situ gelation forming a Schiff base product. Jayakrishnan and

coworkers, elaborately demonstrated the efficacy of the gel as a tissue adhesive, hemostat

and medium in addition to its cytocompatibility and degradability.89 Artzi

and coworkers studied changes in adhesion of dextran with polyamidoamine

on varying the tissue chemistry. However, the adhesion strength of this adhesive was low

compared to other dextran based adhesives.90 were also synthesized by

reaction between aldehyde dextran and ε-poly(L-lysine) (LYDEX). They were shown to have high bonding strength and degradability along with low cytotoxicity. Hyon and coworkers successfully applied these adhesives in lung surgeries, laparoscopic partial nephrectomy, ocular reconstruction, cartilage or bone regeneration, cardiovascular surgery and as a drug or gene carrier agent.91-103 High aldehyde concentrations are

detrimental to tissue and may result in tissue inflammation or tissue necrosis. It is

therefore necessary to monitor the aldehyde concentration in the adhesives to avoid

cytocompatibility issues.

The bonding strengths could also be improved by incorporation of a vinyl or

photocrosslinkable group in the adhesive structure. When irradiated with light these

photocrosslinkable groups undergo crosslinking, increasing the cohesive as well as the

adhesive strength of the network. Matsuda and coworkers developed a photocurable gel

19 using vinylated protein or polysaccharide or a combination of both, which adhered to tissue on photo irradiation. In vivo studies using styrene derivatized gelatin showed satisfactory wound closure on dog thoracic aorta and rat liver tissue on irradiation with visible light. The maximum adhesive strength in vitro was 157.1 ± 14.6 g/cm2 (15.4 ± 1.4 kPa) and the adhesive was also capable of localized drug delivery at the tumor site following surgery. The radicals were generated during the crosslinking reaction using a radical initiator (camphorquinone) which could possibly result in tissue damage or tissue necrosis.104, 105 Photopolymerization of phenolic derivatized gelatin (conversion of lysine residues into tyrosine) yielded a low swelling hydrogel by dityrosine crosslink formation.

The hydrogel demonstrated elasticity and withstood internal burst pressure upto 60 mmHg while the actual adhesion strength was not mentioned.106 A photocrosslinkable chitosan based tissue adhesive was developed by Ishihara and coworkers by incorporating photoactive azide groups. Chitosan was also modified using lactose moieties in order to impart water solubility. On irradiation with UV light the azide groups were converted to highly reactive nitrene groups, which in turn reacted with the amines in the tissue proteins or chitosan to form azo groups resulting in tissue bonding and crosslinking of the adhesive respectively. The maximum bonding strength on ham slices was 43 g/cm2 (4.2 kPa) and the maximum burst pressure upto 225 ± 25 mmHg was capable of sealing a pinhole in the thoracic aorta. The use of UV light for crosslinking and the presence of toxic functional groups (azide, nitrene and azo groups) are detrimental to the underlying tissue, which calls for extensive biocompatibility evaluation.107-110 A bone adhesive with sustained adhesion under wet conditions was

20 developed from a combination of photocurable poly(ethylene glycol) dimethacrylate

(PEGDMA) and isocyanate end capped hydrophilic statistical copolymer of

poly(ethylene oxide) and poly(propylene oxide) (NCO-sP(EO-stat-PO)). fillers like gypsum (CaSO4·2H2O), newberyite (MgHPO4·3H2O) and struvite

(MgNH4PH4·6H2O) were added to improve the consistency of the final adhesive for easy

handling, strong mechanical properties, improved adhesion and porosity to impart

degradability to the matrix in addition to cellular infiltration and angiogenesis. Addition

of NCO-sP(EO-stat-PO) did not have a positive effect on the 3 point bending strength of

the adhesives but preserved the shear adhesion strength in wet conditions.111

Photocurable adhesives occasionally require UV irradiation, which is harmful to the

neighboring healthy tissue. Also, radical initiators or reactive species generated during

irradiation could be detrimental to the underlying tissue resulting in tissue necrosis. In

certain cases, if the light does not penetrate the depth of the adhesive, only surface curing

will occur resulting in a stiff, cured top layer over a soft uncrosslinked adhesive layer

with poor adherence to the tissue surface.

Adhesives were also functionalized with phenolic and/or thiol groups to promote tissue

interaction as well as intermolecular crosslinking using oxidizing agents like periodate,

ferric ions or enzymatic oxidation. Hydrogels based on hydroxyphenyl propionic acid and

tyramine conjugated gelatin crosslinked within 30 s with adhesion strengths considerably

higher than the fibrin glue. These in situ enzymatic crosslinking hydrogels degraded

rapidly in vitro, which left insufficient time for tissue regeneration which could adversely affect wound healing (Figure 1.7).112

21

Reproduced from ref 112 with permission of The Royal Society of Chemistry. http://dx.doi.org/10.1039/c3tb00578j.

Figure 1.7. Synthesis of gelatin-hydroxyphenyl propionic acid (GH) and gelatin- hydroxyphenyl propionic acid - tyramine (GHT) conjugates. Enzymatic or peroxide crosslinking yield gels with 2-3 times higher adhesion strength compared to fibrin glue.

A protein based adhesive synthesized by crosslinking of caffeic acid functionalized gelatin with thiol functionalized keratin was attempted by Gnanamani and coworkers.

Periodate oxidation triggers intra- and inter-molecular cross-linking of the adhesive via

Michael addition, biaryl formation, disulfide bridging and imine formation (Figure 1.8).

The adhesive demonstrated maximum adhesion strength of 16.6 kPa in vitro and 1.67

MPa in vivo with gradual degradation, excellent biocompatibility and rapid wound healing.113

22

Reprinted from Thirupati Kumara Raja, S.; Thiruselvi, T.; Sailakshmi, G.; Ganesh, S.; Gnanamani, A. Rejoining of cut wounds by engineered gelatin-keratin glue. Biochim. Biophys. Acta, Gen. Subj., 1830, 4030-4039, Copyright 2013, with permission from Elsevier.

Figure 1.8. Mechanism of periodate based crosslinking between caffeic acid functionalized gelatin and thiol functionalized keratin.

Bao and coworkers synthesized an in situ forming polysaccharide/polypeptide hydrogel by simple Michael addition reaction between thiolated chitosan (CSS) and maleimide functionalized ε-polylysine (Figure 1.9). The adhesive undergoes fairly rapid crosslinking with the least gelation time of 15 s at physiological conditions with excellent cytocompatibility and hemostatic property. The maximum adhesion strength measured was 87.5 kPa on simulated living tissue.114

23

Reprinted (adapted) from Nie, W.; Yuan, X.; Zhao, J.; Zhou, Y.; Bao, H. Rapidly in-situ forming chitosan/ε-polylysine hydrogels for adhesive sealants and hemostatic materials. Carbohydr. Polym., 96, 342-348, Copyright 2013, with permission from Elsevier.

Figure 1.9. Thiolated chitosan (CSS) and maleimide functionalized ε-polylysine (EPLM) undergo rapid crosslinking by Michael addition reaction.

Webb and coworkers developed acrylate end functionalized poloxamine adhesives as a low swelling alternative to the commonly used PEG adhesives. These adhesives gelled by reverse thermal gelation of thermosensitive poloxamine copolymers and by Michael type addition with a thiol containing crosslinker. The adhesives showed stronger adhesion (25

24 kPa) compared to 4-arm PEG on rat skin, however, the cell adhesion on native hydrogels was poor in the absence of cell interaction ligands like fibronectin and RGD peptide.115

Hwang and coworkers synthesized a chitin nanofiber/gallic acid based crystalline tunicate mimetic tissue adhesive. This hydrogel showed strongest adhesion under wet conditions when crosslinked with periodate (~ 215 kPa). Ferric crosslinking (~ 98 kPa) demonstrated lower adhesion strengths than periodate but had an advantage of a possible self-healing mechanism (Figure 1.10).116

Reprinted from Oh, D. X.; Kim, S.; Lee, D.; Hwang, D. S. Tunicate mimetic nanofibrous hydrogel adhesive with improved wet adhesion. Acta Biomater., 20, 104-112, Copyright 2015, with permission from Elsevier.

Figure 1.10. Gallic acid conjugated chitin nanofiber precursors crosslink by two methods:

Periodate (NaIO4) – induced covalent hydrogel (color change from white to brown) and

25 FeCl3 – induced non-covalent, iron-complex hydrogel (color change from white to red

via purple).

Bianco-Peled and coworkers developed a biomimetic tissue adhesive inspired by wet

adhesion of brown alga Fucus serratus. The adhesive, composed of phloroglucinol,

alginate and calcium ions, was capable of adhering to a variety of substrates including

porcine tissue.117-119 Homo- and copolymers of polypeptides crosslinked with transglutaminase have also demonstrated adhesive characteristics.120 Lei and coworkers

studied the effect of topological structures, side group chemistry, temperature variation

viz. hydrophobicity or hydrophilicity and cure time of hyperbranched poly(amino acid)s.

The thermoresponsive poly(amino acid)s based on a combination of dopamine, arginine,

cysteine and lysine acrylamide (Figure 1.11) were crosslinked by catechol oxidation or

Michael addition reactions. The hyperbranched polymers had higher wet adhesion

strength compared to their linear counterparts and increasing the functionality on the

copolymer also increased its adhesion strength. The globular structure of hyperbranched

polymers limits chain entanglement which allows more functional groups to interact with

the surface. The elastic modulus and adhesion strength were higher at physiological

temperature than at room temperature. With wet adhesion strengths of 138 kPa,

degradability and non-cytotoxicity, this adhesive has potential to be used in biomedical

applications.121

26

Reproduced from ref 121 with permission of The Royal Society of Chemistry. http://dx.doi.org/10.1039/c5py01844g.

Figure 1.11. Thermoresponsive hyperbranched poly(amino acid)s functionalized with

DOPA, arginine, cysteine and/or lysine-acrylamide, demonstrate strong wet adhesion.

Other studies focused on isocyanate or urethane functionalized adhesives for rapid crosslinking. Yang and coworkers developed tissue adhesives based on urethane modified dextrans. This adhesive displayed adhesion strength of 2.99 ± 0.12 MPa which was much higher than commercial Tisseel glue (0.05 MPa). Urethane functionalized oxidized dextran when crosslinked with gelatin demonstrated an increase in the adhesion strength upto 4.16 ± 0.72 MPa. However, hydrogel swelling was an unavoidable shortcoming of this adhesive.122, 123 Matsuda and coworkers developed an isocyanate end

capped polyether copolymer of poly(ethylene glycol) and poly(propylene glycol) based

27 tissue adhesives with impressive in vivo adhesion. Fluorinated aliphatic isocyanate

demonstrated lower cytotoxicity compared to aromatic counterparts however, with slow

bioresorption.124 To build upon Matsuda’s work, Ikada and coworkers synthesized

isocyanate end functionalized polyester copolymers, which cured in the presence of

water. These adhesives demonstrated faster degradation compared to the polyether

copolymers but also resulted in acute tissue inflammation at the application site.125

Bochyńska and coworkers developed isocyanate end functionalized water curable block

copolymer adhesives from trimethylene carbonate for meniscal tissue repair. The

isocyanate groups react with amines in the tissue protein leading to adhesion on tissue.

PEG and trimethylolpropane ethoxylate were used as initiators to synthesize linear and 3

armed adhesives respectively (Figure 1.12). The branched adhesive demonstrated higher adhesion strength (0.68 MPa) compared to the linear counterpart (0.35 MPa) and commercial Dermabond (0.4 MPa). In the following work, hyperbranched polymers were

synthesized by a polycondensation reaction between citric acid and the linear copolymers

to increase the number of reactive groups. However, compared to the previous work, no

significant increase in the number of isocyanate groups was observed and the adhesion

strengths were much lower (20-80 kPa) in addition to longer curing times. Addition of an amine crosslinker (spermidine) or catalysts (2,2-dimorpholinodiethylether (DMDEE) and

1,4-diazabicyclo [2.2.2] octane (DABCO)) reduced the curing time to 2 h from about 8-

24 h. Usage of DABCO resulted in strongest adhesion (64.4 ± 14.3 kPa) in the shortest time (2 h). Presence of collagenase did not seem to improve the adhesion strengths. The

28 cell viabilities on these adhesives were rather low and in vivo cytocompatibility evaluation is necessary to deem these adhesives safe for internal use.126-129

Reproduced from ref 126 with permission from John Wiley and Sons. Copyright 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim, Germany.

Figure 1.12. Hexamethylene diisocyanate modified trimethylene carbonate (TMC) and trimethylolpropane ethoxylate (TMPE) hydrogels. a) PEG-(TMCm-HDI)2; b) TMPE-

(TMCm-HDI)3, (m = 1 or 2).

A biodegradable tissue adhesive based on isocyanate functionalized 1,2-ethylene glycol bis(dilactic acid) (ELA-NCO) and a consisting of free amine and/or hydroxyl groups as a chain elongation agent was developed by Schmitz and coworkers.

Incorporation of chitosan chloride as a chain extender significantly improved the adhesion strength compared to fibrin glue on bovine muscle tissue.130, 131 Ates and coworkers developed polyurethane based adhesives for application in soft tissue

29 adhesion. The authors have reported the effect of incorporation of chlorogenic acid and

xylose in these polyurethanes. The xylose incorporated polymers showed stronger

adhesion (94.0 ± 2.8 kPa) compared to the chlorogenic acid incorporated polymers (40.09

± 5.08kPa) on soft tissue.132, 133 A bone adhesive developed by reinforcing a polyurethane

matrix with nanocrystals of hydroxyapatite showed improved adhesive performance on

wet bone compared to commercial bone cement.4

Tissue adhesives employing different techniques like laser welding, layer-by-layer

assembly, temperature dependent hardening were also developed. Thoroughly dried,

deacetylated chitosan strips applied as tissue adhesives using laser welding demonstrated

adhesion strength of 14.7 ± 4.7 kPa on sheep intestinal tissue presumably by diffusion in

the tissue. Modifications like addition of genipin as a crosslinker, integration with small

intestine submucosa and addition of Rose Bengal dye were also attempted by the authors

but did not bring about an appreciable change in adhesion strength. In addition to sealing

intestinal pinhole defects, the adhesive was also suitable for sealing corneal defects and

nerve anastomoses.134-139 In another study layer-by-layer (LbL) assembly technique was used to fabricate flexible, mechanically robust free standing films by interdiffusion of poly(allylamine hydrochloride)-dextran and hyaluronic acid by Sun and coworkers. The

LbL assembled polyelectrolytes interact with silanol groups on the glass surface by hydrogen bonding and electrostatic interactions. Drug loaded films demonstrated a burst release with ~ 67% of the drug released within 1 h and strong adhesion strength of 2.69 ±

0.62 MPa on bovine periosteum. Adhesion studies on animal tissue are required for

further application of these adhesives.140

30 Biomaterials like lactide and caprolactone are especially attractive because of their

biocompatibility and degradable nature which makes them one of the best options for

adhesive building blocks. Cohn and Lando developed branched oligomers of lactide and

caprolactone extending from a core. The combination of lactide and caprolactone

imparted temperature dependent changes in the rheological properties such that the

polymer possessed low viscosity at the application temperature but hardened at

physiological temperature after application. The adhesive failure strengths of these block

copolymers were higher (~ 6 N/cm) than the commercial cyanoacrylate glues (~ 1 N/cm)

on polyamide substrate.141 Another polysaccharide based biocompatible adhesive

composed of chondroitin sulfate (CS) crosslinked with bone marrow (BM) showed

improved adhesion and biocompatibility compared to the PEG counterpart. The presence

of bone marrow, supplied growth factors at the wound site and aided in bone or meniscal

regrowth, however, it also increased the risk of virus transmission. A trade-off between

CS and BM concentration resulted in either improved adhesion or cell infiltration along

with tissue growth.142-144 A strong surgical sealant with low swelling, non-toxic

degradation products and inherent anti-microbial properties was developed from s- triazine based hyperbranched epoxy with poly(amido amine) hardener (Figure 1.13). This sealant demonstrated strong mechanical properties, slow degradation and minimal/no irritation to the rat tissue but longer curing times could limit its application for minor surgeries involving anesthesia.145

31

Reproduced from ref 145 with permission of The Royal Society of Chemistry. http://dx.doi.org/10.1039/c5tb00753d.

Figure 1.13. Synthetic protocol of s-triazine based hyperbranched epoxy resin. L – linear unit, D – dendritic unit, T – terminal unit.

1.5. POLY(ETHYLENE GLYCOL) (PEG) BASED HYDROGEL ADHESIVES

PEG is a hydrophilic, biocompatible polymer widely used as a biomaterial in tissue

engineering. However, PEG lacks biodegradability and therefore is often modified with

degradable functionalities or copolymerized with degradable polymers. PEG based tissue

adhesives have gained popularity because of ease of modification/functionalization,

bioconjugation, drug delivery, non-immunogenity146 and water solubility. In previous studies, modified PEG was combined with or protein based adhesives described in the previous section. For example, Matsuda and coworkers developed a UV and visible light curable, elastomeric and degradable adhesive gel derived from photoreactive gelatin (functionalized with benzophenone or a xanthene dye) and poly(ethylene glycol) diacrylate (PEGDA). These gels demonstrated adhesion strength of

12 kPa and a burst pressure of 150.7 ± 34.4 mmHg with satisfactory hemostatic and

32 anastomotic performance.147, 148 However, UV-light is detrimental to tissue and cannot penetrate the thick adhesive layer resulting in only surface crosslinking. To overcome these issues the authors further modified this gel structure to develop a visible light curable adhesive composed of styrenated gelatin, PEGDA and carboxylated camphorquinone. A formulation of 35wt% styrenated gelatin, 5wt% PEGDA and

0.05wt% carboxylated camphorquinone demonstrated an adhesive strength of ~ 140 g/cm2 (13.73 kPa) and complete degradation in 4 weeks without any local

inflammation.149 Wang and coworkers developed a one pot synthesis for controlled

homopolymerization of hyperbranched PEG-diacrylate (HPEGDA) with low swelling

and slower degradation time. On UV curing the hydrogel showed good adhesion on

various biological tissues with adhesion strengths ranging between ~ 40 kPa on porcine

skin to ~ 60 kPa on bovine pericardium.150 Alsberg and coworkers synthesized a dual

crosslinking hydrogel based on oxidized methacrylated alginate and 8-arm PEG amine.

The crosslinking mechanisms include imine formation and photocrosslinking of

methacrylate groups. The mechanical properties, swelling, degradation and cytotoxicity

were strongly dependent on the degree of oxidation of the alginate. However, the

adhesion strengths for single and dual crosslinked hydrogels did not show an appreciable

difference (13-15 kPa).151 Zilinski and Kao developed an interpenetrating network (IPN)

of UV curable PEG-diacrylate and gelatin loaded with anti-inflammatory drug. Although,

the bonding strengths were very weak the drug aided in lowering the tissue inflammatory

response at the site of application.152 Yang and coworkers modified a previously

developed urethane methacrylated dextran adhesive by photocrosslinking with three arm

33 PEG-DOPA. These hydrogels demonstrated adhesion strength of 4.0 ± 0.6 MPa and an impressive burst pressure of 620 mmHg but suffer from the drawback of excess swelling.153 Hubbell and coworkers synthesized in situ polymerizing hydrogels based on

DL-lactic acid and glycolic acid polymerized in the presence of PEG and end capped with photopolymerizable acrylate groups. The uncrosslinked polymer adhered well to tissue surface as a consequence of entanglement with the tissue proteins. However, following gelation the polymer demonstrated anti-adhesive properties. Despite these adhesive properties, follow up studies must be done to examine the possibility of necrosis and inflammation on the tissue as a consequence of long wavelength UV radiation.154-156

One strategy to avoid tissue exposure to UV irradiation was incorporation of functional moieties for Schiff base or Michael addition reactions. Kurosawa and coworkers synthesized a crosslinkable polymeric micelle adhesive via Schiff base formation. An aldehyde terminated PEG-PLA block copolymer underwent gelation in the presence of amine terminated polymer or amines present on the tissue surface (Figure 1.14). The adhesive strength achieved by Schiff base formation and mechanical bonding with uneven tissue surface was comparable to fibrin glue.157

34

Reproduced from ref 157 with permission from John Wiley and Sons. Copyright 2006 Wiley Periodicals, Inc.

Figure 1.14. Aldehyde terminated PEG-PLA block copolymeric micelles crosslink with

polyamine and adheres to tissue surface by Schiff base reaction.

Edelman and coworkers synthesized an adhesive based on aldehyde functionalized dextran and amine functionalized 8-arm PEG with tissue specific adhesion properties

(Figure 1.15). The resulting hydrogel was biocompatible, non-cytotoxic, degradable and demonstrated adhesive strength comparable to cyanoacrylate glue. This adhesive was also capable of sealing a corneal incision without fluid leakage.158-163

35

Reproduced from ref 158 with permission from John Wiley and Sons. Copyright 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim, Germany.

Figure 1.15. Crosslinking reaction between aldehyde functionalized dextran and amine

functionalized PEG by Schiff base formation to form an adhesive hydrogel.

Elisseeff and coworkers developed a chondroitin-sulfate succinimidyl succinate (CS-

NHS) crosslinked with PEG-(NH2)6 to form an adhesive hydrogel. The hydrogel adheres

covalently to tissues by reaction with amine groups and crosslinks via reaction with PEG-

(NH2)6. The gel properties could be altered by varying gelation conditions like humidity,

pH and stoichiometry. In addition to maintaining biological activity, the NHS modified

CS-PEG also demonstrated degradability and adhesion strength almost 10 times more

than that of fibrin glue.142 Kaplan and coworkers combined silk from Bombyx mori

(silkworm) with thiol and maleimide functionalized 4-arm PEG to integrate mechanical

strength and biocompatibility respectively. The adhesive system showed an increase in

adhesion strength (~ 50% increase) compared with commercial sealant CoSeal on increasing silk content.164 In another study the silk fibroin from Bombyx mori was

conjugated with PEG for water solubility and dopamine to develop water soluble

36 adhesives. The adhesion strengths of these conjugates were not only enhanced by crosslinking via oxidation but also by silk β-sheet structure which reinforced the material strength.165 Chenault developed a degradable adhesive hydrogel based on polyglycerol aldehyde and water dispersible, multi-arm amine with burst pressures ranging between

5.5 to 27.6 kPa.166 A two component biocompatible and degradable hydrogel made from

PEG-dimethacrylate and thiolated chitosan by Michael reaction was developed by El-

Newehy and coworkers. The lap shear strength on rat skin varied between 16.32 ± 0.76 kPa to 48.73 ± 0.56 kPa.167 A degradable, biocompatible, in situ gelling adhesive was designed by Yuan and coworkers especially for nerve anastomosis. Thiolated chitosan

(CSS) crosslinks with dual functionalized (catechol and maleimide) polylysine via

Michael addition reaction (Figure 1.16) resulting in adhesion strength capable of tolerating 0.185 N load and effective nerve regeneration. In a preceding study when only maleimide was used as the crosslinker the maximum adhesion strength obtained was 148 kPa within a gelling time of 5 s. Catechol incorporation did not have a significant influence on gelling time while it did contribute towards slight improvement in adhesion.

These gels undergo considerable swelling which could potentially impact the wound healing process.168, 169

37

Reproduced with permission from Zhou, Y.; Zhao, J.; Sun, X.; Li, S.; Hou, X.; Yuan, X.; Yuan, X. Rapid gelling chitosan/polylysine hydrogel with enhanced bulk cohesive and interfacial adhesive force: Mimicking features of epineurial matrix for peripheral nerve anastomosis. Biomacromolecules, 2016, 17, 622-630. Copyright 2016 American Chemical society.

Figure 1.16. Thiolated chitosan (CSS) and catechol-maleimide functionalized polylysine crosslink within 10 s via Michael addition. Interactions like hydrogen bonding, electrostatic interaction, π-π interaction and π-cation interaction contribute to the strong bulk cohesive force.

Adhesive nanocomposites have an advantage of improved mechanical properties. To this end Lee and coworkers developed an injectable nanocomposite adhesive from dopamine modified 4-arm PEG and Laponite (nanosilicate). Laponite incorporation not only improved the mechanical properties, curing rate and adhesion strength (7.9 ± 1.8 kPa) of the composite but also demonstrated increased cell viability with minimal tissue inflammation (Figure 1.17).170 38

Reproduced with permission from ref 170. Copyright 2014 American Chemical Society.

Figure 1.17. Laponite incorporated, dopamine modified, 4-arm PEG showed improved

adhesion and cohesion via: (A) polymerization and di-DOPA crosslink formation by

periodate oxidation, (B) interfacial crosslinking with amine groups on the tissue surface,

(C) reversible bonding with Laponite.

In another study a degradable, hemostatic tissue adhesive called TAPE (tannic acid and

poly(ethylene glycol)) was synthesized using tannic acid and PEG which showed

impressive adhesion under wet conditions. The adhesive is formed by simply mixing the

two components at acidic pH which makes it viable for large scale production. The adhesion strength of TAPE was dependent on the number of PEG arms and the end functional group (PEG-NH2 > PEG-OH > PEG-SH). The strong adhesion was attributed to hydrogen bonding present throughout the network with highest adhesion strength of

171 0.18 MPa when PEG-NH2 was used.

39 1.6. BIOMIMETIC TISSUE ADHESIVES

1.6.1. MUSSEL INSPIRED ADHESIVES

The adhesive property of mussels specifically Mytillus edulis has been studied for

decades and is particularly of interest because of its ability to attach to virtually any

surface via byssal threads secreted from the foot.172-174 This adhesive is reversible and

capable of withstanding strong water currents as well as fluctuations in temperature and

salinity. The mussel usually consists of four main components: acid

mucopolysaccharides acting as a primer, adhesive proteins consisting mainly of

polyphenolic proteins rich in 3,4-Dihydroxyphenylalanine (L-DOPA) and lysine, fibrous proteins which acts as an attachment thread between mussel and the substrate and finally, polyphenoloxidase to promote intermolecular crosslinking.175, 176

Preliminary immunological studies showed that the mussel adhesive proteins are poor

antigens and have great potential to be used for biomedical purposes specifically for

biological tissue adhesives.177 Early efforts of mimicking mussel adhesives involved

developing a protein as a non-specific adhesive for cell attachment and growth,178

followed by studies confirming the suitability of mussel adhesive proteins for cell viability, attachment, growth and proliferation.179 The adhesive properties of mussel

adhesive protein (MAP) extracted from the Mytillus edulis showed

satisfactory bonding on various substrates like stainless steel, pig duodenal mucosa,

porcine small intestinal submucosa and porcine skin.180-183 Other efforts involved use of recombinant DNA technology, peptide synthesis, fragment condensation technique and

40 gene cloning to develop synthetic mimics of mussel adhesive.6, 176, 177 Yamamoto and

coworkers synthesized a series of homopolymers and copolymers based on L-DOPA and

L-glutamic acid184,185; followed by synthesis of a series of poly(amino acid)s and

polypeptides to study their bonding strength on metals in aqueous and organic solvent

systems. Poly(Lys).HBr demonstrated the highest tensile strength of 123 kg/cm2 (12.06

MPa) on iron while gelatin demonstrated the highest compressive shear strength of 21 kg/cm2 (2.06 kPa) on aluminum oxide substrates in water. For the organic solvent system

poly(DL-methionine) showed the highest tensile strength of 49 kg/cm2 (4.8 kPa) on iron

and a compressive strength of 22 kg/cm2 (2.16 kPa) in dichloromethane.186 Inspired from

the mussel adhesives the group also synthesized a series of mimics of sequential187-189 or

random copolymers190,191 and polypeptides: polydipeptide, polytripeptide192,

polyoctapeptide193 and polydecapeptide194 and studied their bonding strengths. An

important conclusion from these studies was that the mussel adhesive mimics were

capable of bonding various materials and the bonding strength increased on addition of a

crosslinker like , hydrogen peroxide or basic aqueous solution. The bonding

strengths were also found to increase with increasing DOPA content, copolymer solution

concentration, copolymer molecular weight and curing temperatures.195 Out of all the amino acids present in the mussel adhesive proteins lysine and tyrosine were identified to be of utmost importance for bonding and crosslinking characteristics.192,195,196 The

authors also studied the in vivo adhesive properties of polytripeptide – poly(Gly-Tyr-Lys)

(exhibited highest adhesion strength on pigskin in vitro) on white pigs and compared the

results with a commercial Tisseel® glue. No acute inflammation or tissue reactivity was

41 observed in the area near the incision site. Histology results suggested that the poly(Gly-

Tyr-Lys)-tyrosinase system had less immune reactivity (58 ± 9%) than Tisseel®.196

These studies on the adhesion properties of the mussel adhesive protein (MAP) and their synthetic polypeptide mimics led to a plethora of studies on synthetic polymeric adhesive mimics which will be discussed in detail in the following section.

Yamada and coworkers evaluated the efficacy of a deacetylated chitosan, dopamine and tyrosinase solution as a water resistant adhesive. Tyrosinase oxidizes dopamine into o-quinone which in turn triggered crosslinking within chitosan as well as promoted adhesion on the glass surface. The maximum bonding strength under dry conditions was

400 kPa and under water was ~ 450 kPa. Their studies also concluded that the water resistant adhesion was not specific to the enzyme catalyzed oxidation since glutaraldehyde crosslinking provided similar results with slightly lower bonding strengths. Addition of PEG prolonged the enzyme activity which decreased the crosslinking time significantly in order to achieve similar bonding strengths as in the previous studies.197-199 Ziegler and coworkers synthesized a dual component bone adhesive based on deacetylated chitosan and oxidized starch/dextran which crosslinked via Schiff base formation. Collagen, a primary component of bone contains amine groups which interact with aldehyde groups in the adhesive to promote bonding. Inspired by mussel adhesion the starch/dextran was also functionalized with L-DOPA to incorporate catechol groups to further strengthen crosslinking as well as adhesion. The maximum bonding strength of this glue on bovine cortical bone was ~ 0.41 MPa without L-DOPA and incorporation of L-DOPA groups did not contribute to an appreciable change in

42 adhesion strength.200 An injectable, thermo-sensitive tissue adhesive made from catechol functionalized chitosan and thiol functionalized pluronic F127 demonstrated an adhesion strength of 14.98 ± 3.53 kPa on mouse skin. The degradable adhesive possessed good mechanical integrity and sealing properties while the biocompatibility was not evaluated.201, 202 A dopamine functionalized gelatin macromer, dual crosslinked with Fe3+

(rapid crosslinker) and genipin (long term crosslinker) (Figure 1.18) showed strong

adhesion on porcine skin and cartilage along with degradability and tissue compatibility

in vivo and in vitro.203

Reprinted from Fan, C.; Fu, J.; Zhu, W.; Wang, D.-A. A mussel inspired double crosslinked tissue adhesive intended for internal medical use. Acta Biomater., 33, 51-63. Copyright 2016, with permission from Elsevier.

Figure 1.18. Double crosslinked tissue adhesive derived from dopamine functionalized

gelatin macromer crosslinked by Fe3+ (rapid crosslinker) and genipin (GP) (long term crosslinker).

A dopamine grafted, polysaccharide (alginate and hyaluronic acid) based degradable membrane was developed as a tissue adhesive by Paoletti and coworkers. This membrane

showed enhanced adhesion in vitro and in vivo on pig intestine tissue and was expected to

demonstrate improved fibroblast activity.204

43 Messersmith and coworkers synthesized a range of PEG based DOPA functionalized

copolymers for tissue adhesive applications while eliminating the use of strong oxidizing

agents. Huang et al. synthesized DOPA and DOPA methyl ester (DME) functionalized

PEO-PPO-PEO block copolymers for biomedical applications like tissue adhesives and

drug delivery. These polymers were capable of forming hydrogels with thermally

triggered self-assembly which eliminates the need for strong oxidizing agents. The

gelling temperature was found to be dependent on the copolymer concentration and

molecular weight. Viscometry measurements indicated that DOPA modified copolymers

were significantly more mucoadhesive than the unmodified copolymers.205 In another study, Lee et al. demonstrated the synthesis of N-methacrylated DOPA monomer and its copolymerization with PEG-diacrylate to form a hydrogel. The photopolymerization technique again eliminated the use of strong oxidizing agents for gelation and the moduli of the gels were sufficient for biomedical application. However, the phenolic nature of

DOPA served as a radical scavenger which hindered the photocrosslinking and prolonged the gelation time resulting in lower gel moduli. Furthermore, UV crosslinked gels outperformed the visible light crosslinked gels, in which case the effect of UV irradiation on tissue is concerning.206 Another study in this group involved a thermo-sensitive,

injectable DOPA modified hyaluronic acid (HA-D)/thiol end capped pluronic F127 (Plu-

SH) lightly crosslinked composite gel with Michael type catechol-thiol addition reaction.

These gels show rapid, reversible sol-gel transition at physiological temperature and enhanced adhesion to tissue. In vivo and in vitro studies showed that the gels form robust

44 structure and strong adhesion to the neighboring tissue owing to the unreacted and

oxidized catechol groups.207

Lee et al. synthesized a series of DOPA modified poly(ethylene glycol)s (PEG-DOPA)

(Figure 1.19) capable of rapid gelation in the presence of crosslinking agents and

208 optimized conditions. Burke et al. used these PEG-DOPA4 polymers for tissue

adhesive applications. They utilized temperature sensitive to sequester DOPA

oxidizing agents like periodate in the PEG-DOPA4 solution. These liposomes were

triggered at physiological temperature to release the oxidizing agents resulting in gelation

of the polymer solution. The crosslinked polymer demonstrated adhesion strength of ~ 35

± 12.5 kPa on porcine skin.209

Reproduced with permission from Lee, B. P.; Dalsin, J. L.; Messersmith, P. B. Synthesis and gelation of dopa-modified poly(ethylene glycol) hydrogels. Biomacromolecules, 2002, 3, 1038-1047. Copyright 2002 American Chemical Society.

Figure 1.19. Chemical structure of a series of PEG-DOPA adhesives synthesized by Lee et al.

45 Monomethoxy terminated PEG (mPEG) end-functionalized with 1-3 DOPA amino

acids (mPEG-DOPA) showed rapid, irreversible adsorption on TiO2 surfaces. The strong

adhesion of mPEG-DOPA on TiO2 occured by displacement of the TiO2 surface hydroxyl

groups via formation of charge transfer complexes and provided excellent resistance to

non-specific protein adsorption. This is especially important for bone surgeries involving

titanium implants, stents, catheters and intraocular lenses.210 Lee et al. also synthesized a

photopolymerizable triblock copolymer functionalized with DOPA. In this work a

glycine functionalized methacrylated PEG-PLA monomer (G-PPM) was copolymerized

with DOPA functionalized NCA monomer by ring opening polymerization or with N-

Boc-DOPA by simple carbodiimide coupling to obtain ~ 83% DOPA functionalized

triblock copolymer (Figure 1.20). These water soluble copolymers were photocured into

hydrogels using di-tert-butyl dicarbonate (Boc2O), 2,2′-dimethoxy-2-

phenylacetonephenone (DMPA) as the photoinitiator and 1-vinyl-2-pyrrolidone as the solvent for hydrophobic DMPA. The hydrophilic nature of the polymer ensured that the

DOPA groups remained in the aqueous environment without interfering with the crosslinking process which occured in the hydrophobic environment. The adhesion studies proved that DOPA modified gels showed higher Wadh (work of adhesion)

compared to their unmodified forms on Ti surfaces. The Wadh was found to decrease on oxidizing the gels with NaIO4, thus, proving their hypothesis that catecholic form of

DOPA is essential for adhesion to metallic surfaces.211 Similarly, DOPA functionalized diblock (PS-PEO) and triblock (PMMA-PMAA-PMMA) copolymers showed strong underwater adhesion on TiO2 surfaces and porcine skin by membrane inflation method. It

46 is however, important to determine the biocompatibility and degradability of these adhesives before considering their medical application.212

Reprinted with permission from Lee, B. P.; Chao, C.-Y.; Nunalee, F. N.; Motan, E.; Shull, K. R.; Messersmith, P. B. Rapid gel formation and adhesion in photocurable and biodegradable block copolymers with high dopa content. Macromolecules, 2006, 39, 1740-1748. Copyright 2006 American Chemical Society.

Figure 1.20. Illustration of the photopolymerization of Y-PPM into an adhesive hydrogel.

Y-PPM is the DOPA functionalized copolymer of methacrylated PEG-PLA block copolymers.

The in vivo adhesive performance of catechol-derivatized PEG (cPEG) was studied in a murine model of extrahepatic islet transplantation by Brubaker et al. with promising results. The adhesive invoked minimal inflammatory response and maintained an intact interface with the supporting tissue for upto one year. The application of cPEG as sealants for fetal membrane repair was also explored in vitro, ex vivo and in vivo in a rabbit model. The mussel mimetic cPEG was non-cytotoxic, accomplished leak proof closure, conformed to the shape of the underlying membrane and the bonding strengths were comparable to commercial fibrin glue. Although the degradation properties of this adhesive were not studied, it was an inception for medical application of this biomimetic tissue adhesive.213-217 An enzymatically degradable, mussel inspired tissue adhesive

47 hydrogel with an enzyme cleavable site was also developed by Brubaker et al. A DOPA

end-functionalized 4-arm PEG incorporated with an enzyme elastase peptide substrate

(Ala-Ala dipeptide) in the backbone of the polymer was designed as a macromonomer

(cAAPEG) (Figure 1.21). The catechol groups undergo intermolecular crosslinking

leading to rapid gelation (20-30 s) and adhesion to tissue surfaces under oxidizing

conditions. The adhesion strength of this adhesive was 30.4 ± 3.39 kPa, considerably

higher than commercial fibrin glue. Although, the adhesive demonstrated slow

degradation in vivo with minimal inflammatory response, there is opportunity to improve

the degradation rate by incorporating a longer elastase substrate peptide sequence.218

Reproduced with permission from ref 218. Copyright 2011 American Chemical Society. (http://pubs.acs.org/doi/full/10.1021/bm201261d)

Figure 1.21. Periodate mediated oxidative cross-linking yields rapid gelation and tissue adhesion of enzymatically degradable cAAPEG macromonomer. In the presence of enzyme neutrophil elastase, the Ala-Ala dipeptide linker (blue) is cleaved to provide

degradation sites. Black arrowheads indicate continuation of the cross-linked hydrogel matrix.

48 Following the work of Yu and Deming195, Yin and coworkers synthesized a degradable copolypeptide of DOPA and L-lysine through ring opening polymerization of NCA monomers. Adhesion strengths were evaluated on steel, aluminum, PS and glass. The copolypeptide (DOPA/lysine - 1:4) formed the strongest bond on steel which varied with the type of crosslinking agent and increased with increase in copolypeptide molecular weight and the DOPA content. The average strengths on porcine bone and porcine skin under dry conditions were 0.295 MPa and 0.208 MPa respectively for a cure time of 12 h.

Under wet conditions the average bonding strength on porcine bone decreased to 0.155

MPa while the adhesion on porcine skin failed.219

Attempts were also made to develop recombinant mussel proteins for application as bioadhesives. Lim and coworkers developed MAP based encapsulated coacervates as smart tissue adhesives with drug carrier ability. In this study an adhesive was formed by complex coacervation between cationic recombinant hybrid MAPs (fp-131 or fp-151) and

the anionic hyaluronic acid (HA). The bulk adhesive strengths of coacervates were twice

as strong compared to the protein itself on aluminum substrates.220 Cha and coworkers used the recombinant expression method to obtain rfp-1 MAP (AKPSYPPTYK) for

3+ hydrogel formation by coordination (Fe ) or covalent crosslinking (NaIO4). The

hydrogel system showed maximum adhesion strengths of ~ 130 kPa and ~ 200 kPa when

3+ crosslinked with Fe and NaIO4 respectively. The difficulty in synthesizing the rfps in

bulk limit the clinical application of these adhesive hydrogels.221 In another study they

engineered a residue specific DOPA incorporated recombinant mussel adhesive protein

(dfp-3 and dfp-5) with DOPA content upto 23 mol%. The recombinant protein showed

49 strong dry and underwater adhesion along with significant water resistance.222 They also

developed a light-activated, mussel protein based (LAMBA) hydrogel using

2+ a photo-oxidative reaction in the presence of blue light involving Ru(II)bpy3 as the

activator and sodium persulfate (SPS) as the oxidizing agent with recombinant MAP.

LAMBA demonstrated strong adhesion to wet porcine skin and also promoted wound

healing in addition to wound closure in a rat model.223 Lu and coworkers developed a

hybrid molecular adhesive by fusing Mfps found in DOPA from mussel adhesive with

the CsgA proteins found in the amyloid based adhesives in E. coli (monomeric - CsgA-

Mfp3, Mfp5-CsgA and copolymer constructs - (CsgA-Mfp3)-co-(Mfp5-CsgA)) (Figure

1.22). The molecular hybrid self-assembled in which the β-sheet amyloid protein formed the core while the disordered Mfps were exposed on the exterior. The (CsgA-Mfp3)-co-

(Mfp5-CsgA) copolymer demonstrated impressive adhesion energy of 20.9 mJ/m2, which

made it a strong competitor for application in medical adhesives.224

50

Reprinted with permission from Macmillan Publishers Ltd.: Nat Nano (ref 224), Copyright 2014. http://www.nature.com/nnano/index.html

Figure 1.22. (a-c) – Genetically engineered molecular hybrids of mussel adhesive

proteins (Mfps) and amyloid based adhesive proteins in E. coli (CsgA), (d) – Adhesive

molecular hybrids self-assemble with β-sheet amyloid protein forming the core and the

Mfps flanking the exterior.

In a comparative study of bulk adhesion, the recombinant fp-3 protein demonstrated

stronger adhesion compared to the recombinant fp-5 protein without an oxidant.

However, the recombinant technique suffers from low modification yield when tyrosine

is converted to DOPA using tyrosinase.225 Recently, Waite and coworkers extracted the mussel foot protein, Mfp-3S which is capable of -liquid coacervation variable with

51 buffer pH, ionic strength and temperature. This protein showed strong adsorption on

hydroxyapatite surfaces which can be further explored for dental or orthopaedic adhesive

application.226 Liu and coworkers reported synthesis of an elastin like polypeptide (Figure

1.23) with tunable phase transition in which the tyrosine residues were modified into

DOPA for underwater adhesion applications. Importantly this study showed that DOPA

residues did not affect the adhesion in dry conditions but contributed significantly

towards underwater adhesion when compared to the unmodified protein. Adhesion

strength of DOPA modified protein was >2 MPa in dry conditions and ~ 0.24 MPa in

humid conditions. The recombinant protein also demonstrated impressive adhesion strength of ~ 3 kPa when tested underwater. The final protein yield (before tyrosinase modification) was around 220 mg/L of culture which potentially could limit large scale synthesis of this adhesive.7

Reprinted from Brennan, M. J.; Kilbride, B. F.; Wilker, J. J.; Liu, J. C. Biomaterials, 124, 116-125. Copyright 2017, with permission from Elsevier.

Figure 1.23. Design strategy of DOPA modified recombinant elastin-like polypeptide

(ELP) synthesized by Liu and coworkers.

52 A biodegradable mussel inspired adhesive polymer composed of DOPA,

polycaprolactone (PCL) and PEG was synthesized by Lee and coworkers. The copolymer

demonstrated adhesive strengths 10 times that of commercial fibrin glue on commercial

hernia repair biologic meshes. The lap shear and burst pressure performance of the

adhesive polymer on bovine pericardium also outperformed the commercial fibrin glue.

The lap shear adhesive strength of the adhesive coated biologic scaffolds were 106 ± 22.9

kPa on bovine pericardium and 73.4 ± 24.4 kPa on porcine dermal tissue with possible application for Achilles tendon repair.227, 228 Chung and Grubbs synthesized DOPA

functionalized terpolymer adhesives. A terpolymer of acrylic acid (AA), acrylic acid N-

hydroxysuccinimide ester (AANHS) and N-methacryloyl-3,4-dihydroxy-L-phenylalanine

(MDOPA) (Poly(AA0.7-co-AANHS0.15-co-MDOPA0.15)) underwent rapid covalent crosslinking (~ 30 s) with a thiol terminated 3-armed PEG (PEG-SH) via NHS-thiol

condensation (Figure 1.24). Adhesion strength of Poly(AA0.7-co-AANHS0.15-co-

MDOPA0.15) crosslinked with PEG-SH demonstrated 450% stronger adhesion compared

to uncrosslinked Poly(AA-co-AANHS); comparable to the commercial cyanoacrylate

(Super Glue) on wet porcine skin.229 A photocrosslinkable bioadhesive was developed from photocurable monomer (ethylene glycol acrylate methacrylate dopamine (EGAMA-

DOPA)) and UV crosslinking agent (poly(vinyl alcohol) (UV-PVA)). The adhesion

strength of EGAMA-DOPA increased significantly on addition of UV-PVA, however,

the possibility of tissue damage on UV irradiation still persists.230

53

Reprinted with permission from Chung, H.; Grubbs, R. H. Rapidly cross-linkable dopa containing terpolymer adhesives and peg-based cross-linkers for biomedical applications. Macromolecules, 2012, 45, 9666-9673. Copyright 2012 American Chemical Society.

Figure 1.24. NHS-thiol condensation based cross-linking of poly(AA-co-AANHS-co-

MDOPA) and thiol terminated 3-armed poly(ethylene glycol); AA – acrylic acid,

AANHS – acrylic acid N-hydroxysuccinimide ester and MDOPA - N-methacryloyl-3,4-

dihydroxy-L-phenylalanine.

Yang and coworkers developed a one-step synthesis of injectable citrate based mussel

inspired bioadhesives (iCMBAs) composed of citric acid, PEG and dopamine building

blocks for wet tissue adhesion. The adhesive is synthesized by a simple one step poly-

condensation reaction between the building blocks (Figure 1.25) and crosslinked using

sodium periodate. Citric acid contributed to degradability while enhancing

hemocompatibility and hydrophilicity. The lap shear adhesion strengths varied between

33.4 ± 8.9 kPa and 123.2 ± 13.2 kPa corresponding to variation in composition and were

at least 2 fold stronger than commercial fibrin glue (15.4 ± 2.8 kPa). The iCMBAs were

non-cytotoxic in vitro and were successfully applied for wound closure and instant

54 bleeding control with tissue regeneration ability in vivo. Guo et al. further modified these injectable citrate-based mussel-inspired bioadhesives (iCMBAs) into anti-bacterial and anti-fungal iCMBAs (AbAf iCs). Incorporation of 10-undecylenic acid contributed to anti-fungal properties while silver nitrate (SN) or sodium (meta) periodate (PI) used as crosslinkers contributed to the anti-bacterial properties of the adhesives. On porcine small intestine submucosa PI crosslinked adhesives showed higher adhesion strengths than SN crosslinked adhesives and the adhesion strength in both cases was considerably higher than commercial fibrin glue. The AbAf iCs demonstrated fast degradation, excellent bacterial and fungal inhibition performance and good cytocompatibility in vitro. In the following work two different prepolymers functionalized with azide and alkyne functionalities were synthetized (Figure 1.25). This modification facilitated dual crosslinking by periodate induced oxidation of DOPA and Cu catalyzed azide-alkyne cycloaddition. The dual crosslinking strategy in addition to alkyne functionalized gelatin showed marked improvement in cohesion and in turn on adhesive strength (223.11 ±

15.94 kPa) on wet tissue. The adhesive showed impressive anti-bacterial and anti-fungal properties along with degradability and cytocompatibility; but the application of Cu catalyst directly on tissue is concerning and further studies are necessary to evaluate the long term effect.231-233 Zhu and coworkers modified a previously synthesized citrate based adhesive232 by incorporating PEO blocks to impart hydrophilicity. The copolymer of citric acid, PEO, 1,8-octanediol and dopamine showed water solubility for octanediol:PEO = 0.3:0.7 (molar ratio). This copolymer was interesting from a tissue

55 adhesive point of view because of the strong adhesion, degradability and modulus

comparable to soft tissue.234

a) Reprinted from Mehdizadeh, M.; Weng, H.; Gyawali, D.; Tang, L.; Yang, J. Injectable citrate-based mussel-inspired tissue bioadhesives with high wet strengths for sutureless wound closure. Biomaterials, 33, 7972-7983, Copyright 2012, with permission from Elsevier. b) Reprinted from Guo, J.; Kim, G. B.; Shan, D.; Kim, J. P.; Hu, J.; Wang, W.; Hamad, F. G.; Qian, G.; Rizk, E. B.; Yang, J. Click chemistry improved wet adhesion strength of mussel-inspired citrate-based antimicrobial bioadhesives. Biomaterials, 112, 275-286, Copyright 2017, with permission from Elsevier.

Figure 1.25. a) iCMBA pre-polymer synthesis by polycondensation reaction between

citric acid, PEG and dopamine. b) iCMBAs further modified with azide and alkyne

functionalities to facilitate dual crosslinking by catechol oxidation and click reaction.

del Campo and coworkers synthesized a biocompatible underwater self-curing, self-

healing, surface reactive and photodegradable adhesive material based on nitrodopamine

(PEG-ND4) (Figure 1.26). The hydrogel could be crosslinked by both periodate

(covalent) and Fe3+ (metal co-ordination). The photodegradability of the hydrogel which

was dependant on the light exposure and cross-linker concentration was confirmed by quartz crystal microbalance (QCM-D). The hydrogel demonstrated strong bonding and

56 light activated debonding underwater with non-cytotoxic property. The Fe3+ crosslinked

PEG-ND4 hydrogels showed complete self-healing behavior. Although the tissue adhesion characteristics of these gels were not studied, it is well suited for wound closure application.235

Reproduced from ref 235 with permission from John Wiley and sons. Copyright 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim, Germany.

Figure 1.26. Nitrodopamine modified, catechol functionalized 4-arm PEG adheres by

oxidative and metal coordination crosslinking. Nitrodopamine groups are responsible for

debonding on photoirradiation.

Wang and coworkers synthesized a hyperbranched poly(dopamine-co-acrylate) copolymer from dopamine and a triacrylate monomer. In addition to strong wet adhesion

(stronger than commercial fibrin glue) on porcine skin this hyperbranched copolymer also demonstrated degradation and non-cytotoxic properties.236 A photocrosslinkable

57 terpolymer of dopamine acrylamide (DAM), N-isopropylacrylamide (NIPAAm) and polyethylene glycol-triacrylate (PEG-TA) showed thermoresponsive swelling, strong lap shear adhesion and moisture resistance on gelatin surface along with low cytotoxicity on mouse fibroblast (L929) cells.237 Becker and coworkers synthesized a catechol

functionalized poly(ester-urea) copolymer for tissue adhesion application (Figure 1.27a).

The copolymer showed adhesive strengths (~ 9 kPa) comparable to commercial fibrin

glue on porcine skin with a strong potential for further development in medical adhesives.

The polymer was further modified by incorporation of poly(propylene glycol) (PPG)

(Figure 1.27b) to solubilize it in a clinically relevant solvent like ethanol without much

effect on the adhesion strength (~ 10.6 ± 2.1 kPa) on wet porcine skin.238, 239

a) Reproduced with permission from Zhou, J.; Defante, A. P.; Lin, F.; Xu, Y.; Yu, J.; Gao, Y.; Childers, E.; Dhinojwala, A.; Becker, M. L. Adhesion properties of catechol-based biodegradable amino acid-based poly(ester urea) copolymers inspired from mussel proteins. Biomacromolecules, 2015, 16, 266-274. Copyright 2015 American Chemical Society. b) Reproduced with permission from Zhou, J.; Bhagat, V.; Becker, M. L. Poly(ester urea)-based adhesives: Improved deployment and adhesion by incorporation of poly(propylene glycol) segments. ACS Appl. Mater. Interfaces, 2016, 8, 33423-33429. Copyright 2016 American Chemical Society.

58 Figure 1.27. Chemical structure of catechol modified poly(ester-urea). a) Catechol modified copolymer based on tyrosine and leucine amino acids. b) Catechol modified terpolymer based on serine and leucine amino acid with poly(propylene glycol) (PPG) groups incorporated in the backbone for ethanol solubility.

A degradable, catechol functionalized polyester based adhesive was synthesized by

Agarwal and coworkers, by a radical ring opening copolymerization on glycidyl

methacrylate (GMA), oligo(ethylene glycol) methacrylate (OEGMA) and 2-methylene-

1,3-dioxepane (MDO). The adhesive demonstrated strongest adhesion strength of 13.13 ±

1.74 kPa when Fe(acac)3 was used as a crosslinker, however with poor yields and

considerable cytotoxicity.240 Wilker and coworkers developed a degradable poly((3,4-

dihydroxymandelic acid)-co-(lactic acid)) adhesive with adhesion strengths of 2.6 ± 0.4

MPa under dry conditions and 1.0 ± 0.3 MPa under wet conditions, after cross-linking on aluminum substrate. Adhesion studies performed on sanded steel and teflon resulted in strengths of 1.7 ± 0.5 MPa and 0.32 ± 0.05 MPa respectively. This adhesive holds great promise for application in tissue adhesion, however, more studies on biocompatibility need to be performed.241 In a series of studies, Wilker and coworkers synthesized 3,4-

dihydroxystyrene based biomimetic polymers and studied the effect of molecular weight,

curing conditions and DOPA content on the adhesion strength.242-244 In a recent study,

this biomimetic polymer demonstrated impressive underwater adhesion (~ 3 MPa)

compared to various commercial adhesives (~ 1-0.5 MPa).245 In vitro cell studies have

deemed this polymer cytocompatible, however, lack of degradability limits its biological

application.246 Wu and coworkers reported a pH, glucose and dopamine triple responsive,

59 self-healing hydrogel tissue adhesive based on reversible covalent complexation between

4-arm-PEG-DA (dopamine functionalized 4-arm PEG) and 4-arm-PEG-PBA (phenyl boronic acid functionalized 4-arm PEG) (Figure 1.28). The as formed hydrogel showed adhesion strength of 5.2 ± 0.28 kPa on porcine tissue which remained nearly unchanged for a self-healed hydrogel (5.07 ± 0.35 kPa).247

Reproduced from ref 247 with permission of The Royal Society of Chemistry. http://dx.doi.org/10.1039/ c7py00519a.

Figure 1.28. Hydrogel formation or crosslinking reaction between 4-arm PEG-DA

(dopamine functionalized 4-arm PEG) and 4-arm-PEG-PBA (phenyl boronic acid functionalized 4-arm PEG).

A mussel inspired tissue adhesive based on polyvinylpyrrolidone (PVP) backbone synthesized by Wan and coworkers is particularly noteworthy because of its stronger underwater adhesion compared to dry adhesion. The adhesion strength was strongly dependent on crosslinker to catechol ratio, polymer molecular weight and the catechol content in the polymer. In this study, a polymer of molecular weight ~ 10 kDa, catechol content ~ 16 mol% and FeCl3:catechol = 1:1 demonstrated the highest underwater

60 adhesion strength of 1.33 MPa. The authors postulated that the strong adhesion strength

is a result of a combination of interaction of catechol and amide groups with the glass

surface. Also, the compatibility of the PVP backbone with water allows diffusion of water molecules facilitating penetration of FeCl3 resulting in complexation. The strong

adhesion characteristics, however, need to be supported with studies on biological

substrates for further application in medical adhesives.248 Lei and coworkers synthesized

thermoresponsive Polypeptide-Pluronic-Polypeptide triblock copolymers (Figure 1.29)

and studied their adhesion in wet environments and their hemostatic ability. The block

copolymers were functionalized with different functionalities like catechol, guanidyl,

sulfhydryl and acryloyl to study their effect on adhesion strength. The copolymers were

biocompatible, biodegradable and showed impressive adhesion under wet and humid

conditions on porcine skin and bone in addition to hemostatic properties and improved in

vivo bone healing.249

Reproduced with permission from Lu, D.; Wang, H.; Li, T. E.; Li, Y.; Dou, F.; Sun, S.; Guo, H.; Liao, S.; Yang, Z.; Wei, Q.; Lei, Z. Mussel-inspired thermoresponsive polypeptide-pluronic copolymers for versatile surgical adhesives and hemostasis. ACS Appl. Mater. Interfaces, 2017, 9, 16756-16766. Copyright 2017 American Chemical Society.

61 Figure 1.29. Structures of PPDAC (Pluronic L-31-poly[(DOPA)-co-(Arg-co-Cys)]) and

PPDAL (Pluronic L-31-poly[(DOPA)-co-(Arg-co-Ac-Lys)]) adhesives. Different functional groups result in different interactions: (a), (b) Covalent interaction of catechol groups with tissue surface; (c) di-DOPA crosslinking and DOPA polymerization; (d) electrostatic interaction between guadinium ions (Gu+) and oxoanions on the tissue; (e) disulfide crosslink formation; (f) covalent interaction between catechol and sulfhydryl groups; (g) thiol-ene click reaction.

A number of groups have adopted different strategies like extraction of mussel proteins, recombinant hybrids of proteins or DOPA/catechol functionalization of polysaccharides and synthetic polymers to develop mimics of mussel adhesive. These adhesives work well under dry conditions, however, a significant drop in adhesive strength is generally observed under humid conditions or underwater. Waite and coworkers have reasoned that water tends to form a weak boundary layer at the interface, crazes into the interface, triggers hydrolysis resulting in swelling or plasticizing the adhesive which eventually causes adhesive failure.250 Mussels have figured out a way to

form tenacious bonds with any substrate irrespective of the presence of water, but the

synthetic mimics discussed here are far from the actual mussel adhesive in terms of

adhesion strength. Also, use of UV irradiation, photoinitiators and strong oxidizing

agents could damage the healthy cells around the application site and therefore are only

suitable for topical applications currently. A compilation of mussel inspired mimics, their

crosslinking conditions, test substrates and the maximum reported adhesion strengths are

presented in Table 1.3.

62

Table 1.3. Polymeric mussel inspired adhesives with their crosslinking conditions and maximum reported adhesion strengths.

Maximum adhesion strength Adhesion Cure condition & Adhesive Test substrate Crosslinker Reference test time Reported Standardized

i Surface Porcine duodenal 2 Schnurrer et MAP (130 kDa) \ Films dried under N 2 9 mN/cm 90 Pa 181 Tensiometer mucosa al.

MAP End-to-end Porcine skin \ Humid, 48 h 0.95 ± 0.19 MPa Ninan et al.182

ii 183 MAP Lap-shear Porcine SIS V5+ (250 mM) 1 h 462 ± 46 kPa Ninan et al. Iron 32 kg/cm2 (x=5) 3.14 MPa 1 x iii Tensile shear 23 °C, 3 days, Yamamoto et Copoly(Tyr Lys ) Tyrosinase 60% RH 191 strength 2 al. Alumina 32 kg/cm (x=2) 3.14 MPa Tatehata et Polytripeptide (Gly- Shear Porcine skin Tyrosinase 37 °C, 30 min 118 gf/cm2 11.56 kPa iv 192 Tyr-Lys)n adhesive test al. Poly(Lys.HBr4- Tensile bond 195 v Aluminum Tyrosinase 35 °C, 1 day 4.7 MPa Yu et al. DOPA1) strength

In air, 24 h 400 kPa Deacetylated Yamada et chitosan, dopamine Lap shear Glass slides Tyrosinase al.198 Underwater, 24 h ~ 450 kPa

vi 209 PEG-DOPA4 Lap shear Porcine skin Periodate 37 °C, 24 h 35.1 ± 12.5 kPa Burke et al.

Tensile HA/Pluronic bonding Mouse skin \ R.T., 5 min 7.18 ± 0.93 kPa Lee et al.207 hydrogelvii strength

Catechol-Ala-Ala- decellularized Brubaker et Lap shear \ R.T., 2 h 30.4 ± 3.39 kPa 218 PEG (cAAPEG)viii porcine al.

63

Porcine skin 0.21 ± 0.10 MPa Dry, 25 °C, 12 h

Poly((Lys.HBr)x- Porcine bone 0.29 ± 0.21 MPa 219 Lap-shear Ferric citrate Wang et al. (DOPA)y) Porcine bone Wet, 37 °C, 12 h 0.15 ± 0.01 MPa

Deacetylated chitosan; oxidized Bovine cortical Hoffman et and DOPA End-to-end \ Wet, 37 °C, 3 h 0.31 MPa 200 bone al. functionalized dextran mfp-131 + HA 4.00 ± 0.53 MPa Lap-shear Aluminum \ R.T., 24 h Lim et al.220 mfp-151 + HA 3.17 ± 0.51 MPa (1) R.T., 2 h, 107 ± 24.7 kPa PEG-Dopamine- Lap shear (2) PBS, 37 °C, 1 h Murphy et PCLix Bovine NaIO 227 4 al. Burst test pericardium \ ~ 600 mmHg

Poly(AA-co-AANHS- Chung et Lap shear Porcine skin thiol-PEG R.T., 10 min ~ 11.8 kPa 229 co-MDOPA)x al.

sodium (meta) 168.15 ± 17.02 periodate kPa AbAf iCs Lap shear Porcine SIS Guo et al.231 Humid chamber, 2 h ~ 79.04 ± 9.28 Silver nitrate kPa

xi Porcine, acellular sodium (meta) Mehdizade et iCMBA Lap shear Humid chamber, 2 h 123.2 ± 13.2 kPa 232 SIS periodate (PI) al.

CuSO4, sodium Porcine, acellular 223.11 ± 15.94 233 Click iCs Lap shear L-ascorbate; Humid chamber, 2 h Guo et al. SIS kPa Periodate

64

230 EGAMA-DOPA Lap shear Glass UV-PVA 25 °C, 24 h 0.32 MPa Xue et al. Surface Forces Mfp-3S Mica \ R.T., 1 h F/R = -20 mN/m Wei et al.226 Apparatus (SFA)

Poly(dopamine-co- Horseradish Zhang et Lap shear Porcine skin R.T., 1 day 76 ± 13.4 kPa 236 acrylate) (PDA) peroxide (HRP) al. dry, 6 h (NIPAAm:DAM = 2.27 ± 0.33 MPa xii 20 wt% gelatin 0:10) DAM-NIPAAm Lap shear PEG-TA Ai et al.237 solution 80% humidity, (NIPAAm:DAM = 0.31 ± 0.07 MPa 10:0)

2 h, in buffer soln FeCl 3 (pH = 8.2) ~ 130 kPa rfp-1 (MAP) Lap shear Porcine skin Kim et al.221 30 min air dry, NaIO 4 2 h in PBS ~ 200 kPa

DOPA incorporated recombinant MAP SFA Mica \ Underwater F/R = -9.4 mN m-1 Yang et al.222 (dfp-3)

Irradiation with LAMBAxiii Lap shear Porcine skin dental curing R.T., 2 h, PBS 72.2 ± 3.7 kPa Jeon et al.223 lamp for 60 s

Mfps-CsgA hybrid Atomic force Mica substrate; Zhong et \ \ F/R = 197.5 mN/m 224 protein microscopy Silica probe tip al. Sodium rfp-3 Lap shear Aluminum 37 °C, 4 h 2.28 MPa Yang et al.225 periodate

65

PEU (poly(CA-Tyr- Aluminum xv 65 °C, 24 h 2.46 ± 1.03 MPa 238 co-Leu))xiv Lap shear Bu4N(IO4) Zhou et al. Porcine skin R.T., 4 h 9 kPa

Aluminum 65 °C, 24 h 3.2 ± 0.8 MPa PEU (poly(CA-Ser- Bhagat et co-Leu-co-PPG))xvi Lap shear Bu4N(IO4) al.239 Porcine skin Wet, R.T., 4 h 10.6 ± 2.1 kPa

POEC-d

(octanediol, PEO, Sodium Lap shear Porcine skin humid chamber, 2 h 33.7 kPa Ji et al.234 citric acid, periodate dopamine)

DCTA (gelatin Porcine skin 24.7 ± 3.3 kPa 3+ 203 macromer, Fe , Lap shear FeCl3 + genipin 37 °C, 2 h Fan et al. genipin) Articular cartilage 194.4 ± 20.7 kPa

Dopamine-alginate and HA Bernkop- Scognamiglio Porcine intestine \ R.T., 16 h > 300 min 204 polysaccharide Schnürch et al. membrane

DOPA Porcine skin 13.13 ± 1.74 kPa xvii 240 functionalized Lap shear Fe(acac) 3 R.T., 30 s Shi et al. polyester Aluminum 218.4 ± 16.0 kPa

R.T., 30 min; 2.6 ± 0.4 MPa 37 °C, 24 h Aluminum Poly((3,4- 37 °C, 24 h, 1.0 ± 0.3 MPa dihydroxymandelic Lap shear Bu N(IO ) Underwater, 24 h Jenkins et acid)-co-(lactic 4 4 al.241 acid)) Sanded steel 1.7 ± 0.5 MPa R.T., 30 min; 37 °C,

24 h PTFE 0.32 ± 0.05 MPa

4-arm-PEG-DA and Lap shear Porcine skin \ R.T., 30 min 5.2 ± 0.28 kPa Shan et al.247 4-arm-PEG-PBAxviii

66

Polypeptide- Lap shear Porcine skin 106 kPa xix 249 Pluronic- HRP , H2O2 R.T., 24-25 h Lu et al. Polypeptide Tensile shear Porcine bone 675 kPa

Catechol Sodium 165 functionalized silk Lap shear Aluminum shims R.T., 24 h 120 kPa Burke et al. periodate fibroin

i MAP – mussel adhesive protein ii SIS – small intestine submucosa iii Tyr – tyrosine, Lys – lysine iv Gly - glycine v DOPA – L-3,4-dihydroxyphenylalanine vi PEG – poly(ethylene glycol) vii HA – hyaluronic acid viii Ala – alanine ix PCL - polycaprolactone x AA – acrylic acid; AANHS – acrylic acid N-hydroxysuccinimide ester, MDOPA – N-methacryloyl-3,4-dihydroxy-L-phenylalanine xi iCMBA – injectable citrate based mussel inspired bioadhesives xii DAM – dopamine acrylamide; NIPAAm – N-isopropylacrylamide xiii LAMBA – light activated mussel protein based bioadhesive xiv PEU – poly(ester urea); CA – catechol xv Bu4N(IO4) – tetrabutylammonium periodate xvi Ser – serine; PPG – poly(propylene glycol) xvii Fe(acac) – iron(III) acetylacetonate xviii PBA – phenyl boronic acid; DA - dopamine xix HRP – horseradish peroxide 67 1.6.2. GECKO INSPIRED TISSUE ADHESIVES

In one of the first studies, Karp and coworkers designed a nanopatterned gecko inspired biocompatible and biodegradable elastomeric tissue adhesive by using poly(glycerol-co- sebacate acrylate) (PGSA) (Figure 1.30). The adhesion strengths on porcine intestinal tissue were dependent on nanopillar parameters, the polymer composition and oxidized dextran coating thickness. The oxidized dextran promotes adhesion with tissue by Schiff base formation and at the same time forms hemiacetal with hydroxyl groups from the glycerol subunit of PGSA. The maximum adhesion strength on porcine intestinal tissue was ~ 4.8x104 N/m2.251

Reproduced with permission from ref 251, Copyright 2008 National Academy of Sciences

Figure 1.30. Gecko inspired poly(glycerol-co-sebacate acrylate) (PGSA) tissue adhesive.

PGSA prepolymer is nanomolded by UV irradiation, followed by spin coating with dextran aldehyde (DXTA). SEM image showed excellent pattern transfer fidelity.

68 1.6.3. SANDCASTLE WORM INSPIRED TISSUE ADHESIVES

An acrylate based mimic of sandcastle worm glue was developed by Stewart and coworkers based on the principle of coacervation, electrostatic interaction as well as oxidative curing. The adhesion strength on wet porcine bone was 1/3 of the strength of natural glue and ~ 37% of the cyanoacrylate glue (control). In a continuation of this study, a two component mimic consisting of poly(MAEP85-dopamide15) and amine modified gelatin with divalent cations (Ca2+ and Mg2+) showed temperature dependent

coacervation which could be altered by changing the cation ratio (Figure 1.31). This

model adhesive was successfully applied in craniofacial reconstruction while

demonstrating both in vitro and in vivo adhesion, non-cytotoxicity towards cells along

with aligned tissue reconstruction without inflammatory response.252-254

Reprinted from ref 252 with permission from John Wiley and Sons. Copyright 2009 Wiley-VCH Verlag GmbH & KGaA, Weinheim, Germany.

69 Figure 1.31. Synthetic adhesive mimic of P. californica glue. 1: Structure of Pc3 analog;

2: Structure of Pc1 analog. Model of pH dependent coacervation and adhesion. (a) At

acidic pH 4, polyphosphates (black) and polyamines (grey) form colloidal polyelectrolyte

complexes (PECs) with a net positive charge, (b) At slightly basic pH of 8.2, the

extended polyphosphates form network with polyamines and divalent cations with a net

negative charge, (c) On oxidation 3,4-dihydroxyphenol (D) initiates covalent crosslinking

and surface interaction via quinones (Q).

Kaur et.al. added a second polymerizable phase of polyethylene glycol-diacrylate

(PEG-da) to poly(MOEP-co-DMA), poly(acrylamide-co-aminopropyl methacrylamide) and Ca2+ coacervates. The highest bonding strength was 973 ± 263 kPa for 17.7 wt%

PEG-da on aluminum substrates with good underwater adhesion.255 Wan and coworkers

functionalized a polyoxetane backbone with 5 mol% catechol moieties and 25 mol% bis-

phosphoric acid groups to mimic sandcastle worm adhesive. A maximum bonding

strength of 0.35 MPa was achieved under humid conditions when Fe3+ was used as a

curing agent. This study also highlighted that, an optimum concentration of adhesive

groups and crosslinker is crucial to achieve strong bonding strengths.256 Miserez and coworkers synthesized a two component polypeptide mimic of the sandcastle worm. In this mimic a lysine containing positively charged polypeptide was functionalized with

DOPA and a negatively charged polypeptide was functionalized with phosphorylated serine and tyrosine. On mixing, the two polypeptides formed complex coacervates under neutral charge conditions.257

70 1.6.4. MIMETIC ADHESIVES

Nishida and coworkers synthesized a polypeptide by ring opening polymerization of

NCA monomers to mimic barnacle (Balanus hameri) adhesives. The synthetic mimic

demonstrated highest tensile shear strength of 15.2 kg/cm2 and compressive shear

strength of 19.2 kg/cm2 on iron substrates.258 In their subsequent efforts, they synthesized

three different model peptides AA-5, AA-10 and AA-17 with different amino acid compositions. AA-17 crosslinked with tyrosine and α-chymotrypsin demonstrated strong bonds with iron surfaces (~ 37 kg/cm2). All the model peptides showed poor adhesion on

bovine bone while tyrosine crosslinked AA-17 showed the strongest bond ~ 3.7 kg/cm2

(362.85 kPa).259 In another study to mimic the barnacle adhesive, a polyacrylamide

based copolymer with hydroxyl and hexyl groups for surface interaction and tetra alanine

groups for crosslinking via hydrogen bonding (hydrophobic interaction) was developed.

The copolymer gel showed strong adhesion on PMMA substrate with tensile adhesive

strength of 402 kPa.260

1.6.5. CADDISFLY INSPIRED TISSUE ADHESIVES

Stewart and coworkers described an electrostatically driven coacervate formation by

using alternating anionic and cationic block copolymers. As a mimic of caddisfly silk

these block copolymers are functionalized with amine, phosphate, divalent cations and

also dihydroxyl aromatic groups for oxidative crosslinking.261 Becker and coworkers

developed a caddisfly mimic adhesive of phosphate functionalized, amino acid based

poly(ester urea) copolymer (Figure 1.32). These adhesives demonstrated strong adhesion

71 of ~ 439 ± 203 kPa on bovine bone when crosslinked via electrostatic interactions with

Ca2+.262

Reprinted with permission from Bhagat, V.; O’Brien, E.; Zhou, J.; Becker, M. L. Caddisfly inspired phosphorylated poly(ester urea)-based degradable bone adhesives. Biomacromolecules, 2016, 17, 3016-3024. Copyright 2016 American Chemical Society.

Figure 1.32. Caddisfly adhesive mimic of phosphate functionalized poly(ester urea) copolymer based on serine and valine amino acids.

72

CHAPTER II

MATERIALS AND INSTRUMENTS

2.1. MATERIALS

All chemicals were used as received unless noted otherwise.

Acetic acid (≥ 99.7%), calcium hydride (95%) (CaH2), 3,4-dihydroxyhydrocinnamic acid (98%), 4-dimethylaminopyridine (99%) (DMAP), diphenylphosphoryl chloride

(99%) (DPPC), L-valine (99%), 4M HCl/Dioxane solution, L-leucine (99%), 1,8-

octanediol (98%), p-toluene sulfonic acid monohydrate (≥ 98.5%) (pTSA), palladium on carbon (extent of loading: 10 wt% loading, matrix activated carbon support) (Pd/C), PtO2

(surface area ≥ 75 m2/g), pyridine (anhydrous, 99.8%), poly(propylene glycol) bis(2-

aminopropyl ether) (average Mn ~ 400 Da) (PPG), tetrabutylammonium periodate (97%)

(Bu4N(IO4)), triethylamine (≥ 99%) (TEA) and triisopropylsilane (98%) (TIPS) were

purchased from Sigma Aldrich. Sodium carbonate (99.5%) (Na2CO3) was purchased

from Fisher Scientific. N-Boc-O-Bzl-L-Serine (98%) was purchased from Ark Pharm,

Inc. N,N’-diisopropylcarbodiimide (98%) (DIC), ethylcarbodiimide hydrochloride (EDC)

and trifluoroacetic acid (99%) (TFA) were purchased from Oakwood Chemicals.

Triphosgene (98%) was purchased from TCI America.

73 Organic used were ACS grade and obtained from Sigma-Aldrich or Fisher

Scientific. Anhydrous chloroform and anhydrous DCM were obtained by distilling the respective ACS grade solvents after stirring overnight with CaH2. Toluene, ethanol and

DMF were obtained from Sigma-Aldrich and used without further purification.

Silica gel for column chromatography was purchased from Sorbent (Norcross, GA),

with porosity 60 Å, size 200-400 mesh, surface area 450-550 m2/g, bulk density 0.5 g/ml

and pH 6.0-7.0.

2.2. INSTRUMENTS

2.2.1. NUCLEAR MAGNETIC RESONANCE (NMR).

1H, 13C and 31P NMR spectra of monomers and polymers were recorded on a Varian

NMR Spectrophotometer at 300 or 500 MHz for 1H NMR, 125 or 75 MHz for 13C NMR

and 202 MHz for 31P NMR. Chemical shifts (δ) were reported in ppm and referenced to

1 13 residual solvent resonances ( H NMR, DMSO-d6: δ = 2.50 ppm and C NMR, DMSO-

31 d6: δ = 39.50 ppm). P NMR was referenced to 85% H3PO4 as an external standard (δ =

0 ppm). Abbreviations of multiplicities are denoted as s-singlet, d-doublet, m-multiplet,

q-quartet, dd-double doublet, td-triple doublet. Coupling constants are reported in hertz

(Hz).

2.2.2. ELECTROSPRAY IONIZATION-MASS SPECTROMETRY (ESI-MS).

ESI-MS spectrum of the monomer was recorded on Bruker HCTultra II quadrupole ion

trap (QIT) mass spectrometer (Billerica, MA) equipped with an ESI source.

74 2.2.3. ATTENUATED TOTAL REFLECTION INFRARED SPECTROSCOPY (ATR-

IR).

ATR-IR spectra of the polymers were recorded on Shimadzu Miracle 10 ATR-IR

equipped with a quartz crystal window. The spectra were recorded between 4000 to 400

cm-1. A small piece of polymer, enough to cover the crystal window was used for the

measurements.

2.2.4. SIZE EXCLUSION CHROMATOGRAPHY (SEC).

Molecular weights and PDI (ÐM) of the polymers were determined by Size exclusion

chromatography on TOSOH ECOSEC HLC-8320GPC using DMF (0.1 M LiBr salt solution) as eluent at a flow rate of 0.5 mL/min at 50 °C equipped with a refractive index

detector.

2.2.5. Differential Scanning Calorimetry (DSC).

The glass transition temperature (Tg) of polymers was determined by Differential

Scanning Calorimetry (DSC, TA Q200) at a scanning rate of 10 °C/min for 5 cycles from

-50 °C to 60 °C. A small piece of polymer about 5 mg-8 mg sealed in a hermetic pan was used as the sample.

2.2.6. THERMOGRAVIMETRIC ANALYSIS (TGA).

The decomposition temperatures of polymers (Td) were determined using

Thermogravimetric Analysis (TGA, TA Q500) at a heating rate of 10 °C/min from 25 °C

to 600 °C.

75 2.2.7. ULTRAVIOLET-VISIBLE SPECTROSCOPY (UV-VIS).

UV-vis spectra were recorded on a SynergyMX plate reader (Bio-Tek Instrument, Inc.).

Data was analyzed by BioTek’s Gen5TM Data Analysis Software.

2.2.8. CONTACT ANGLE.

Contact angles were determined for surface energy measurements using the

Owens/Wendt method. of known surface tension (water, glycerol, ethylene

glycol, propylene glycol, and formamide) were used for the contact angle measurements.

Polymer thin films were spun coat on ozone treated silicon wafers using 1% (w/v)

solution of the polymers in ethanol at 2000 rpm for 1 min. The samples were dried at 70

°C under vacuum overnight. Measurements (n = 3) were performed at room temperature

on a Ramé-Hart contact angle goniometer and standard deviation of the mean was

calculated from independent measurements.

2.2.9. INSTRON.

Lap shear adhesion tests on aluminum adherends were carried out on an Instron 5567

universal testing machine equipped with a 1000 N load cell by pulling the adherends at a

rate of 1.3 mm/min. The adhesion strength was measured by dividing the maximum force

(N) required with the lap area (m2).

2.2.10. TEXTURE ANALYZER.

The lap-shear measurements on porcine skin and bovine bone substrates were

performed on a Texture Analyzer (TA.XT.Plus) equipped with a 5 kg load cell by pulling 76 the adherends at a rate of 1.3 mm/min. The adhesion strength was measured by dividing the maximum force (N) required with the lap area (m2).

77

CHAPTER III

POLY(ESTER UREA) BASED ADHESIVES: IMPROVED DEPLOYMENT AND

ADHESION BY INCORPORATION OF POLY(PROPYLENE GLYCOL) SEGMENTS

V Bhagat ||, J Zhou ||, ML Becker, ACS Appl. Mater. Interfaces, 2016, 8(49), 33423-33429

3.1. OUTLINE

The adhesive nature of mussels arises from the catechol moiety in 3,4- dihydroxyphenylalanine (DOPA) amino acid; one of the many proteins that contribute to the unique adhesion properties of mussels. Inspired from these properties, many biomimetic adhesives have been developed over the last few years in an attempt to replace adhesives like fibrin, cyanoacrylate and epoxy glues. In the present work we have synthesized ethanol soluble but water insoluble catechol functionalized poly(ester urea) random copolymers that help to facilitate delivery and adhesion in wet environments.

Poly(propylene glycol) units incorporated into the polymer backbone imparts ethanol solubility to these polymers making them clinically relevant. A catechol to crosslinker ratio of 10:1 with a curing time of 4 h exceeded the performance of commercial fibrin glue (4.8 ± 1.4 kPa) with adhesion strength of 10.6 ± 2.1 kPa. These adhesion strengths are significant considering the adhesion studies were performed under wet conditions.

78 3.2. INTRODUCTION

Mussels are aquatic organisms with the ability to attach to a wide variation of surfaces.

Mussel adhesives are unique in their ability to attach to a variety of substrates underwater

and the impressive tensile strength of the adhesive bond.173, 174 The adhesive discs at the

tip of the mussel byssus consist of polyphenolic proteins rich in 3,4-

dihydroxyphenylalanine (DOPA) and hydroxyproline. The mussel byssus adhesives

attach rapidly, are reversible, survive variations in temperature and are generally

unaffected by the presence of water175, 250; all of which are desirable in synthetic tissue

adhesives. Although the dynamic role of DOPA in mussel adhesion is not completely

understood, studies have shown its participation in both adhesive and cohesive bonding.

The reduced form of DOPA imparts adhesive bonding while the oxidized form is

responsible for cohesive bonding.263 Various crosslinking mechanisms proposed for

cohesive bonding include diDOPA crosslink formation upon oxidation by periodate or

peroxides264, 265, metal ion via reactions266, 267 and radical coupling reactions.268, 269

In the last decade, tissue adhesives based on fibrin, polyethylene glycol (PEG),

cyanoacrylate, epoxy, chitosan, gelatin and urethane have gained considerable clinical

interest and use; however the applications are limited due to cytotoxicity, poor

mechanical properties and tissue damage caused by the crosslinking mechanism.3, 10, 25, 270

Recently, biomimetic surrogates, including mussel inspired adhesives have demonstrated

the potential to surpass the performance of other synthetic adhesives with minimal side effects. Mussel inspired tissue adhesives based on water soluble co-polypeptides121, 195, 79 coacervates based on mussel adhesive protein (fp-151-RGD) and hyaluronic acid271,

polystyrene242, 272, polyethylene glycol (PEG)210, 213, 218, 267, polyacrylate211, 229, 273, 274,

polyacrylamide237, 275-277, polyurethane278 and polyester240 have all been synthesized and studied for their adhesive properties. These adhesives however, are typically water soluble and tend to rapidly dissolve in the surrounding body fluids during deployment. It is difficult to sequester them at the application site until they are crosslinked. It is

therefore important to design synthetic biomimetic adhesives that are insoluble in water

but soluble in clinically relevant solvents such as ethanol.

Poly(ester urea)s are a class of polymers consisting of both ester and urea bonds

responsible for degradability and strength, respectively.279 Our previous studies on in vitro degradation have shown poly(ester urea)s to be degradable over time without local acidification making them suitable candidates for tissue adhesive application.280, 281 In our

previous work, the synthesis and characterization of catechol functionalized poly(ester-

urea) copolymers as mimics of mussel adhesive and their potential as soft tissue

adhesives was demonstrated. The maximum adhesion strength obtained by lap shear tests

on porcine skin was ~ 9 kPa which is comparable to commercially available fibrin glue.

238 In spite of the high adhesion strength of the polymers, their clinical application was

limited as they were only soluble in high boiling organic solvents which are detrimental

to the tissue. Ethanol is a clinically relevant delivery mechanism as it and isopropanol are

sometimes used to dehydrate surfaces locally in surgical situations where adhesives are

being applied. For this purpose the present work summarizes synthesis and characterization of ethanol soluble, water insoluble catechol functionalized poly(ester

80 urea) random copolymers. Poly(propylene glycol) (PPG, 20 mol%) units incorporated

into the polymer backbone imparts ethanol solubility to these polymers making them

clinically relevant. PEUs with different PPG composition were synthesized and their

solubility was studied in both ethanol and water. The solubility studies showed that below

20 mol% PPG incorporation the polymers were partially ethanol soluble, however, above

20 mol% PPG; the polymers showed water soluble behavior. 20 mol% PPG units in the

backbone chain were found to provide optimum composition for the polymers to be

ethanol soluble while at the same time water insoluble.

3.3. EXPERIMENTAL SECTION

3.3.1. Synthesis of di-p-toluenesulfonic acid salt of bis(L-leucine)-1,8-octanyl diester

(M1). A mixture of 1,8-octanediol (10.00 g, 0.068 mol), p-toluenesulfonic acid monohydrate (31.07 g, 0.163 mol), L-leucine (20.46 g, 0.156 mol) and toluene (200 mL) was heated to reflux for 20 h. A product precipitates from the reaction mixture after cooling to ambient temperature. The solid product was washed with diethyl ether and purified by recrystallization (3 times) in water to obtain 42.9 g (yield ~ 88%) white

1 . H NMR (300 MHz, DMSO-d6): 0.89 (d, J = 5.86 Hz, 12H), 1.28 (br. s., 8H),

1.51-1.65 (m, 8H), 1.66-1.78 (m, 2H), 2.29 (s, 6H), 3.98 (t, J = 7.03 Hz, 2H), 4.07-4.23

(m, 4H), 7.12 (dd, J = 8.49, 0.59 Hz, 4H), 7.42-7.54 (m, 4H), 8.30 (br. s., 6H). 13C NMR

(125 MHz, DMSO-d6): 21.22, 22.37, 22.58, 24.26, 25.58, 28.32, 28.89, 51.09, 66.06,

125.93, 128.51, 138.18, 145.94, 170.38.

81

Scheme 3.1. Synthesis of di-p-toluenesulfonic acid salt of bis(L-leucine)-1,8-octanyl diester (M1).

3.3.2. Synthesis of bis-N-Boc-O-benzyl(L-serine)-1,8-octanyl diester. 1,8-octanediol

(1.55 g, 10.6 mmol), N-Boc-O-benzyl-L-serine (7.50 g, 25.4 mmol) and DPTS (0.62 g,

2.12 mmol) were stirred in anhydrous CH2Cl2 (100 mL) under N2 and lowered in an ice

bath. When the reaction mixture cools down to 0 °C, DIC (4.6 mL, 29.6 mmol) was

injected in the reaction flask and stirred overnight. The solution was filtered and

concentrated to obtain an oily liquid as crude product. The product was dissolved in

CHCl3, washed twice with 5% HCl solution, brine once, dried over Na2SO4 and

concentrated under vacuum. A light yellow oil (13.7 g, yield 92%) was obtained after

purification with column chromatography on silica gel with EtOAc/hexane (10/90, v/v)

1 as eluent. H NMR (500 MHz, CDCl3): 1.22-1.36 (m, 8H), 1.40-1.51 (m, 18H), 1.56-1.66

(m, 4H), 3.70 (dd, J = 9.41, 3.06 Hz, 2H), 3.89 (dd, J = 9.17, 2.81 Hz, 2H), 4.09-4.20 (m,

4H), 4.43 (d, J = 8.80 Hz, 2H), 4.47-4.57 (m, 4H), 5.40 (d, J = 8.56 Hz, 2H), 7.23-7.38

13 (m, 10H). C NMR (125 MHz, DMSO-d6): 25.71, 28.32, 29.03, 54.08, 65.49, 70.14,

73.23, 79.80, 127.50, 128.36, 137.61, 155.46, 170.66.

3.3.3. Synthesis of di-hydrochloric acid salt of bis-O-benzyl(L-serine)-1,8-octanyl diester (M2). bis-N-Boc-O-benzl(L-serine)-1,8-octanyl diester (13.7 g, 19.5 mmol) was 82 stirred in 4.0 M HCl/dioxane solution (110 mL) under inert gas for 12 h. The solution

was freeze dried to obtain a light yellow solid as product. The product was washed three

times with diethyl ether and dried under vacuum to yield 10.4 g (yield ~ 93%) of light

1 yellow powder as pure product. H NMR (500 MHz, DMSO-d6): 1.13-1.31 (m, 8H),

1.46-1.57 (m, 4H), 3.81-3.92 (m, 4H), 4.05-4.21 (m, 4H), 4.29 (t, J = 3.30 Hz, 2H), 4.46

(d, J = 12.23 Hz, 2H), 4.58 (d, J = 12.23 Hz, 2H), 7.25-7.40 (m, 8H), 8.79 (br. s., 6H).

13 C NMR (125 MHz, DMSO-d6): 25.53, 28.40, 28.92, 52.87, 66.12, 66.82, 67.86, 72.89,

127.99, 128.11, 128.67, 137.87, 168.18.

Scheme 3.2. Synthesis of di-hydrochloric acid salt of bis-O-benzyl(L-serine)-1,8-octanyl diester (M2).

3.3.4. Synthesis of Poly(benzyl serine-co-leucine-co-poly(propylene glycol))

(Poly(bzlSer-co-Leu-co-PPG)) via Interfacial Polymerization. M1 (7.50 g, 10.44 mmol),

M2 (2.00 g, 3.50 mmol), poly(propylene glycol) bis(2-aminopropyl ether) (1.40 g, 3.50 mmol) and sodium carbonate (3.90 g, 36.5 mmol) were weighed in a 1 L three-neck

83 round-bottom flask. About 75 mL of DI water was poured in the flask and the solution

was stirred using a mechanical stirrer. The reaction solution was stirred for 1 h in a 35 °C

water bath. The water bath was replaced with an ice bath and additional sodium

carbonate (2.03 g, 19.14 mmol) dissolved in 15 mL water was added. After about 30 min,

triphosgene (2.10 g, 7.08 mmol) dissolved in 20 mL CHCl3 was quickly added to the

flask under vigorous stirring. After stirring for 30 min, the ice bath was removed and an

additional aliquot of triphosgene solution (0.35 g, 1.15 mmol in 10 mL chloroform) was

added dropwise for 30 min. After stirring the reaction mixture for 2 h, the organic phase

was precipitated in boiling water. 6.00 g of white polymer was obtained (yield 89%) after

drying. (ÐM = 1.8, Mn = 42.3 kDa, Mw = 76.3 kDa)

3.3.5. Synthesis of Poly(serine-co-leucine-co-poly(propylene glycol)) (Poly(Ser-co-

Leu-co-PPG)). Poly(bzlSer-co-Leu-co-PPG) (5.80 g in 50 mL DMF) was subjected to hydrogenation (60 psi) at 40 °C for 24 h with Pd/C (0.90 g) catalyst. The catalyst was removed by filtration and the concentrated polymer solution was precipitated in water.

5.78 g (yield 96%) of white polymer was obtained as pure product. (ÐM = 1.9, Mn = 19.8

kDa, Mw = 37.4 kDa)

3.3.6. Synthesis of Poly(acetonide protected catechol-serine-co-leucine-co- poly(propylene glycol)) (Poly(CA-AN-Ser-co-Leu-co-PPG)). A solution of Poly(Ser- co-Leu-co-PPG) (2.00 g), acetonide protected 3,4-dihydroxyhydrocinnamic acid (0.46 g,

2.1 mmol) and DPTS (0.123 g, 0.42 mmol) dissolved in 15 mL DMF under N2 was

cooled to 0 °C with an ice bath. EDC (0.48 g, 2.52 mmol) was injected in the flask and

84 the reaction was stirred overnight. The reaction solution was precipitated in water (pH =

8-9, adjusted by addition of NaHCO3) to obtain a solid product which was further

dissolved in CHCl3 and precipitated in cold ether. A white solid (1.8 g, yield 59%) was

obtained as pure polymer after drying. (ÐM = 1.7, Mn = 26.0 kDa, Mw = 43.5 kDa)

3.3.7. Synthesis of Poly(catechol-serine-co-leucine-co-poly(propylene glycol))

(Poly(CA-Ser-co-Leu-co-PPG)). Poly(CA-AN-Ser-co-Leu-co-PPG) (1.00 g) was dissolved in 10 mL degassed CHCl3 and a mixture of 4 mL trifluoroacetic acid, 0.2 mL

H2O and 5 drops of triisopropylsilane was added to the flask under argon and stirred for 2

h at room temperature. The solution was concentrated and precipitated in cold ether to

obtain 0.83 g of polymer.

3.3.8. Determination of catechol content in Poly(CA-Ser-co-Leu-co-PPG). The

content of the catechol functional groups in the PEU was determined by UV-visible spectroscopy. As the catechol group has a characteristic π-π* absorption at 283 nm, a standard curve was obtained using N-(3,4-dihydroxyphenethyl)methacrylamide as a control molecule (Figure 3.1). THF was used as the solvent.

Figure 3.1. Structure of N-(3,4-dihydroxyphenethyl)methacrylamide.

3.3.9. In vitro cell viability. Poly(CA-Ser-co-Leu-co-PPG) films were spun coat on 12 mm diameter glass coverslips from a 1% solution of PEU in ethanol. The spun coat

85 coverslips were dried under vacuum at 80 °C overnight. The samples were carefully transferred to a 12 well plate and about 1 mL of media was added to each well. NIH 3T3 mouse fibroblast cells were detached from the culture flask using trypsin/EDTA solution and the cell density was determined using a hemocytometer. About 250 μL aliquot of cell solution was added to each sample well (cell density – 6000 cells/cm2). The well plates were mildly agitated to ensure uniform cell distribution on the sample surface. The samples were incubated for 24 h before the viability test.

The cell viability of NIH3T3 cells on PEU samples was measured using a

LIVE/DEAD® Viability/Cytotoxicity Kit, for mammalian cells (Life Technologies). The staining solution is prepared by dissolving calcein-AM (4 mM, 5 μL) and ethidium homodimer (10 μL) in 10 mL DPBS in dark. The sample wells were rinsed with DPBS once and incubated in 0.5 mL staining solution at 37 °C for 10 min. Images were taken on an Olympus fluorescence microscope equipped with a Hamamatsu orca R2CCD camera,

FITC and TRITC filters at 4X magnification using CellSENS software. Calcein-AM stains live cells green while the dead cells are stained red by ethidium homodimer. The images were analyzed in ImageJ software using cell counter plugin.

3.3.10. Statistical analysis. Statistical analysis was performed using ANOVA with

Tukey post hoc analysis with a significance level of p < 0.05 in Minitab statistical analysis software.

86 3.4. RESULTS AND DISCUSSION

3.4.1. Polymer synthesis and characterization. Scheme 3.3. shows the synthetic route

for the catechol functionalized poly(ester urea) (PEU) incorporating poly(propylene

glycol) (PPG) units. Poly(bzlSer-co-Leu-co-PPG) with benzyl protected serine monomer, leucine monomer and poly(propylene glycol) units was prepared by interfacial polymerization between the respective monomers and triphosgene. The benzyl groups were removed via hydrogenation in the presence of Pd/C catalyst to obtain free hydroxyl groups. The hydroxyl groups were further functionalized in a post polymerization modification strategy involving a simple DIC coupling between the hydroxyl groups on the copolymer and acetonide protected 3,4-dihydroxyhydrocinnamic acid. The acetonide group is acid labile and the deprotection was carried out in the presence of trifluoroacetic acid (TFA) for 2 h to obtain Poly(CA-Ser0.2-co-Leu0.6-co-PPG0.2). In the previous work,

catechol functionalized tyrosine and leucine based copolymers were reported as soft

tissue adhesives.238 However, their insolubility in clinically relevant solvents limits their

applicability. Poly(propylene glycol) imparts ethanol solubility to polymers and is less

toxic compared to poly(ethylene glycol) (PEG).282 In this work, we incorporated 20

mol% PPG units in the polymer backbone chain to promote solubility in ethanol making

these adhesives suitable for medical applications.

87

Scheme 3.3. Synthesis route of Poly(CA-Ser0.2-co-Leu0.6-co-PPG0.2).

88

Figure 3.2 is a compilation of 1H NMR spectra for each of the polymers at different

steps during the synthesis. The peak around δ ~ 1.14 ppm (Figure 3.2(A-D)) corresponds to the methyl groups on the PPG backbone unit confirming its incorporation in the polymer. The protons on benzyl protecting groups have chemical shifts in the aromatic region around δ ~ 7.25 – 7.33 ppm (Figure 3.2(A)) which disappears after hydrogenolysis

(Figure 3.2(B)) confirming complete and clean deprotection. Figure 3.2(C) corresponds to the acetonide protected 3,4-dihydroxyhydrocinnamic acid functionalized copolymer.

Figure 3.2. 1H NMR spectra of (A) Poly(bzlSer-co-Leu-co-PPG) (B) Poly(Ser-co-Leu-co-

PPG) (C) Poly(CA-AN-Ser-co-Leu-co-PPG) (D) Poly(CA-Ser-co-Leu-co-PPG).

89 The methylene groups on the catechol functionality have chemical shifts between δ ~

2.5 - 2.7 ppm. The chemical shifts of the aromatic ring in the catechol group are between

6.57 – 6.67 ppm and the methyl groups on the acetonide protecting group have a

chemical shift around δ ~ 1.59 ppm. This confirms the successful post polymer

modification by the protected catechol moiety. Figure 3.2(D) represents the 1H NMR spectra of the final catechol functionalized polymer poly(CA-Ser-co-Leu-co-PPG). The peak at δ ~ 1.59 ppm disappears after deprotection with TFA confirming complete deprotection of the acetonide groups. The catechol functionalization was also confirmed by 13C NMR spectroscopy (Figure 3.3).

Figure 3.3. 13C NMR of (A) Poly(Ser-co-Leu-co-PPG) (B) Poly(CA-AN-Ser-co-Leu-co-

PPG) and (C) Poly(CA-Ser-co-Leu-co-PPG).

90 The peak at δ ~ 25.9 ppm is characteristic of the methyl protons of the acetonide protecting group (Figure 3.3(B)) which disappears after deprotection with TFA (Figure

3.3(C)). FT-IR spectra (Figure 3.4) also prove the successful functionalization with

catechol group. The peak around 1500 cm-1 (Figure 3.4(B)) is the characteristic of the methyl groups on the acetonide protecting group which disappears after the deprotection

(Figure 3.4(C)).

Figure 3.4. FT-IR spectra of (A) Poly(Ser-co-Leu-co-PPG) (b) Poly(CA-AN-Ser-co-Leu- co-PPG) (c) Poly(CA-Ser-co-Leu-co-PPG).

The copolymer was also characterized for its physical properties including number

average and weight average molecular masses (Mn and Mw respectively) and molecular mass distribution (ÐM) by SEC, decomposition temperature (Td) by TGA and glass

transition temperature (Tg) by DSC. The molecular mass of the polymer drops after the

benzyl deprotection and shows an increase following the protected catechol

91 derivatization. The poly(CA-Ser-co-Leu-co-PPG) is highly adhesive in nature and to avoid adhesion to the high surface area SEC columns, molecular mass measurements of the deprotected polymers were not performed. The molecular mass distributions of these polymers are slightly less than 2 (typical for step growth polymerization) as the polymer undergoes slight fractionation when precipitated in water during the purification step. The decomposition temperatures of the copolymers were above 150 °C, ensuring their stability at physiological temperatures.

Table 3.1. Physical properties of PPG containing poly(ester urea)s.

(a) Mn Mw Td Tg Polymer ÐM (kDa) (kDa) (°C) (°C)

Poly(bzlSer-co-Leu-co-PPG) 42.3 76.3 1.8 254 12

Poly(Ser-co-Leu-co-PPG) 19.8 37.4 1.9 239 18

Poly(CA-AN-Ser-co-Leu-co-PPG) 26.0 43.5 1.7 234 11

Poly(CA-Ser-co-Leu-co-PPG) \ \ \ 187 10

(a) Tg determined from DSC.

3.4.2. Determination of the catechol content in Poly(CA-Ser-co-Leu-co-PPG). UV-

Visible spectroscopy (UV-VIS) was used to quantify the catechol content in the copolymer. UV-Vis is highly sensitive even at micromolar to picomolar concentrations and significantly more accurate than 1H NMR spectroscopy. The protons from the PPG

unit have broad resonance which overlaps with the water peak from DMSO. This overlap

leads to significant errors in the integration values leading to inaccurate measurement of

the catechol content. The catechol group has a characteristic absorption at 283 nm.

92 Absorption of N-(3,4-dihydroxyphenethyl) methacrylamide at different concentrations in

THF was measured. A plot of absorption vs. concentration provides the standard curve for the accurate measurement of catechol content (Figure 3.5, Inset).

Catechol content in the copolymer was determined from the absorption value at 283 nm from polymer solution in THF. The polymer solution (0.47 mg/mL) has an absorption value of 0.952 (Figure 3.5). From the standard curve, the catechol content in the copolymer was determined to be 0.34 mM. Determination of the catechol content is important to figure out the optimum concentration of cross-linker required for curing during the adhesion measurements.

Figure 3.5. UV-vis spectrum of Poly(CA-Ser-co-Leu-co-PPG) showing a peak around

283 nm characteristic of the π-π* absorption of the catechol group. Inset shows the calibration curve constructed from N-(3,4-dihydroxyphenethyl)methacrylamide and used to determine the concentration of catechol groups in the terpolymer.

93 3.4.3. Lap-shear adhesion measurements. Lap shear adhesion tests were first performed on aluminum substrates (Figure 3.6(A)). The lap-shear adhesion strength of

Poly(CA-Ser-co-Leu-co-PPG) was determined before and after crosslinking with

Bu4N(IO4) (Figure 3.6(B)). Poly(Ser-co-Leu-co-PPG) was used as a control polymer and showed an adhesion strength of 0.7 ± 0.3 MPa. The incorporation of catechol groups did not lend to an appreciable change in the adhesion strength 1.4 ± 0.6 MPa. However, after crosslinking with periodate, the adhesion strength increases considerably to 3.2 ± 0.8

MPa compared to the control and its uncross-linked counterpart. Lap-shear adhesion strength is a combination of both adhesive and cohesive strengths and an optimum balance between the two is important to achieve the maximum adhesion strength.

Crosslinking improves the cohesive strength within the bulk of the polymer resulting in an overall increase in the adhesion strength of the copolymer. This result clearly indicates that the catechol groups improve the adhesion properties dramatically after cross-linking.

Figure 3.6. (A) Illustration of the lap-shear adhesion test272 (B) Comparison of adhesion

strength on aluminum substrates; Ser-PEU: Poly(Ser-co-Leu-co-PPG) – Control; CA-

94 - PEU: Poly(CA-Ser-co-Leu-co-PPG); CA-PEU+IO4 : Poly(CA-Ser-co-Leu-co-PPG) crosslinked with Bu4N(IO4). * represents p < 0.05 for samples compared with the control

(n = 10).

To demonstrate the clinical applicability of catechol functionalized copolymers, lap-

shear adhesion tests were also performed on porcine skin under wet conditions (Figure

3.7). Poly(Ser-co-Leu-co-PPG) (Ser-PEU) was used as a control polymer for this study

along with commercially available fibrin glue. The adhesion strengths were compared

before and after crosslinking with periodate ions. For the porcine skin study, 4 different

- CA:crosslinker (IO4 ) ratios were chosen with 2 different crosslinking times (30 min and

4 h). The adhesion strength of Ser-PEU was very low about 1.5 ± 0.5 kPa after curing for

30 min. Increasing the curing time to 4 h did not bring an appreciable change in its adhesion strength ~ 2.0 ± 0.5 kPa. For Poly(CA-Ser-co-Leu-co-PPG) (CA-PEU) without crosslinker, the adhesion strength after 30 min of curing was ~ 2.0 ± 0.7 kPa which showed a slight increase (~ 2.5 ± 0.6 kPa) after curing for 4 h. In both the cases, adhesion strength was much lower compared to fibrin glue which was ~ 3.5 ± 1.7 kPa after 30 min of curing and 4.8 ± 1.4 kPa after 4 h of curing. When the samples were crosslinked; for

- CA:IO4 ratio of 10:1 and 10:2 the measured adhesion strengths were ~ 1.7 ± 0.9 kPa and

~ 1.2 ± 0.3 kPa respectively after curing for 30 min, which were much lower than the

- fibrin glue. However, when the CA:IO4 ratio increased to 10:3 and 10:4, the adhesion

strengths observed were ~ 3.4 ± 1.3 kPa and ~ 3.8 ± 0.8 kPa respectively and comparable

- to the adhesion strength of fibrin glue at 30 min. These results suggest that CA:IO4 ratios

of 10:1 and 10:2 are insufficient to obtain high adhesion strengths for a curing time of 30

95 min. Ratios of 10:3 and 10:4 resulted in higher adhesion strengths as the increased

crosslinker content accelerated the crosslinking reaction.

Figure 3.7. Lap shear adhesion strengths on porcine skin substrates. Ser-PEU: Poly(Ser-

co-Leu-co-PPG); CA-PEU: Poly(CA-Ser-co-Leu-co-PPG); CA:IO4-: Poly(CA-Ser-co-

Leu-co-PPG) crosslinked with Bu4N(IO4). * indicates p < 0.05 for 30 min and 4 h curing

of each sample. ** indicates p < 0.05 for 30 min samples and *** indicates p < 0.05 for 4

h samples. (n ≥ 5).

For the curing time of 4 h, it is observed that the adhesion strength drops as the

- - CA:IO4 ratio increases. The CA:IO4 ratio of 10:1 has the highest adhesion strength of ~

10.6 ± 2.1 kPa which is superior to the adhesion strength of the fibrin glue. The ratios of

10:2 and 10:3 exhibit similar adhesion strengths of ~ 8.7 ± 2.7 kPa and ~ 8.4 ± 3.0 kPa 96 respectively and slightly higher than the fibrin glue. The ratio 10:4 shows adhesion

strength of 6.5 ± 1.2 kPa which is comparable to the fibrin glue after 4 h of curing. The trend observed at 4 h of crosslinking is contrary to that for 30 min. For 4 h of curing at higher ratios, majority of the catechol groups crosslink rapidly while very few are available for surface bonding decreasing the overall adhesion strength. Therefore, to obtain maximum adhesion strength it is necessary to optimize the CA:crosslinker ratio and the curing time.

3.4.4. In vitro cell viability studies. The cell viabilities of NIH 3T3 fibroblast cells on

blank glass coverslips, poly(Ser-co-Leu-co-PPG) (Ser-PEU), poly(CA-Ser-co-Leu-co-

PPG) (CA-PEU) and periodate crosslinked poly(CA-Ser-co-Leu-co-PPG) (xlinked PEU) are shown in Figure 3.8 (A), (B), (C) and (D), respectively. Figure 3.8(E) shows a comparison of normalized %cell viability of fibroblasts on different substrates. The normalized %cell viabilities for all the substrates are summarized in Table 3.2. The fact that cell viability of our crosslinked polymer is equivalent to glass (control), proves that

PPG containing catechol functionalized PEUs and their periodate crosslinked counterparts are non-toxic to cells and are ideal candidates for biomaterial applications.

Table 3.2. Normalized %cell viabilities of fibroblast cells on PEU substrates.

Crosslinked Substrate Glass Ser-PEU CA-PEU CA-PEU

Normalized Cell 100.0 ± 0.3 95.6 ± 3.3 98.6 ± 7.1 99.4 ± 2.9 Viability (%)

97

Figure 3.8. NIH 3T3 fibroblast cell viability on PEU films. Cells stained green are live

cells and the cells stained red are dead. (A) Blank glass substrate - Control, (B) Poly(Ser-

co-Leu-co-PPG) (Ser-PEU) – Control, (C) Poly(CA-Ser-co-Leu-co-PPG) (CA-PEU) and

(D) Poly(CA-Ser-co-Leu-co-PPG) crosslinked with Bu4N(IO4) (xlinked CA-PEU). Scale

bar: 1mm. (E) Normalized %Cell Viability of NIH 3T3 cells on PEU films. Cell viability

was calculated from a total of 10 images and 3 replicates of each sample were used. No

significant difference in cell viability was observed among PEU samples (p > 0.05) and

the glass control.

3.5. CONCLUSION

Catechol functionalized poly(ester urea) copolymers that incorporate poly(propylene

glycol) units were synthesized. Most of the biomimetic adhesives are either water soluble

or have solubility in high boiling solvents. High boiling solvents are detrimental to tissue

while it is difficult to limit the application of water soluble adhesives in the desired

region because water tends to spread. A clinically relevant improvement in our

copolymer is the ethanol solubility which is relevant for tissue applications. About 20 98 mol% PPG units impart solubility in ethanol making this copolymer clinically relevant.

The copolymer exhibits impressive adhesion strength on wet porcine skin. A balance

between the catechol:crosslinker ratio and the curing time is necessary to achieve high

adhesion strength. The catechol:crosslinker ratio of 10:1 with a curing time of 4 h show

superior performance ~ 10.6 ± 2.1 kPa compared to commercial fibrin glue which is ~

4.8 ± 1.4 kPa. The catechol functionalized copolymers show in vitro normalized cell viability of 98.6 ± 7.1% and their crosslinked form shows cell viability of 99.4 ± 2.9% demonstrating that the catechol functionalized poly(ester urea)s and their crosslinked counterparts are non-toxic to cells.

99

CHAPTER IV

CADDISFLY INSPIRED PHOSPHORYLATED POLY(ESTER UREA)-BASED

DEGRADABLE BONE ADHESIVES

V Bhagat, E O’Brien, J Zhou, ML Becker, Biomacromolecules, 2016, 17(9), 3016-3024

4.1. OUTLINE

Bone and tissue adhesives are essential in surgeries for wound healing, hemostasis,

tissue reconstruction and drug delivery. However, there are very few degradable

materials with high adhesion strengths that degrade into bioresorbable byproducts.

Caddisfly adhesive silk is interesting due to the presence of phosphoserines which are thought to afford adhesive properties. In this work phosphoserine-valine poly(ester urea) copolymers with 2% and 5% phosphoserine content were synthesized to mimic caddisfly adhesive silk. Significantly the materials are ethanol soluble and water insoluble making them clinically relevant. Their physical properties were quantified and the adhesion properties were studied on aluminum and bovine bone substrates before and after crosslinking with Ca2+ ions. The adhesive strength of the phosphorylated copolymer on a

bone substrate after crosslinking with Ca2+ was 439 ± 203 kPa, comparable to

commercially available PMMA bone cement (530 ± 133 kPa).

100 4.2. INTRODUCTION

Bone and tissue adhesives are essential in surgeries for their assistance in wound

healing, hemostasis, tissue reconstruction and drug delivery.270 Classical structural

alternatives in bone surgeries include use of metallic plates, pins and screws as support

medium. These options while safe and serviceable suffer from aseptic loosening, poor

anchorage to the bone, cause irritation to the neighboring soft tissue and discomfort to the

patient and also need to be replaced or removed after the bone regeneration. Poly(methyl

methacrylate) (PMMA) bone cement is also commonly used in bone surgeries however,

it is not adhesive in nature, lacks chemical interaction, cause significant heat generation

and shrinkage. Attempts to use synthetic glues like cyanoacrylate, polyurethane, epoxy

resin and calcium or magnesium phosphate ceramic bone cement as bone adhesives have

failed either due to lack of interaction with the bone surface or poor strength.11

Alternative, bone adhesives with degradable properties and high adhesion strengths are

currently an attractive clinical target.

Phosphate based compounds have been used as adhesion promoters for decades in underwater , dental applications, bone implants, fillers and metal substrates.283-288

Several studies have demonstrated adhesion or bonding strength and osteoconductive

potential of phosphate functionalized polymeric bone grafts which show significant

improvement in bone bonding and healing.253, 285, 289-296 Recently Breucker et.al.

developed a facile route to phosphorus functionalized polyurethane aqueous dispersions.

Quartz crystal microbalance (QCM) studies showed enhanced affinity of the dispersions

towards hydroxyapatite and stainless steel surfaces suggesting their application as 101 potential bone adhesives.297 Clearly, phosphate containing polypeptides or polymers

demonstrate improved adhesion behavior on hydroxyapatite or bone surface and show

promise for next generation bone adhesives.

Marine organisms like mussels, barnacles, , sandcastle worms and caddisflies

have evolved to synthesize their own underwater adhesive with strong bonding

characteristics.298 Studies on caddisfly adhesive silk have confirmed the presence of phosphoserine (pSer) residues in the form of (pSX)n motifs where pS is phosphoserine

and X is usually a more hydrophobic amino acid like valine or isoleucine. Elemental

analysis also showed the presence of divalent cations like Ca2+ and Mg2+ which undergo

strong electrostatic interaction with the phosphate groups in pSer to impart strength to the

fibers. The adhesive nature of caddisflies is attributed to the heavily phosphorylated

regions in the adhesive filament.299-303 Recent studies by Wang et.al. describe an

alternative model of caddisfly silk adhesion consisting of a crosslinked dityrosine

peripheral layer and an evenly distributed PEVK like protein throughout the fiber core.

The dityrosines crosslinked in the presence of reactive oxygen species were thought to

impart strength and adhesion to the fibers.304, 305 However, the role of phosphate groups

as adhesion promoters in caddisfly silk was not completely ruled out. In this work,

inspired from the caddisfly adhesive silk, a phosphate functionalized poly(ester urea) was

created as a bone adhesive.

Poly(ester urea)s (PEU)s are a class of polymers well suited for biomaterial

applications because of their attractive properties including degradation into metabolic

components, tunable mechanical properties, wide range of functionality and nontoxicity 102 in vitro and in vivo.238, 281, 306-308 PEU copolymers based on pSer and valine amino acid, as found in caddisfly silk were synthesized and characterized for their physical properties.

The pSer content incorporated in the copolymer was 2% and 5%. The solubility of these polymers in ethanol makes them more clinically relevant and shows strong possibility for further development as bone adhesives. Their adhesion strengths were studied before and after crosslinking with Ca2+ by lap shear adhesion on aluminum substrates and end-to-end

adhesion on bovine bone substrates.

4.3. EXPERIMENTAL SECTION

4.3.1. Synthesis of di-p-toluenesulfonic acid salt of bis(L-Valine)-1,8-octanyl diester

(M1). L-Valine (27.5 gm, 0.234 mol), 1,8-octanediol (15 gm, 0.102 mol), p- toluenesulfonic acid monohydrate (46.57 gm, 0.190 mol) were weighed in a 500 mL round bottom flask along with toluene equipped with a dean stark trap, condenser and a magnetic stir bar. The solution was heated to reflux for 48 h under stirring. After the product cooled down to ambient temperature it was dissolved in hot water and stirred with carbon black to remove any colored impurities. The monomer was further subjected to 3 recrystallizations in water to yield 40 gm (yield = 65%) of white powder as pure

1 product. H NMR (300 MHz, DMSO-d6): 0.95 (dd, J = 6.96 Hz, 10.48 Hz, 12H; CH3),

1.21 (d, J = 45.71 Hz, 8H; CH2), 1.54 (m, 4H, CH2), 2.02-2.20 (m, 2H, CH), 2.31 (d, J =

19.91 Hz, 6H; CH3), 3.90 (d, J = 4.56 Hz, 2H; CH), 3.98-4.40 (m, J = 6.53 Hz, 10.83 Hz,

17.18 Hz, 4H; CH2), 7.29 (dd, J = 7.97 Hz, 110.42 Hz, 4H; CH), 8.23 (br s, 6H, NH).

103

Scheme 4.1. Synthesis of di-p-toluenesulfonic acid salt of bis(L-valine)-1,8-octanyl diester (M1).

4.3.2. Synthesis of 1,8-octanediol-L-valine poly(ester urea) (Poly(1-Val-8)). A solution polymerization was carried out by dissolving the di-p-toluenesulfonic acid salt of bis(L-Valine)-1,8-octanyl-diester (M1) (13.74 g, 20 mmol, 1 eq.) and triethylamine (12.55

ml, 90 mmol, 4.5 eq.) in 40 mL of chloroform in a 500 mL three neck round bottom flask

equipped with a magnetic stir bar and a pressure equalizing addition funnel. Triphosgene

(2.37 g, 8 mmol, 0.4 eq.) was then dissolved in 15 mL of freshly distilled chloroform and

added dropwise under nitrogen at room temperature. The reaction continued for 12 h after

which an additional aliquot of triphosgene (0.45 g, 1.5 mmol, 0.08 eq.) dissolved in 15

mL of chloroform was added to the flask dropwise. The reaction was stirred for an

additional 8 h. The polymer was purified by precipitation in hot water and dried under

1 vacuum. H NMR (500 MHz, DMSO-d6): 0.86 (m, 12H, CH3), 1.26 (s, 8H, CH2), 1.54 (s,

4H, CH2), 1.98 (dd, J = 6.69 Hz, 11.73 Hz, 2H; CH), 3.78-4.21 (m, 6H, CH2, CH), 6.37

13 (d, J = 8.89 Hz, 2H; NH). C NMR (500 MHz, DMSO-d6): 19.39, 25.69, 28.55, 28.97,

30.91, 58.13, 64.52, 157.96 (NH-C=O), 172.84 (O-C=O). ATR-IR (ν): 1690-1650 cm-1

104 (s, C=O urea stretch), 1750-1735 cm-1 (s, C=O ester stretch), 3500-3300 cm-1 (m, N-H

urea stretch). ÐM = 3.2, Mn = 4.3 kDa, Mw = 13.9 kDa.

Scheme 4.2. Synthesis of 1,8-octanediol-L-valine poly(ester urea) (Poly(1-Val-8)).

4.3.3. Synthesis of bis-N-boc-O-benzyl(L-serine)-1,8-octanyl diester (M2). N-boc-O-

benzyl-L-serine (15 gm, 50.79 mmol), 1,8-octanediol (3.09 gm, 21.16 mmol) and DPTS

(1.06 gm, 4.23 mmol) were dissolved in 30 mL of anhydrous CH2Cl2 under N2. The

reaction flask was then immersed in an ice bath and DIC (7.62 mL, 48.67 mmol) was

quickly injected into the flask under N2 at 0 °C. The reaction was left stirring overnight under inert atmosphere while the temperature gradually increased to room temperature.

The reaction solution was filtered to remove the urea crystals and concentrated under vacuum. The oily product was dissolved in CHCl3 and washed with 5% HCl solution

twice and once with DI water, dried over anhydrous Na2SO4, filtered and concentrated in

vacuo. A light yellow colored oil (13.38 gm, yield = 89%) was obtained by silica gel column chromatography with hexane/ethyl acetate (2.65/1, v/v) as eluents. 1H NMR (300

MHz, DMSO-d6): 1.20 (s, 4H, CH2), 1.37 (s, 22H, CH3), 1.50 (m, 4H, CH2), 3.53-3.77

(m, 4H, CH2), 3.85-4.13 (m, 4H, CH2), 4.24 (dd, J = 5.55 Hz, 12.80 Hz, 2H; CH), 4.46 (s,

4H, CH2), 7.09 (d, J = 7.92Hz, 2H; CH), 7.19-7.40 (m, 12H, Ar H).

105 4.3.4. Synthesis of bis-N-boc(L-serine)-1,8-octanyl diester (M3): M2 (11 gm, 15.69 mmol) was dissolved in 30 mL of ethanol in a hydrogenation bottle. 1.3 gm of Pd/C catalyst was added to the polymer solution and subjected to hydrogenolysis under 60 psi

H2 pressure with slight heating (~ 35 °C) for 48 h. The solution was centrifuged at 5000 rpm for 30 min. The supernatant was filtered, concentrated and dried under vacuum to

1 obtain colorless oil as product (8.5 gm, yield = 77%). H NMR (300 MHz, DMSO-d6):

1.20 (s, 4H, CH2), 1.37 (s, 22H, CH3), 1.53 (m, 4H, CH2), 3.65 (t, J = 5.44 Hz, 4; CH2),

3.8-4.19 (m, 4H, CH2), 6.91 (d, J = 8.02 Hz, 2H; CH).

4.3.5. Synthesis of bis-N-boc-O-diphenylphosphate(L-serine)-1,8-octanyl diester

(M4). M3 (9 gm, 3.40 mmol) was dissolved in about 15 mL of pyridine under nitrogen and immersed in an ice bath for 15 min. Diphenylphosphoryl chloride (10.21 gm, 8 mL,

6.72 mmol) was slowly added to the reaction solution under nitrogen. The reaction was

stirred overnight while the reaction temperature gradually increases to room temperature.

About 30 mL of CHCl3 was added to the reaction flask to dissolve the product. The

solution was washed once with 100 mL of DI water, 100 mL of 1M HCl and finally 2x

100 mL of DI water. The organic phase was dried over anhydrous Na2SO4 and further

dried under vacuum until a yellowish solid product was obtained. The yellow solid was

subsequently recrystallized 3 times in isopropanol to obtain white powder as pure product

1 (8.8 gm, yield = 52%). H NMR (300 MHz; DMSO-d6): 1.17 (s, 4H, CH2), 1.35 (s, 22H,

CH3), 1.47 (m, 4H, CH2), 3.99 (t, J = 6.36 Hz, 6H; CH2), 4.42 (d, J = 6.11 Hz, 4H; CH2),

7.23 (dd, J = 7.61 Hz, 15 Hz, 8H; Ar H), 7.4 (t, J = 7.85Hz, 12H; Ar H).

106

Scheme 4.3. Synthesis of dihydrochloride salt of bis-O-diphenylphosphate(L-serine)-1,8- octanyl diester from N-boc-O-benzyl(L-serine) and 1,8-octanediol.

4.3.6. Synthesis of dihydrochloride salt of bis-O-diphenylphosphate(L-serine)-1,8- octanyl diester (M5). 4M HCl/Dioxane solution (40 mL) was added to M4 (8.8 gm) and stirred overnight under nitrogen. The solution was concentrated and freeze dried to remove solvent. The light yellow solid obtained after freeze drying was washed with ethyl acetate and ether and dried to get a white powder as pure product (8.8 gm, yield =

1 100%). H NMR (300 MHz, DMSO-d6): 1.16 (t, J = 7.05 Hz, 4H; CH2), 1.43 (m, 4H,

CH2), 4.02 (t, J = 6.20 Hz, 4H; CH2), 4.68 (s, 6H, CH2), 7.24 (dd, J = 7.94 Hz, 15.66 Hz,

107 8H, Ar H), 7.41(t, J = 7.29 Hz, 12H; Ar H), 8.96 (br s, 6H, NH). ESI-MS (m/z):

Theoretical: 784.74; Calculated: [M+H]+ = 785.2

4.3.7. Synthesis of Poly(serine diphenylphosphate-co-valine) (Poly(SerDPP-co-Val).

The copolymer was synthesized via a solution polymerization similar to P(1-Val-8).

Typically, both the monomers (20 mmol, 1 eq.) and triethylamine (90 mmol, 4.5 eq.)

were dissolved in 40 mL of freshly distilled CHCl3 under N2 in a 500 mL three neck

round bottom flask equipped with a magnetic stirrer. Triphosgene (8 mmol, 0.4 eq.) was

dissolved in 15 mL of distilled CHCl3 and added dropwise to the flask under nitrogen at

room temperature. The reaction continued for 12 h and then additional aliquot of

triphosgene (1.5 mmol, 0.075 eq in 10 mL of chloroform) was added dropwise to the

flask. The reaction continued for another 8 h after which the reaction solution was

transferred to a separatory funnel and precipitated dropwise in hot water. After the water

cools to ambient temperature, the polymer was washed in water at room temperature

overnight and finally dried under vacuum to obtain white polymer as product.

4.3.8. 5% Poly(SerDPP-co-Val). Theoretical monomer feed ratio- pSer/Val (M5/M1) =

1 15/85: H NMR (500 MHz, DMSO-d6): Actual Monomer ratio = 5/95, δ = 0.85 (dd, J =

6.76 Hz, 13.63 Hz, 195H; CH3), 1.26 (d, J = 11.12 Hz, 154H; CH2), 1.54 (m, 72H, CH2),

1.98 (dd, J = 6.56 Hz, 12.58 Hz, 31H; CH), 3.93-4.14 (m, 137H, CH2, CH), 6.37 (d, J =

8.91 Hz, 38H; NH), 7.06-7.18 (m, 26H, Ar H), 7.19-7.34 (m, 12H, Ar H). 13C NMR (500

MHz, DMSO-d6): 19.4, 25.69, 28.55, 30.90, 58.15, 64.53, 120.34, 123.08, 129.31, 157.97

31 (NH-C=O), 172.85 (O-C=O). P NMR (500 MHz, DMSO-d6, 85% H3PO4 external

108 reference): 11.58 ppm. ATR-IR (ν): 1690-1650 cm-1 (s, C=O urea stretch), 1750-1735

cm-1 (s, C=O ester stretch), 3500-3300 cm-1 (m, N-H urea stretch), ~ 1040 cm-1 (w, P=O

-1 stretch), ~ 675 cm (w, P-O stretch). ÐM = 1.4, Mn = 7.7 kDa, Mw = 10.5 kDa.

4.3.9. 2% Poly(SerDPP-co-Val). Theoretical monomer feed ratio- pSer/Val (M5/M1) =

10/90: 1H NMR spectra resonances are similar to 5% poly(SerDPP-co-Val) with different integration values. Actual Monomer ratio = 2/98. 13C and 31P NMR spectra and ATR-IR

spectra corresponds with that of 5% poly(SerDPP-co-Val). ÐM = 1.4, Mn = 6.5 kDa, Mw

= 9.3 kDa.

4.3.10. Synthesis of Poly(phospho serine-co-valine) (Poly(pSer-co-Val). The diphenyl protecting groups were deprotected by hydrogenolysis in the presence of PtO2 catalyst to

synthesize phosphate functionalized copolymers as mimics of caddisfly silk. Typically,

poly(SerDPP-co-Val) (1.00 g) was dissolved in 50 mL CHCl3: 40% TFA/AcOH (4:1)

solution. PtO2 (1.1 eq. of SerDPP) as a catalyst was added to the polymer solution in a

hydrogenation bomb reactor and the reaction was carried out in the presence of H2 (60

psi) at room temperature for 2 h. The solution was filtered and concentrated under

vacuum. The pure polymer was obtained as white solid after precipitation in hot water

and dried under vacuum.

1 4.3.11. 5% Poly(pSer-co-Val). 0.9 g, yield = 90%. H NMR (500 MHz, DMSO-d6): 0.85

(dd, J = 6.76 Hz, 13.61 Hz, 194H; CH3), 1.26 (s, 155H, CH2), 1.54 (m, 72H, CH2), 1.98

(td, J = 6.72 Hz, 13.39 Hz, 35H; CH), 3.87-4.18 (m, 102H, CH2, CH), 4.37 (t, J = 6.48

13 Hz, 2H; CH2), 6.37 (d, J = 8.91 Hz, 26H, NH). C NMR (500 MHz, DMSO-d6): 19.4,

109 25.69, 28.55, 30.90, 58.15, 64.53, 157.97 (NH-C=O), 172.85 (O-C=O). 31P NMR (500

- MHz, DMSO-d6, 85% H3PO4 external reference): -1.20 ppm. ATR-IR (ν): 1690-1650 cm

1 (s, C=O urea stretch), 1750-1735 cm-1 (s, C=O ester stretch), 3500-3300 cm-1 (m, N- H

urea stretch), ~ 1040 cm-1 (w, P=O stretch), ~ 710 cm-1 (w, P-O stretch).

4.3.12. 2% Poly(pSer-co-Val). 1 g, yield = 100%. 1H NMR spectra resonances are

similar to 5% poly(pSer-co-Val) with different integration values. 13C and 31P NMR

spectra and ATR-IR spectra corresponds with that of 5% poly(pSer-co-Val).

4.3.13. Surface energy measurements. Contact angles of five different liquids were

measured to calculate surface energy for each of the polymers using the Owens/Wendt

method. Liquids of known surface tension (water, glycerol, ethylene glycol, propylene

glycol and formamide) were used for the contact angle measurements.281 Polymer thin

films were spun coat on ozone treated silicon wafers using 1% (w/v) solution of the

polymers in ethanol at 2000 rpm for 1 min. The samples were dried at 70 °C under

vacuum overnight. Measurements (n = 3) were performed at room temperature for each

sample and standard deviation of the mean was calculated from independent

measurements.

4.3.14. Lap shear adhesion test on aluminum substrate. Lap shear adhesion tests were

performed on aluminum substrates according to ASTM D1002 standard. Aluminum

adherends 1.6 mm in thickness were cut into rectangular substrates 75 mm long x 12.5

mm wide with a 6.5 mm diameter hole drilled 12.5 mm from one end into each adherend.

For adhesion tests polymer solution in ethanol (30 μL, 300 mg in 1 mL ethanol) was

110 applied to one end of the adherend. Another adherend was placed over it in a lap shear

configuration with an overlap area of 1.56 cm2. When crosslinkers were utilized, polymer solution (25 μL, 300 mg in 1 mL ethanol) and crosslinker solution (5 μL, 0.3 eq. per pSer group in 1 mL acetone) were mixed and applied to one end of the adherend. The adherends were pressed together and allowed to cure for 1 h at room temperature and 24 h at 75 °C and 1 h at room temperature before testing. Lap shear adhesion was performed using an Instron 5567 instrument with a load cell of 1000 N. The adherends were pulled apart at a speed of 1.3 mm/min until failure occurred. The adhesion strength (Pa) is obtained by dividing the maximum load at failure (N) by the area of adhesion (m2).

4.3.15. End-to-End adhesion test on bovine bone substrate. To study the adhesion

characteristics of the poly(pSer-co-Val) we performed end-to-end adhesion experiments on cortical bovine bone. Bovine bone samples were obtained from a local grocery store.

The bones were cut into rectangular pieces of ~ 2 cm long x 0.6 cm width x 0.4 cm thick on a bone saw. Subsequently, the test ends of the bone samples were sanded using a 320 grit sandpaper (3M Pro Grade Precision, X-fine) and the bones were kept in PBS solution

1 h prior to adhesive application. The polymer solution (30 μL, 300 mg in 1 mL ethanol) was applied on one end of the bone and the second bone was placed end-to-end over it.

For crosslinked specimens, polymer solution (25 μL, 300 mg in 1 mL ethanol) and crosslinker solution (5 μL, 0.3 eq. per pSer group in 1 mL acetone) was applied to one bone and the second bone was placed in an end-to-end configuration. The bone samples were clipped together and wrapped in PBS soaked gauze for 24 h. The samples were then placed in an incubator at 37 °C, 95% humidity and 5% CO2 for 2 h before adhesion tests. 111 Adhesion tests were performed on a Texture Analyzer (TA.XT.Plus) equipped with a 5 kg load cell. Bone samples were pulled apart at a speed of 1.3 mm/min until failure occurred. The adhesion strength (Pa) is obtained by dividing the maximum load at failure

(N) by the area of adhesion (m2).

4.3.16. Cell studies

4.3.16.1. In vitro cell viability study. Cell viability and spreading of MC3T3 cells was studied on the phosphate functionalized polymers at day 1 and day 3 respectively after seeding. Polymer films were spun coat from a 3% (w/v) solution in ethanol at 2000 rpm for 1 min on 12 mm glass coverslips. The films were dried overnight at 80 °C under vacuum and carefully transferred to 12 well plates. Cells were rinsed with PBS prior to detaching with 1 mL of 0.05% trypsin/EDTA solution at 37 °C, 95% humidity, 5% CO2 for 5 min. The trypsin/EDTA solution was deactivated by adding 5 mL of media and cells were collected by centrifugation at 4 °C, 3000 rpm for 1 min. The media/trypsin solution was aspirated without disturbing the cell pellet and cells were resuspended in 5 mL fresh media. The cell density was counted using hemocytometer with trypan blue staining.

Cells were seeded at a density of 6000 cells/cm2. The well plates were mildly agitated to ensure even distribution of cells over the samples before incubation.

Cell viability was assessed by Live/Dead assay (Life Technologies). About 5 μL of

Calcein AM (4 mM) and 10 μL of ethidium homodimer were added to 10 mL of DPBS to prepare the staining solution. Media was aspirated from all samples and rinsed with

DPBS prior to adding 0.5 mL of staining solution to each well. The well plates were

112 covered in aluminum foil and incubated at 37 °C for 10 min before imaging. Images were

taken on an Olympus fluorescence microscope equipped with a Hamamatsu orca R2CCD

camera, FITC and TRITC filters at 4X magnification using CellSENS software. Cells

stained green were considered live and the cells stained red were considered dead. Cells

were counted in ImageJ software using cell counter plugin.

4.3.16.2. Cell spreading assay. For the spreading studies, cells were prefixed in 1

mL media and 1 mL 3.7% paraformaldehyde (PFA) in CS buffer solution for 5 min at 37

°C on a dry block. After aspiration, the cells were fixed in 2 mL 3.7% PFA solution in CS

buffer for 5 min at 37 °C. The samples were then rinsed 3 times with 2 mL CS buffer

followed by addition of 1.5 mL of Triton X-100 in CS buffer (0.5% v/v) in each well to permeabilize the cells for 10 min at 37 °C. The samples were rinsed 3 times with CS buffer. 2 mL of freshly prepared 0.1 wt% NaBH4 in CS buffer was then added to each well for 10 min at room temperature to quench the aldehyde fluorescence, followed by aspiration and incubation in 5% donkey serum for 20 min at room temperature to block the non-specific binding. After aspiration the samples were incubated in 300 μL vinculin primary antibody Mouse in CS buffer (v/v = 1:200) at 4 °C overnight. The samples were

then rinsed 3 times with 1% donkey serum and stained with 50 μL rhodamine phalloidin

(v/v = 1:40) and Alexa Fluor 488 secondary antibody Mouse (v/v = 1:200) solution on

wax paper for 1 h at room temperature in dark. After washing the samples 3 times with

CS buffer, the nuclei were stained using DAPI in CS buffer (6 μL/10 mL) for 20 min at

room temperature in dark. The samples were rinsed 3 times with CS buffer to remove

excess staining and viewed under Olympus fluorescence microscope equipped with a 113 Hamamatsu orca R2CCD camera with FITC, TRITC and DAPI filters at 20X

magnification.

4.4. RESULTS AND DISCUSSION

4.4.1. Synthesis and characterization of polymers. The 1H NMR spectra of

phosphoserine monomer is shown in Figure 4.1. The aromatic peaks from the benzyl

protecting groups between ~ 7.2 – 7.4 ppm (Figure 4.1(a)) disappear after deprotection

(Figure 4.1(b)). Phosphate functionalization was confirmed by the presence of aromatic

peaks between ~ 7.0 – 7.5 ppm corresponding to the phenyl protecting groups (Figure

4.1(c)). Disappearance of the peak ~ 1.4 ppm and appearance of a new peak ~ 8.9 ppm

confirmed the successful deprotection of Boc groups (Figure 4.1(d)). Scheme 4.1

represents the synthesis of Poly(pSer-co-Val). The first step of polymer synthesis involves solution polymerization of di-p-toluenesulfonic acid salt of bis(L-Valine)-1,8-

octanyl diester (1-Val-8) monomer and a diphenyl protected phosphoserine monomer, to

produce poly(serine diphenylphosphate-co-valine) (Poly(SerDPP-co-Val)). The second step involves hydrogenolysis to deprotect phenyl groups on poly(SerDPP-co-Val) to obtain the phosphate functionalized copolymer, poly(phosphoserine-co-valine)

(poly(pSer-co-Val)).

114

Figure 4.1. 1H NMR spectra of phosphoserine monomer synthesis. (a) bis-N-boc-O-

benzyl(L-serine)-1,8-octanyl diester (M2), (b) bis-N-boc(L-serine)-1,8-octanyl diester

(M3), (c) bis-N-boc-O-diphenylphosphate(L-serine)-1,8-octanyl diester (M4), (d) dihydrochloride salt of bis-O-diphenylphosphate(L-serine)-1,8-octanyl diester (M5).

In this study, copolymers with 2% and 5% pSer content were synthesized and noted as

2% poly(pSer-co-Val) and 5% poly(pSer-co-Val), respectively. The actual amount of serine monomer incorporated into the polymer was found to be less than the feed amount as measured by 1H NMR spectroscopy. A plausible justification is the presence of four bulky phenyl groups on the monomer. This tends to lower the reactivity of the serine monomer resulting in low incorporation of serine content. From the 1H NMR spectra of

5% poly(pSer-co-Val), the phenyl protecting groups show peaks in the aromatic region δ

115 = 7.06-7.34 ppm. The successful deprotection of these phenyl groups after

hydrogenolysis was confirmed by the loss of aromatic peaks. The proton environments

on the methylene attached directly to the protected phosphate group have a resonance at δ

~ 4.0 ppm which shifts slightly downfield to δ ~ 4.37 ppm after deprotection (Figure

4.2(a), (b)).

Scheme 4.4. Synthesis of phosphate functionalized PEU: (a) Copolymerization of di-p-

toluenesulfonic acid salt of bis(L-valine)-1,8-octanyl-diester (M1) and diphenylphosphoserine monomer (M5) in the presence of triphosgene to obtain

Poly(SerDPP-co-Val), (b) Deprotection of phenyl protecting groups by hydrogenolysis to obtain Poly(pSer-co-Val).

The deprotection of diphenyl groups was also confirmed from 13C NMR spectra. The

aromatic peaks at δ ~ 120-130 ppm in 5% poly(SerDPP-co-Val) disappear after deprotection in 5% poly(pSer-co-Val) (Figure 5.19). 31P NMR spectra also proves

complete deprotection of the phenyl groups. 5% poly(SerDPP-co-Val) has a

116 characteristic phosphorus peak at δ = -11.58 ppm which shifts to δ = -1.20 ppm after deprotection (Figure 4.2(c), (d)).

Figure 4.2. 1H NMR spectra of (a) 5% Poly(SerDPP-co-Val), (b) 5% Poly(pSer-co-Val);

31P NMR spectra of (c) 5% Poly(SerDPP-co-Val), (d) 5% Poly(pSer-co-Val), referenced to 85% H3PO4 as external standard. Inset shows magnification from δ ~ 7.1-7.3 ppm and

δ ~ 4.0 – 4.5 ppm; characteristic peaks of the diphenyl protecting groups disappear after

deprotection. A triplet at δ ~ 4.37 ppm is characteristic of the proton environment on the

methylene group attached to the deprotected phosphate group, which is not prominent

before deprotection.

The attenuated total reflection infrared spectroscopy (ATR-IR), show the presence and

conservation of phosphate groups before and after deprotection of the diphenyl groups.

The P-O-H bond has a characteristic IR peak in the range of 700–500 cm-1 while P=O

bond has a characteristic peak between 1200–1100 cm-1. The P-O characteristic peak for poly(SerDPP-co-Val) is seen around ~ 675 cm-1 which shows a slight shift to ~ 710 cm-1

after deprotection. The P=O bond shows a characteristic peak ~ 1040 cm-1 before and

117 after deprotection. None of these two characteristic phosphate stretching peaks are prominent in the 1-Val-8 polymer (poly(1-Val-8)) (Figure 4.3). The 1H NMR spectra for

2% copolymer resembles that of the 5% copolymer except the integration values are different (Figure 5.15, Figure 5.16). The 13C and 31P NMR spectra for the 2% copolymers also have similar δ (ppm) as that of 5% copolymers.

Figure 4.3. ATR-IR spectra of Poly(1-Val-8), Poly(SerDPP-co-Val) and Poly(pSer-co-

Val). The peaks corresponding to the P=O (1040 cm-1) and P-O stretching (710 - 675 cm-

1) are highlighted and labelled.

PEUs were characterized to determine their physical properties like decomposition temperature (Td) by TGA, glass transition temperature (Tg) by DSC, number average molecular weight (Mn), weight average molecular weight (Mw) and polydispersity (ÐM)

118 by SEC. Table 4.1 summarizes the physical properties of the polymers. The molecular masses of poly(pSer-co-Val) were not determined to prevent blocking of the SEC columns because of the adhesive nature of the polymer. The hydrogenolysis reaction in the presence of PtO2 catalyst is highly selective and site specific so we did not expect polymer degradation to occur during this reaction.

Table 4.1. Physical properties of the poly(ester urea)s.

Surface a) Mn Mw b) Polymer Tg (°C) Td (°C) ÐM Energy (kDa) (kDa) 2 c) γP (mJ/m )

Poly(1-Val-8) 19 285 4.3 13.9 3.2 29.0 ± 0.6

2% Poly(SerDPP-co-Val) 8 266 6.5 9.3 1.4 32.4 ± 1.0

5% Poly(SerDPP-co-Val) 5 269 7.7 10.5 1.4 37.0 ± 1.7

2% Poly(pSer-co-Val) 2 175 \ \ \ 35.8 ± 0.7

5% Poly(pSer-co-Val) 3 161 \ \ \ 33.5 ± 1.5

a) Tg determined by DSC; b) ÐM determined from SEC after purification by precipitation in water; c) Surface energy determined by The Owens/Wendt method from contact angle of 5 different liquids on the polymer film.

Poly(pSer-co-Val)s show slightly lower glass transition temperatures than poly(SerDPP-co-Val). The presence of the bulky phenyl groups on the protected polymer makes the chains stiff and hinders their movement while after deprotection the chains become more flexible and flow freely showing a slight drop in their respective Tg (Figure

4.4(a)). Both the protected and deprotected polymers possess decomposition temperatures above 150 °C (Figure 4.4(b)). The molecular masses of all the polymers are comparable which avoids deviation in adhesion strength due to molecular mass effects. The

119 molecular mass distribution (ÐM) is slightly lower than what is expected of a step growth

polymerization reaction for poly(SerDPP-co-Val) because during purification the polymer undergoes fractionation narrowing the ÐM.

Figure 4.4. (a) DSC curves to determine the glass transition temperatures (Tg). Tg of the

phosphate functionalized polymers do not show significant change after deprotection

because of the low functionality on the polymer backbone chains. (b) Thermogravimetric

analysis curves to determine the decomposition temperature (Td) of the polymers. A

slight decrease in the decomposition temperature is observed after deprotection of the

diphenyl groups.

Surface energies of these polymers were determined from contact angles of 5 different liquids of known surface energies to calculate the polar and dispersive components of the polymer surface energy by the following equation:

(1+ ) . ……………………………………………….(1) = + 𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐 𝛾𝛾 𝑝𝑝 𝐿𝐿𝐿𝐿 𝐿𝐿 𝐿𝐿 𝛾𝛾 𝑝𝑝 2 𝑑𝑑 𝑆𝑆 𝑑𝑑 √𝛾𝛾 𝑑𝑑 √𝛾𝛾 √ 𝛾𝛾𝐿𝐿 𝑆𝑆 √𝛾𝛾𝐿𝐿

120 Where, – Contact angle between polymer and the liquid; – Liquid surface

𝐿𝐿𝐿𝐿 𝐿𝐿 tension; 𝜃𝜃 – Dispersive component of liquid surface tension; – Polar𝛾𝛾 component of

𝑑𝑑 𝑝𝑝 liquid surface𝛾𝛾𝐿𝐿 tension; – Polar component of the polymer surface𝛾𝛾 𝐿𝐿 energy; –

𝑝𝑝 𝑑𝑑 𝛾𝛾 𝛾𝛾 𝑆𝑆 Dispersive component of𝑆𝑆 polymer surface energy. The total surface energy of the

polymer (mJ/m2) is the sum of the polar and dispersive components calculated from

𝑃𝑃 Equation 𝛾𝛾(1).

Table 4.2. Polymer Surface Energies.

2% 5% 2% 5% Components Poly(1-Val-8) Poly(SerDPP Poly(SerDPP Poly(pSer- Poly(pSer- -co-Val) -co-Val) co-Val) co-Val)

9.6 ± 0.4 12.6 ± 0.7 13.0 ± 1.2 21.2 ± 0.5 19.4 ± 1.0 𝑝𝑝 𝛾𝛾𝑆𝑆 19.4 ± 0.4 19.8 ± 0.7 24.0 ± 1.2 14.6 ± 0.5 14.1 ± 1.1 𝑑𝑑 𝑆𝑆 𝛾𝛾 29.0 ± 0.6 32.4 ± 1.0 37.0 ± 1.7 35.8 ± 0.7 33.5 ± 1.5

𝛾𝛾𝑃𝑃 Surface energies calculated by Owens-Wendt method

Total surface energies of the polymers along with their polar and dispersive

components are tabulated in Table 4.2. Both the 2% and 5% poly(pSer-co-Val) have polar and dispersive components for surface energy. The presence of polar component is indicative of surface polar groups that play a significant role in surface adhesion. The surface energies of the 2% and 5% polymers are comparable with each other and are higher than poly(1-Val-8). Since the percentage of functional groups on the polymers was small we did not see an appreciable change in surface energies between the 2% and 5%

functionalized polymers.

121 4.4.2. Lap shear adhesion test on aluminum substrate. Lap shear adhesion is a common method used to determine the strength of adhesive materials and their mode of failure. Adhesive strength of the polymer was studied under lap shear configuration at room temperature on aluminum adherends. Lap shear adhesion test was performed on 2% and 5% poly(pSer-co-Val) with and without crosslinking with Ca2+. Poly(1-Val-8) was used as one of the controls to prove the significance of phosphate groups in adhesion.

Commercially available PMMA bone cement was used as another control for comparison with poly(pSer-co-Val). The results of lap shear adhesion test are summarized in Figure

4.5.

Figure 4.5. (a) Lap shear adhesion on aluminum adherends at room temperature.

Adhesion strengths were calculated from average of 10 replicates (n=10) and reported

122 with standard errors. Aluminum adherends show cohesive failure for all samples, (b)

poly(1-Val-8), (c) PMMA bone cement, (d) 2% poly(pSer-co-Val), (e) 5% poly(pSer-co-

Val), (f) 2% poly(pSer-co-Val) crosslinked with 0.3 eq. Ca2+, (g) 5% poly(pSer-co-Val)

crosslinked with 0.3 eq. Ca2+.

The adhesion strength of poly(1-Val-8) was 0.04 ± 0.16 MPa and the PMMA bone cement was 0.04 ± 0.01 MPa. The low adhesion strength for bone cement is expected because PMMA does not have adhesive properties. PMMA bone cements have high modulus and are essentially used as fillers for bone and teeth. Poly(pSer-co-Val) show improved adhesion strength compared to the control samples which could arise from electrostatic or hydrogen bonding interactions between phosphate groups and the aluminum surface. The adhesion strengths of 2% and 5% poly(pSer-co-Val) are 0.92 ±

0.18 MPa and 0.77 ± 0.09 MPa respectively (Figure 4.5). Previous studies on caddisfly silk have shown that crosslinking with Ca2+ impart strength to the fiber and the crystalline

β-sheet structure collapses if the divalent cations are removed suggesting the significance

of Ca2+ in the system301, 302. Our hypothesis is that phosphate groups on the polymer

backbone chain will interact with Ca2+ in the crosslinking agent resulting in physical

crosslinking which will improve cohesive forces within the polymer bulk further

increasing the adhesive strength compared to their uncrosslinked counterparts (Figure

4.6). Adhesion strength is the result of combined effects of both adhesive and cohesive

forces within a system. To test our hypothesis we chose calcium iodide as the source of

Ca2+ for our system.

123

Figure 4.6. Phosphate groups ( ) on Poly(pSer-co-Val) interact with positive charges on the bone surface promoting adhesion to the bone. Ca2+ from calcium iodide (cross-linking

agent) interact with phosphate groups in the bulk of the polymer giving rise to cohesive

forces.

Both 2% and 5% Ca2+ crosslinked poly(pSer-co-Val) showed an increase in the adhesion strengths compared to their uncrosslinked counterparts. The adhesion strengths of crosslinked 2% poly(pSer-co-Val) and crosslinked 5% poly(pSer-co-Val) are 1.17 ±

0.19 MPa and 1.14 ± 0.02 MPa respectively (Figure 4.5). Our results demonstrate the strong affinity of the phosphate functionalized polymers on the aluminum substrates.

Cohesive failure occurs when the adhesive is stuck on both the adherends after failure. If the adhesive cleanly detaches from one adherend and sticks to the other after failure, it is called adhesive failure. The mode of failure is important in determining the commercial potential of an adhesive. For most adhesives, cohesive failure is desirable. Interestingly, all the samples exhibit cohesive failure except bone cement which failed adhesively

(Figure 4.5(b) to (g). This result is not surprising considering the fact that bone cements

124 are not adhesive in nature but only serve as a support medium for the bone. The results of

the adhesion studies on metal suggests that incorporation of even small percentage of

phosphate groups on the polymer backbone show significant improvement in the

adhesion strength on metal substrate. Also, crosslinking of the phosphate groups with

divalent cations further enhances the adhesion strength via electrostatic interaction to

some extent.

4.4.3. End-to-end adhesion test on bovine bone. We also tested the adhesion strengths

of poly(pSer-co-Val) on bovine bone samples to demonstrate its potential application as a

bone adhesive. Bones have an array of positive and negative charges on the surface. The

basic building block of bone is hydroxyapatite with the chemical structure of

Ca5(PO4)3(OH). Figure 4.7(a) summarizes the adhesion strengths of the polymers on

bovine bone. The controls for this study were the same as that for the metal substrate.

Poly(1-Val-8) had absolutely no adhesion to the bone surface. All the samples failed

before performing the tests. Poly(pSer-co-Val) show increased adhesion strengths compared to Poly(1-Val-8) control. 2% poly(pSer-co-Val) showed adhesion strength of

190 ± 70 kPa and 5% poly(pSer-co-Val) showed adhesion strength of 399 ± 101 kPa.

An increase in adhesion strength was observed after adding 0.3 eq. of Ca2+ as

crosslinking agent. The adhesion strength of crosslinked 2% poly(pSer-co-Val) increased to 211 ± 77 kPa while that of crosslinked 5% poly(pSer-co-Val) increased to 439 ± 203 kPa. The increase in adhesion strength of the phosphate functionalized polymer compared to that of the control polymer proves the significance of the presence of phosphate

125 groups. It is also notable that incorporation of a small percentage of phosphate functionality brings an appreciable change in adhesion. The adhesion strength of 5% poly(pSer-co-Val) is comparable to commercially available bone cement (530 ± 133 kPa) which suggests that poly(pSer-co-Val) has strong potential for further development into bone adhesives. All the samples showed cohesive failure except for the bone cement which showed adhesive failure.

Figure 4.7. (a) End-to-end adhesion on bovine bone at room temperature. Adhesion strengths were calculated from average of 3 replicates (n=3) and reported with standard errors, (b) Schematic of end-to-end adhesion on bovine bone sample (Left) and end-to- end adhesion test on texture analyzer (Right), (c) Image showing cohesive failure of 5% poly(pSer-co-Val) crosslinked with 0.3 eq. of Ca2+.

4.4.4. In vitro cell viability and spreading assay. The cell viability of MC3T3 cells was calculated using at least 40 images for each sample and normalized to 2% Poly(pSer-

126 co-Val) (Figure 4.8(a)-(f)). The viability of cells on glass, poly(1-Val-8), 2% and 5% poly(pSer-co-Val), 2% and 5% crosslinked poly(pSer-co-Val) was (84 ± 9)%, (94 ± 8)%,

(100 ± 3)%, (97 ± 5)%, (97 ± 7)% and (98 ± 4)% respectively (Figure 4.9(a)). The low cell viability on glass could be attributed to handling and seeding errors. High cell viabilities (>95%) on the functionalized polymers prove that the phosphate functionality is non-toxic to the cells. The cytoskeletal structure or spreading behavior of MC3T3 cells on all samples was similar (Figure 4.8(g)-(l)). The cells were stained blue (DAPI) for nuclei, green (Alexa Fluor 488) for focal adhesion points and red (rhodamine phalloidin) for actin filaments. The aspect ratio of the cells on all samples were fairly close (~ 2-3), which proves that the functionalized polymers behaved similar to the controls (Figure

4.9(b)).

Figure 4.8. Cell viability ((a)-(f)) and spreading analysis of MC3T3 cells ((g)-(l)) on day

1 and day 3 respectively. (a) and (g) Glass substrate, (b) and (h) poly(1-Val-8), (c) and (i)

2% poly(pSer-co-Val), (d) and (j) 5% poly(pSer-co-Val), (e) and (k) 2% poly(pSer-co-

Val) crosslinked with 0.3 eq. Ca2+, (f) and (l) 5% poly(pSer-co-Val) crosslinked with 0.3 eq. Ca2+. Cell viability on different samples was studied by live-dead assay from atleast

40 images per sample (4x magnification, scale bar ~ 200 μm). Live cells were stained

127 green by calcein AM and dead cells were stained red by ethidium homodimer. For

spreading studies cells were stained with rhodamine phalloidin – actin filaments (red),

alexa fluor 488 secondary antibody – focal adhesion points (green) and dapi – nuclei

(blue). Images were taken at 20x magnification (Scale bar ~ 50 μm) and aspect ratios were calculated from 30 cells per sample.

Figure 4.9. (a) Normalized cell viability of MC3T3 cells on polymers, (b) Comparison of

aspect ratios on different polymers. MC3T3 cells show similar spreading behavior on all

samples.

4.5. CONCLUSION

Phosphate functionalized PEU copolymers were designed and created to mimic the

properties of caddisfly adhesive silk. These copolymers are ethanol soluble which

provides a suitable delivery mechanism and makes them suitable for medical

applications. The copolymers showed maximum adhesion strength of 1.17 ± 0.19 MPa on

metal substrates after crosslinking with Ca2+. The adhesive strengths of copolymers on

128 bone samples were significant (439 ± 203 kPa) and comparable to commercially available PMMA bone cement (530 ± 133 kPa). The phosphate functionalized copolymers demonstrated improved and significant adhesion strength compared to the valine polymer analogs demonstrating that the phosphate groups play a key role in promoting adhesion. In all cases the copolymers showed cohesive failure. The phosphate functionalized copolymers have significant potential as orthopedic adhesives, scaffold materials for spinal cord injury and orthopedic repairs in the presence of growth peptides like OGP or BMP-2. Another added advantage is that PEUs are degradable in vitro and in vivo. Our future efforts will focus on improving the pSer content in the copolymers, studying the effect of varying the pSer:Ca2+ ratio, curing kinetics and adhesive properties on translationally relevant substrates.

129

CHAPTER V

SUMMARY AND FUTURE SCOPE

Sutures, staples and metallic grafts are an integral part of surgery and also the gold standard for wound closure. However, the pain and discomfort caused by these invasive techniques have led to an urgent need for development of tissue adhesives for surgical settings. Biomedical adhesives have the potential to replace invasive surgical closure techniques like sutures, staples and insertion of metallic bone grafts. The history of medical adhesives dates back to as early as the 1700s with a multitude of developments to date. Adhesives like fibrin glue, cyanoacrylate glue and gelatin-resorcinol formaldehyde/glutaraldehyde (GRFG) glue have been involved in a number of studies and clinical trials resulting in their use as commercial adhesives for medical applications.

Fibrin glue is the only glue which can be passed off as a hemostat and sealant as well as an adhesive and finds application in a number of different surgical procedures.

Application of fibrin glue, although deemed biocompatible and degradable can suffer from risk of virus transmission and poor adhesion under wet conditions. Gelatin-resorcin- formaldehyde/glutaraldehyde glue has demonstrated adhesion strengths stronger than fibrin glue but the application of aldehyde containing materials on tissue is potentially toxic. Cyanoacrylate glue and its variants have consistently shown strongest adhesion on wet tissue among the different class of adhesives; however, their toxicity limits their use

130

only to topical applications. Although these adhesives have found use in a medical or

surgical setting, they are being used cautiously or confined to topical applications. Their

shortcomings have driven the need to develop alternate tissue adhesives with improved

adhesion mechanisms and cytocompatibility. Another class of adhesives based on

polysaccharides, semi-synthetic polymers and poly(ethylene glycol) (PEG) backbone was

a promising alternative to the above mentioned adhesives. Polysaccharide or protein

based tissue adhesives usually covalently attach to tissues (by Schiff base formation)

resulting in strong adhesion under dry conditions. The amine, hydroxyl or carboxylic acid

functionalities on these adhesives form covalent bonds with the tissue surface or result in

intermolecular crosslinks via NHS- or carbodiimide chemistry, Schiff base formation,

Michael addition reaction, biaryl formation, imine formation or π-π interaction.

Incorporation of isocyanate groups or photocrosslinkable vinyl groups assists in improving the cohesive strength. The use of crosslinking agents, toxic reaction products

(urea), photoirradiation and photoinitiators are detrimental to the tissue and alternative methods need to be explored for further application.

Biomimetic adhesives are another class of tissue adhesives inspired from examples of adhesion in nature and are rapidly gaining momentum. A range of different organisms and animals like mussels, caddisflies, spiders, geckos, barnacles, tunicates and sandcastle worms demonstrate strong adhesion under dry and/or wet conditions. Our knowledge of underwater adhesion is still quite limited and considerable efforts are being invested to study adhesion in natural systems. Deeper understanding of the interplay of

131 environmental and chemical factors, chemistries and mechanisms of natural adhesion will

open numerous possibilities for further advancement in biomimetic tissue adhesives.

Currently, biomimetic tissue adhesives involve coacervate formation or

functionalization with an adhesive group like DOPA, catechol or phosphates. Although,

these strategies have shown satisfactory adhesion under dry conditions, they often tend to

fail under humid and/or wet conditions. Mussel adhesion is the most commonly studied

adhesive system owing to its strong reversible adhesion underwater. The presence of

polyphenolic proteins rich in L-DOPA in conjunction with adhesive primers, lysine, fibrous proteins and polyphenoloxidase promote strong underwater adhesion in mussels.

Cationic species like Fe3+, co-localized with foot proteins in the mussel plaque, form

bis/tris-DOPA complexes at basic pH. This complexation strategy in addition to enzyme

catalyzed cross-linking improves intermolecular bonding which further strengthens the

adhesion strength of the mussel plaque.

Poly(CA-Tyr-co-Leu) copolymer previously synthesized in our group was only soluble

in potentially toxic solvents like dimethylformamide (DMF) or dioxane. To improve its

solubility in tissue friendly solvents like ethanol, 20 mol% poly(propylene glycol) (PPG)

was incorporated in the backbone chain. The poly(CA-Ser-co-Leu-co-PPG) copolymer demonstrated adhesion strength of ~ 2.5 ± 0.6 kPa on wet porcine skin. The adhesion strength can be further improved by addition of an oxidizing agent like periodate.

Periodates oxidize catechols into semiquinones or quinones catalyzing the formation of di-DOPA crosslinks in the bulk leading to intermolecular cross-linking. Addition of

132 periodate in a ratio (catechol:periodate) of 10:1 resulted in the strongest bonding of ~

10.6 ± 2.1 kPa on wet porcine skin.

In the case of caddisflies, major structural components of their adhesive silk are H-

fibroin and L-fibroin silk covalently linked in a 1:1 ratio through disulfide bonds.

Extensive phosphorylation of the H-fibroins and presence of an approximately

stoichiometric ratio of metals to phosphorus is a peculiar property of the caddisfly silk.

Calcium is the most abundant metal at a molar ratio of 0.88 (calcium:phosphate) in the

native fiber. Divalent and trivalent cations like Ca2+ and Fe3+ present in the caddisfly

adhesive system interact with the phosphate groups and are responsible for stiffness,

tensile strength and energy dissipating qualities of the fibers. The phosphate groups are present in the form of (pSX)n (pS – phosphoserine and X – valine or isoleucine) repeating

units and comprise 30% of the H-fibroin sequence. The electrostatic interaction between

the charged phosphate groups and the substrate drives the underwater adhesion in

caddisfly silk. Taking inspiration from these two systems, we synthesized a mussel

mimetic catechol functionalized poly(CA-Ser-co-Leu-co-PPG) and phosphate functionalized poly(pSer-co-Val) mimic of caddisfly adhesive for application as a tissue adhesive.

AFM and QCM studies have demonstrated the strong binding affinity between phosphate and calcium owing to electrostatic interactions. Since the main component of bone is calcium phosphate/hydroxyapatite, a phosphate functionalized polymer would demonstrate strong adhesion on bone surface. Also, currently available PMMA bone cement lacks adhesion, chemical interaction and causes significant heat generation upon 133 crosslinking. A phosphate functionalized polymer on the other hand has strong

interaction with the bone surface and can also aid in bone regrowth. Poly(pSer-co-Val)

demonstrated adhesion strength of ~ 399 ± 101 kPa. Addition of calcium iodide as a

cross-linking agent showed a slight increase in bonding ~ 439 ± 203 kPa on wet bovine

bone.

Poly(1-Phe-6) has been studied as a potential bone graft material in a sheep model with

superior performance. Our future efforts will be focused on studying the adhesion

between bovine bone and poly(1-Phe-6) using poly(pSer-co-Val) as an adhesive. Post- polymer modification/functionalization strategies are widely used to incorporate functionality or bioactive species on a polymer backbone in the biomaterials field. Click reactions are the most commonly employed reactions to carry out post-polymerization modification. For poly(pSer-co-Val), the low reactivity of the phosphate functionalized monomer results in a loss of reaction stoichiometry. In the future, post-polymerization modification strategies can be employed by using clickable phosphate groups to get the desired functionalization with strong control over the stoichiometry. In this work, the maximum phosphate functionalization achieved by monomer functionalization route was

~ 5% which can be improved considerably via post-polymerization modification using a click reaction. Higher phosphate functionalization should provide more charged sites for electrostatic interactions resulting in higher adhesion strengths on bone substrates. Effect of co-monomer structure on the adhesion properties of the polymer is quite interesting to study. Future efforts will also be focused on studying the effect from a variation of co-

monomers like arginine or isoleucine on the adhesion properties of the polymers. Spinal

134 cord injuries (SCI) usually result in cellular trauma resulting in cell swelling and lysis

leading to an increase in extracellular Ca2+. Increased extracellular glutamate levels cause

an influx of Ca2+ into neurons leading to neuronal apoptosis.309 The phosphate groups in poly(pSer-co-Val) can utilize these excess Ca2+, thus reversing the neuronal calcium

uptake and decreasing calcium dependent neuronal death. For SCI application QCM-D

studies can be used to determine the amount of calcium uptake and calculate an optimum

phosphate concentration to remove excess calcium at the site of injury.

Recent studies have looked at incorporating a range of crosslinking chemistries to

strike a balance between the adhesive and cohesive strengths. In spite of such extensive

research on tissue adhesives, we have been unsuccessful in developing adhesive mimics

capable of rivaling natural systems that would be scalable, non-toxic, biocompatible, easy

to use and degradable. There are still unknowns about the effect of backbone chemistry,

polymer hydrophilicity/hydrophobicity, combination of different amino acids and

charged side groups on the adhesion strengths. It is also necessary to widen the scope of

adhesives to study drug delivery, tissue grafts, wound healing and tissue reconstruction

via addition of peptides. In addition, long term studies and clinical trials are essential

before these adhesives can be realized in medical or surgical applications.

135

BIBLIOGRAPHY

1. Weiser, T. G.; Haynes, A. B.; Molina, G.; Lipsitz, S. R.; Esquivel, M. M.; Uribe- Leitz, T.; Fu, R.; Azad, T.; Chao, T. E.; Berry, W. R.; Gawande, A. A. Estimate of the global volume of surgery in 2012: An assessment supporting improved health outcomes. Lancet 2015, 385 Suppl 2, S11. 2. Weiser, T. G.; Regenbogen, S. E.; Thompson, K. D.; Haynes, A. B.; Lipsitz, S. R.; Berry, W. R.; Gawande, A. A. An estimation of the global volume of surgery: A modelling strategy based on available data. Lancet 2008, 372, 139-144. 3. Lauto, A.; Mawad, D.; Foster, L. J. R. Adhesive biomaterials for tissue reconstruction. J. Chem. Technol. Biotechnol. 2008, 83, 464-472. 4. Schreader, K. J.; Bayer, I. S.; Milner, D. J.; Loth, E.; Jasiuk, I. A polyurethane- based nanocomposite biocompatible bone adhesive. J. Appl. Polym. Sci. 2013, 127, 4974-4982. 5. Wistlich, L.; Rucker, A.; Schamel, M.; Kubler, A. C.; Gbureck, U.; Groll, J. A bone glue with sustained adhesion under wet conditions. Adv. Healthcare Mater. 2017, 6, 1600902. 6. Waite, J. H. Nature's underwater adhesive specialist. Int. J. Adhes. Adhes. 1987, 7, 9-14. 7. Brennan, M. J.; Kilbride, B. F.; Wilker, J. J.; Liu, J. C. A bioinspired elastin-based protein for a cytocompatible underwater adhesive. Biomaterials 2017, 124, 116- 125. 8. Donkerwolcke, M.; Burny, F.; Muster, D. Tissues and bone adhesives - historical aspects. Biomaterials 1998, 19, 1461-1466. 9. Heiss, C.; Kraus, R.; Schluckebier, D.; Stiller, A.-C.; Wenisch, S.; Schnettler, R. Bone adhesives in trauma and orthopedic surgery. European Journal of Trauma 2006, 32, 141-148. 10. Shah, N. V.; Meislin, R. Current state and use of biological adhesives in orthopedic surgery. Orthopedics 2013, 36, 945-956. 11. Farrar, D. F. Bone adhesives for trauma surgery: A review of challenges and developments. Int. J. Adhes. Adhes. 2012, 33, 89-97. 12. Petersen, B.; Barkun, A.; Carpenter, S.; Chotiprasidhi, P.; Chuttani, R.; Silverman, W.; Hussain, N.; Liu, J.; Taitelbaum, G.; Ginsberg, G. G. Tissue adhesives and fibrin glues. Gastrointestinal Endoscopy 2004, 60, 327-333.

136

13. Ryou, M.; Thompson, C. C. Tissue adhesives: A review. Techniques in Gastrointestinal Endoscopy 2006, 8, 33-37. 14. Spotnitz, W. D. Fibrin sealant: The only approved hemostat, sealant, and adhesive-a laboratory and clinical perspective. ISRN Surgery 2014, 2014, 203943. 15. Reece, T. B.; Maxey, T. S.; Kron, I. L. A prospectus on tissue adhesives. The American Journal of Surgery 2001, 182, S40-S44. 16. Lerner, R.; Binur, N. S. Current status of surgical adhesives. J. Surg. Res. 1990, 48, 165-181. 17. Duarte, A. P.; Coelho, J. F.; Bordado, J. C.; Cidade, M. T.; Gil, M. H. Surgical adhesives: Systematic review of the main types and development forecast. Prog. Polym. Sci. 2012, 37, 1031-1050. 18. Ferreira, P.; Gil, H.; Alves, P., An overview in surgical adhesives. In Recent advances in adhesions research; Nova Science Publishers, Inc.: Hauppauge, N.Y, 2013; pp 59-85. 19. Annabi, N.; Tamayol, A.; Shin, S. R.; Ghaemmaghami, A. M.; Peppas, N. A.; Khademhosseini, A. Surgical materials: Current challenges and nano-enabled solutions. Nano Today 2014, 9, 574-589. 20. Peng, H. T.; Shek, P. N. Novel wound sealants: Biomaterials and applications. Expert Rev. Med. Devices 2010, 7, 639-659. 21. Mehdizadeh, M.; Yang, J. Design strategies and applications of tissue bioadhesives. Macromol. Biosci. 2013, 13, 271-288. 22. Modaresifar, K.; Azizian, S.; Hadjizadeh, A. Nano/biomimetic tissue adhesives development: From research to clinical application. Polymer Reviews 2016, 56, 329-361. 23. Chivers, R. A.; Wolowacz, R. G. The strength of adhesive-bonded tissue joints. Int. J. Adhes. Adhes. 1997, 17, 127-132. 24. Stewart, R. J.; Wang, C. S.; Shao, H. Complex coacervates as a foundation for synthetic underwater adhesives. Adv. Colloid Interface Sci. 2011, 167, 85-93. 25. Bre, L. P.; Zheng, Y.; Pego, A. P.; Wang, W. X. Taking tissue adhesives to the future: From traditional synthetic to new biomimetic approaches. Biomater. Sci. 2013, 1, 239-253. 26. Cronkite, E. P.; Lozner, E. L.; Deaver, J. M. Use of thrombin and fibrinogen in skin grafting: Preliminary report. J. Am. Med. Assoc. 1944, 124, 976-978. 27. Alving, B. M.; Weinstein, M. J.; Finlayson, J. S.; Menitove, J. E.; Fratantoni, J. C. Fibrin sealant: Summary of a conference on characteristics and clinical uses. Transfusion (Paris) 1995, 35, 783-790. 28. Martinowitz, U.; Saltz, R. Fibrin sealant. Curr. Opin. Hematol. 1996, 3, 395-402.

137 29. Jackson, M. R. Fibrin sealants in surgical practice: An overview. Am. J. Surg. 2001, 182, 1-7. 30. Kjaergard, H. K.; Weis-Fogh, U. S. Important factors influencing the strength of autologous fibrin glue; the fibrin concentration and reaction time – comparison of strength with commercial fibrin glue. Eur. Surg. Res. 1994, 26, 273-276. 31. Sierra, D. H.; Feldman, D. S.; Saltz, R.; Huang, S. A method to determine shear adhesive strength of fibrin sealants. J. Appl. Biomater. 1992, 3, 147-151. 32. Lin, C.; Ritter, J. A. Effect of synthesis ph on the structure of carbon xerogels. Carbon 1997, 35, 1271-1278. 33. Albes, J. M.; Krettek, C.; Hausen, B.; Rohde, R.; Haverich, A.; Borst, H.-G. Biophysical properties of the gelatin-resorcinformaldehyde/glutaraldehyde adhesive. The Annals of Thoracic Surgery 1993, 56, 910-915. 34. Walker, J. D.; Kratz, J. M.; Basler, C. G.; Meck, L. P.; Stratton, J. R.; Kribbs, S. B.; Crawford, F. A.; Spinale, F. G. Fate of gelatin–resorcinol– formaldehyde/glutaraldehyde adhesive on femoral vessel morphology. J. Surg. Res. 1997, 71, 73-78. 35. Bonchek, L. I.; Braunwald, N. S. Experimental evaluation of a cross-linked gelatin adhesive in gastrointestinal surgery. Ann. Surg. 1967, 165, 420-424. 36. Nomori, M. D. H.; Horio, M. D. H. Gelatin-resorcinol–formaldehyde- glutaraldehyde glue-spread stapler prevents air leakage from the lung. The Annals of Thoracic Surgery 1997, 63, 352-355. 37. Kodama, K.; Doi, O.; Higashiyama, M.; Yokouchi, H. Pneumostatic effect of gelatin-resorcinol formaldehyde-glutaraldehyde glue on thermal injury of the lung: An experimental study on rats. Eur. J. Cardiothorac. Surg. 1997, 11, 333-337. 38. Nomori, H.; Horio, H.; Morinaga, S.; Suemasu, K. Gelatin-resorcinol– formaldehyde-glutaraldehyde glue for sealing pulmonary air leaks during thoracoscopic operation. The Annals of Thoracic Surgery 1999, 67, 212-216. 39. Shiono, M. Surgery for acute aortic dissection using gelatin-resorcin-formalin glue: Perspective from 10 years of follow-up at a single center. J. Artif. Organs 2008, 11, 19-23. 40. Matsuda, S.; Iwata, H.; Se, N.; Ikada, Y. Bioadhesion of gelatin films crosslinked with glutaraldehyde. J. Biomed. Mater. Res. 1999, 45, 20-27. 41. Bae, S. K.; Sung, T.-H.; Kim, J.-D. A soft-tissue gelatin bioadhesive reinforced with a proteinoid. J. Adhes. Sci. Technol. 2002, 16, 361-372. 42. Sung, H.-W.; Huang, D.-M.; Chang, W.-H.; Huang, R.-N.; Hsu, J.-C. Evaluation of gelatin hydrogel crosslinked with various crosslinking agents as bioadhesives: In vitro study. J. Biomed. Mater. Res. 1999, 46, 520-530.

138 43. Berchane, N. S.; Andrews, M. J.; Kerr, S.; Slater, N. K.; Jebrail, F. F. On the mechanical properties of bovine serum albumin (bsa) adhesives. J. Mater. Sci. Mater. Med. 2008, 19, 1831-1838. 44. Leonard, F.; Kulkarni, R. K.; Brandes, G.; Nelson, J.; Cameron, J. J. Synthesis and degradation of poly (alkyl α-cyanoacrylates). J. Appl. Polym. Sci. 1966, 10, 259- 272. 45. Petersen, B.; Barkun, A.; Carpenter, S.; Chotiprasidhi, P.; Chuttani, R.; Silverman, W.; Hussain, N.; Liu, J.; Taitelbaum, G.; Ginsberg, G. G.; Technology Assessment Committee, A. S. f. G. E. Tissue adhesives and fibrin glues. Gastrointestinal Endoscopy 2004, 60, 327-333. 46. Eiferman, R. A.; Snyder, J. W. Antibacterial effect of cyanoacrylate glue. Arch. Ophthalmol. 1983, 101, 958-960. 47. Kull, S.; Martinelli, I.; Briganti, E.; Losi, P.; Spiller, D.; Tonlorenzi, S.; Soldani, G. Glubran2 surgical glue: In vitro evaluation of adhesive and mechanical properties. J. Surg. Res. 2009, 157, 15-21. 48. Hee Park, D.; Bum Kim, S.; Ahn, K.-D.; Yong Kim, E.; Jun Kim, Y.; Keun Han, D. In vitro degradation and cytotoxicity of alkyl 2-cyanoacrylate polymers for application to tissue adhesives. J. Appl. Polym. Sci. 2003, 89, 3272-3278. 49. Mizrahi, B.; Stefanescu, C. F.; Yang, C.; Lawlor, M. W.; Ko, D.; Langer, R.; Kohane, D. S. Elasticity and safety of alkoxyethyl cyanoacrylate tissue adhesives. Acta Biomater. 2011, 7, 3150-3157. 50. Jaffe, H.; Wade, C. W. R.; Hegyeli, A. F.; Rice, R.; Hodge, J. Synthesis and bioevaluation of alkyl 2-cyanoacryloyl glycolates as potential soft tissue adhesives. J. Biomed. Mater. Res. 1986, 20, 205-212. 51. Jaffe, H.; Wade, C. W. R.; Hegyeli, A. F.; Rice, R. M.; Hodge, J. Synthesis and bioevaluation of a rapidly biodegradable tissue adhesive: 1,2‐isopropylidene glyceryl 2‐cyanoacrylate. J. Biomed. Mater. Res. 1986, 20, 213-217. 52. Tseng, Y.-C.; Hyon, S.-H.; Ikada, Y. Modification of synthesis and investigation of properties for 2-cyanoacrylates. Biomaterials 1990, 11, 73-79. 53. Pritykin, L. M.; Lakiza, O. V.; Niazashvili, G. A.; Karmazin, V. B.; Klimentova, N. V.; Mager, K. A.; Tutorskii, I. A.; Vakula, V. L. Effect of the composition of copolymers of 1,1,2-trichlorobutadiene-1,3 with ethyl- and allyl-α-cyanoacrylates on their adhesional properties. Polym. Sci. U.S.S.R. 1991, 33, 930-937. 54. Petrov, C.; Serafimov, B.; Kotzev, D. L. Strength, deformation and relaxation of joints bonded with modified cyanoacrylate adhesives. Int. J. Adhes. Adhes. 1988, 8, 207-210.

139 55. Linden, C. L.; Shalaby, S. W. Performance of modified cyanoacrylate composition as tissue adhesives for soft and hard tissues. J. Biomed. Mater. Res. 1997, 38, 348- 355. 56. Han, M. G.; Kim, S. Controlled degradation of poly(ethyl cyanoacrylate-co-methyl methacrylate) (peca-co-pmma) copolymers. Polymer 2009, 50, 1270-1280. 57. Lim, J. I.; Lee, Y.-K.; Shin, J.-S.; Lim, K.-J. Cyanoacrylate adhesives curable to flexible polymeric materials by poly(l-lactide-co-ε-caprolactone). Mater. Lett. 2010, 64, 2438-2440. 58. Lim, J. I.; Kim, J. H.; Park, H.-K. The adhesive properties of prepolymerized allyl 2-cyanoacrylate/poly l-3,4-dihydroxyphenylalanine for use as bioglue. Mater. Lett. 2012, 81, 251-253. 59. Leggat, P. A.; Smith, D. R.; Kedjarune, U. Surgical applications of cyanoacrylate adhesives: A review of toxicity. ANZ J. Surg. 2007, 77, 209-213. 60. Citrin, P.; Doscher, W.; Wise, L.; Margolis, I. B. Control of needle hole bleeding with ethylcyanoacrylate glue (krazy glue). Journal of Vascular Surgery 1985, 2, 488-490. 61. Barbalinardo, R. J.; Citrin, P.; Franco, C. D.; Hobson, R. W., 2nd. A comparison of isobutyl 2-cyanoacrylate glue, fibrin adhesive, and oxidized regenerated cellulose for control of needle hole bleeding from polytetrafluoroethylene vascular prostheses. Journal of Vascular Surgery 1986, 4, 220-223. 62. Roth, J.; Shtokman, J.; Shamir, M. H.; Nissan, M.; Shchetinkov, L.; Trejo, L. L.; Rochkind, S. Regeneration of the transected rat sciatic nerve after suturing or adhesion with cyanoacrylate glue. J. Neurosurg. 2011, 114, 245-252. 63. Chan, C. C.; Cheong, T. H.; Lee, H. S.; Wang, Y. T.; Poh, S. C. Case of occupational asthma due to glue containing cyanoacrylate. Ann. Acad. Med. Singapore 1994, 23, 731-733. 64. Bruze, M.; Björkner, B.; Lepoittevin, J.-P. Occupational allergic contact dermatitis from ethyl cyanoacrylate. Contact Dermatitis 1995, 32, 156-159. 65. Otani, Y.; Tabata, Y.; Ikada, Y. A new biological glue from gelatin and poly (l- glutamic acid). J. Biomed. Mater. Res. 1996, 31, 157-166. 66. Otani, Y.; Tabata, Y.; Ikada, Y. Adhesion to soft tissues by gelatin-polyanion hydrogels. The Journal of Adhesion 1996, 59, 197-205. 67. Otani, Y.; Tabata, Y.; Ikada, Y. Rapidly curable biological glue composed of gelatin and poly(l-glutamic acid). Biomaterials 1996, 17, 1387-1391. 68. Otani, Y.; Tabata, Y.; Ikada, Y. Effect of additives on gelation and tissue adhesion of gelatin–poly(l-glutamic acid) mixture. Biomaterials 1998, 19, 2167-2173. 69. Iwata, H.; Matsuda, S.; Mitsuhashi, K.; Itoh, E.; Ikada, Y. A novel surgical glue composed of gelatin and n-hydroxysuccinimide activated poly(l-glutamic acid): 140 Part 1. Synthesis of activated poly(l-glutamic acid) and its gelation with gelatin. Biomaterials 1998, 19, 1869-1876. 70. Cohen, B.; Pinkas, O.; Foox, M.; Zilberman, M. Gelatin–alginate novel tissue adhesives and their formulation–strength effects. Acta Biomater. 2013, 9, 9004- 9011. 71. Shefy-Peleg, A.; Foox, M.; Cohen, B.; Zilberman, M. Novel antibiotic-eluting gelatin-alginate soft tissue adhesives for various wound closing applications. Int. J. Polym. Mater. Polym. Biomater. 2014, 63, 699-707. 72. Cohen, B.; Shefy-Peleg, A.; Zilberman, M. Novel gelatin/alginate soft tissue adhesives loaded with drugs for pain management: Structure and properties. J. Biomater. Sci. Polym. Ed. 2014, 25, 224-240. 73. Foox, M.; Raz-Pasteur, A.; Berdicevsky, I.; Krivoy, N.; Zilberman, M. In vitro microbial inhibition, bonding strength, and cellular response to novel gelatin- alginate antibiotic-releasing soft tissue adhesives. Polym. Adv. Technol. 2014, 25, 516-524. 74. Pinkas, O.; Zilberman, M. Effect of hemostatic agents on properties of gelatin– alginate soft tissue adhesives. J. Biomater. Sci. Polym. Ed. 2014, 25, 555-573. 75. Cohen, B.; Panker, M.; Zuckerman, E.; Foox, M.; Zilberman, M. Effect of calcium phosphate-based fillers on the structure and bonding strength of novel gelatin- alginate bioadhesives. J. Biomater. Appl. 2014, 28, 1366-1375. 76. Taguchi, T.; Saito, H.; Uchida, Y.; Sakane, M.; Kobayashi, H.; Kataoka, K.; Tanaka, J. Bonding of soft tissues using a novel tissue adhesive consisting of a citric acid derivative and collagen. Mater. Sci. Eng., C 2004, 24, 775-780. 77. Taguchi, T.; Saito, H.; Aoki, H.; Uchida, Y.; Sakane, M.; Kobayashi, H.; Tanaka, J. Biocompatible high-strength glue consisting of citric acid derivative and collagen. Mater. Sci. Eng., C 2006, 26, 9-13. 78. Aoki, H.; Taguchi, T.; Saito, H.; Kobayashi, H.; Kataoka, K.; Tanaka, J. Rheological evaluation of gelatin gels prepared with a citric acid derivative as a novel cross-linker. Mater. Sci. Eng., C 2004, 24, 787-790. 79. Saito, H.; Taguchi, T.; Kobayashi, H.; Kataoka, K.; Tanaka, J.; Murabayashi, S.; Mitamura, Y. Physicochemical properties of gelatin gels prepared using citric acid derivative. Mater. Sci. Eng., C 2004, 24, 781-785. 80. Taguchi, T.; Saito, H.; Iwasashi, M.; Sakane, M.; Kakinoki, S.; Ochiai, N.; Tanaka, J. Development of a novel glue consisting of naturally-derived biomolecules: Citric acid and human serum albumin. J. Nanosci. Nanotechnol. 2007, 7, 742-747. 81. Taguchi, T.; Saito, H.; Iwasashi, M.; Sakane, M.; Ochiai, N. Biodegradable adhesives composed of human serum albumin and organic acid-based crosslinkers with active ester groups. J. Bioact. Compat. Polym. 2009, 24, 546-559.

141 82. Matsuda, M.; Ueno, M.; Endo, Y.; Inoue, M.; Sasaki, M.; Taguchi, T. Enhanced tissue penetration-induced high bonding strength of a novel tissue adhesive composed of cholesteryl group-modified gelatin and disuccinimidyl tartarate. Colloids Surf., B 2012, 91, 48-56. 83. Yoshizawa, K.; Taguchi, T. Bonding behavior of hydrophobically modified gelatin films on the intestinal surface. J. Bioact. Compat. Polym. 2014, 29, 560-571. 84. Sanders, L.; Stone, R.; Webb, K.; Mefford, T.; Nagatomi, J. Mechanical characterization of a bifunctional tetronic hydrogel adhesive for soft tissues. J. Biomed. Mater. Res., Part A 2015, 103, 861-868. 85. Yuan, L.; Wu, Y.; Fang, J.; Wei, X.; Gu, Q.; El-Hamshary, H.; Al-Deyab, S. S.; Morsi, Y.; Mo, X. Modified alginate and gelatin cross-linked hydrogels for soft tissue adhesive. Artif. Cells, Nanomed., Biotechnol. 2017, 45, 76-83. 86. Lu, H. S. M., Dextran-based polymer tissue adhesive for medical use. In Google Patents: 2016. 87. Mo, X.; Iwata, H.; Ikada, Y. A tissue adhesives evaluated in vitro and in vivo analysis. J. Biomed. Mater. Res., Part A 2010, 94A, 326-332. 88. Tardy, M.; Tiollier, J.; Tayot, J. L. Biocompatible bioresorbable and non-toxic adhesive composition for surgical use. US5618551 A, April 8, 1997. 89. Balakrishnan, B.; Soman, D.; Payanam, U.; Laurent, A.; Labarre, D.; Jayakrishnan, A. A novel injectable tissue adhesive based on oxidized dextran and chitosan. Acta Biomater. 2017, 53, 343-354. 90. Oliva, N.; Shitreet, S.; Abraham, E.; Stanley, B.; Edelman, E. R.; Artzi, N. Natural tissue microenvironmental conditions modulate adhesive material performance. Langmuir 2012, 28, 15402-15409. 91. Nakajima, N.; Sugai, H.; Tsutsumi, S.; Hyon, S.-H. Self-degradable bioadhesive. Key Eng. Mater. 2007, 342-343, 713-716. 92. Araki, M.; Tao, H.; Nakajima, N.; Sugai, H.; Sato, T.; Hyon, S.-H.; Nagayasu, T.; Nakamura, T. Development of new biodegradable hydrogel glue for preventing alveolar air leakage. The Journal of Thoracic and Cardiovascular Surgery 2007, 134, 1241-1248. 93. Naitoh, Y.; Kawauchi, A.; Kamoi, K.; Soh, J.; Okihara, K.; Hyon, S.-H.; Miki, T. Hemostatic effect of new surgical glue in animal partial nephrectomy models. Urology 2013, 81, 1095-1100. 94. Hyon, S.-H.; Nakajima, N.; Sugai, H.; Matsumura, K. Low cytotoxic tissue adhesive based on oxidized dextran and epsilon-poly-l-lysine. J. Biomed. Mater. Res., Part A 2014, 102, 2511-2520.

142 95. Matsumura, K.; Nakajima, N.; Sugai, H.; Hyon, S.-H. Self-degradation of tissue adhesive based on oxidized dextran and poly-l-lysine. Carbohydr. Polym. 2014, 113, 32-38. 96. Takaoka, M.; Nakamura, T.; Sugai, H.; Bentley, A. J.; Nakajima, N.; Fullwood, N. J.; Yokoi, N.; Hyon, S.-H.; Kinoshita, S. Sutureless amniotic membrane transplantation for ocular surface reconstruction with a chemically defined bioadhesive. Biomaterials 2008, 29, 2923-2931. 97. Takaoka, M.; Nakamura, T.; Sugai, H.; Bentley, A. J.; Nakajima, N.; Yokoi, N.; Fullwood, N. J.; Hyon, S.-H.; Kinoshita, S. Novel sutureless keratoplasty with a chemically defined bioadhesive. Invest. Ophthalmol. Vis. Sci. 2009, 50, 2679-2685. 98. Tsujita, H.; Brennan, A. B.; Plummer, C. E.; Nakajima, N.; Hyon, S.-H.; Barrie, K. P.; Sapp, B.; Jackson, D.; Brooks, D. E. An ex vivo model for suture-less amniotic membrane transplantation with a chemically defined bioadhesive. Curr. Eye Res. 2012, 37, 372-380. 99. Kazusa, H.; Nakasa, T.; Shibuya, H.; Ohkawa, S.; Kamei, G.; Adachi, N.; Deie, M.; Nakajima, N.; Hyon, S. H.; Ochi, M. Strong adhesiveness of a new biodegradable hydrogel glue, lydex, for use on articular cartilage. J. Appl. Biomater. Funct. Mater. 2013, 11, e180-186. 100. Yamamoto, T.; Fujibayashi, S.; Nakajima, N.; Sugai, H.; Hyon, S.-H.; Nakamura, T. Biodegradable adhesive (lydex) with hydroxyapatite granules. Key Eng. Mater. 2008, 361-363, 575-578. 101. Kamitani, T.; Masumoto, H.; Kotani, H.; Ikeda, T.; Hyon, S.-H.; Sakata, R. Prevention of retrosternal adhesion by novel biocompatible glue derived from food additives. The Journal of Thoracic and Cardiovascular Surgery 2013, 146, 1232- 1238. 102. Takeda, T.; Shimamoto, T.; Marui, A.; Saito, N.; Uehara, K.; Minakata, K.; Miwa, S.; Nakajima, N.; Ikeda, T.; Hyon, S.-H.; Sakata, R. Topical application of a biodegradable disc with amiodarone for atrial fibrillation. The Annals of Thoracic Surgery 2011, 91, 734-739. 103. Togo, Y.; Takahashi, K.; Saito, K.; Kiso, H.; Huang, B.; Tsukamoto, H.; Hyon, S. H.; Bessho, K. Aldehyded dextran and epsilon -poly(l-lysine) hydrogel as nonviral gene carrier. Stem Cells Int. 2013, 2013, 634379. 104. Okino, H.; Nakayama, Y.; Tanaka, M.; Matsuda, T. In situ hydrogelation of photocurable gelatin and drug release. J. Biomed. Mater. Res. 2002, 59, 233-245. 105. Matsuda, T.; Nakayama, Y.; Tsutsui, N., Photo-curing tissue adhesive. In Google Patents: 2001. 106. Vuocolo, T.; Haddad, R.; Edwards, G. A.; Lyons, R. E.; Liyou, N. E.; Werkmeister, J. A.; Ramshaw, J. A. M.; Elvin, C. M. A highly elastic and adhesive gelatin tissue

143 sealant for gastrointestinal surgery and colon anastomosis. J. Gastrointest. Surg. 2012, 16, 744-752. 107. Ono, K.; Saito, Y.; Yura, H.; Ishikawa, K.; Kurita, A.; Akaike, T.; Ishihara, M. Photocrosslinkable chitosan as a biological adhesive. J. Biomed. Mater. Res. 2000, 49, 289-295. 108. Ishihara, M.; Ono, K.; Saito, Y.; Yura, H.; Hattori, H.; Matsui, T.; Kurita, A. Photocrosslinkable chitosan: An effective adhesive with surgical applications. Int. Congr. Ser. 2001, 1223, 251-257. 109. Ono, K.; Ishihara, M.; Ozeki, Y.; Deguchi, H.; Sato, M.; Saito, Y.; Yura, H.; Sato, M.; Kikuchi, M.; Kurita, A.; Maehara, T. Experimental evaluation of photocrosslinkable chitosan as a biologic adhesive with surgical applications. Surgery 2001, 130, 844-850. 110. Ishihara, M.; Ono, K.; Sato, M.; Nakanishi, K.; Saito, Y.; Yura, H.; Matsui, T.; Hattori, H.; Fujita, M.; Kikuchi, M.; Kurita, A. Acceleration of wound contraction and healing with a photocrosslinkable chitosan hydrogel. Wound Repair Regen. 2001, 9, 513-521. 111. Wistlich, L.; Rücker, A.; Schamel, M.; Kübler, A. C.; Gbureck, U.; Groll, J. A bone glue with sustained adhesion under wet conditions. Adv. Healthcare Mater. 2017, 6, 1600902-n/a. 112. Lee, Y.; Bae, J. W.; Oh, D. H.; Park, K. M.; Chun, Y. W.; Sung, H.-J.; Park, K. D. In situ forming gelatin-based tissue adhesives and their phenolic content-driven properties. J. Mater. Chem. B 2013, 1, 2407-2414. 113. Thirupathi Kumara Raja, S.; Thiruselvi, T.; Sailakshmi, G.; Ganesh, S.; Gnanamani, A. Rejoining of cut wounds by engineered gelatin–keratin glue. Biochim. Biophys. Acta, Gen. Subj. 2013, 1830, 4030-4039. 114. Nie, W.; Yuan, X.; Zhao, J.; Zhou, Y.; Bao, H. Rapidly in situ forming chitosan/ε- polylysine hydrogels for adhesive sealants and hemostatic materials. Carbohydr. Polym. 2013, 96, 342-348. 115. Cho, E.; Lee, J. S.; Webb, K. Formulation and characterization of poloxamine- based hydrogels as tissue sealants. Acta Biomater. 2012, 8, 2223-2232. 116. Oh, D. X.; Kim, S.; Lee, D.; Hwang, D. S. Tunicate-mimetic nanofibrous hydrogel adhesive with improved wet adhesion. Acta Biomater. 2015, 20, 104-112. 117. Bitton, R.; Berglin, M.; Elwing, H.; Colin, C.; Delage, L.; Potin, P.; Bianco-Peled, H. The influence of halide-mediated oxidation on -born adhesives. Macromol. Biosci. 2007, 7, 1280-1289. 118. Bitton, R.; Bianco-Peled, H. Novel biomimetic adhesives based on algae glue. Macromol. Biosci. 2008, 8, 393-400.

144 119. Bitton, R.; Josef, E.; Shimshelashvili, I.; Shapira, K.; Seliktar, D.; Bianco-Peled, H. Phloroglucinol-based biomimetic adhesives for medical applications. Acta Biomater. 2009, 5, 1582-1587. 120. Labroo, V.; Busby, S. J. Transglutaminase cross-linkable polypeptides and methods relating thereto. US5939385 A, August 17, 1999. 121. Lu, D.; Zhang, Y.; Li, T. e.; Li, Y.; Wang, H.; Shen, Z.; Wei, Q.; Lei, Z. The synthesis and tissue adhesiveness of temperature-sensitive hyperbranched poly(amino acid)s with functional side groups. Polym. Chem. 2016, 7, 1963-1970. 122. Li, H.; Niu, R.; Yang, J.; Nie, J.; Yang, D. Photocrosslinkable tissue adhesive based on dextran. Carbohydr. Polym. 2011, 86, 1578-1585. 123. Wang, T.; Nie, J.; Yang, D. Dextran and gelatin based photocrosslinkable tissue adhesive. Carbohydr. Polym. 2012, 90, 1428-1436. 124. Matsuda, T.; Nakajima, N.; Itoh, T.; Takakura, T. Development of a compliant surgical adhesive derived from novel fluorinated hexamethylene diisocyanate. ASAIO Trans. 1989, 35, 381-383. 125. Kobayashi, H.; Hyon, S.-H.; Ikada, Y. Water-curable and biodegradable prepolymers. J. Biomed. Mater. Res. 1991, 25, 1481-1494. 126. Bochyńska, A. I.; Sharifi, S.; van Tienen, T. G.; Buma, P.; Grijpma, D. W. Development of tissue adhesives based on amphiphilic isocyanate-terminated trimethylene carbonate block copolymers. Macromol. Symp. 2013, 334, 40-48. 127. Bochyńska, A. I.; Van Tienen, T. G.; Hannink, G.; Buma, P.; Grijpma, D. W. Development of biodegradable hyper-branched tissue adhesives for the repair of meniscus tears. Acta Biomater. 2016, 32, 1-9. 128. Bochyńska, A. I.; Hannink, G.; Janssen, D.; Buma, P.; Grijpma, D. W. Development of a fast curing tissue adhesive for meniscus tear repair. J. Mater. Sci. Mater. Med. 2016, 28, 1. 129. Bochyńska, A. I.; Hannink, G.; Verhoeven, R.; Grijpma, D. W.; Buma, P. Evaluation of novel biodegradable three-armed- and hyper-branched tissue adhesives in a meniscus explant model. J. Biomed. Mater. Res., Part A 2017, 105, 1405-1411. 130. Sternberg, K.; Rohm, H. W.; Lurtz, C.; Wegmann, J.; Odermatt, E. K.; Behrend, D.; Michalik, D.; Schmitz, K.-P. Development of a biodegradable tissue adhesive based on functionalized 1,2-ethylene glycol bis(dilactic acid). I. J. Biomed. Mater. Res., Part B 2010, 94B, 318-326. 131. Rohm, H. W.; Lurtz, C.; Wegmann, J.; Odermatt, E. K.; Behrend, D.; Schmitz, K.- P.; Sternberg, K. Development of a biodegradable tissue adhesive based on functionalized 1,2-ethylene glycol bis(dilactic acid). Ii. J. Biomed. Mater. Res., Part B 2011, 97B, 66-73.

145 132. Balcioglu, S.; Parlakpinar, H.; Vardi, N.; Denkbas, E. B.; Karaaslan, M. G.; Gulgen, S.; Taslidere, E.; Koytepe, S.; Ates, B. Design of xylose-based semisynthetic polyurethane tissue adhesives with enhanced bioactivity properties. ACS Appl. Mater. Interfaces 2016, 8, 4456-4466. 133. Ates, B.; Koytepe, S.; Karaaslan, M. G.; Balcioglu, S.; Gulgen, S.; Demirbilek, M.; Denkbas, E. B. Chlorogenic acid containing bioinspired polyurethanes: Biodegradable medical adhesive materials. Int. J. Polym. Mater. Polym. Biomater. 2015, 64, 611-619. 134. Lauto, A.; Hook, J.; Doran, M.; Camacho, F.; Poole-Warren, L. A.; Avolio, A.; Foster, L. J. R. Chitosan adhesive for laser tissue repair: In vitro characterization. Lasers Surg. Med. 2005, 36, 193-201. 135. Lauto, A.; Stoodley, M.; Marcel, H.; Avolio, A.; Sarris, M.; McKenzie, G.; Sampson, D. D.; Foster, L. J. R. In vitro and in vivo tissue repair with laser- activated chitosan adhesive. Lasers Surg. Med. 2007, 39, 19-27. 136. Lauto, A.; Foster, L. J.; Avolio, A.; Sampson, D.; Raston, C.; Sarris, M.; McKenzie, G.; Stoodley, M. Sutureless nerve repair with laser-activated chitosan adhesive: A pilot in vivo study. Photomed. Laser Surg. 2008, 26, 227-234. 137. Lauto, A. Integration of extracellular matrix with chitosan adhesive film for sutureless tissue fixation. Lasers Surg. Med. 2009, 41, 366-371. 138. Shahbazi, J.; Marçal, H.; Watson, S.; Wakefield, D.; Sarris, M.; Foster, L. J. R. Sutureless sealing of penetrating corneal wounds using a laser-activated thin film adhesive. Lasers Surg. Med. 2011, 43, 490-498. 139. Barton, M.; Morley, J. W.; Stoodley, M. A.; Ng, K.-S.; Piller, S. C.; Duong, H.; Mawad, D.; Mahns, D. A.; Lauto, A. Laser-activated adhesive films for sutureless median nerve anastomosis. J. Biophotonics 2013, 6, 938-949. 140. Zhang, J.; Chen, D.; Li, Y.; Sun, J. Layer-by-layer assembled highly adhesive microgel films. Polymer 2013, 54, 4220-4226. 141. Cohn, D.; Lando, G. Tailoring lactide/caprolactone co-oligomers as tissue adhesives. Biomaterials 2004, 25, 5875-5884. 142. Strehin, I.; Nahas, Z.; Arora, K.; Nguyen, T.; Elisseeff, J. A versatile ph sensitive chondroitin sulfate–peg tissue adhesive and hydrogel. Biomaterials 2010, 31, 2788- 2797. 143. Simson, J.; Crist, J.; Strehin, I.; Lu, Q.; Elisseeff, J. H. An orthopedic tissue adhesive for targeted delivery of intraoperative biologics. J. Orthop. Res. 2013, 31, 392-400. 144. Simson, J. A.; Strehin, I. A.; Allen, B. W.; Elisseeff, J. H. Bonding and fusion of meniscus fibrocartilage using a novel chondroitin sulfate bone marrow tissue adhesive. Tissue Eng., Part A 2013, 19, 1843-1851.

146 145. Barua, S.; Chattopadhyay, P.; Karak, N. S-triazine-based biocompatible hyperbranched epoxy adhesive with antibacterial attributes for sutureless surgical sealing. J. Mater. Chem. B 2015, 3, 5877-5885. 146. Zhu, J. Bioactive modification of poly(ethylene glycol) hydrogels for tissue engineering. Biomaterials 2010, 31, 4639-4656. 147. Nakayama, Y.; Matsuda, T. Photocurable surgical tissue adhesive glues composed of photoreactive gelatin and poly(ethylene glycol) diacrylate. J. Biomed. Mater. Res. 1999, 48, 511-521. 148. Nakayama, Y.; Sakaguchi, Y.; Masuda, S.; Okano, T.; Matsuda, T. Design of photocurable tissue adhesive consisting of photoreactive gelatin and poly(ethylene glycol) diacrylate: Application for arterial surgery. Jpn J Artif Organs 1997, 26, 225-231. 149. Li, C.; Sajiki, T.; Nakayama, Y.; Fukui, M.; Matsuda, T. Novel visible-light- induced photocurable tissue adhesive composed of multiply styrene-derivatized gelatin and poly(ethylene glycol) diacrylate. J. Biomed. Mater. Res., Part B 2003, 66B, 439-446. 150. Zhang, H.; Zhao, T.; Duffy, P.; Dong, Y.; Annaidh, A. N.; O'Cearbhaill, E.; Wang, W. Hydrolytically degradable hyperbranched peg-polyester adhesive with low swelling and robust mechanical properties. Adv. Healthcare Mater. 2015, 4, 2260- 2268. 151. Jeon, O.; Samorezov, J. E.; Alsberg, E. Single and dual crosslinked oxidized methacrylated alginate/peg hydrogels for bioadhesive applications. Acta Biomater. 2014, 10, 47-55. 152. Zilinski, J. L.; Kao, W. J. Tissue adhesiveness and response of in situ photopolymerizable interpenetrating networks containing methylprednisolone acetate. J. Biomed. Mater. Res., Part A 2004, 68, 392-400. 153. Li, C.; Wang, T.; Hu, L.; Wei, Y.; Liu, J.; Mu, X.; Nie, J.; Yang, D. Photocrosslinkable bioadhesive based on dextran and peg derivatives. Mater. Sci. Eng., C 2014, 35, 300-306. 154. Sawhney, A. S.; Pathak, C. P.; Hubbell, J. A. Bioerodible hydrogels based on photopolymerized poly(ethylene glycol)-co-poly(alpha-hydroxy acid) diacrylate macromers. Macromolecules 1993, 26, 581-587. 155. Sawhney, A. S.; Pathak, C. P.; van Rensburg, J. J.; Dunn, R. C.; Hubbell, J. A. Optimization of photopolymerized bioerodible hydrogel properties for adhesion prevention. J. Biomed. Mater. Res. 1994, 28, 831-838. 156. Hill-West, J. L.; Chowdhury, S. M.; Sawhney, A. S.; Pathak, C. P.; Dunn, R. C.; Hubbell, J. A. Prevention of postoperative adhesions in the rat by in situ photopolymerization of bioresorbable hydrogel barriers. Obstet. Gynecol. 1994, 83, 59-64.

147 157. Murakami, Y.; Yokoyama, M.; Okano, T.; Nishida, H.; Tomizawa, Y.; Endo, M.; Kurosawa, H. A novel synthetic tissue-adhesive hydrogel using a crosslinkable polymeric micelle. J. Biomed. Mater. Res., Part A 2007, 80A, 421-427. 158. Artzi, N.; Shazly, T.; Crespo, C.; Ramos, A. B.; Chenault, H. K.; Edelman, E. R. Characterization of star adhesive sealants based on peg/dextran hydrogels. Macromol. Biosci. 2009, 9, 754-765. 159. Chenault, H. K.; Bhatia, S. K.; DiMaio, W. G.; Vincent, G. L.; Camacho, W.; Behrens, A. Sealing and healing of clear corneal incisions with an improved dextran aldehyde-peg amine tissue adhesive. Curr. Eye Res. 2011, 36, 997-1004. 160. Shazly, T. M.; Artzi, N.; Boehning, F.; Edelman, E. R. Viscoelastic adhesive mechanics of aldehyde-mediated soft tissue sealants. Biomaterials 2008, 29, 4584- 4591. 161. Artzi, N.; Shazly, T.; Baker, A. B.; Bon, A.; Edelman, E. R. Aldehyde-amine chemistry enables modulated biosealants with tissue-specific adhesion. Adv. Mater. 2009, 21, 3399-3403. 162. Bhatia, S. K.; Arthur, S. D.; Chenault, H. K.; Figuly, G. D.; Kodokian, G. K. Polysaccharide-based tissue adhesives for sealing corneal incisions. Curr. Eye Res. 2007, 32, 1045-1050. 163. Artzi, N.; Zeiger, A.; Boehning, F.; bon Ramos, A.; Van Vliet, K.; Edelman, E. R. Tuning adhesion failure strength for tissue-specific applications. Acta Biomater. 2011, 7, 67-74. 164. Serban, M. A.; Panilaitis, B.; Kaplan, D. L. Silk fibroin and polyethylene glycol- based biocompatible tissue adhesives. J. Biomed. Mater. Res., Part A 2011, 98A, 567-575. 165. Burke, K. A.; Roberts, D. C.; Kaplan, D. L. Silk fibroin aqueous-based adhesives inspired by mussel adhesive proteins. Biomacromolecules 2016, 17, 237-245. 166. Chenault, H. K. Tissue adhesive and sealant comprising polyglycerol aldehyde. US8404779 B2, March 26, 2013. 167. Zeng, Z.; Mo, X.-m.; He, C.; Morsi, Y.; El-Hamshary, H.; El-Newehy, M. An in situ forming tissue adhesive based on poly(ethylene glycol)-dimethacrylate and thiolated chitosan through the michael reaction. J. Mater. Chem. B 2016, 4, 5585- 5592. 168. Zhou, Y.; Zhao, J.; Sun, X.; Li, S.; Hou, X.; Yuan, X.; Yuan, X. Rapid gelling chitosan/polylysine hydrogel with enhanced bulk cohesive and interfacial adhesive force: Mimicking features of epineurial matrix for peripheral nerve anastomosis. Biomacromolecules 2016, 17, 622-630.

148 169. Zhou, Y.; Nie, W.; Zhao, J.; Yuan, X. Rapidly in situ forming adhesive hydrogel based on a peg-maleimide modified polypeptide through michael addition. J. Mater. Sci. Mater. Med. 2013, 24, 2277-2286. 170. Liu, Y.; Meng, H.; Konst, S.; Sarmiento, R.; Rajachar, R.; Lee, B. P. Injectable dopamine-modified poly(ethylene glycol) nanocomposite hydrogel with enhanced adhesive property and bioactivity. ACS Appl. Mater. Interfaces 2014, 6, 16982- 16992. 171. Kim, K.; Shin, M.; Koh, M.-Y.; Ryu, J. H.; Lee, M. S.; Hong, S.; Lee, H. Tape: A medical adhesive inspired by a ubiquitous compound in plants. Adv. Funct. Mater. 2015, 25, 2402-2410. 172. Smyth, J. D. A technique for the histochemical demonstration of oxidase and its application to -shell formation in helminths and byssus formation in mytilus. Q. J. Microsc. Sci. 1954, s3-95, 139. 173. Yonge, C. M. On the primitive significance of the byssus in the bivalvia and its effects in evolution. J. Mar. Biol. Assoc. U.K. 1962, 42, 113-125. 174. Waite, J. H.; Tanzer, M. L. Polyphenolic substance of mytilus edulis: Novel adhesive containing l-dopa and hydroxyproline. Science 1981, 212, 1038. 175. Waite, J. H. Adhesion in byssally attached bivalves. Biological Reviews 1983, 58, 209-231. 176. Silverman, H. G.; Roberto, F. F. Understanding marine mussel adhesion. Mar. Biotechnol. (N. Y.) 2007, 9, 661-681. 177. Sáez, C.; Pardo, J.; Gutierrez, E.; Brito, M.; Burzio, L. O. Immunological studies of the polyphenolic proteins of mussels. Comp. Biochem. Physiol. B 1991, 98, 569- 572. 178. Benedict, C. V.; Picciano, P. T., Adhesives from marine mussels. In Adhesives from renewable resources; American Chemical Society: 1989; Vol. 385, pp 465-483. 179. Hwang, D. S.; Gim, Y.; Kang, D. G.; Kim, Y. K.; Cha, H. J. Recombinant mussel adhesive protein mgfp-5 as cell adhesion biomaterial. J. Biotechnol. 2007, 127, 727- 735. 180. Hansen, D. C.; Luther, G. W.; Waite, J. H. The adsorption of the adhesive protein of the blue mussel mytilus edulis l onto type 304l stainless steel. J. Colloid Interface Sci. 1994, 168, 206-216. 181. Schnurrer, J.; Lehr, C.-M. Mucoadhesive properties of the mussel adhesive protein. Int. J. Pharm. 1996, 141, 251-256. 182. Ninan, L.; Monahan, J.; Stroshine, R. L.; Wilker, J. J.; Shi, R. Adhesive strength of marine mussel extracts on porcine skin. Biomaterials 2003, 24, 4091-4099.

149 183. Ninan, L.; Stroshine, R. L.; Wilker, J. J.; Shi, R. Adhesive strength and curing rate of marine mussel protein extracts on porcine small intestinal submucosa. Acta Biomater. 2007, 3, 687-694. 184. Yamamoto, H.; Hayakawa, T. Synthesis and conformational study of poly(0,0′- dicarbobenzoxy-l-β-3,4-dihydroxyphenyl-α-alanine). 1977, 16, 1593- 1607. 185. Yamamoto, H.; Hayakawa, T. Improved synthetic method and conformation studies of polymers and copolymers of l-β-3,4-dihydroxyphenyl-α-alanine with l-glutamic acid. Polymer 1978, 19, 1115-1117. 186. Yamamoto, H. Bonding strength of synthetic poly(amino acid)s on metals. Nippon Kagaku Kaishi 1986, 1986, 90-92. 187. Yamamoto, H.; Hayakawa, T. Synthesis of sequential polypeptides containing l-β- 3,4-dihydroxyphenyl-α-alanine (dopa) and l-glutamic acid. Biopolymers 1979, 18, 3067-3076. 188. Yamamoto, H.; Hayakawa, T. Synthesis of sequential polypeptides containing l-β- 3,4-dihydroxyphenyl-α-alanine (dopa) and l-lysine. Biopolymers 1982, 21, 1137- 1151. 189. Yamamoto, H.; Hayakawa, T. Conformational studies of sequential polypeptides containing l-β-(3,4-dihydroxyphenyl)-α-alanine (dopa) and l-lysine. Macromolecules 1983, 16, 1058-1063. 190. Nagai, A.; Yamamoto, H. Insolubilizing studies of water-soluble poly(lys tyr) by tyrosinase. Bull. Chem. Soc. Jpn. 1989, 62, 2410-2412. 191. Yamamoto, H.; Kuno, S.; Nagai, A.; Nishida, A.; Yamauchi, S.; Ikeda, K. Insolubilizing and adhesive studies of water-soluble synthetic model proteins. Int. J. Biol. Macromol. 1990, 12, 305-310. 192. Tatehata, H.; Mochizuki, A.; Kawashima, T.; Yamashita, S.; Yamamoto, H. Model polypeptide of mussel adhesive protein. I. Synthesis and adhesive studies of sequential polypeptides (x-tyr-lys)n and (y-lys)n. J. Appl. Polym. Sci. 2000, 76, 929-937. 193. Yamamoto, H.; Sakai, Y.; Ohkawa, K. Synthesis and wettability characteristics of model adhesive protein sequences inspired by a marine mussel. Biomacromolecules 2000, 1, 543-551. 194. Yamamoto, H. Synthesis and adhesive studies of marine polypeptides. J. Chem. Soc., Perkin Trans. 1 1987, 613-618. 195. Yu, M.; Deming, T. J. Synthetic polypeptide mimics of marine adhesives. Macromolecules 1998, 31, 4739-4745.

150 196. Tatehata, H.; Mochizuki, A.; Ohkawa, K.; Yamada, M.; Yamamoto, H. Tissue adhesive using synthetic model adhesive proteins inspired by the marine mussel. J. Adhes. Sci. Technol. 2001, 15, 1003-1013. 197. Kumar, G.; Bristow, J. F.; Smith, P. J.; Payne, G. F. Enzymatic gelation of the natural polymer chitosan. Polymer 2000, 41, 2157-2168. 198. Yamada, K.; Chen, T.; Kumar, G.; Vesnovsky, O.; Topoleski, L. D. T.; Payne, G. F. Chitosan based water-resistant adhesive. Analogy to mussel glue. Biomacromolecules 2000, 1, 252-258. 199. Yamada, K.; Aoki, T.; Ikeda, N.; Hirata, M. Application of enzymatically gelled chitosan solutions to water-resistant adhesives. J. Appl. Polym. Sci. 2007, 104, 1818-1827. 200. Hoffmann, B.; Volkmer, E.; Kokott, A.; Augat, P.; Ohnmacht, M.; Sedlmayr, N.; Schieker, M.; Claes, L.; Mutschler, W.; Ziegler, G. Characterisation of a new bioadhesive system based on polysaccharides with the potential to be used as bone glue. J. Mater. Sci. Mater. Med. 2009, 20, 2001-2009. 201. Ryu, J. H.; Lee, Y. H.; Kong, W. H.; Kim, T. G.; Park, T. G. Bio-inspired tissue adhesive chitosan/pluronic composite hydrogel. J. Controlled Release 2011, 152 Suppl 1, e236-237. 202. Ryu, J. H.; Lee, Y.; Kong, W. H.; Kim, T. G.; Park, T. G.; Lee, H. Catechol- functionalized chitosan/pluronic hydrogels for tissue adhesives and hemostatic materials. Biomacromolecules 2011, 12, 2653-2659. 203. Fan, C.; Fu, J.; Zhu, W.; Wang, D.-A. A mussel-inspired double-crosslinked tissue adhesive intended for internal medical use. Acta Biomater. 2016, 33, 51-63. 204. Scognamiglio, F.; Travan, A.; Borgogna, M.; Donati, I.; Marsich, E.; Bosmans, J. W. A. M.; Perge, L.; Foulc, M. P.; Bouvy, N. D.; Paoletti, S. Enhanced bioadhesivity of dopamine-functionalized polysaccharidic membranes for general surgery applications. Acta Biomater. 2016, 44, 232-242. 205. Huang, K.; Lee, B. P.; Ingram, D. R.; Messersmith, P. B. Synthesis and characterization of self-assembling block copolymers containing bioadhesive end groups. Biomacromolecules 2002, 3, 397-406. 206. Lee, B. P.; Huang, K.; Nunalee, F. N.; Shull, K. R.; Messersmith, P. B. Synthesis of 3,4-dihydroxyphenylalanine (dopa) containing monomers and their co- polymerization with peg-diacrylate to form hydrogels. J. Biomater. Sci. Polym. Ed. 2004, 15, 449-464. 207. Lee, Y.; Chung, H. J.; Yeo, S.; Ahn, C.-H.; Lee, H.; Messersmith, P. B.; Park, T. G. Thermo-sensitive, injectable, and tissue adhesive sol-gel transition hyaluronic acid/pluronic composite hydrogels prepared from bio-inspired catechol-thiol reaction. Soft Matter 2010, 6, 977-983.

151 208. Lee, B. P.; Dalsin, J. L.; Messersmith, P. B. Synthesis and gelation of dopa- modified poly(ethylene glycol) hydrogels. Biomacromolecules 2002, 3, 1038-1047. 209. Burke, S. A.; Ritter-Jones, M.; Lee, B. P.; Messersmith, P. B. Thermal gelation and tissue adhesion of biomimetic hydrogels. Biomedical Materials 2007, 2, 203-210. 210. Dalsin, J. L.; Lin, L.; Tosatti, S.; Vörös, J.; Textor, M.; Messersmith, P. B. Protein resistance of titanium oxide surfaces modified by biologically inspired mpeg−dopa. Langmuir 2005, 21, 640-646. 211. Lee, B. P.; Chao, C.-Y.; Nunalee, F. N.; Motan, E.; Shull, K. R.; Messersmith, P. B. Rapid gel formation and adhesion in photocurable and biodegradable block copolymers with high dopa content. Macromolecules 2006, 39, 1740-1748. 212. Guvendiren, M.; Brass, D. A.; Messersmith, P. B.; Shull, K. R. Adhesion of dopa- functionalized model membranes to hard and soft surfaces. J. Adhes. 2009, 85, 631- 645. 213. Brubaker, C. E.; Kissler, H.; Wang, L.-J.; Kaufman, D. B.; Messersmith, P. B. Biological performance of mussel-inspired adhesive in extrahepatic islet transplantation. Biomaterials 2010, 31, 420-427. 214. Bilic, G.; Brubaker, C.; Messersmith, P. B.; Mallik, A. S.; Quinn, T. M.; Haller, C.; Done, E.; Gucciardo, L.; Zeisberger, S. M.; Zimmermann, R.; Deprest, J.; Zisch, A. H. Injectable candidate sealants for fetal membrane repair: Bonding and toxicity in vitro. Am. J. Obstet. Gynecol. 2010, 202, 85.e81-85.e89. 215. Haller, C. M.; Buerzle, W.; Kivelio, A.; Perrini, M.; Brubaker, C. E.; Gubeli, R. J.; Mallik, A. S.; Weber, W.; Messersmith, P. B.; Mazza, E.; Ochsenbein-Koelble, N.; Zimmermann, R.; Ehrbar, M. Mussel-mimetic tissue adhesive for fetal membrane repair: An ex vivo evaluation. Acta Biomater. 2012, 8, 4365-4370. 216. Kivelio, A.; Dekoninck, P.; Perrini, M.; Brubaker, C. E.; Messersmith, P. B.; Mazza, E.; Deprest, J.; Zimmermann, R.; Ehrbar, M.; Ochsenbein-Koelble, N. Mussel mimetic tissue adhesive for fetal membrane repair: Initial in vivo investigation in rabbits. Eur. J. Obstet. Gynecol. Reprod. Biol. 2013, 171, 240-245. 217. Haller, C. M.; Buerzle, W.; Brubaker, C. E.; Messersmith, P. B.; Mazza, E.; Ochsenbein-Koelble, N.; Zimmermann, R.; Ehrbar, M. Mussel-mimetic tissue adhesive for fetal membrane repair: A standardized ex vivo evaluation using elastomeric membranes. Prenatal Diagn. 2011, 31, 654-660. 218. Brubaker, C. E.; Messersmith, P. B. Enzymatically degradable mussel-inspired adhesive hydrogel. Biomacromolecules 2011, 12, 4326-4334. 219. Wang, J.; Liu, C.; Lu, X.; Yin, M. Co-polypeptides of 3,4-dihydroxyphenylalanine and l-lysine to mimic marine adhesive protein. Biomaterials 2007, 28, 3456-3468.

152 220. Lim, S.; Choi, Y. S.; Kang, D. G.; Song, Y. H.; Cha, H. J. The adhesive properties of coacervated recombinant hybrid mussel adhesive proteins. Biomaterials 2010, 31, 3715-3722. 221. Kim, B. J.; Oh, D. X.; Kim, S.; Seo, J. H.; Hwang, D. S.; Masic, A.; Han, D. K.; Cha, H. J. Mussel-mimetic protein-based adhesive hydrogel. Biomacromolecules 2014, 15, 1579-1585. 222. Yang, B.; Ayyadurai, N.; Yun, H.; Choi, Y. S.; Hwang, B. H.; Huang, J.; Lu, Q.; Zeng, H.; Cha, H. J. In vivo residue-specific dopa-incorporated engineered mussel bioglue with enhanced adhesion and water resistance. Angew. Chem. 2014, 126, 13578-13582. 223. Jeon, E. Y.; Hwang, B. H.; Yang, Y. J.; Kim, B. J.; Choi, B.-H.; Jung, G. Y.; Cha, H. J. Rapidly light-activated surgical protein glue inspired by mussel adhesion and structural crosslinking. Biomaterials 2015, 67, 11-19. 224. Zhong, C.; Gurry, T.; Cheng, A. A.; Downey, J.; Deng, Z.; Stultz, C. M.; Lu, T. K. Strong underwater adhesives made by self-assembling multi-protein nanofibres. Nat. Nanotechnol. 2014, 9, 858-866. 225. Yang, B.; Kang, D. G.; Seo, J. H.; Choi, Y. S.; Cha, H. J. A comparative study on the bulk adhesive strength of the recombinant mussel adhesive protein fp-3. 2013, 29, 483-490. 226. Wei, W.; Tan, Y.; Martinez Rodriguez, N. R.; Yu, J.; Israelachvili, J. N.; Waite, J. H. A mussel-derived one component adhesive coacervate. Acta Biomater. 2014, 10, 1663-1670. 227. Murphy, J. L.; Vollenweider, L.; Xu, F.; Lee, B. P. Adhesive performance of biomimetic adhesive-coated biologic scaffolds. Biomacromolecules 2010, 11, 2976- 2984. 228. Michael, B.; Laura, V.; John, L. M.; Fangmin, X.; Arinne, L.; William, D. L.; Bruce, P. L. Biomechanical properties of achilles tendon repair augmented with a bioadhesive-coated scaffold. Biomedical Materials 2011, 6, 015014. 229. Chung, H.; Grubbs, R. H. Rapidly cross-linkable dopa containing terpolymer adhesives and peg-based cross-linkers for biomedical applications. Macromolecules 2012, 45, 9666-9673. 230. Xue, J.; Wang, T.; Nie, J.; Yang, D. Preparation and characterization of a photocrosslinkable bioadhesive inspired by marine mussel. J. Photochem. Photobiol. B 2013, 119, 31-36. 231. Guo, J.; Wang, W.; Hu, J.; Xie, D.; Gerhard, E.; Nisic, M.; Shan, D.; Qian, G.; Zheng, S.; Yang, J. Synthesis and characterization of anti-bacterial and anti-fungal citrate-based mussel-inspired bioadhesives. Biomaterials 2016, 85, 204-217.

153 232. Mehdizadeh, M.; Weng, H.; Gyawali, D.; Tang, L.; Yang, J. Injectable citrate-based mussel-inspired tissue bioadhesives with high wet strength for sutureless wound closure. Biomaterials 2012, 33, 7972-7983. 233. Guo, J.; Kim, G. B.; Shan, D.; Kim, J. P.; Hu, J.; Wang, W.; Hamad, F. G.; Qian, G.; Rizk, E. B.; Yang, J. Click chemistry improved wet adhesion strength of mussel-inspired citrate-based antimicrobial bioadhesives. Biomaterials 2017, 112, 275-286. 234. Ji, Y.; Ji, T.; Liang, K.; Zhu, L. Mussel-inspired soft-tissue adhesive based on poly(diol citrate) with catechol functionality. J. Mater. Sci. Mater. Med. 2015, 27, 1-9. 235. Shafiq, Z.; Cui, J.; Pastor-Perez, L.; San Miguel, V.; Gropeanu, R. A.; Serrano, C.; del Campo, A. Bioinspired underwater bonding and debonding on demand. Angew. Chem. Int. Ed. Engl. 2012, 51, 4332-4335. 236. Zhang, H.; Bré, L. P.; Zhao, T.; Zheng, Y.; Newland, B.; Wang, W. Mussel- inspired hyperbranched poly(amino ester) polymer as strong wet tissue adhesive. Biomaterials 2014, 35, 711-719. 237. Ai, Y. F.; Nie, J.; Wu, G.; Yang, D. Z. The dopa-functionalized bioadhesive with properties of photocrosslinked and thermoresponsive. J. Appl. Polym. Sci. 2014, 131, 41102. 238. Zhou, J.; Defante, A. P.; Lin, F.; Xu, Y.; Yu, J.; Gao, Y.; Childers, E.; Dhinojwala, A.; Becker, M. L. Adhesion properties of catechol-based biodegradable amino acid- based poly(ester urea) copolymers inspired from mussel proteins. Biomacromolecules 2015, 16, 266-274. 239. Zhou, J.; Bhagat, V.; Becker, M. L. Poly(ester urea)-based adhesives: Improved deployment and adhesion by incorporation of poly(propylene glycol) segments. ACS Appl. Mater. Interfaces 2016, 8, 33423-33429. 240. Shi, Y.; Zhou, P.; Jérôme, V.; Freitag, R.; Agarwal, S. Enzymatically degradable polyester-based adhesives. ACS Biomaterials Science and Engineering 2015, 1, 971-977. 241. Jenkins, C. L.; Siebert, H. M.; Wilker, J. J. Integrating mussel chemistry into a bio- based polymer to create degradable adhesives. Macromolecules 2017, 50, 561-568. 242. White, J. D.; Wilker, J. J. Underwater bonding with charged polymer mimics of marine mussel adhesive proteins. Macromolecules 2011, 44, 5085-5088. 243. Meredith, H. J.; Jenkins, C. L.; Wilker, J. J. Enhancing the adhesion of a biomimetic polymer yields performance rivaling commercial glues. Adv. Funct. Mater. 2014, 24, 3259-3267.

154 244. Meredith, H. J.; Wilker, J. J. The interplay of modulus, strength, and ductility in adhesive design using biomimetic polymer chemistry. Adv. Funct. Mater. 2015, 25, 5057-5065. 245. North, M. A.; Del Grosso, C. A.; Wilker, J. J. High strength underwater bonding with polymer mimics of mussel adhesive proteins. ACS Appl. Mater. Interfaces 2017, 9, 7866-7872. 246. Brennan, M. J.; Meredith, H. J.; Jenkins, C. L.; Wilker, J. J.; Liu, J. C. Cytocompatibility studies of a biomimetic copolymer with simplified structure and high-strength adhesion. J. Biomed. Mater. Res., Part A 2016, 104, 983-990. 247. Shan, M.; Gong, C.; Li, B.; Wu, G. A ph, glucose, and dopamine triple-responsive, self-healable adhesive hydrogel formed by phenylborate-catechol complexation. Polym. Chem. 2017, 8, 2997-3005. 248. Li, A.; Mu, Y.; Jiang, W.; Wan, X. A mussel-inspired adhesive with stronger bonding strength under underwater conditions than under dry conditions. Chem. Commun. 2015, 51, 9117-9120. 249. Lu, D.; Wang, H.; Li, T. e.; Li, Y.; Dou, F.; Sun, S.; Guo, H.; Liao, S.; Yang, Z.; Wei, Q.; Lei, Z. Mussel-inspired thermoresponsive polypeptide–pluronic copolymers for versatile surgical adhesives and hemostasis. ACS Appl. Mater. Interfaces 2017, 9, 16756-16766. 250. Waite, J. H.; Andersen, N. H.; Jewhurst, S.; Sun, C. J. Mussel adhesion: Finding the tricks worth mimicking. J. Adhes. 2005, 81, 297-317. 251. Mahdavi, A.; Ferreira, L.; Sundback, C.; Nichol, J. W.; Chan, E. P.; Carter, D. J. D.; Bettinger, C. J.; Patanavanich, S.; Chignozha, L.; Ben-Joseph, E.; Galakatos, A.; Pryor, H.; Pomerantseva, I.; Masiakos, P. T.; Faquin, W.; Zumbuehl, A.; Hong, S.; Borenstein, J.; Vacanti, J.; Langer, R.; Karp, J. M. A biodegradable and biocompatible gecko-inspired tissue adhesive. Proc. Natl. Acad. Sci. U. S. A. 2008, 105, 2307-2312. 252. Shao, H.; Bachus, K. N.; Stewart, R. J. A water-borne adhesive modeled after the sandcastle glue of p. Californica. Macromol. Biosci. 2009, 9, 464-471. 253. Shao, H.; Stewart, R. J. Biomimetic underwater adhesives with environmentally triggered setting mechanisms. Adv. Mater. 2010, 22, 729-733. 254. Winslow, B. D.; Shao, H.; Stewart, R. J.; Tresco, P. A. Biocompatibility of adhesive complex coacervates modeled after the sandcastle glue of phragmatopoma californica for craniofacial reconstruction. Biomaterials 2010, 31, 9373-9381. 255. Kaur, S.; Weerasekare, G. M.; Stewart, R. J. Multiphase adhesive coacervates inspired by the sandcastle worm. ACS Appl. Mater. Interfaces 2011, 3, 941-944.

155 256. Li, A.; Jia, M.; Mu, Y.; Jiang, W.; Wan, X. Humid bonding with a water-soluble adhesive inspired by mussels and sandcastle worms. Macromol. Chem. Phys. 2015, 216, 450-459. 257. Zhang, L.; Lipik, V.; Miserez, A. Complex coacervates of oppositely charged co- polypeptides inspired by the sandcastle worm glue. J. Mater. Chem. B 2016, 4, 1544-1556. 258. Yamamoto, H.; Nagai, A.; Okada, T.; Nishida, A. Synthesis and adhesive studies of barnacle model proteins. Mar. Chem. 1989, 26, 331-338. 259. Yamamoto, H.; Nagai, A. Polypeptide models of the arthropodin protein of the barnacle balanus balanoides. Mar. Chem. 1992, 37, 131-143. 260. Nishida, J.; Higaki, Y.; Takahara, A. Synthesis and characterization of barnacle adhesive mimetic towards underwater adhesion. Chem. Lett. 2015, 44, 1047-1049. 261. Stewart, R. J., Adhesive complex coacervates produced from electrostatically associated block copolymers and methods for making and using the same. In Google Patents: 2013. 262. Bhagat, V.; O’Brien, E.; Zhou, J.; Becker, M. L. Caddisfly inspired phosphorylated poly(ester urea)-based degradable bone adhesives. Biomacromolecules 2016, 17, 3016-3024. 263. Wilker, J. J. Biomaterials: Redox and adhesion on the rocks. Nat. Chem. Biol. 2011, 7, 579-580. 264. Burzio, L. A.; Waite, J. H. Cross-linking in adhesive quinoproteins: Studies with model decapeptides. Biochemistry 2000, 39, 11147-11153. 265. McDowell, L. M.; Burzio, L. A.; Waite, J. H.; Schaefer, J. Rotational echo double resonance detection of cross-links formed in mussel byssus under high-flow stress. J. Biol. Chem. 1999, 274, 20293-20295. 266. Fullenkamp, D. E.; Barrett, D. G.; Miller, D. R.; Kurutz, J. W.; Messersmith, P. B. Ph-dependent cross-linking of catechols through oxidation via fe3+ and potential implications for mussel adhesion. RSC Advances 2014, 4, 25127-25134. 267. Holten-Andersen, N.; Harrington, M. J.; Birkedal, H.; Lee, B. P.; Messersmith, P. B.; Lee, K. Y. C.; Waite, J. H. Ph-induced metal-ligand cross-links inspired by mussel yield self-healing polymer networks with near-covalent elastic moduli. Proc. Natl. Acad. Sci. U. S. A. 2011, 108, 2651-2655. 268. Wilker, J. J. Marine bioinorganic materials: Mussels pumping iron. Curr. Opin. Chem. Biol. 2010, 14, 276-283. 269. Wilker, J. J. The iron-fortified adhesive system of marine mussels. Angew. Chem. Int. Ed. 2010, 49, 8076-8078. 270. MacGillivray, T. E. Fibrin sealants and glues. J. Card. Surg. 2003, 18, 480-485.

156 271. Hwang, D. S.; Zeng, H.; Srivastava, A.; Krogstad, D. V.; Tirrell, M.; Israelachvili, J. N.; Waite, J. H. Viscosity and interfacial properties in a mussel-inspired adhesive coacervate. Soft Matter 2010, 6, 3232-3236. 272. Matos-Pérez, C. R.; White, J. D.; Wilker, J. J. Polymer composition and substrate influences on the adhesive bonding of a biomimetic, cross-linking polymer. J. Am. Chem. Soc. 2012, 134, 9498-9505. 273. Wu, J.; Zhang, L.; Wang, Y.; Long, Y.; Gao, H.; Zhang, X.; Zhao, N.; Cai, Y.; Xu, J. Mussel-inspired chemistry for robust and surface-modifiable multilayer films. Langmuir 2011, 27, 13684-13691. 274. Guvendiren, M.; Messersmith, P. B.; Shull, K. R. Self-assembly and adhesion of dopa-modified methacrylic triblock hydrogels. Biomacromolecules 2008, 9, 122- 128. 275. Nishida, J.; Kobayashi, M.; Takahara, A. Gelation and adhesion behavior of mussel adhesive protein mimetic polymer. J. Polym. Sci., Part A: Polym. Chem. 2013, 51, 1058-1065. 276. Sparks, B. J.; Hoff, E. F. T.; Hayes, L. P.; Patton, D. L. Mussel-inspired thiol–ene polymer networks: Influencing network properties and adhesion with catechol functionality. Chem. Mater. 2012, 24, 3633-3642. 277. Nishida, J.; Kobayashi, M.; Takahara, A. Light-triggered adhesion of water-soluble polymers with a caged catechol group. ACS Macro Letters 2013, 2, 112-115. 278. Sun, P.; Wang, J.; Yao, X.; Peng, Y.; Tu, X.; Du, P.; Zheng, Z.; Wang, X. Facile preparation of mussel-inspired polyurethane hydrogel and its rapid curing behavior. ACS Applied Materials and Interfaces 2014, 6, 12495-12504. 279. Katsarava, R.; Beridze, V.; Arabuli, N.; Kharadze, D.; Chu, C. C.; Won, C. Y. Amino acid-based bioanalogous polymers. Synthesis, and study of regular poly(ester amide)s based on bis(α-amino acid) α,ω-alkylene diesters, and aliphatic dicarboxylic acids. J. Polym. Sci., Part A: Polym. Chem. 1999, 37, 391-407. 280. Gao, Y.; Childers, E. P.; Becker, M. L. L-leucine-based poly(ester urea)s for vascular tissue engineering. ACS Biomaterials Science and Engineering 2015, 1, 795-804. 281. Yu, J.; Lin, F.; Lin, P.; Gao, Y.; Becker, M. L. Phenylalanine-based poly(ester urea): Synthesis, characterization, and in vitro degradation. Macromolecules 2014, 47, 121-129. 282. LaKind, J. S.; McKenna, E. A.; Hubner, R. P.; Tardiff, R. G. A review of the comparative mammalian toxicity of ethylene glycol and propylene glycol. Crit. Rev. Toxicol. 1999, 29, 331-365.

157 283. Fu, B.; Sun, X.; Qian, W.; Shen, Y.; Chen, R.; Hannig, M. Evidence of chemical bonding to hydroxyapatite by phosphoric acid esters. Biomaterials 2005, 26, 5104- 5110. 284. Gonzalez, I.; Mestach, D.; Leiza, J. R.; Asua, J. M. Adhesion enhancement in waterborne acrylic latex binders synthesized with phosphate methacrylate monomers. Prog. Org. Coat. 2008, 61, 38-44. 285. Zeller, A.; Musyanovych, A.; Kappl, M.; Ethirajan, A.; Dass, M.; Markova, D.; Klapper, M.; Landfester, K. Nanostructured coatings by adhesion of phosphonated polystyrene particles onto titanium surface for implant material applications. ACS Appl Mater Interfaces 2010, 2, 2421-2428. 286. Staller, C.; Schumacher, K. H.; Schlarb, B.; Centner, A.; Hartz, O., Use of phosphate group-containing polymer dispersions as adhesives. In Google Patents: 2004. 287. Monge, S.; Canniccioni, B.; Graillot, A.; Robin, J. J. Phosphorus-containing polymers: A great opportunity for the biomedical field. Biomacromolecules 2011, 12, 1973-1982. 288. Zhang, S.; Zou, J.; Zhang, F.; Elsabahy, M.; Felder, S. E.; Zhu, J.; Pochan, D. J.; Wooley, K. L. Rapid and versatile construction of diverse and functional nanostructures derived from a polyphosphoester-based biomimetic block copolymer system. J. Am. Chem. Soc. 2012, 134, 18467-18474. 289. Kamei, S.; Tomita, N.; Tamai, S.; Kato, K.; Ikada, Y. Histologic and mechanical evaluation for bone bonding of polymer surfaces grafted with a phosphate- containing polymer. J. Biomed. Mater. Res. 1997, 37, 384-393. 290. van Blitterswijk, C. A.; Bakker, D.; Hesseling, S. C.; Koerten, H. K. Reactions of cells at implant surfaces. Biomaterials 1991, 12, 187-193. 291. Bonfield, W. Hydroxyapatite-reinforced polyethylene as an analogous material for bone replacement. Ann. N.Y. Acad. Sci. 1988, 523, 173-177. 292. Suzuki, S.; Whittaker, M. R.; Grøndahl, L.; Monteiro, M. J.; Wentrup-Byrne, E. Synthesis of soluble phosphate polymers by raft and their in vitro mineralization. Biomacromolecules 2006, 7, 3178-3187. 293. Brendan, M. W.; Kasper, F. K.; Antonios, G. M. Phosphorous-containing polymers for regenerative medicine. Biomedical Materials 2014, 9, 025014. 294. Abou Neel, E. A.; Salih, V.; Revell, P. A.; Young, A. M. Viscoelastic and biological performance of low-modulus, reactive calcium phosphate-filled, degradable, polymeric bone adhesives. Acta Biomater. 2012, 8, 313-320. 295. Datta, P.; Chatterjee, J.; Dhara, S. Electrospun nanofibers of a phosphorylated polymer—a bioinspired approach for bone graft applications. Colloids Surf., B 2012, 94, 177-183.

158 296. Lipik, V.; Zhang, L. H.; Miserez, A. Synthesis of biomimetic co-polypeptides with tunable degrees of phosphorylation. Polym. Chem. 2014, 5, 1351-1361. 297. Breucker, L.; Landfester, K.; Taden, A. Phosphonic acid-functionalized polyurethane dispersions with improved adhesion properties. ACS Appl Mater Interfaces 2015, 7, 24641-24648. 298. Stewart, R. J. Protein-based underwater adhesives and the prospects for their biotechnological production. Appl. Microbiol. Biotechnol. 2011, 89, 27-33. 299. Yonemura, N.; Mita, K.; Tamura, T.; Sehnal, F. Conservation of silk genes in trichoptera and lepidoptera. J. Mol. Evol. 2009, 68, 641-653. 300. Wang, Y.; Sanai, K.; Wen, H.; Zhao, T.; Nakagaki, M. Characterization of unique heavy chain fibroin filaments spun underwater by the caddisfly stenopsyche marmorata (trichoptera; stenopsychidae). Mol. Biol. Rep. 2010, 37, 2885-2892. 301. Addison, J. B.; Ashton, N. N.; Weber, W. S.; Stewart, R. J.; Holland, G. P.; Yarger, J. L. Beta-sheet nanocrystalline domains formed from phosphorylated serine-rich motifs in caddisfly larval silk: A solid state nmr and xrd study. Biomacromolecules 2013, 14, 1140-1148. 302. Addison, J. B.; Weber, W. S.; Mou, Q.; Ashton, N. N.; Stewart, R. J.; Holland, G. P.; Yarger, J. L. Reversible assembly of beta-sheet nanocrystals within caddisfly silk. Biomacromolecules 2014, 15, 1269-1275. 303. Stewart, R. J.; Wang, C. S. Adaptation of caddisfly larval to aquatic habitats by phosphorylation of h-fibroin serines. Biomacromolecules 2010, 11, 969-974. 304. Wang, C. S.; Ashton, N. N.; Weiss, R. B.; Stewart, R. J. Peroxinectin catalyzed dityrosine crosslinking in the adhesive underwater silk of a casemaker caddisfly larvae, hysperophylax occidentalis. Insect Biochem. Mol. Biol. 2014, 54, 69-79. 305. Wang, C.-S.; Pan, H.; Weerasekare, G. M.; Stewart, R. J. Peroxidase-catalysed interfacial adhesion of aquatic caddisworm silk. Journal of The Royal Society Interface 2015, 12. 306. Kartvelishvili, T.; Tsitlanadze, G.; Edilashvili, L.; Japaridze, N.; Katsarava, R. Amino acid based bioanalogous polymers. Novel regular poly(ester urethane)s and poly(ester urea)s based on bis(l-phenylalanine) α,ω-alkylene diesters. Macromol. Chem. Phys. 1997, 198, 1921-1932. 307. Zimmermann, J.; Loontjens, T.; Scholtens, B. J.; Mulhaupt, R. R. The formation of poly(ester-urea) networks in the absence of isocyanate monomers. Biomaterials 2004, 25, 2713-2719. 308. Stakleff, K. S.; Lin, F.; Smith Callahan, L. A.; Wade, M. B.; Esterle, A.; Miller, J.; Graham, M.; Becker, M. L. Resorbable, amino acid-based poly(ester urea)s crosslinked with osteogenic growth peptide with enhanced mechanical properties and bioactivity. Acta Biomater 2013, 9, 5132-5142.

159 309. McKay, C. A.; Pomrenke, R. D.; McLane, J. S.; Schaub, N. J.; DeSimone, E. K.; Ligon, L. A.; Gilbert, R. J. An injectable, calcium responsive composite hydrogel for the treatment of acute spinal cord injury. ACS Appl. Mater. Interfaces 2014, 6, 1424-1438.

160

APPENDIX

161

APPENDIX

5.1. Figures of 1H, 13C and 31P NMR Spectra of Monomers and Polymers

5.1.1. Synthesis of 4-(dimethylamino)pyridinium 4-toluenesulfonate (DPTS).

Hydrated p-toluenesulfonic acid (pTSA, 12.00 g, 63.1 mmol) was dried by azeotropic

distillation of a toluene solution using a Dean-Stark trap. 4-(dimethylamino)pyridine

(8.50 g, 49.4 mmol) dissolved in hot toluene was added to the above solution at about 60

°C. After stirring for 1 h, the solution was cooled to room temperature, filtered, washed

with toluene and vacuum dried. After recrystallization with dichloromethane for 3 times,

the product was obtained as white crystals (16.03 g, Yield = 86%). 1H NMR (500 MHz,

DMSO-d6): 2.28 (s, 3H), 3.17 (s, 6H), 6.97 (d, J = 7.82 Hz, 2H), 7.11 (d, J = 7.83 Hz,

2H), 7.47 - 7.51 (m, 2H), 8.18 - 8.23 (m, 2H), 13.19 (br. s., 1H). 13C NMR (125 MHz,

DMSO-d6): 21.23, 107.43, 125.95, 128.51, 138.10, 139.57, 146.15, 157.42.

162

Figure 5.1. 1H NMR spectrum of 4-(dimethylamino)pyridinium 4-toluenesulfonate

(DPTS).

5.1.2. Synthesis of acetonide-protected 3,4-dihydroxyhydrocinnamic acid.

Scheme 5.1. Synthesis of acetonide-protected 3,4-dihydroxyhydrocinnamic acid.

3,4-dihydroxyhydrocinnamic acid (6 g, 32.94 mmol) was dissolved in ~ 45 mL anhydrous acetone under N2. The solution was cooled to 0 °C and PCl3 (2.3 mL, 26.34 mmol) was added dropwise over 30 min. The reaction is continued at 0 °C for 6 h.

Acetone was removed under vacuum and the product was redissolved in diethyl ether:water solution (1:1:: v/v, 150 mL). The organic phase was separated and washed 5 times with water, dried over Na2SO4 and dried to obtain a brown colored solid. The product was further purified by silica gel column chromatography using ethyl acetate as

163 1 eluent to obtain 2.14 g (Yield = 29%) light grey solid. H NMR (500 MHz, CDCl3): 1.65

(s, 6H), 2.55 - 2.63 (m, 2H), 2.79 - 2.88 (m, 2H), 6.55 - 6.66 (m, 3 H), 9.34 (br. s., 1H).

13 C NMR (125 MHz, CDCl3): 25.79, 30.72, 36.70, 108.01, 108.53, 117.60, 120.37,

133.61, 145.80, 147.48, 179.38.

Figure 5.2. 1H NMR spectrum of acetonide protected 3,4-dihydroxyhydrocinnamic acid.

Figure 5.3. 13C NMR spectrum of acetonide protected 3,4-dihydroxyhydrocinnamic acid.

164

Figure 5.4. 1H NMR spectrum of di-p-toluenesulfonic acid salt of bis(L-leucine)-1,8- octanyl diester (M1).

Figure 5.5. 13C NMR spectrum of di-p-toluenesulfonic acid salt of bis(L-leucine)-1,8- octanyl diester (M1).

165

Figure 5.6. 1H NMR spectrum of bis-N-Boc-O-benzyl(L-serine)-1,8-octanyl diester.

Figure 5.7. 13C NMR spectrum of bis-N-Boc-O-benzyl(L-serine)-1,8-octanyl diester.

166

Figure 5.8. 1H NMR spectrum of di-hydrochloric acid salt of bis-O-benzyl(L-serine)-1,8- octanyl diester (M2).

Figure 5.9. 13C NMR spectrum of di-hydrochloric acid salt of bis-O-benzyl(L-serine)-1,8- octanyl diester (M2).

167

Figure 5.10. 1H NMR spectra of PPG-PEU Polymer (a) Poly(bzlSer-co-Leu-co-PPG), (b)

Poly(Ser-co-Leu-co-PPG), (c) Poly(CA-AN-Ser-co-Leu-co-PPG), (d) Poly(CA-Ser-co-

Leu-co-PPG).

Figure 5.11. 1H NMR spectra of di-p-toluenesulfonic acid salt of bis(L-valine)-1,8-

octanyl diester (M1). 168

Figure 5.12. 1H NMR spectra of Poly(1-Val-8).

Figure 5.13. 13C NMR spectra of Poly(1-Val-8).

169

Figure 5.14. 1H NMR spectra of dihydrochloride salt of bis-O-diphenylphosphate(L- serine)-1,8-octanyl diester (M5).

170 Figure 5.15. 1H NMR spectra of 2% Poly(SerDPP-co-Val). R group denotes diphenyl protected phosphate groups. Inset shows aromatic peaks from the diphenyl protecting groups.

Figure 5.16. 1H NMR spectra of 2% Poly(pSer-co-Val). R’ denotes deprotected phosphate groups. Inset (a) shows disappearance of the aromatic peaks between 7.0 –

7.25 ppm confirming deprotection of the diphenyl groups; (b) Appearance of a triplet

around ~ 4.37 ppm corresponds to the proton environment on methylene group attached

to the deprotected phosphate group.

171

Figure 5.17. 13C NMR spectra of 2% Poly(SerDPP-co-Val).

Figure 5.18. 1H NMR spectra of: (a) 5% Poly(SerDPP-co-Val), (b) 5% Poly(pSer-co-

Val). 172

Figure 5.19. 13C NMR spectra of: (a) 5% Poly(SerDPP-co-Val), (b) 5% Poly(pSer-co-

Val).

173