Amino Acid-Based Polymeric Scaffold Fabrication And

AMINO ACID-BASED POLYMERIC SCAFFOLD FABRICATION AND

MODIFICATION FOR BONE REGENERATION APPLICATIONS

A Dissertation

Presented to

The Graduate Faculty of The University of Akron

In Partial Fulfillment

of the Requirements for the Degree

Doctor of Philosophy

Shan Li

May, 2018

i AMINO ACID-BASED POLYMERIC SCAFFOLD FABRICATION AND

MODIFICATION FOR BONE REGENERATION APPLICATIONS

Shan Li

Dissertation

Approved: Accepted:

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

Committee Member Dean of the College Dr. Yu Zhu Dr. Eric J. Amis

Committee Member Dean of the Graduate School Dr. Darrell H. Reneker Dr. Chand Midha

Committee Member Date Dr. Toshikazu Miyoshi

Committee Member Dr. Rebecca Kuntz Willits

ii ABSTRACT

Bone tissue engineering has evolved into an inter-disciplinary field of chemistry, engineering, and biology to regenerate defective tissues or organs since its emergence. Polymeric scaffolds represent one of the key components in bone tissue engineering and are widely used due to the low cost, unlimited supply, biointegrity, biodegradability, bioresorbability, tunable mechanical properties, non-toxicity and processability. It is designed to act as a 3D template to provide mechanical support and guide cells to form new tissue by mimicking ECM. Amino acid based poly(ester urea)s (PEUs) are high modulus, biodegradable, and non-toxic thermoplastic polymers, which have been synthesized and characterized by the Becker lab for their applications in tissue engineering. However, their inherent radiolucent and bioinert properties limit clinical application. Efforts have been made to modify PEUs with radiopacity for in vivo detection with X-rays and osteoinductivity via growth factor delivery for bone inducement.

The advances in new technologies bring 3D printing to the scaffold fabrication.

Unlike the traditional fabrication methods, 3D printing adds materials layer by layer under the guide of a computer, which saves the cost of mold and ensures the high reproducibility of the scaffold. Via CAD design, the architecture of scaffolds could be controlled to meet the criteria for cell penetration and tissue growth. Especially when coupled with medical imaging such as computerized tomography (CT) and magnetic resonance imaging (MRI), the patient specific scaffolds could be 3D printed, which is useful in the clinical aspect.

iii Unfortunately, one challenge with 3D printing is the delivery of growth factors since their structure and activity will be affected under the 3D printing processing conditions such as photochemical crosslinking or high temperature. Here the post-

printing PEU scaffold surface modification with OGP [10-14] and BMP-2 [73-92] via

copper-catalyzed azide alkyne cycloaddition (CuAAC) is reported to resolve this

problem. The in vitro hMSCs osteogenic differentiation study demonstrated the

enhancement effect of both peptides in terms of ALP activity, gene expression, protein

expression, and calcium deposition. The in vivo study with a 8 mm rat cranial critical

size defect model confirms the results from in vitro that more new bone was formed in

the peptide functionalized samples using µ-CT 3D scanning, H&E staining and

Goldner’s trichrome staining. Additionally, under X-ray, the iodinated PEUs rendered

clear images showing 3D architecture while the non-iodine functionalized groups

were undetectable.

ACKNOWLEDGEMENTS

This dissertation would not have been possible without the help from other

people. First and foremost I would like to express the deepest appreciation to my

advisor, Dr. Matthew L. Becker, for all his constant support, valuable help, and patient

guidance during my graduate study. He has been a very good example as researcher and advisor and it has been my great honor to work with him. I am also very grateful to my dissertation committee: Dr. Yu Zhu, Dr. Darrell H. Reneker, Dr. Abraham Joy

and Dr. Rebecca Kuntz Willits for their comments and help in the final stage of work.

I appreciate the help from Dr. Rebecca Kuntz Willits in the animal study and

the student in College of Polymer Science and Engineering, especially Dr. Fei Lin, Dr.

Yanyi Xu, Dr. Jukuan Zheng, Dr. Kun Yang, Jiayi Yu, Karissa Hagen, Derek Luong and all my group members for their collaboration, discussion, and encouragement in my research.

Finally, special thanks go to my family. For my parents and grandparents who raised me with love and encouragement. For my little brother who gave me understanding. For my husband, Tian Liang, for his love and support.

TABLE OF CONTENTS

Page

TABLE OF CONTENTS ...... vi LIST OF FIGURES ...... x LIST OF SCHEMES ...... xvi LIST OF TABLES ...... xvii CHAPTER I. INTRODUCTION ...... 1 1.1 Bone defect repair ...... 1 1.2 Scaffold ...... 3 1.2.1 Synthetic flexibility ...... 5 1.2.2 Chemical functionality ...... 6 1.2.3 Mechanical properties ...... 9 1.2.4 Degradation property ...... 10 1.3 Scaffold fabrication by 3D printing...... 11 1.4 Scaffolds modification ...... 17 1.4.1 Radiopacity ...... 17 1.4.2 Osteoinductivity...... 21 1.4.2.1 Bone morphogenic proteins (BMPs) ...... 21 1.4.2.2 Osteogenic growth peptide (OGP) ...... 25 II. MATERIALS AND INSTRUMENTS ...... 29 2.1 Materials ...... 29 2.2 Instruments ...... 31 III. RADIOPAQUE, IODINE FUNCTIONALIZED PHENYLALANINE-BASED POLY(ESTERUREA)S ...... 36 3.1 Abstract ...... 36 3.2 Introduction ...... 37

vi 3.3 Experimental Section ...... 40 3.3.1 Materials ...... 40 3.3.2 Characterization of chemical structure and thermal properties ...... 41 3.3.3 Synthesis of di-p-toluene sulfonic acid salt of bis-L-phenylalanine-1,6- hexanediol-diester (1-PHE-6 monomer) and di-p-toluene sulfonic acid salt of bis- 4-I-L-phenylalanine-1,6-hexanediol-diester (1-iPHE-6 monomer) 41 3.3.4 Synthesis of bis-L-phenylalanine-1,6-hexanediol-diester PEU (poly(1-PHE- 6)), bis-4-I-L-phenylalanine-1,6-hexanediol-diester PEU (poly(1-iPHE-6)), copolymers of 1-iPHE-6 monomer and 1-PHE-6 monomer (1:4 molar ratio, poly(1- iPHE-6)0.24-co-poly(1-PHE-6)0.76) and co-poly(ester urea) of 1-iPHE-6 monomer and 1-PHE-6 monomer (3:4 molar ratio, poly(1- iPHE-6)0.44-co- poly(1-PHE-6)0.56)...... 43 3.3.5 PEU films and 3D porous scaffold preparation and characterization ...... 46 3.3.6 In vitro cell viability and spreading characterization...... 47 3.4 Results and discussion ...... 49 3.4.1 Synthesis of L-phenylalanine-based and 4-I-L-phenylalanine-based poly(ester urea)s ...... 49 3.4.2 Thermal properties of PEUs ...... 53 3.4.3 Mechanical properties of bulk PEU films ...... 55 3.4.4 Radiopacity of PEUs...... 56 3.4.5 PEU 3D porous scaffolds analysis ...... 59 3.4.6 Cell viability and spreading assay ...... 62 3.5 Conclusion ...... 64 3.6 Acknowledgement ...... 65 IV. ENHANCED OSTEOGENIC ACTIVITY OF POLY(ESTER UREA) SCAFFOLDS USING FACILE POST-3D PRINTING PEPTIDE FUNCTIONALIZATION STRATEGIES ...... 66 4.1 Abstract ...... 66 4.2 Introduction ...... 67 4.3 Experimental Section ...... 70 4.3.1 Materials ...... 70 4.3.2 Characterization of chemical structure and thermal properties ...... 70 4.3.3 Synthesis of di-p-toluene sulfonic acid salt of bis-L-phenylalanine-1,6- hexanediol-diester (1-PHE-6 monomer) ...... 71

vii 4.3.4 Synthesis of di-hydrochloride acid salt of bis-4-propargyl-L-tyrosine-1,6- hexanediol-diester (1-pTYR-6 monomer) ...... 71 4.3.5 Synthesis of homopolymer of 1-PHE-6 monomer (poly(1-PHE-6)) ...... 72 4.3.6 Synthesis of copolymer of 1-pTYR-6 monomer and 1-PHE-6 monomer (2:98 molar ratio, poly[(1-pTYR-6)0.02-co-(1-PHE-6)0.98]) ...... 72 4.3.7 Synthesis of azide-derivatized peptide ...... 73 4.3.8 3D printed PEU porous scaffold preparation and characterization ...... 73 4.3.9 Scaffold surface peptide immobilization ...... 75 4.3.10 Human mesenchymal stem cell (hMSC) culture ...... 75 4.3.11 Immunohistochemical staining and alizarin red staining ...... 76 4.3.12 Alkaline phosphatase (ALP) activity assay ...... 77 4.3.13 Quantitative real time reverse transcription polymerase chain reaction (real time RT-PCR) ...... 78 4.3.14 Statistics ...... 79 4.4 Result ...... 79 4.4.1 Polymer synthesis ...... 79 4.4.3 Filament geometry ...... 84 4.4.4 Printing of scaffolds ...... 84 4.4.5 Scaffold surface functional group density ...... 84 4.4.6 Effects of immobilized peptide on hMSCs osteogenic differentiation ...... 86 4.5 Discussion ...... 91 4.6 Conclusion ...... 96 4.7 Acknowledgement ...... 96 V. CRITICAL SIZED CRANIAL DEFECT REPAIR USING 3D PRINTED RADIOPAQUE POLY(ESTER UREA) SCAFFOLD MODIFIED WITH BIOMIMETIC PEPTIDES THAT ENHANCE BONE REGENERATION ...... 98 5.1 Abstract ...... 98 5.2 Introduction ...... 99 5.3 Experimental Section ...... 103 5.3.1 Materials ...... 103 5.3.2 Characterization of chemical structure and thermal properties ...... 103 5.3.3 Synthesis of di-p-toluene sulfonic acid salt of bis-L-phenylalanine-1,6- hexanediol-diester (1-PHE-6 monomer)...... 104

viii 5.3.4 Synthesis of di-p-toluene sulfonic acid salt of bis-4-I-L-phenylalanine-1,6- hexanediol-diester (1-iPHE-6 monomer) ...... 104 5.3.5 Synthesis of di-hydrochloride acid salt of bis-4-propargyl-L-tyrosine-1,6- hexanediol-diester (1-pTYR-6 monomer) ...... 105 5.3.6 Synthesis of copolymer of 1-PHE-6 monomer and 1-iPHE-6 monomer (89:11 M ratio, poly[(1-iPHE-6)0.11-co-(1-PHE-6)0.89]) ...... 105 5.3.7 Synthesis of terpolymer of 1-pTYR-6 monomer, 1-iPHE-6 monomer, and 1-PHE-6 monomer (2:11:87 M ratio, poly[(1-pTYR-6)0.02-co-(1-iPHE- 6)0.11-co- (1-PHE-6)0.87])……………………………………………………106 5.3.8 Synthesis of azide-derivatized peptide ...... 106 5.3.9 PEU porous scaffold 3D printing and surface functionalization ...... 107 5.3.10 Human mesenchymal stem cell (hMSC) culture ...... 108 5.3.11 In vitro hMSC osteogenic differentiation on 3D printed peptide functionalized scaffolds ...... 108 5.3.12 In vivo rat critical size defect recovery study using 3D printed scaffolds ...... 111 5.3.12.1 Animal surgeries ...... 111 5.3.12.2 Micro-CT evaluation ...... 112 5.3.12.3 Histology analysis ...... 113 5.3.13 Statistics ...... 114 5.4 Result ...... 114 5.4.1 Polymer synthesis ...... 114 5.4.2 Thermal properties of poly(ester urea)s ...... 118 5.4.3 3D printed scaffold characterization ...... 120 5.4.4 In vitro hMSCs osteogenic differentiation on peptide functionalized scaffolds ...... 121 5.4.5 In vivo rat cranial critical size defect recovery ...... 124 5.5 Discussion ...... 129 5.6 Conclusion ...... 133 5.7 Acknowledgement ...... 133 VI. SUMMARY ...... 134 REFERENCES ...... 137 APPENDIX ...... 167

ix LIST OF FIGURES

Figure Page

1.1. Bone remodeling cycle. (A): Bone lining cells resting on the bone surface; (B): Bone resorption stage: under the stimulus of micro-damage or other biomechanical stimuli, the bone lining cells detach from the bone surface and active osteoclasts promote bone matrix digestion. (C): Reversal phase: monocytes differentiate into macrophages to remove the debris. (D): Bone formation stage: MSCs-derived osteoblasts secrete organic bone matrix proteins and then (E) mineralize to form calcified bone. Figure reproduced with permission from ref. 7. Copyright © 2016 Revotech Press...... 2

1.2. Chemical structure of PEUs synthesized from amino acid and diols, showing the characteristic urea and ester bonds of PEUs...... 5

1.3. Synthesis of PEUs derived from different combination of amino acids (L-analine, L-2-aminobutyric acid, L-isoleucine or L-phenylalanine) and diols (1,6- hexanediol, 1,8-octanediol, 1,10-decanediol, 1,12-dodecanediol). Figure reproduced with permission from ref. 25. Copyright © 2017 American Chemical Society ...... 6

1.4. PEUs with pendent “clickable” groups (alkyne, azide, alkene, tyrosine−phenol, and ketone groups) via the modification of tyrosine amino acid building blocks. Biological molecules could be attached covalently by click reaction to expand the material application in vitro and in vivo. Figure reproduced with permission from ref. 26. Copyright © 2013 American Chemical Society...... 7

1.5. Synthesis and functionalization of monomers and polymers through the starting materials to introduce the reactive handle for future click reaction (amino acid and diol) and OGP peptide conjugation via click reaction. Figure reproduced with permission from ref. 28. Copyright © 2015 American Chemical Society .... 8

1.6. The process from medical imaging acquisition (usually CT or MRI) followed by the translation to a 3D CAD model and generation a visualized motion program, then to the produce of a human scale ear by 3D printing of the cell-laden hydrogels together with PCL. Figure reproduced with permission from ref. 48. Copyright © 2016 Nature America, Inc ...... 13

1.7. The process from medical imaging acquisition to the produce of 3D printed human scale mandible bone. (a): 3D CAD model of mandible bone defect generated from CT image data. (b): Visualized motion program for the 3D

x printing of the mandible Bone defect construct using CAM software. Green, blue and red lines indicate the printing path of different materials. (c): 3D printing process. (d): Photograph of the 3D printed mandible bone defect construct; (e): Alizarin Red S staining for calcium deposition to confirm the osteogenic differentiation of the construct. Figure reproduced with permission from ref. 48. Copyright © 2016 Nature America, Inc ...... 14

1.8. The produce of 3D printed calvarial bone defect construct by 3D printing. (a): Visualized motion program for the calvarial bone defect construct printing (top) and photograph of the 3D printed construct (bottom). (b): Scanning electron microscope images of the 3D printed calvarial bone defect construct showing the micro-structure. (c): implantation of the printed bone construct. Figure reproduced with permission from ref. 48. Copyright © 2016 Nature America, Inc……………………..14

1.10. The attenuation of X-rays when interacting with materials: (a) photoelectric effect, (b) Compton scattering, and (c) electron pair effect...... 20

1.11. BMPs signaling pathway. BMPs bind to the type II and type I receptors to form a heteromeric complex, the phosphorylate of the type I receptor activates the Smad pathway through the phosphorylation of Smad1/5/8 receptors, followed by the association with Smad4 and translocation to the nucleus to stimulate the gene expression. I-Smads inhibit receptor activation. BMPs can also signal via the non- Smad pathways, for example, MAPK pathway by regulating the Smads receptor activation and translocation to the nucleus. Figure reproduced with permission from ref. 67. Copyright © 2012 Federation of European Biochemical Societies. Published by Elsevier B.V...... 24

1.12. Peptide sequence of OGP. It consists of two domains called accessory domain and active domain. The accessory domain is responsible for the binding of peptide to OGP binding protein, mainly α2-macroglobulin (α2M) and the active domain drives the cell proliferation and osteogenic differentiation. Figure reproduced with permission from ref. 84. Copyright © 2015 Wiley Periodic als, Inc ...... 25

1.13. OGP signaling pathway for the proliferation of osteoblastic cells (MC3T3-E1). After disassociation with α2M, the active domain [10-14] is proteolytically cleaved from OGP and then bound to Gi protein to activate the downstream mitogen-activated protein (MAP) kinase signaling pathway and DNA synthesis to drive the proliferation of osteoblast cell lines. Figure reproduced with permission from ref. 84. Copyright © 2015 Wiley Periodic als, Inc ...... 26

xi 1 3.1. H-NMR (DMSO-d6) of PEUs. (a) homopolymer of 1-iPHE-6 monomer (poly(1- iPHE-6), which has ring signals at 6.95 and 7.6 ppm, characteristic of a para- substituted aromatic ring. (b) copolymer of 1-iPHE-6 monomer/1-PHE-6 monomer at a 3:4 molar ratio (44% poly(1-iPHE-6) and 56% poly(1-PHE-6)); (c) copolymer of 1- iPHE-6 monomer/1-PHE-6 monomer at a 1:4 molar ratio (24% poly(1-iPHE-6) and 76% poly(1-PHE-6) in the copolymer). They all possess characteristic peaks of both poly(1-PHE-6) and poly(1-iPHE-6). (d) homopolymer of 1-PHE-6 monomer (poly(1- PHE-6)), possessing proton resonances characteristic of the benzyl group, 7.1 to 7.3 ppm...... 51

3.2. FT-IR of PEUs. (a) iodinated phenylalanine-based poly(1-iPHE-6); (b) copolymer of 44% poly(1-iPHE-6) and 56% poly(1-PHE-6); (c) copolymer of 24% poly(1-iPHE-6) and 76% poly(1-PHE-6); (d) phenylalanine-based poly(1- PHE-6). All spectra show the characteristic ester and urea peaks. For iodinated polymers, (a), (b) and (c), the characteristic C-I stretching signal at 1007 cm-1 increased with greater iodine content...... 52

3.3. Stress-strain curves of poly(1-PHE-6), poly(1-iPHE-6)0.24-co-poly(1-PHE-6)0.76 and poly(1-iPHE-6)0.44-co-poly(1-PHE-6)0.56 measured by dynamic mechanical analysis with a strain rate of 2.5 %/min at room temperature. Three samples were tested for each polymer film. The elastic moduli were obtained in the linear region of the stress-strain curve and the average value of three samples was calculated. Incorporation of iodine in poly(1-PHE-6) makes the normally brittle PEU more ductile. However, with increasing iodine content, the PEUs again become brittle. Poly(1-iPHE-6) homopolymer for example was too brittle to be measured by DMA. The elastic moduli of PEUs decreased following iodine modification……………………………………………………………………55

3.4. Micro-CT images of an aluminum wedge 0.5–2.5 mm in 0.5 mm steps (a-e) and PEU films with different iodine content (f-i) with 0.5 mm thickness. Radiopacity of PEUs increases with increasing iodine content. The poly(1-iPHE-6) film (i) has comparable radio contrast to that of the aluminum reference with a thickness of 1 mm (b), the poly(1-iPHE-6)0.44-co-poly(1-PHE-6)0.56 film (h) has comparable radio contrast as that of the aluminum reference with a thickness of 0.5 mm (a), and the radiopacity of poly(1-iPHE-6)0.24-co-poly(1-PHE-6)0.76 film is lower than that of the 0.5 mm thick aluminum reference, but is much higher than that of the poly(1-PHE-6) film (f)………………………………………..57

3.5. Reconstruction slices of Micro-CT 3D scanning of porous scaffolds with different iodine content under the same scanning conditions. The images show the cross- section of the scaffold throughout the sample. It is difficult to see the poly(1- PHE-6) scaffold (a) structure because of the poor radiopacity. Poly(1-iPHE- 6)0.24-co-poly(1- PHE-6)0.76 (b) and poly(1-iPHE-6)0.44-co-poly(1-PHE-6)0.56) (c) show the internal structure (pore size, pore type and interconnectivity) of the scaffolds. The porosity of poly(1-PHE-6), (poly(1-iPHE-6)0.24-co-poly(1-PHE- 6)0.76 and poly(1-iPHE-6)0.44-co- poly(1-PHE-6)0.56) scaffolds were calculated to xii be (90±1.6)%, (85±0.4)% and (88±0.5)%, respectively...... 59

3.6. MC3T3 cell viability on PEU films (n=3). 10 images were used for calculating cell viability on each sample. (a): poly(1-PHE-6) film; (b): poly(1-iPHE-6)0.24- co- poly(1-PHE-6)0.76 film; (c): poly(1-iPHE-6)0.44-co-poly(1-PHE-6)0.56) film. (d): Comparison of cell viability on PEU films with different iodine content shows no significant difference in cell viability ...... 62

3.7. MC3T3 cell spreading on PEU films (n=3). 20 images were used for quantification of cell aspect ratio and cell area for each sample. (a) and (d): poly(1- PHE-6) film; (b) and (e): poly(1-iPHE-6)0.24-co-poly(1-PHE-6)0.76 film; (c)and (f): poly(1-iPHE-6)0.44-co-poly(1-PHE-6)0.56) film. (g): Comparison of cell aspect ratio on PEU films with different iodine content. (h): Comparison of cell area on PEU films with different iodine content, which shows that cells are well spread on PEU films, supporting cell viability results...... 63

1 4.1. H-NMR spectra (DMSO-d6) of poly(1-PHE-6) and poly[(1-pTYR-6)0.02-co-(1- * PHE-6)0.98]. The resonance b (δ = 4.35-4.39 ppm) is set as the reference peak to calculate the polymer composition. Poly(1-PHE-6): homopolymer of 1-PHE-6 monomer; poly[(1-pTYR-6)0.02-co-(1-PHE-6)0.98]: random copolymer of 1- pTYR-6 monomer and 1-PHE-6 monomer at a molar ratio of 2:98...... 80

4.2. UV-visible Spectra of PEUs. The peak at 257 nm is assigned to phenylalanine absorption due to π−π* transition from aromatic ring. n, π-hyperconjugation from o- propargyl substitution red shifts the π→π* transition from 257 nm to longer wavelength (278 nm). The composition can be easily confirmed by a ratio of the two transitions...... 81

4.3. 3D printed scaffold fabrication procedure. A Dynisco LCR 7000 Capillary Rheometer (a) was used to extrude continuous PEU filament with diameter of 1.75 mm (b). After fused deposition modeling (FDM) 3D printing, porous scaffolds (8 mm in diameter, and 1 mm in thickness) were obtained. (c): Cartesio 3D printer; (d): A schematic of how a FDM 3D printer works; (e): A PEU scaffold being printed. (f): The design of architecture within a scaffold; (g): A photograph of 3D printed PEU scaffold; (h and i): Micro-CT 3D reconstruction images of a 3D printed PEU scaffold. The filament diameter is measured to be around 200 µm and pore size is approximately 400 µm...... 83

4.4. Quantification of the surface functional groups on the scaffolds by UV-visible spectroscopy (a) and fluorescence spectroscopy (c). (b): Calibration curve for poly[(1- pTYR-6)0.02-co-(1-PHE-6)0.98] (UV-visible absorbance at 257 nm). (d): Calibration curve for Chromeo 488 azide fluorescence emission at 511 nm. Black curve: Propargyl PEU (after EtO) + dye without Cu catalyst; Blue curve:

xiii Propargyl PEU (after EtO) + dye with Cu catalyst; Red curve: Propargyl PEU (before EtO) + dye with Cu catalyst. Concentrations of polymer and dye are calculated from calibration curves (insets)...... 86

4.5. The summary of osteogenic differentiation of hMSCs within 3D printed PEU scaffolds. (A-C): Real time RT-PCR amplification of RUNX2 (early stage marker), BSP (middle stage marker) and OCN (late stage marker) on 2, 3 and 4 weeks in vitro with GAPDH was endogenous control and hMSCs as calibrator sample. Data represent mean ±SEM of 4 determinations. * indicates p value < 0.05, and ** indicates p value < 0.01. (D and E): Immunofluorescence staining for cell nuclei (blue) and BSP (green), OCN (red) and RUNX2 (green) after 2 and 4 weeks in vitro...... 89

4.6. ALP expression (normalized by DNA in sample) of hMSCs within scaffolds after 2 and 4 weeks in vitro. Consistent with the real time RT-PCR results, at 2 weeks, the OGP functionalized scaffolds show higher ALP expression than the poly(1- PHE-6) control scaffold. * indicates p value < 0.05...... 90

4.7. Bright field histological images for calcium deposition by alizarin red staining after 2 and 4 weeks culture in vitro. Fibers are outlined by dotted line. These images clearly demonstrate that the presence of tethered OGP or BMP-2 enhanced hMSCs osteogenic differentiation and mineralization...... 91

1 5.1. H-NMR spectra (DMSO-d6) of poly[(1-iPHE-6)0.11-co-(1-PHE-6)0.89] and * poly[(1-pTYR-6)0.02-co-(1-iPHE-6)0.11-co-(1-PHE-6)0.87]. The resonance b (δ = 4.29-4.42 ppm) is set as the reference peak to calculate the polymer composition. Poly[(1- iPHE-6)0.11-co-(1-PHE-6)0.89]: random copolymer of 1-PHE-6 monomer and 1-iPHE- 6 monomer at a molar ratio of 10:90; poly[(1-pTYR-6)0.02-co-(1- iPHE-6)0.11-co-(1-PHE-6)0.87]: random terpolymer of 1-pTYR-6 monomer, 1- iPHE-6 monomer and 1- PHE-6 monomer at a molar ratio of 2:10:88………115

5.2. UV-visible spectra of PEUs. The peak at 257 nm corresponds to L-phenylalanine absorption due to π−π* transition from aromatic ring (B band), which in the propargyl functionalized PEU red shifts to 278 nm due to n, π-hyperconjugation with o- propargyl substitution. The strong absorbance at 238 nm results from the red shift of π−π* transition (K band) due to n, π-hyperconjugation with iodine atom ...... 115

5.3. 3D porous scaffold printing and characterization. (A): a schematic of the working principle of FDM printing; (B): 3D printed PEU scaffold; (C-E): µ-CT 3D scanning results of scaffolds: poly(1-PHE-6) control (C), poly[(1-iPHE-6)0.11-co- (1-PHE-6)0.89] scaffold (D), and poly[(1-pTYR-6)0.02-co-(1-iPHE-6)0.11-co-(1- PHE-6)0.87] scaffold (E). From top to bottom: shadow projection of scaffolds under X-ray, 2D reconstructed cross-section of scaffolds; and 3D reconstructed xiv images. The iodinated PEU scaffolds show higher radiocontrast than the poly(1- PHE-6) control...... 120

5.4. Effect of OGP or BMP-2 peptide on the protein expression of osteogenic markers, RUNX2 (early stage marker), BSP (middle stage marker) and OCN (late stage marker) in hMSCs cultured in 3D printed scaffolds in vitro. Immunofluorescence staining after 2 weeks and 4 weeks with cell nuclei stained with blue color, BSP with green color, OCN with red color and RUNX2 with green color. The introduction of peptide increased the expression of osteoblast protein markers ...... 121

5.5. Effect of OGP or BMP-2 peptide on the ALP activity (normalized by total DNA) of hMSCs cultured in 3D printed scaffolds after 2 and 4 weeks in vitro. Significant difference was detected at early stage of osteoblast differentiation (2 weeks) on the peptide attached samples. (N=3, each sample were tested three times). Groups that do not share a letter are significantly different...... 123

5.6. Bone formation of each group after 4 and 12 weeks. Visible in the µ-CT 3D results, new bone grew inside the pores. Non-iodine functionalized scaffolds showed no contrast with the tissue, so they were invisible under X-ray after implantation. However, from the iodinated PEU groups, the scaffold can be distinguished from the surrounding tissue and bone. In the µ-CT 2D cross- section and histology images stained by H&E and Goldner’s trichrome, new bone grew from the periphery and the dural side. Goldner’s trichrome stained uncalcified osteoid red and mineralized bone green. Quantification results from µ-CT (b) and Goldner’s trichrome (c) demonstrate that BMP-2 enhances new bone growth in the cranial defect. Groups that do not share a letter are significantly different… ...... 124

5.7. Bone formation of each group after 4 and 12 weeks. Visible in the µ-CT 3D results, new bone grew inside the pores. Non-iodine functionalized scaffolds showed no contrast with the tissue, so they were invisible under X-ray after implantation. However, from the iodinated PEU groups, the scaffold can be distinguished from the surrounding tissue and bone. In the µ-CT 2D cross- section and histology images stained by H&E and Goldner’s trichrome, new bone grew from the periphery and the dural side. Goldner’s trichrome stained uncalcified osteoid red and mineralized bone green. Quantification results from µ-CT (b) and Goldner’s trichrome (c) demonstrate that BMP-2 enhances new bone growth in the cranial defect. Groups that do not share a letter are significantly different… ...... 127

xv 5.8 The histological images of PEU + BMP-2 scaffold at 12 weeks post-surgery stained with H&E (a-d) and Goldner’s trichrome (e-h), showing detailed new bone growth information. (a): H&E image of the defect site. New bone binds with the host bone. (b-d): scaffold pores are filled with new bone and fibrous connective tissue. New bone is formed in two configurations: lamellar bone under the scaffold and woven bone in the pores of scaffold (b and d). (e): Goldner’s trichrome image of the defect site. (f-h) shows that the fibrous connective tissue is composed of unmineralized collagen (osteoid) and mineralized collagen. S: scaffold; HB: host bone; NB: new bone; FT: fibrous connective tissue; BV: blood vessel; and Ob: osteoblast...... 128

xvi LIST OF SCHEMES

Scheme Page

3.1. The two-step synthesis of amino acid-based poly(ester urea)s (PEUs). R could be an H or I atom, which results in two distinct monomers (1-PHE-6 monomer and 1- iPHE-6 monomer). Homopolymerization of the two monomers results in poly(1-PHE-6) and poly(1-iPHE-6). Copolymerization of these two monomers with different feed ratios results in poly(1-iPHE-6)-co- poly(1-PHE-6)s with varied iodine content………………………………50

4.1. Interfacial polymerization of amino acid-based poly(ester urea)s (PEUs)……..79

xvii LIST OF TABLES

Table Page

1.1 Tunable mechanical properties of PEUs with different amino acid and diol ...... 10

1.2. Comparison between different 3D printing techniques ...... 16

3.1. PEUs composition from 1H-NMR and FT-IR ...... 53

3.2. Characterization summary of PEUs ...... 54

3.3. Compression modulus of PEU porous scaffolds ...... 61

4.1. Characterization summary of PEUs ...... 82

5.1. Characterization summary of PEUs ...... 117

xvii

CHAPTER I

INTRODUCTION

1.1 Bone defect repair

Bone defect, by definition, is the loss/fracture of bone. It presents a main

health concern for people all over the world and a challenging project for bone

salvage surgeons.1 Bone has the ability to self-heal due to its constant remodeling,

which includes the bone resorption and formation processes.2-4 The bone remodeling process includes five stages: resting phase, resorption phase, reversal phase, bone formation and mineralization, of which the resorption phase and bone formation phase receive the majority of research focus. In the bone resorption process, pre-osteoclasts, originating from hematopoietic stem cells, differentiate into osteoclasts, which then attach to the bone surface to create a local acidic environment. The old bone matrix is digested by proteolytic enzymes and the secreted acid, and then the growth factors

2+ 3- and bone related mineral ions, such as Ca and PO4 , are released. In the bone

formation process, the growth factors that are released from the bone matrix in the

resorption process induce mesenchymal stem cells to differentiate into osteoblasts for

bone organic matrix formation (osteoid) and subsequent mineralization in which

4-7 calcium phosphate crystals are deposited precisely within the bone matrix.

Figure 1.1. Bone remodeling cycle. (A): Bone lining cells resting on the bone surface; (B): Bone resorption stage: under the stimulus of micro-damage or other biomechanical stimuli, the bone lining cells detach from the bone surface and active osteoclasts promoting bone matrix digestion. (C): Reversal phase: monocytes differentiate into macrophages to remove the debris. (D): Bone formation stage: MSCs-derived osteoblasts secrete organic bone matrix proteins and then (E) mineralize to form calcified bone. Figure reproduced with permission from ref. 7. Copyright © 2016 Revotech Press. Small bone defects can be repaired spontaneously through a “four-step-model” which includes bone remodeling.8 However, if defects surpass a critical size, which is

defined by the smallest size of bone defect that cannot heal spontaneously, then

surgical intervention is needed to regenerate new bone in order to restore the bone

function, also called bone grafting.9,10 According to a market research, the 2014 global bone grafts and substitutes market was worth 2.35 billion USD and is expected to reach 3.48 billion USD in 2023 due to the increasing demand and elevating industry standards. The ideal bone graft should meet the following requirements: non-

toxic, osteoconductive, osteoinductive, low cost, easy to obtain, and similar to bone

2 with respect to morphology and mechanical properties.9,11-13 Osteoconduction means that bone is able to grow on the graft surface whereas osteoinduction is the process of inducing immature cells for osteoblast differentiation and bone formation.14 Bone grafts are categorized into autograft, allograft, xenograft and bone graft substitutes, based on their source.9,15,16 Autografts are harvested from the patient’s own bone. It is osteoconductive and osteoinductive, and there is no risk of disease transmission or immunogenicity.9 However, the amount of bone that can be removed from the patient is limited and there are many problems associated with bone harvest including extended surgery time, blood loss, bone defects at the donor site, hematoma and infection.16 Allografts resolve these problems by using bones from other human donors, deceased or living. It is osteoconductive, osteoinductive and more available than autografts. However, bone tissues from other people may transmit bacteria, viruses, or disease. It may also induce host immune rejection, leading to the failure of bone healing.9 While allografts are harvested from human donors, xenografts are harvested from animal donors making them highly available and low cost. However, there are severe risks of disease transmission and immune rejection.9 Most of the listed problems could be eliminated by a tissue engineering treatment, which drives

17 the bone graft research to synthetic bone grafts, also called scaffold.

1.2 Scaffold

Tissue engineering takes advantage of cells isolated from patients, 3D scaffolds and growth factors to restore the function or shape of lost/damaged tissues.

Since cells alone cannot form the 3D shape of tissues, scaffolds are used to act as

3 templates to direct cell proliferation and differentiation under certain biological

18 cues.

Over the last three decades, the research of tissue engineering has been

focused on scaffolds, especially polymeric scaffolds, which are classified as natural or

synthetic materials.18,19 When choosing scaffold materials, the following properties

should be taken into consideration: biointegrity; biodegradability; bioresorbability;

mechanical properties, toxicity and processability.15,20,21 The first criterion of tissue

engineered scaffolds is biointegrity. Cells can migrate and attach on the surface of

scaffolds and then penetrate into the scaffold, eventually leading to proliferation and

differentiation. There should be limited immune reaction and inflammatory response

after implantation. The ultimate aim of tissue engineering is to regenerate new body

tissue over a period of time to replace the implanted scaffolds. It is ideal that the

degradation time of the scaffolds is similar to the growth rate of new tissue. The

degradation by-products of scaffolds should be non-toxic and cleared through normal metabolic processes. The mechanical properties of scaffolds should be consistent with

15 the original tissue at the implantation site.

One class of such scaffold material is amino acid containing poly(ester urea)s

(PEUs). The first report of the synthesis of PEUs without the use of diisocyanates was

in 1997 by Katsarava et al. via active polycondensation.22,23 Since the material

properties depend on the polymer composition, chemical structure, and interactions

between polymer chains, PEUs hold the following properties due to the amino acid

and ester building blocks: (1) chemical functionality; (2) synthetic flexibility; (3)

tunable mechanical properties; (4) degradation under physiological conditions.

Figure 1.2. Chemical structure of PEUs synthesized from amino acids and diols, showing the characteristic urea and ester bonds of PEUs.

1.2.1 Synthetic flexibility

Figure 1.2 shows the chemical structure of PEUs, which are prepared via an

interfacial polycondensation method from amino acids and diols. As there are different combinations of amino acids and diols, PEUs could be simply designed according to structure-property relationships. Gao Reported the use of L-leucine as a starting material for vascular tissue engineering since the aliphatic side chain provided flexibility for PEU and the hydrogen bonding network in the urea groups afforded the mechanical strength.24 Peterson summarized the synthesis and shape memory

properties of PEUs from L-analine, L-2-aminobutyric acid, L-isoleucine or L-

phenylalanine and 1,6-hexanediol, 1,8-octanediol, 1,10-decanediol, 1,12-dodecanediol

25 as demonstrated in Figure 1.3.

Figure 1.3. Synthesis of PEUs derived from different combination of amino acids (L- analine, L-2-aminobutyric acid, L-isoleucine or L-phenylalanine) and diols (1,6- hexanediol, 1,8-octanediol, 1,10-decanediol, 1,12-dodecanediol). Figure reproduced with permission from ref. 25. Copyright © 2017 American Chemical Society.

1.2.2 Chemical functionality

Functional groups such as hydroxyl, amine, carboxyl, thiol, and alkyne groups

could be introduced into PEUs easily by choosing the functional group bearing amino

acids or diols as starting materials. This enables further post bioactive species

conjugation or other chemistry modifications to polymeric device for surface property

control. Lin et al. reported the successful synthesis of PEUs carrying pendent

“clickable” groups (alkyne, azide, alkene, tyrosine−phenol, and ketone groups) via the modification of tyrosine amino acid building blocks.26 Following the electrospinning

of PEUs, the presence of “clickable” groups on the fiber surface was confirmed by the

fluorescence dye with corresponding reactive groups using bio-orthogonal reactions.

Bioactive peptides were subsequently attached to the nanofiber surface in the same

manner to yield bioactivity. It is also reported by Zhou and his coworker that the

pendant catechol groups were introduced to tyrosine-derived PEUs through the amino

acid modification to prepare biomimetic adhesives.27 Comparable adhesive strength to

the commercially available fibrin glue was achieved by the cross-linking of catechol

pendant groups in the PEUs. In the study conducted by Policastro, alkene groups for

6 polymer crosslinking and alkyne groups for peptide conjugation were incorporated

into PEUs by the diol and amino acid building blocks, respectively, in a

copolymerization method.28 In vitro cell studies demonstrated that the OGP promoted

hMSCs osteogenic differentiation and in vivo animal studies revealed that the tissue

and scaffolds were well integrated without any inflammatory response and promoted

osteogenesis along with angiogenesis in the OGP functionalized group.

Figure 1.4. PEUs with pendent “clickable” groups (alkyne, azide, alkene, tyrosine−phenol, and ketone groups) via the modification of tyrosine amino acid building blocks. Biological molecules could be attached covalently by click reaction to expand the material application in vitro and in vivo. Figure reproduced with permission from ref. 26. Copyright © 2013 American Chemical Society.

Figure 1.5. Synthesis and functionalization of monomers and polymers through the starting materials to introduce the reactive handle for future click reactions (amino acid and diol) and OGP peptide conjugation via click reactions. Figure reproduced with permission from ref. 28. Copyright © 2015 American Chemical Society.

8 1.2.3 Mechanical properties

As scaffolds materials, it is essential for PEUs to withstand applied stress and

strain over time. The mechanical properties of polymers are dependent on the polymer

composition, structure, and the resulting chain interaction (branching, cross-linking, polarity, hydrogen bonding and crystallinity). For example, L-phenylalanine-based

poly(1-PHE-6) showed elastic modulus as high as 6.1 GPa due to the strong hydrogen

bonding and semi-crystalline structure within the network. By replacing the L-

phenylalanine with L-leucine, the elastic modulus decreased from 6.1 GPa to 4.4

GPa.29 The higher elastic modulus of poly(1-PHE-6) than poly(1-LEU-6) could be explained by the fact that the rigid aromatic side chain hindered the movement of polymer chains and the polar aromatic ring increased the interchain interactions. By increasing the ester segment length (from 1,6-hexanediol, 1,8-octanediol, 1,10-

decanediol, to 1,12-dodecanediol) in the polymer chain, the elastic modulus

decreased.30 Since the ester part contributes to polymer chain flexibility, longer ester

segment increases the chain flexibility and decreases the elastic modulus. Similarly,

Gao et al. found that the elastic modulus of L-leucine based poly(1-LEU-6), poly(1-

LEU-8), poly(1-LEU-10), and poly(1-LEU-12) was 1208 MPa, 651 MPa, 134 MPa, and 12 MPa, respectively. 31 Table 1.1 summarized the elastic modulus of L- phenylalanine and L-leucine based PEUs with varying diol chain length.

9 Table 1.1 Tunable mechanical properties of PEUs with different amino acid and diol.

Component 1,6-hexanediol 1,8-octanediol 1,10-decanediol 1,12-dodecanediol

L-phenylalanine 3100 MPa 2700 MPa 1200 MPa 500 MPa

L-leucine 1208 MPa 651 MPa 134 MPa 12 MPa

By introducing branches, the modulus of PEUs could also be tuned. Yu et al. reported the synthesis of a series of L-phenylalanine-based PEUs with controlled branching density in a copolymerization method using a triol as branching agent.32

The branched PEUs possessed a decreased elastic modulus when compared to the linear PEU (0% branching) because of the reduced polymer chain interaction. By increasing the ratio of branching unit incorporated, the typical polymer chain packing was interrupted, leading to suppressed crystallinity and chain interaction, which therefore decreased elastic modulus. Overall, by choosing different amino acids and diol/triol, the mechanical properties of PEUs could be tuned for different applications from bone grafts to blood vessels.

1.2.4 Degradation property

In tissue engineering, degradation is necessary as it circumvents the issues related with the new tissue formation site and later implant retrieval.33,34 For degradable materials, cells penetrate into the scaffolds and secrete extracellular matrix

(ECM) while polymer scaffolds can vanish to leave space for new tissue.

Concomitantly another desirable property of biomaterials is controllable degradation, which could be adjusted according to the growth of newly formed tissues. All

10 degradable polymers undergo either surface erosion or bulk erosion according to a

theoretical model developed by Göpferich et al.35 The erosion mechanism depends on

the competition of water diffusivity inside the polymeric matrix and the polymer

backbone degradation rate. If the water diffusion into the matrix is faster than the

polymer backbone degradation rate, the polymeric matrix will undergo bulk erosion;

conversely, if the polymer degrades before water penetrates into the matrix,

degradation will be confined to the surface. The degradation property of PEUs could

be tuned by changing the water uptake property and chain degradation. For example,

PEUs degraded faster with longer ester chain segment incorporation.30 This could be

explained by the fact that longer ester chain segment increased the chain flexibility,

leading to easier water penetration into the polymer matrix.

1.3 Scaffold fabrication by 3D printing

Besides scaffold material, another key challenge in orthopedic surgery, tissue engineering and regenerative medicine research is how to fabricate 3D scaffolds with fine structures that allow for bone or more complicated tissue regeneration.36

Scaffolds should have an interconnected porous structure to ensure cell penetration,

nutrition supply, waste removal and new tissue formation can all function

simultaneously.37 In order to fabricate such scaffolds in a clinically and commercially

viable way, a number of conventional methods such as porogen (salt) leaching,

emulsion freeze drying, gas-foaming and phase separation have been explored.13,38-40

In a porogen leaching method, a polymer solution is cast onto salts and then after

solvent evaporation, the salts are leached from the scaffolds. This method prepares

scaffolds with controlled pore size, pore shape, porosity and surface area to volume

11 ratio; however, it involves the use of toxic solvents and also takes time for the solvent

to evaporate. Salt particles may be retained in the scaffolds, so this method can only

be used to prepare scaffolds with thin walls such that the soluble salt can be removed

from within the structure.39 Emulsion means the heterogeneous mixture of one liquid

dispersed in another immiscible liquid in droplets. Via the emulsion freeze drying

method, inter-connected pores with high porosity can be achieved. However, the pore

structure is hard to control and the pore size is usually less than 200 μm, which is not

suitable for new bone growth. The phase-separation technique takes advantage of the fact that poor solvent or cooling induces the phase separation of a homogeneous polymer solution.38,41 Similar to the emulsion freeze drying method, scaffolds with

inter-connected pores and high porosity can be prepared but the pore structure is

difficult to control. Using high pressure CO2, macroporous sponges can be fabricated,

in which a polymer/gas system is formed to create pores.38,42 However, this method

38 often results in closed pores which prevent cell penetration and tissue regeneration.

All the fabrication methods described above do not allow for the production of

3D porous scaffolds in any size with controlled pore size, structure and porosity. This

could be resolved by the recently emerged breakthrough technology, 3D printing, also

called additive manufacturing, which is a process that produces scaffolds layer by

layer directly from a computer model.37,40,43 It was developed by Sachs et al. in the

early 1990s at MIT and expanded its applications to consumer goods, medical

implants, automotive industry, aerospace industry, architecture industry and mold

industry.40,44,45 In the medical field, the key to tissue regeneration may come from a

3D printer since it allows on-site and on-demand fabrication of medical implants with

specifically designed structures to mimic the living tissue, especially when coupled

12 with medical imaging such as CT or MRI scanning.46,47 Hence 3D printing is a potential and powerful technology in tissue engineering and regenerative medicine.

Kang et al. reported the 3D printing of vascularized cellular scaffolds with clinically relevant size, shape and structural integrity. The mechanical stability was provided by the biodegradable polymers integrated, and shape of the scaffold was achieved using the imaging data as a computer model, followed by its translation into a motion program to control the movement of printing nozzle. The printed microchannel facilitated the nutrition transportation. Several constructs including ear

48 cartilage, mandible bone and calvarial bone were printed successfully.

Figure 1.6. The process from medical imaging acquisition (usually CT or MRI) followed by the translation to a 3D CAD model and the generation of a visualized motion program, to the produce of a human scale ear by 3D printing of the cell-laden hydrogels together with PCL. Figure reproduced with permission from ref. 48. Copyright © 2016 Nature America, Inc.

Figure 1.7. The process from medical imaging acquisition to the produce of 3D printed human scale mandible bone. (a): 3D CAD model of mandible bone defect generated from CT image data. (b): Visualized motion program for the 3D printing of the mandible Bone defect construct using CAM software. Green, blue and red lines indicate the printing path of different materials. (c): 3D printing process. (d): Photograph of the 3D printed mandible bone defect construct; (e): Alizarin Red S staining for calcium deposition to confirm the osteogenic differentiation of the construct. Figure reproduced with permission from ref. 48. Copyright © 2016 Nature America, Inc.

Figure 1.8. The produce of 3D printed calvarial bone defect construct by 3D printing. (a): Visualized motion program for the calvarial bone defect construct printing (top)

14 and photograph of the 3D printed construct (bottom). (b): Scanning electron microscope images of the 3D printed calvarial bone defect construct showing the micro-structure. (c): implantation of the printed bone construct. Figure reproduced with permission from ref. 48. Copyright © 2016 Nature America, Inc. There are several 3D printing techniques used in the biomedical area including inkjet printing, fused deposition modeling (FDM), laser-assisted printing (LAP), and stereolithography (SLA).49,50 Those techniques are summarized in Table 1.2.45 Here only FDM printing is described.

Fused deposition modeling was developed by Stratasys in 1991 and has become one of the most widely used 3D printing techniques in the industry because of the simplicity, a variety of materials available, low cost and wide-range applications.47

Figure 1.8 shows the principle of FDM printing.51 Thermoplastic filament is used as feeding material into the printer, melting inside the heating chamber and extruded out from the nozzle by a continuous driving force. The platform moves in X and Y directions to control the shape in each layer and after the deposition of the first layer, the platform lowered down for the deposition of the second layer on top of prior layer.

This process is repeated until the completion of the scaffolds. All the movement in

XYZ directions is controlled by computer design. Theoretically, any thermoplastic can be FDM printed, which makes FDM applicable for polymer processing.

15 Table 1.2. Comparison between different 3D printing techniques

Advantages Disadvantages Material choice inkjet simple printing process, low resolution and thermo/pH/

material flexible, low cost, discontinuous printing resulting photo-sensitive high speed in structural integrity problem,

nozzle clogging

FDM simple printing process, medium resolution, nozzle thermal sensitive

material flexible, low cost, clogging, high temperature

high speed, continuous printing

printing and good

mechanical integrity

LAP high resolution high cost, limited material photosensitive

choice, material wasting and

discontinuous printing resulting

in structural integrity problem

SLA high resolution, continuous high cost, material wasting photosensitive

printing with no artificial limited material choice

interfaces and good

mechanical integrity

1.4 Scaffolds modification

1.4.1 Radiopacity

Owing to desirable properties as scaffold materials and their potency in the scaffold fabrication via 3D printing technique, PEUs have been extensively explored by our lab as an implantation material. However, within this context, one problem arises: how can we “see” the scaffold after implantation? The fact that PEUs are invisible under X-ray significantly limits their application in the medical field, where

X-ray based techniques such as X-radiography and CT are widely used noninvasively to detect the position and analyze the morphology of implants. Therefore, it is clinically significant for PEUs to have radio contrast. Radio contrast comes from the

17 difference in the radiopacity (attenuation of X-ray) between objects that need to be distinguished.

In theory, when photons interact with materials, there are three main interaction mechanisms based on the energy level of incident photons and the atom

number of the materials.52 When the photon energy is low and interacting material

atom number is high, energy will be absorbed by the elevation of electrons in the

material to a higher energy level. The photons transfer their energy to the electrons,

which is called photoelectric (photon-electron) interaction. It is predominant when the

electron binding energy is slightly less than the photon energy. When the photon

energy is intermediate and the interacting material atom number is low, a portion of

the photon energy is absorbed and a new photon with reduced energy is produced.

Since the direction of the new photon is different from the incident photon, this

process is called Compton scattering interaction. When the photon energy is very high,

for example X-ray radiation, it is too high to be absorbed by the electron transition

between states. X-ray photons interact with the electron by knocking it completely out

of the atom. This is why X-rays are a form of ionizing radiation. In this process, X-ray

photons can give all the energy to an electron (photoionization) or partial energy to an

electron and produce a photon with reduced energy and different direction (Compton

scattering interaction) or if the X-ray photon energy is even higher, it can create

an electron positron pair near the nucleus. Therefore, the attenuation of X-rays comes from three phenomenons: photoelectric effect, Compton scattering, and electron pair effect, which is correlated with the atomic number of materials to the fourth power.53

Atom mass plays a significant role in the radiopacity. Since PEUs are composed of the same basic elements such as hydrogen, carbon, nitrogen and oxygen as body tissue, their radiopacity is similar, which results in no radio contrast between the

18 polymer scaffolds and their surrounding tissues. Efforts have been made to impart

polymers with enhanced radiopacity to have better contrast. By introducing a high

atomic mass element such as iodine, barium, bismuth, bromine, zirconium, strontium

or tungsten physically or chemically to the polymer, scaffolds are easily detected via

X-ray imaging after implantation. For example, Wang et al. reported the synthesis of

iodinated polyurethanes (IPUs) with radiopaque property using N-(1,3-

dihydroxypropan-2-yl)-2,3,5-triiodobenzamide as a chain extender to covalently incorporate the heavy iodine atoms.54 The IPU with 16 wt% iodine content had

radiopacity equal to the aluminum plate. In another attempt, iron oxide (Fe3O4)

nanoparticles were blended with biodegradable PLLA and this composite was

fabricated into X-ray imaging enhanced bone screws for implantation into rabbit

femoral condyles.55 Hong et al. compared the radiopacity of hydrogels containing

three radiopaque agents iopamidol (a clinically used nonionic low-osmolar iodinated

contrast agent), gold powder, and Ta2O5 microparticle quantitatively using CT

imaging. The hydrogels with 10% (w/v) metals showed even stronger radio contrast

56 than typical bones.

Figure 1.10. The attenuation of X-rays when interacting with materials: (a) photoelectric effect, (b) Compton scattering, and (c) electron pair effect.

20 1.4.2 Osteoinductivity

As the bioinert property of polymers, osteoinductivity is necessary to induce bone growth with biomimetic bone grafts.57 Osteogenic growth factors are frequently used in bone tissue engineering to regulate cellular activities such as cell attachment, proliferation, osteogenic differentiation, bone matrix formation and mineralization.

Usually, they are classified into the following groups: bone morphogenic proteins

(BMPs), osteogenic growth peptide (OGP), fibroblastic growth factors (FGFs), transforming growth factor beta (TGF-β), platelet derived growth factor (PDGF) and insulin-like growth factors (IGF).58,59 Here emphasis is given to BMPs and OGP.

1.4.2.1 Bone morphogenic proteins (BMPs)

BMPs are a group of growth factors that stimulate cartilage and bone formation in the body. In 1965, Urist MR discovered that the implantation of mineralized bone matrix induced new bone formation due to the naturally present bone inducing proteins, which were named as BMPs.60 Over the next three decades, effort was spent on the isolation and purification of BMPs.61,62 Till now, there are more than 20 BMP family members have been identified and characterized.63,64 BMPs typically form homo or hetero-dimers through disulfide bonds.65 BMP dimmers interact with the cell surface by binding to the type II and type I BMP receptors to form a heteromeric complex. The proximity between the two receptors allows the phosphorylation/activation of the type I receptor by type II receptor.66 Following phosphorylation, the type I receptors in turn phosphorylate the intracytoplasmic signaling Smad1/5/8 receptors, propagating the signal into the cell. Then, the phosphorylated Smad1/5/8 associates with Smad4 to form Smad4/phosphorylated 21 Smad1/5/8 complex, which enters into the cell nucleus to regulate the bone related

gene expression such as runt-related transcription factor 2 (RUNX2).65,67 In addition to the Smad pathway, non-Smad pathways, including mitogen activated protein

(MAPKs p38, ERK, and JNK) kinase, phosphoinositide 3 kinase/AKT (PI3K/AKT ) and protein kinase C (PKC) signaling pathways, and small Rho-like GTPases, can also been initiated directly by activated BMP receptors to modulate the downstream cellular responses.67-69 Although the Smad pathway plays a fundamental role in BMP

signaling, the non-Smad pathways, especially MAP kinase pathway, contribute to the

signal diversity and fine-tuning.

Among the BMP family, BMP-2 is most widely used to stimulate osteoblast differentiation and bone formation. The most notable is recombinant human BMP-2

(rhBMP-2), which has been approved by United States Food and Drug Administration

(FDA) for some specific clinical applications.70 It has been reported that the

incorporation of rhBMP-2 into mPCL–TCP/collagen scaffold stimulated significant

bone recovery in a rat calvarial defect at 4 and 15 weeks. Most notably at 15 weeks,

the rhBMP-2 treated group showed total closure of the defect.71 However, the high

cost, unstable property, short half-lives and possible complications, such as ectopic

bone formation and increased swelling at the defect site of current BMP-2 treatment have limited its application.72,73 To resolve these problems, the activity of BMP-2 can

be mimicked by a BMP-2 amino acid peptide fragment, which can be immobilized to

the scaffold. Saito et al. developed a new BMP-2 P-4 fragment (H-Lys-Ile-Pro-Lys-

Ala-Ser-Ser-Val-Pro-Thr-Glu-Leu-Ser-Ala-Ile-Ser-Thr-Leu-Tyr-Leu-OH), a 20

amino acid synthetic peptide corresponding to residues 73-92 of the BMP-2 knuckle

epitope and investigated its osteogenic activity.74,75. This BMP-2 mimetic peptide

promoted the osteoblastic differentiation of C3H10T1/2 cells and induced ectopic

22 bone-like calcification in rat calf muscle within 3 weeks.74 This could be explained by

the result that BMP [73-92] peptide bound to both BMP type IA and type II receptors,

so it may be one of the binding sites on BMP-2.74 Further study by Atsuhiro Saito

demonstrated that this BMP-2 mimetic peptide prolonged ectopic calcification in rat

calf muscle 76 and accelerated bone repair in the rat tibial bone defect model when it

was immobilized on alginate gel.77 Since then, this BMP [73-92] peptide has attracted

considerable interest in the osteoblast differentiation and bone formation application.

It has been covalently grafted to alginate on the surface of mesoporous silica

nanoparticles and promoted in vitro osteogenic differentiation and mineralization of

bone mesenchymal stem cell (BMSCs) characterized by the alkaline phosphatase

(ALP) activity, protein expression and calcium deposition and the in vivo ectopic

bone formation after intramuscular implantation in rats.78 Moore found a synergistic

enhancement of proliferation activity and osteogenic differentiation of human bone

marrow stromal cell (hBMSC) on the BMP [73-92] and RGD dually functionalized

gradient substrate at densities greater than 130 pmol cm−2 with 65 pmol cm−2 for each

peptide.79 Recently, a catechol-bearing BMP [73-92] peptide bioconjugate was

synthesized and bound to the surface of poly(propylene fumarate)-bioglass composite substrate via the coordinative bonds between the catechol functional groups and the oxides in bioglass. After 4 weeks culture of human mesenchymal stem cells (hMSCs)

on the substrates, the BMP [73-92] functionalized group showed enhanced cell adhesion, spreading, proliferation, osteogenic differentiation and mineralization

80 compared with the PPF/bioglass only control.

Figure 1.11. BMPs signaling pathway. BMPs bind to the type II and type I receptors to form a heteromeric complex, the phosphorylate of the type I receptor activates the Smad pathway through the phosphorylation of Smad1/5/8 receptors, followed by the association with Smad4 and translocation to the nucleus to stimulate the gene expression. I-Smads inhibit receptor activation. BMPs can also signal via the non- Smad pathways, for example, MAPK pathway by regulating the Smads receptor activation and translocation to the nucleus. Figure reproduced with permission from ref. 67. Copyright © 2012 Federation of European Biochemical Societies. Published by Elsevier B.V.

24 1.4.2.2 Osteogenic growth peptide (OGP)

Figure 1.12. Peptide sequence of OGP. It consists of two domains called accessory domain and active domain. The accessory domain is responsible for the binding of peptide to OGP binding protein, mainly α2-macroglobulin (α2M) and the active domain drives the cell proliferation and osteogenic differentiation. Figure reproduced with permission from ref. 84. Copyright © 2015 Wiley Periodic als, Inc.

Figure 1.13. OGP signaling pathway for the proliferation of osteoblastic cells (MC3T3-E1). After disassociation with α2M, the active domain [10-14] is proteolytically cleaved from OGP and then bound to Gi protein to activate the downstream mitogen-activated protein (MAP) kinase signaling pathway and DNA synthesis to drive the proliferation of osteoblast cell lines. Figure reproduced with permission from ref. 84. Copyright © 2015 Wiley Periodic als, Inc.

Osteogenic growth peptide (H-Ala-Leu-Lys-Arg-Gln-Gly-Arg-Thr-Leu-Tyr-

Gly-PheGly-Gly-OH) is also a bone stimulator containing 14 amino acids, existing naturally in the blood serum.75 In 1990s, Bab et al. found that a growth factor promoting the osteogenic-cell growth was produced by the regenerating rat bone marrow.81,82 After three-step purification by size exclusion, cation-exchange and

reverse-phase chromatoraphies, a novel osteogenic growth peptide was isolated,

which was designated as OGP.83 Its amino acid sequence was determined, which was

surprisingly identical with the C-terminal sequence 89–102 of histone H4. This newly

26 discovered peptide stimulated in vitro osteogenic cell proliferation and ALP activity

83 and in vivo bone formation after exogenous administration.

OGP consists of two domains called the accessory domain [1-9] and the active

domain [10-14]. The accessory domain is responsible for the binding of peptide to

OGP binding protein, mainly α2-macroglobulin (α2M). The active domain drives the cell proliferation and osteogenic differentiation.84 In blood circulation, the OGP

concentration maintains a steady level due to the regulation of OGP binding protein,

which protects OGP from proteolysis and acts as a reservoir. After disassociation with

α2M, the active domain [10-14] is proteolytically cleaved from OGP. It is then bound

to Gi protein to activate the downstream mitogen-activated protein (MAP) kinase signaling pathway and DNA synthesis to drive the proliferation of osteoblast cell lines.84-86 Studies revealed that OGP regulates the osteogenic differentiation via the

activation of RhoA/ROCK pathway.87 RhoA is one of the Rho protein family that is

activated by the components in blood serum including growth factors, many of which

signal through G-protein coupled receptors,88 which in turn stimulates the

downstream effector, Rho-associated protein kinase (ROCK), which is necessary for

87,89 osteogenic differentiation.

OGP [10-14], the active domain, is an OGP derived peptide with the minimal

amino acid sequence and retains the activity of full sequence OGP to stimulate

proliferation, osteogenic differentiation and matrix mineralization of osteoblastic

lineage cells.89,90 In vitro studies demonstrated that both OGP and OGP [10-14]

promoted proliferation activity and osteogenic differentiation of cells including

osteoblastic MC3T3-E1 cells, bone marrow mesenchymal stem cells (MSCs) and hMSCs.28,83,91-93 Since the soluble OGP is easy to be cleared, in tissue engineering,

27 like BMP-2, it is favorable to immobilize it onto scaffold. Moore et al. synthesized both full sequence OGP and OGP [10-14] and immobilized the peptide on the substrate surface with gradient concentration via click chemistry. MC3T3-E1 cell culture showed that the substrate with OGP/OGP [10-14] showed enhanced cell proliferation compared to the control.92 Policastro and coworkers introduced OGP

[10-14] into amino acid based poly(ester urea)s (PEU) porous scaffolds. The hMSCs osteogenic differentiation was enhanced in terms of in vitro ALP activity, calcium deposition, osteoblast related gene expression and protein expression and in vivo

ECM matrix mineralization.28 In the rabbit segmental long bone defect model, OGP was incorporated into a PLGA scaffold to accelerate defect healing. At 8 weeks the defect in the OGP group was replaced by new bone while the control group had only

94 partial new bone formation.

28 CHAPTER II

MATERIALS AND INSTRUMENTS

2.1 Materials

All reagents were purchased from Sigma-Aldrich, Alfa Aesar, Fisher

Scientific or VWR and used as received unless otherwise specified.

Fluorenylmethyloxycarbonyl (FMOC)-protected amino acids and preloaded Wang- resins were purchased from Aapptec (Louisville, KY) and Advanced Chem Tech

(Louisville, KY). Chromeo™ 488 azide was purchased from Santa Cruz

Biotechnology.

All organic solvents were purchased from Sigma-Aldrich as ACS Grade and used as received without any further purification unless otherwise specified.

Chloroform was dried by calcium hydride and distilled before use.

L-phenylalanine: Fisher Scientific, ≥98.5%.

4-Iodo-L-phenylalanine: VWR, ≥95%.

L-tyrosine: Fisher Scientific, ≥99%.

Boc-Tyr-OMe: Sigma Aldrich, 97%.

1,6-hexanediol: Sigma Aldrich, 97%.

P-toluene sulfonic acid monohydrate: Sigma Aldrich, ≥98.5%.

Propargyl bromide solution: Sigma Aldrich, 80 wt.% in toluene, contains 0.3% magnesium oxide as stabilizer.

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

Sodium hydroxide (NaOH): Sigma Aldrich, ≥98%.

Sodium carbonate (Na2CO3): Fisher Scientific, ≥99.5%.

Triphosgene: Alfa Aesar, 98%.

Di-tert-butyl dicarbonate: Sigma Aldrich, 99%.

Hydrogen chloride solution: Sigma Aldrich, 4 M in dioxane.

Anhydrous Magnesium sulfate (MgSO4): Sigma-Aldrich, ≥99.5%

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

Calcium hydride (CaH2): Sigma-Aldrich, 95%.

6-Bromohexanoic acid: Sigma-Aldrich, ≥97%

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

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

(HBTU): Sigma-Aldrich, ≥98.0%.

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

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

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

N-N’-Diisopropylcarbodiimide (DIC): Sigma Aldrich, 99%.

Human mesenchymal stem cell (hMSC) cells and growth medium: Lonza

(Walkersville, MD).

30 SensoLyte® pNPP Alkaline Phosphatase Assay Kit: Anaspec (Fremont, CA).

Cyquant cell proliferation assay kit: Thermo Fisher Scientific.

RNA isolation kits: Qiagen (Valencia, CA).

Real time reverse transcription polymerase chain reaction (real time RT-PCR) kit:

Thermofisher (Applied Biosystems).

10% phosphate buffered formalin solution: Fisher Scientific.

2.2 Instruments

Nuclear Magnetic Resonance (NMR): 1H-NMR (500 MHz) and 13C-NMR

(125 MHz) spectra were obtained using a Varian NMR Spectrophotometer. All

1 chemical shifts were reported in ppm (δ) with solvent resonances ( H-NMR DMSO-d6

13 2.50 ppm; C-NMR DMSO-d6 39.50 ppm). Abbreviations of s, d and m were used to

represent singlet, doublet and multiplet, respectively. Relaxation time of 1H NMR was

1 sec and scan number was 64. Relaxation time of 13C NMR was 5 sec and scan

number was 2000. Data analysis was processed using MestReNova software.

Fourier Transform Infrared (FT-IR): Attenuated total reflectance Fourier-

transform infrared (ATR-FTIR, MIRACLE 10, Shimadzu Corp.) spectroscopy was

performed using a MIRACLE 10, Shimadzu Corp. ATR-FTIR Spectrometer with a

wavenumber (cm-1) detection ranging from 4000-400 cm-1 and a resolution of 4 cm-1.

Size Exclusion Chromatography (SEC): The molecular masses and molecular

mass distributions of each of the polymers were determined by size exclusion

chromatography (SEC) using a TOSOH HLC-8320 gel permeation chromatograph

31 that was calibrated with a series of linear polystyrene standards. Eluograms were

collected using DMF containing 0.1 M LiBr as the eluent at a rate of 0.3 mL/min at

40 °C with a refractive index (RI) detector.

UV-Visible Spectroscopy (UV-Vis): UV-visible spectra were collected using a

Synergy Mx Microplate Reader (BioTek, Winooski, VT) at a wavelength ranging

from 230 nm to 325 nm and resolution of 1 nm.

Fluorescence Spectroscopy: The fluorescence spectra of Chromeo 488 dye attached polymers were recorded using Synergy Mx Microplate Reader (BioTek,

Winooski, VT) ranging from 500 to 700 nm at an excitation wavelength of 485 nm

(λ485) and measuring the emission at 511 nm as the analytic peak.

Mass Spectroscopy: Electrospray ionization (ESI) was carried out using a

HCT Ultra II quadrupole ion trap mass spectrometer (Bruker Daltonics, Billerica, MA)

equipped with an electrospray ionization source. Matrix-Assisted Laser

Desorption/Ionization-Time of Flight (MALDI-ToF Spectroscopy) mass spectrawas performed using a Bruker Ultraflex-III TOF/TOF mass spectrometer (Bruker

Daltonics, Inc., Billerica, MA) equipped with a Nd:YAG laser (355 nm).

Thermal Analysis: Thermogravimetric Analysis (TGA, TA Q500) was used to

measure the thermal decomposition properties of the polymers at a heating rate of

20 °C/min from room temperature to 600 °C under nitrogen atmosphere. The glass

transition temperature, Tg, was determined using Differential Scanning Calorimetry

(DSC, TA Q200) at a scanning rate of 20 °C/min from 0 °C to 200 °C for 3 cycles.

The midpoint of the transition shown in the second heating cycle was used to

determine the Tg.

32 3D printing: To obtain polymer filament devoid of trapped gasses, poly(1-

PHE-6) and poly[(1-pTYR-6)0.02-co-(1-PHE-6)0.98] films (100 mm × 80 mm × 1.0 mm) were vacuum compressed at 160 °C (TMP Technical Machine Products Corp).

The polymer film (3 g) was cut into pieces and fed into a capillary rheometer

(Dynisco, LCR 7000 Series) equipped with a 1.5 mm die. The extrusion temperature was 160 °C and extrusion speed was 16 mm/min. PEU filament (2.4 g) prepared by the capillary rheometer extrusion was used as feedstock material for 3D printing

(CartesioW equipped with 0.2 mm nozzle). The printing temperature was 160 °C and bed temperature was 65 °C. Scaffolds with orthogonally knitted porous structure (8.0 mm in diameter, 1.0 mm in thickness) were printed as designed (Google SketchUp 8) with 2 mm/s printing speed, 20% fill density, rectilinear fill pattern and 0.15 mm layer height. To better fit in the animal surgery with an 8 mm critical size defect, the scaffold design for the in vivo animal study was adjusted to 7.8 mm in diameter.

Mechanical Testing: Young’s modulus was determined using Dynamic

Mechanical Analysis (DMA Q800) with a strain rate of 2.5%/min at room temperature. The sample size used was 10 mm × 2 mm × 0.2 mm (3 samples for each polymer). The slope from the linear region of each stress-strain curve was calculated as the elastic modulus. The compression modulus of 3D scaffolds was obtained using

Instron (Instron 5567 Universal Testing Machine, compression mode) with a compression speed of 0.5 mm/min for 4 min at room temperature. The compression modulus, the slope of the linear region in the compression stress-strain curve, in both dry and wet states was analyzed.

Micro-computed Tomography (µ-CT) anylysis: Radiopacity of the polymers was characterized nondestructively using µ-CT (Skyscan 1172) with film sample

(25mm*5mm*0.5mm) under the following parameters: 100 kV voltage, medium

33 camera, 0.5 mm Al filter and 9.9 µm resolution. An aluminum wedge (0.5-2.5 mm in

0.5 mm step) was used as the contrast standard reference.

The architecture of 3D constructs was also characterized nondestructively by

µ-CT. 3D scanning of the salt leached porous scaffolds was scanned under the

following parameters: 60 kV voltage, large camera, no filter, 30 ms camera exposure

preset time and 18.4 µm resolution. In order to have sufficient contrast for PEU

scaffolds, the following parameters were used: 40 kV voltage, large camera, no filter,

70 ms camera exposure preset time and 15.0 µm resolution. The scanning of the 3D

printed porous scaffolds was performed under the following parameters: 60 kV

voltage, medium camera, no filter, 30 ms camera exposure preset time and 10.0 µm

resolution. Projections from different angles were obtained and reconstructed by the

NRecon program and analyzed by the CTAn program.

In the animal study, following fixation, samples were wrapped by parafilm

and scanned with a µ-CT scanner (Skyscan 1172) vertically to the coronal aspect of

the cranial bone. Acquisition parameters were: 10 µm resolution, medium camera, 0.5

mm aluminum filter, 40 kV, 250 µA, and 0.4° rotation step. The resulting 2D micro-

radiographic images in different angles were reconstructed by NRecon software to

generate a series of 2D cross sectional images of the sample, which were further

analyzed by a CT analyzer program to quantify the new bone growth. The

threshholding range was set from -727 to 11709 Hounsfield units (Hu) to remove the effect of contrast from iodinated scaffold. After threshholding, processing and calculation, the new bone volume was obtained.

Microscopy Imaging: Fluorescence images were obtained using an

IX81microscope (Olympus) equipped with Hamamatsu Orca R2 CCD camera and

34 DAPI, FITC and TRITC fluorescence filters. Hisological images were recorded using

an IX81microscope (Olympus) equipped with QImaging Micropublisher 3 camera

under bright field. Data was analyzed by Olympus cell Sens Dimension software.

Quantification of new bone growth was calculated by color thresholding.

Real time reverse transcription polymerase chain reaction (real time RT-PCR):

RNA for real time RT-PCR characterization was extracted from samples at 2, 3, and 4

week time points by the RNeasy Mini Kit (Qaigen, Valencia, CA) and quantified by

Microplate Reader using a Take3 Multi-Volume Plate. Following the protocol, RNA

was reverse transcripted to complementary DNA (cDNA), which was then stored at

−20 °C for further use. The real time RT-PCR was performed by 7500 Real Time

PCR System (Applied Biosystems) using SYBR Green fluorescence detection system.

Human Glyceraldehyde-3-phosphate dehydrogenase (GAPDH) was used as the

housekeeping gene. cDNA derived from hMSCs without any treatment was set as

calibrator sample for data analysis. The value obtained was expressed as fold changes.

PCR primer sets used were as follows: GAPDH forward: 5’-

GACAGTCAGCCGCATCTT-3’; reverse: 5’-CCATGGTGTCTGAGCGATGT-3’;

RUNX2 forward: 5’-GGACGAGGCAAGAGTTTCAC-3’; reverse: 5’-

CAAGCTTCTGTCTGTGCCTTC-3’; BSP forward: 5’-

CCTGGCACAGGGTATACAGG-3’; reverse: 5’-CTGCTTCGCTTTCTTCGTTT-3’;

OCN forward: 5’-CATGAGAGCCCTCACA-3’; reverse: 5’-

AGAGCGACACCCTAGAC-3’.

35 CHAPTER III

RADIOPAQUE, IODINE FUNCTIONALIZED PHENYLALANINE-BASED

POLY(ESTER UREA)S

This work has been previously published as

Shan Li, Jiayi Yu, Mary Beth Wade, Gina M. Policastro, and Matthew L. Becker

Biomacromolecules, 2015, 16 (2), 615-624.

3.1 Abstract

The synthesis and characterization of iodine-functionalized phenylalanine- based poly(ester urea)s (PEUs) are reported. 4-Iodo-L-phenylalanine and L- phenylalanine were separately reacted with 1,6-hexanediol to produce two monomers, bis-4-I-L-phenylalanine-1,6-hexanediol-diester (1-iPHE-6 monomer) and bis-L- phenylalanine-1,6-hexanediol-diester (1-PHE-6 monomer). By varying the feed ratio of the 1-iPHE-6 and 1-PHE-6 monomers, the copolymer composition was modulated resulting in a wide variation in thermal, mechanical and radiopacity properties. Micro- computed tomography (µ-CT) projections demonstrate that increasing iodine content results in greater X-ray contrast. Compression tests of dry and wet porous scaffolds indicate that the poly(1-iPHE-6)0.24-co-poly(1-PHE-6)0.76 material results in the highest compression modulus. MC3T3 cell viability and spreading studies show PEUs are non-toxic to cells. As most medical device procedures require placement verification via fluoroscopic imaging, materials that possess inherent X-ray contrast are valuable for a number of applications.

36 3.2 Introduction

Surgeons rely on fluoroscopic imaging to track the placement of medical devices and implants in patients. This is especially challenging when the devices are polymeric. Conventional polymers are not easily detected using physiologically relevant X-ray radiography because their radiopacity is similar to that of human tissue as a result of the similar C, H, O and N elemental composition.95-97 Polymeric biomaterials with enhanced radiopacity have been extensively studied in recent years due to their potential applications in implantable orthopedic, prostheses and vascular devices that remain visible to X-ray after implantation.98-106 The ability of an element to attenuate X-rays is correlated with the atomic number of the element to the fourth power.107 Hence, heavy atoms, including iodine, have been utilized to impart radiopacity into polymers and enhance X-ray contrast. Most enhancement strategies focus on two methods to modify the radiopacity of polymers. The first is to make radiopaque blends by incorporating radiopaque additives such as inorganic salts of

108-119 heavy elements (La2O3, BaO, BaSO4, SrO, ZrO2, Ta2O5/SiO2, or SrCO3), or

120-122 organic compounds with heavy atoms (triphenyl bismuth, I4C2B10H8). The majority of commercial radiopaque polymeric medical implants123 are prepared in this way because it is easy to manufacture using extrusion and molding and the contrast can be controlled by adjusting the blending ratio. However, physical blending also possesses some drawbacks in that it is difficult to achieve stable blend dispersions, and the limited compatibility of polymers with radiopaque additives can lead to contrast agent leakage, which can subsequently lead to a decrease in radiopacity and invoke unwanted biological responses and mechanical failures.95,109,110 The second

37 method to enhance contrast is to synthesize polymers possessing covalently bonded

heavy atoms.95,96,124-132

Koole and coworkers110,133 prepared monomers containing covalently bonded

iodine (4-IEMA) and terpolymerized 4-IEMA with 2-hydroxyethyl methacrylate

(HEMA) and methyl methacrylate (MMA). Electron spectroscopy for chemical

analysis (ESCA) demonstrated the stability of this iodinated polymer. In Qu’s

work,134 iodine was incorporated into poly(ether urethane) in a two-step condensation

polymerization by using iodine-containing diol. This iodinated poly(ether urethane) had high radiopacity, good thermal stability and was not cytotoxic. Kohn135 et al.

demonstrated the preparation of iodinated and/or brominated derivatives of dihydroxy

monomers and polymers with different structures. These materials were found to be

degradable and tissue-compatible. Kohn et al. also reported an iodine-modified

poly(desaminotyrosyl-tyrosine ethyl ester carbonate) (pI2DTEc) that was synthesized

using a monomer containing iodine atoms in the 3,5 position of the aromatic rings of

tyrosine.97 Incorporation of iodine atoms had a distinct influence on the mechanical

and protein adsorption properties of the resulting polymers.98 Combinatorial methods

have been used to determine the minimal amount of iodinated polymer needed to have

sufficient X-ray contrast under a variety of translationally relevant imaging conditions.

Such chemical modification introduces radiopacity intrinsically into polymers.

Amino acid-based poly(ester urea)s (PEUs) are finding use in a number of regenerative medicine applications due to the inherent synthetic flexibility, which results in tunable mechanical and degradation properties.136-138 The resulting polymers

are semi-crystalline depending on the amino acid precursors, and the hydrogen

bonding in the urea groups imparts the polymers with strong mechanical properties.

38 The ester and urea bonds allow for both hydrolytic and enzymatic degradation.139 The

final degradation byproducts are amino acids, small diol segments and CO2, which can be removed readily. Unlike the acidic degradation byproducts of polyesters, the carboxyl group in PEU is buffered by the urea linkages at each repeat unit. Therefore, we believe that the lack of inflammation presented in our previous work140 in vivo is a

direct result of the absence of localized acidification during and after PEU

degradation. Furthermore histological analysis of PEUs has shown that they are

nontoxic and are therefore excellent candidates for tissue engineering constructs.140

Significantly, PEUs are synthetically flexible in that there are 20 kinds of naturally

occurring amino acids and Tirrell et al. have successfully used a number of non-

natural amino acids derivatives in a number of applications.141-143 These amino acids,

along with the various diols commercially available, PEUs with vastly different

139 properties can be synthesized.

PEUs can also be chemically modified with bioactive groups to initiate

specific responses in vitro and in vivo. Growth factors and peptides, including

osteogenic growth peptide (OGP), have been used to crosslink PEUs in order to

increase the mechanical properties and bioactivity of the resulting materials.140 Lin et

al. have also reported the chemical modifications of PEUs with pendant clickable

groups to fabricate functional nanofibers.136 However, these materials also lack

radiopacity. Therefore enhancing X-ray contrast is necessary to increase the translational potential of these materials. Contrast enables use of X-ray fluoroscopy to show the clinician the precise location of the devices in vivo effectively and efficiently.97 The level of contrast needed varies with a number of factors including

X-ray flux, tissue coverage and location relative to bone and other internal structures.

Minimizing chemical modifications can reduce the variance in the physical-chemical

39 properties. As such there is a fine balance between ensuring sufficient contrast in a

material while minimizing physical property changes to the polymer.

Herein, the synthesis of a series of iodinated phenylalanine-based copolymers via a two-step step-growth polymerization using 4-iodo-phenylalanine and 1,6-hexane diol as starting materials. The resulting polymers were characterized using a number of chemical, thermal and mechanical methods. Micro-computed tomography (µ-CT)

2D projections of polymers with varied iodine content were compared to established aluminum contrast standards.144 Porous 3D scaffolds were made with varied iodine

content, and were characterized for radiopacity and compression modulus. Cell

viability and spreading tests were carried out to exam the toxicity of PEUs.

3.3 Experimental Section

3.3.1 Materials

4-Iodo-L-phenylalanine (95+%) was purchased from VWR. L-phenylalanine,

1,6-hexanediol, p-toluene sulfonic acid monohydrate, activated carbon black, calcium hydride, sodium carbonate, triphosgene (98.00%), toluene, chloroform, hexafluoro-2-

propanol (HFIP), ethanol and N,N-dimethyl formamide (DMF) were purchased from

Sigma-Aldrich or Alfa Aesar. Chloroform was dried and distilled before use. All other

chemicals were used as received.

40 3.3.2 Characterization of chemical structure and thermal properties

1H-NMR and 13C-NMR spectra of monomers and polymers were obtained

using Varian NMR Spectrophotometer (500 MHz). All chemical shifts were reported

1 13 in ppm (δ) with solvent resonances ( H-NMR DMSO-d6 2.50 ppm; C-NMR DMSO-

d6 39.50 ppm). Abbreviations of s, d and m were used to represent singlet, doublet and multiplet. Fourier transform infrared spectra (FT-IR) of PEUs were characterized using Excalibur Spectrometer FTS 3000. Measurements were conducted by preparing KBr pellet and recording the spectra using 64 scans with 4 cm-1 resolution.

Molecular masses of polymers were obtained from size exclusion chromatography

(SEC) analysis (TOSOH HLC-8320 gel permeation chromatograph) using DMF (with

0.01 M LiBr) as eluent (flow rate 1 mL/min) at 50 oC and a refractive index detector.

Thermogravimetric Analysis (TGA, TA Q500) was used to measure the thermal

properties of PEUs at a heating rate of 20 oC/min from room temperature to 600 oC

under nitrogen atmosphere. The glass transition temperature, Tg, was determined

using Differential Scanning Calorimetry (DSC, TA Q200) at a scanning rate of 20

oC/min from -20 oC to 200 oC for 3 cycles. The midpoint of the transition shown in

the second heating cycle was used to determine Tg.

3.3.3 Synthesis of di-p-toluene sulfonic acid salt of bis-L-phenylalanine-1,6- hexanediol-diester (1-PHE-6 monomer) and di-p-toluene sulfonic acid salt of bis-4-I-

L-phenylalanine-1,6-hexanediol-diester (1-iPHE-6 monomer)

1-PHE-6 and 1-iPHE-6 monomers were synthesized as described

previously.140 1,6-hexanediol (20.00 g, 1.0 equiv., 0.17 mol), L-phenylalanine (64.32

g, 2.3 equiv., 0.39 mol), p-toluene sulfonic acid monohydrate (77.29 g, 2.4 equiv.,

41 0.41 mol) and toluene (500 mL) were mixed in a 1L one-neck round-bottomed flask

using a magnetic stir bar with a dean stark trap. The system was refluxed at 110 °C for

21 h. The crude product was vacuum filtered overnight to remove toluene, decolorized by activated carbon black (4.00 g) and recrystallized from boiling water 4

1 times to yield 105.50 g (yield 82.4%). H-NMR (500 MHz, DMSO-d6): 1.06 (m, 4H),

1.38 (m, 4H), 2.27 (s, 6H), 2.48 (m, DMSO), 2.97-3.15 (m, 4H), 3.29 (s, H2O), 3.98-

4.03 (m, 4H), 4.25-4.28 (m, 2H), 7.09-7.11 (d, 4 H), 7.20-7.30 (m, 10H), 7.41-7.49 (d,

13 4H), 8.36 (s, 6H). C-NMR (500 MHz, DMSO-d6): 20.84, 24.72, 27.65, 36.22,

38.67-39.78 (DMSO-d6), 53.36, 65.48, 125.56, 127.26, 128.24, 128.58, 129.34,

134.73, 138.14, 145.03, 169.08.

The synthesis of 1-iPHE-6 monomer was performed using the same method,

but alcohol was added to the water (1:1) to increase solubility for recrystallization. 1,6-

hexanediol (17.63 g, 1.00 equiv., 0.15 mol), 4-I-L-phenylalanine (100.00 g, 2.3 equiv.,

0.34 mol), p-toluene sulfonic acid monohydrate (68.13 g, 2.4 equiv., 0.36 mol ) and

toluene (1000 mL) were mixed in a 2L one-neck round-bottomed flask using a

magnetic stir bar with a dean stark trap. The system was refluxed at 110 °C for 21 h.

The crude product was vacuum filtered overnight to remove toluene, and was

recrystallized from mixture solvent of alcohol and water (1:1) 4 times to yield 111.40

1 g (yield 74.0%). H-NMR (500 MHz, DMSO-d6): 1.05 (m, 4H) 1.39 (m, 4H) 2.27 (s,

6H) 2.48 (m, DMSO) 2.92-3.11 (m, 4H) 3.32 (s, H2O) 4.02-4.03 (m, 4H) 4.26-4.30

13 (m, 2H), 7.02-7.69(m, 16H). C-NMR (500 MHz, DMSO-d6): 21.25, 25.24, 28.15,

36.05, 39.16-40.83 (DMSO-d6), 53.42, 66.03, 93.84(C-I), 125.95, 128.58, 132.18,

134.94, 137.75, 138.33, 145.73, 169.37.

42 3.3.4 Synthesis of bis-L-phenylalanine-1,6-hexanediol-diester PEU (poly(1-PHE-6)),

bis-4-I-L-phenylalanine-1,6-hexanediol-diester PEU (poly(1-iPHE-6)), co-polymers of 1-iPHE-6 monomer and 1-PHE-6 monomer (1:4 molar ratio, poly(1-iPHE-6)0.24- co-poly(1-PHE-6)0.76) and co-poly(ester urea) of 1-iPHE-6 monomer and 1-PHE-6

monomer (3:4 molar ratio, poly(1-iPHE-6)0.44-co-poly(1-PHE-6)0.56).

Di-p-toluene sulfonic acid salt of bis-L-phenylalanine-1,6-hexanediol-diester

(1-PHE-6 monomer) (30.00 g, 1.0 equiv., 0.04 mol), sodium carbonate (8.83 g, 2.1 equiv., 0.083 mol) and 400 mL distilled water were added to a 3 L 3-neck round bottom flask. The contents were mechanically stirred at 35 °C until the mixture was dissolved. The 35 °C water bath was then replaced with an ice bath. When the reaction temperature reached 0 °C, additional sodium carbonate (4.42 g, 1.05 equiv.,

0.042 mol) was dissolved in 150 mL distilled water and added to the flask.

Triphosgene (4.21 g, 0.35 equiv., 0.014 mol, 98%), dissolved in distilled chloroform

(100 mL), was added to the flask quickly. After 30 minutes, additional triphosgene

(1.00 g, 0.08 equiv., 0.003 mol, 98%), dissolved in distilled chloroform (30 mL), was added to the flask dropwise for 2 h. The crude product was transferred to a separatory funnel and precipitated into boiling water dropwise to obtain polymer 15.99 g (yield

1 92.0%). H-NMR (500 MHz, DMSO-d6): 1.15 (m, 4H) 1.43 (m, 4H) 2.49(DMSO)

2.85-2.94 (m, 4H) 3.29 (s, H2O), 3.94 (m, 4H) 4.35-4.39 (m, 2H) 6.47-6.48 (m, 2 H)

13 7.13-7.26 (m, 10H). C-NMR (500 MHz, DMSO-d6): 25.32, 28.35, 38.15, 39.52-

40.53 (DMSO), 54.50, 64.72, 126.97, 128.65, 129.59, 137.33, 157.09, 172.70.

The same procedure was used to synthesize the other three polymers, except

for the use of different amounts of the monomers in copolymerization.

43 For poly(1-iPHE-6), di-p-toluene sulfonic acid salt of bis-4-I-L-phenylalanine-

1,6-hexanediol-diester (1-iPHE-6 monomer) (8.00 g, 1.0 equiv., 7.94 mmol), sodium carbonate (1.77 g, 2.1 equiv., 16.70 mmol) and 133 mL distilled water were added to a 1 L 3-neck round bottom flask. The contents were mechanically stirred at 35 °C until the mixture was dissolved. The 35 °C water bath was then replaced with an ice bath. When the reaction temperature reached 0 °C, additional sodium carbonate (0.88 g, 1.05 equiv., 8.33 mmol) was dissolved in 50 mL distilled water and added to the flask. Triphosgene (0.84 g, 0.35 equiv., 2.83 mmol, 98%), dissolved in distilled chloroform (33 mL), was added to the flask quickly. After 30 minutes, additional triphosgene (0.20 g, 0.08 equiv., 0.67 mmol, 98%), dissolved in distilled chloroform

(10 mL), was added to the flask dropwise for 2 h. The crude product was transferred to a separatory funnel and precipitated into boiling water dropwise to obtain polymer

1 4.85 g (yield 88.6%). H-NMR (500 MHz, DMSO-d6): 1.17 (m, 4H) 1.44 (m, 4H)

2.49(DMSO) 2.81-2.90 (m, 4H) 3.29 (s, H2O), 3.94-3.97 (m, 4H) 4.33-4.37 (m, 2H)

6.43-6.45 (m, 2 H) 6.93-6.95 (m, 4H) 7.58-7.60 (m, 4H). 13C-NMR (500 MHz,

DMSO-d6): 25.38, 28.39, 37.58, 39.53-40.63 (DMSO), 54.17, 64.85, 92.89, 132.04,

137.39, 156.98, 172.48.

For copolymer of 1-iPHE-6 monomer and 1-PHE-6 monomer at 1:4 in molar ratio (poly(1-iPHE-6)0.24-co-poly(1-PHE-6)0.76), di-p-toluene sulfonic acid salt of bis-

4-I-L-phenylalanine-1,6-hexanediol-diester (1-iPHE-6 monomer) (3.00 g, 1.0 equiv.,

2.98 mmol), di-p-toluene sulfonic acid salt of bis-L-phenylalanine-1,6-hexanediol-

diester (1-PHE-6 monomer) (9.00 g, 1.0 equiv., 11.90 mmol), sodium carbonate (3.31

g, 2.1 equiv., 31.23 mmol) and 133 mL distilled water were added to a 1 L 3-neck

round bottom flask. The contents were mechanically stirred at 35 °C until the mixture

was dissolved. The 35 °C water bath was then replaced with an ice bath. When the

44 reaction temperature reached 0 °C, additional sodium carbonate (1.66 g, 1.05 equiv.,

15.66 mmol) was dissolved in 50 mL distilled water and added to the flask.

Triphosgene (1.58 g, 0.35 equiv., 5.32 mmol, 98%), dissolved in distilled chloroform

(33 mL), was added to the flask quickly. After 30 minutes, additional triphosgene

(0.38 g, 0.08 equiv., 1.27 mmol, 98%), dissolved in distilled chloroform (10 mL), was

added to the flask dropwise for 2 h. The crude product was transferred to a separatory

funnel and precipitated into boiling water dropwise to obtain polymer 6.83 g (yield

1 94.0%). H-NMR (500 MHz, DMSO-d6): 1.16 (m, 4H) 1.43 (m, 4H) 2.49(DMSO)

2.81-2.94 (m, 4H) 3.29 (s, H2O), 3.93-3.95 (m, 4H) 4.35-4.39 (m, 2H) 6.43-6.48 (m, 2

13 H) 6.94-7.60 (m, 9.52H). C-NMR (500 MHz, DMSO-d6): 25.32, 28.35, 38.16,

39.52-40.53 (DMSO), 54.50, 64.72, 92.86, 126.97, 128.65, 129.59, 132.05, 137.33,

157.09, 172.70.

For copolymers of 1-iPHE-6 monomer and 1-PHE-6 monomer at 3:4 in molar ratio (poly(1-iPHE-6)0.44-co-poly(1-PHE-6)0.56), di-p-toluene sulfonic acid salt of bis-

4-I-L-phenylalanine-1,6-hexanediol-diester (1-iPHE-6 monomer) (6.00 g, 1.0 equiv.,

5.95 mmol), di-p-toluene sulfonic acid salt of bis-L-phenylalanine-1,6-hexanediol-

diester (1-PHE-6 monomer) (6.00 g, 1.0 equiv., 7.94 mmol), sodium carbonate (3.09 g,

2.1 equiv., 29.15 mmol) and 133 mL distilled water were added to a 1 L 3-neck round

bottom flask. The contents were mechanically stirred at 35 °C until the mixture was

dissolved. The 35 °C water bath was then replaced with an ice bath. When the

reaction temperature reached 0 °C, additional sodium carbonate (1.55 g, 1.05 equiv.,

14.62 mmol) was dissolved in 50 mL distilled water and added to the flask.

Triphosgene (1.47 g, 0.35 equiv., 4.96 mmol, 98%), dissolved in distilled chloroform

(33 mL), was added to the flask quickly. After 30 minutes, additional triphosgene

(0.35 g, 0.08 equiv., 1.18 mmol, 98%), dissolved in distilled chloroform (10 mL), was

45 added to the flask dropwise for 2 h. The crude product was transferred to a separatory

funnel and precipitated into boiling water dropwise to obtain polymer 7.02 g (yield

1 92.6%). H-NMR (500 MHz, DMSO-d6): 1.16 (m, 4H) 1.43 (m, 4H) 2.49(DMSO)

2.81-2.93 (m, 4H) 3.28 (s, H2O), 3.94-3.95 (m, 4H) 4.32-4.39 (m, 2H) 6.43-6.48 (m, 2

13 H) 6.93-7.60 (m, 9.12H). C-NMR (500 MHz, DMSO-d6): 25.32, 28.36, 38.15,

39.53-40.53 (DMSO), 54.49, 64.73, 92.87, 126.97, 128.65, 129.59, 132.04, 137.39,

157.04, 172.69.

3.3.5 PEU films and 3D porous scaffold preparation and characterization

Films with two sizes (25 mm × 5 mm × 0.5 mm film and 10 mm × 2 mm × 0.2

mm) were prepared by vacuum compression molding as published139 (temperature

150 oC, TMP Technical Machine Products Corp). 3D porous scaffolds were fabricated

as previously described145 by casting polymer solution (2 g polymer in 5 mL DMF)

into sieved salt (weight: 22 g, size: 250 µm-400 µm and theory porosity: 85.6%146).

DMF was removed by vacuum drying at 65 oC for 3 days and the salt was leached in

deionized water for 3 days. The obtained porous scaffolds were dried and cut into 7

mm × 7 mm × 7 mm for characterization (6 samples for each scaffold). In order to

prepare wet samples, dry scaffolds were soaked in PBS buffer for 4 days (3 samples

for each polymer in each condition).

The mechanical properties of the films were measured using Dynamic

Mechanical Analysis (DMA Q800) with a strain rate of 2.5%/min at room

temperature. The sample size used was 10mm × 2mm × 0.2mm (3 samples for each

polymer). The slope from the linear region of each stress-strain curve was calculated as the elastic modulus. The compression modulus of 3D scaffolds was obtained using

46 Instron (Instron 5567 Universal Testing Machine, compression mode) with a

compression speed of 0.5mm/min for 4 min at room temperature. The compression

modulus, the slope of the linear region in the compression stress-strain curve, in both

dry and wet states was analyzed.

Radiopacity of the polymers was characterized nondestructively using X-ray

micro-computed tomography (µ-CT). In µ-CT (Micro-CT, Skyscan 1172) projection,

25 mm × 5 mm × 0.5 mm films were used. The following parameters were adopted

when using µ-CT: 100 kV voltage, medium camera, 0.5 mm Al filter and 9.9 µm

resolution. An aluminum wedge (0.5-2.5 mm in 0.5 mm step) was used as the contrast

standard reference.144 The porosity of scaffolds was also characterized

nondestructively by µ-CT. 3D scanning of scaffolds was carried out under the

following parameters: 60 kV voltage, large camera, no filter, 30 ms camera exposure

preset time and 18.4 µm resolution. In order to have sufficient contrast for PEU

scaffolds, the following parameters were used: 40 kV voltage, large camera, no filter,

70 ms camera exposure preset time and 15.0 µm resolution. In this test, projections

from different angles were obtained and reconstructed by the NRecon program and

analyzed by the CTAn program.

3.3.6 In vitro cell viability and spreading characterization

PEU films were prepared on 12 mm diameter cover glass by spin coating of

5wt% PEU solution (dissolved in HFIP), vacuum dried at 80 oC over night and

carefully moved to wells of a 24-well plate using tweezers. All samples were sterilized in 70% ethanol for 20 minutes, rinsed once in PBS to remove residual ethanol, and then submerged in 1 mL media prior to seeding. Cells were rinsed with

47 PBS and detached from the bottom of the flask using 0.05% Trypsin/EDTA at 37 °C,

95% humidity, 5% CO2 for 5 minutes. Detached cells were then collected into a

conical tube containing equal parts media to trypsin. Cells were centrifuged into a

pellet at 3,000 rpm, 4 °C for 1 minute. The media/trypsin was aspirated and cells were

re-suspended in fresh media. Then cells were counted using a hemacytometer with

trypan blue exclusion. Cells were seeded at a density of 25 cells/mm2 in 100 μL aliquots per sample by dripping into the center of sample wells containing 1mL of media. The well plate was agitated to ensure even dispersion of cells over samples prior to incubation at 37 °C, 95% humidity, 5% CO2 for twenty-four hours.

Cell viability was assessed using a Live/Dead® viability assay (Life

Technologies). 5 μL of calcein AM (4 mM) and 10 μL of ethidium homodimer-1 (2 mM) were added to 10 mL of DPBS as a working solution. Media was aspirated from all samples, samples were rinsed once in DPBS, and then 0.5 mL of the working solution was added to each well. Samples were incubated for 10 minutes at 37 °C before imaging at 4× magnification using CellSENS® imaging software with an

Olympus microscope equipped with a Hamamatsu Orca R2 CCD camera and a filter cube containing FITC and TRITC fluorescence filters. Images were analyzed for live/dead cell counts using Image J (NIH) software with a cell counter plugin. Cells stained green were counted as live and cells that stained red were counted as dead.

Live and dead cell counts for all images per sample were totaled to calculate % viability for each sample.

After twenty-four hours of incubation, cells were fixed first by adding 0.6 mL of 3.7% paraformaldehyde in PBS to 0.4 mL media in each well for 10 minutes, and then in 1 mL of 3.7% paraformaldehyde in PBS for 5 minutes at 37 °C. Cells were then permeabilized using 0.5% tritonX-100 in cytoskeletal stabilization (CS) buffer

48 for 9 minutes at 37 °C. Samples were rinsed three times, 5 minutes each time, in CS

buffer at room temperature. Aldehyde autofluorescence was then quenched using 0.1%

sodium borohydride in CS buffer for 10 minutes at room temperature. Non-specific

staining was blocked using 5% donkey serum in PBS for 20 minutes at 37 °C.

Samples were then rinsed three times, 5 minutes each time, in CS buffer at room

temperature. Cells were stained with 6 μM DAPI (Life Technologies) in CS Buffer for

10 minutes at 37 °C. Cells were then stained to observe cytoskeletal actin using

rhodamine-phalloidin (Life Technologies) (6.6 μM diluted 1:40 in 1% donkey serum)

for 1 hour at 37 °C. After staining cells were rinsed three times in 1% donkey serum

and two times in PBS with no wait. Samples were imaged immediately using

CellSENS® imaging software with an Olympus microscope equipped with a

Hamamatsu Orca R2 CCD camera and a filter cube containing DAPI and TRITC

fluorescence filters. Images were analyzed for cell aspect ratio and cell area using

Image J (NIH) software. Cell aspect ratio was quantified using the cells greatest

length divided by the diameter of the cell across the center of the nucleus. Twenty

cells per image were used to calculate average cell spreading as well as cell area for

each sample (n=3).

3.4 Results and discussion

3.4.1 Synthesis of L-phenylalanine-based and 4-I-L-phenylalanine-based poly(ester urea)s

Chemical structure and copolymer composition were determined using NMR and FT-IR spectroscopy. In Figure 3.1, the urea peak at 6.5 ppm shows the successful synthesis of the four different PEU polymers. For poly(1-PHE-6), the aromatic

49 hydrogen peaks appear around 7.1 to 7.3 ppm. However, substitution of one hydrogen

atom with an iodine atom shifts the other aromatic protons to 6.95 ppm and 7.6 ppm.

For copolymers, the aromatic protons have characteristic poly(1-PHE-6) and poly(1-

iPHE-6) resonances. With the increase of iodine content, the poly(1-iPHE-6)

characteristic peak intensity increases and the poly(1-PHE-6) characteristic peak intensity decreases. The copolymer composition can therefore be calculated by integration of these peaks from 1H-NMR results. By using the hydrogen attached to

tertiary carbons at 4.35-4.39 ppm as the reference peak, the integration of the aromatic

rings should change with different iodine content. For example, for poly(1-PHE-6)

there are 10 hydrogen atoms located on the aromatic ring and 2 are attached to the

tertiary carbon for every repeat unit with a ratio of 5. For poly(1-iPHE-6), the ratio is

R R Cl Cl R SO3H Cl O O Cl SO3 O OH OH Cl O Cl O O H2N + HO H3N O H 3 O NH3 Toluene 3 N O O Interfacial Polymerization * O N * o O 3 n 110 C, 21h SO3 H 2.3 eq. 1 eq. Na2CO3 O o 0 C, water+CHCl3, 2h R= H or I R R

Scheme 3.1. The two-step synthesis of amino acid-based poly(ester urea)s (PEUs). R could be an H or I atom, which results in two distinct monomers (1-PHE-6 monomer and 1-iPHE-6 monomer). Homopolymerization of the two monomers results in poly(1-PHE-6) and poly(1-iPHE-6). Copolymerization of these two monomers with different feed ratios results in poly(1-iPHE-6)-co-poly(1-PHE-6)s with varied iodine content.

For copolymers, the ratio should be between 4 and 5, and indicates the extent

of iodination imparted to the polymer. As listed in the first three columns of Table 3.1,

the NMR normalized integration ratio of poly(1-PHE-6), poly(1-iPHE-6) and the two

copolymers are 4.99, 3.98, 4.76 and 4.56, respectively. The corresponding poly(1-

iPHE-6) content in the polymers are therefore calculated to be 0%, 100%, 24% and

50 44%. This result is consistent with the feed ratio. 13C-NMR (Figure S2) supports these

results. The copolymer composition could also be obtained from FT-IR (Figure 3.2).

1 Figure 3.1. H-NMR (DMSO-d6) of PEUs. (a) homopolymer of 1-iPHE-6 monomer (poly(1-iPHE-6), which has ring signals at 6.95 and 7.6 ppm, characteristic of a para- substituted aromatic ring. (b) copolymer of 1-iPHE-6 monomer/1-PHE-6 monomer at a 3:4 molar ratio (44% poly(1-iPHE-6) and 56% poly(1-PHE-6)); (c) copolymer of 1- iPHE-6 monomer/1-PHE-6 monomer at a 1:4 molar ratio (24% poly(1-iPHE-6) and 76% poly(1-PHE-6) in the copolymer). They all possess characteristic peaks of both poly(1-PHE-6) and poly(1-iPHE-6). (d) homopolymer of 1-PHE-6 monomer (poly(1- PHE-6)), possessing proton resonances characteristic of the benzyl group, 7.1 to 7.3 ppm.

For FT-IR, baseline deduction and normalization of the urea group peak

absorbance at 3398 cm-1 were carried out for all absorbance spectra. The C-I

absorbance peak intensity at 1007 cm-1 increases with increasing iodine content,

which is set as the analytical peak. For poly(1-PHE-6) and poly(1-iPHE-6), the peak intensity at 1007 cm-1 is 0 and 0.29, respectively. Thus, the copolymer intensity at

1007 cm-1 is entirely contributed by the iodinated part. As such, the ratio of the C-I peak intensity of the copolymers to that of poly(1-iPHE-6) (0.29) shows the iodinated composition in the copolymer to be 24% and 41%, respectively, which again confirms

51 the NMR result (Table 3.1). Both NMR and FT-IR spectroscopy demonstrate the

successful synthesis of iodine-functionalized PEUs. The content of iodine can be

easily adjusted via simply changing the feed ratio of the different monomers.

Figure 3.2. FT-IR of PEUs. (a) iodinated phenylalanine-based poly(1-iPHE-6); (b) copolymer of 44% poly(1-iPHE-6) and 56% poly(1-PHE-6); (c) copolymer of 24% poly(1-iPHE-6) and 76% poly(1-PHE-6); (d) phenylalanine-based poly(1-PHE-6). All spectra show the characteristic ester and urea peaks. For iodinated polymers, (a), (b) and (c), the characteristic C-I stretching signal at 1007 cm-1 increased with greater iodine content.

52 Table 3.1. PEUs composition from 1H-NMR and FT-IR

1-iPHE-6 NMR normalized Poly(1-iPHE-6) FT-IR Poly(1-iPHE-6) monomer feed integration ratio * content in polymer normalized content in polymer ratio from NMR * peak height # from FT-IR #

100% 3.98 100% 0.29 100%

43% 4.56 44% 0.12 41%

20% 4.76 24% 0.07 24%

0 4.99 0 0 0

* Reference peak: hydrogen attached to tertiary carbon (Chemical shift: 4.35-4.39 ppm). Analytical peak: hydrogen in aromatic ring (Chemical shift: 6.93-7.60ppm). NMR normalized integration ratio=integration of analytical peak/integration of reference peak. Poly(1-iPHE-6) content in copolymer from NMR= (5 - NMR normalized integration ratio) ×100%. # Reference peak: urea group peak absorbance at 3398 cm-1. Analytical peak: C-I absorbance at 1007 cm-1. Poly(1-iPHE-6) content in copolymer= (FT-IR normalized peak height/ 0.29) ×100%.

3.4.2 Thermal properties of PEUs

PEUs are expected to be used as parts of implantable medical devices, which

generally are fabricated using melt processing. A high degradation temperature is

therefore preferred for the manufacture of implantable devices. The thermal stability

of PEUs was characterized using TGA (Table 3.2). The TGA results show that the 5%

weight loss temperatures for poly(1-PHE-6), poly(1-iPHE-6)0.24-co-poly(1-PHE-6)0.76,

53 poly(1-iPHE-6)0.44-co-poly(1-PHE-6)0.56 and poly(1-iPHE-6) are 277, 287, 297 and

303 oC, respectively. This suggests that iodine incorporation increases the thermal stability of PEUs. The glass transition temperatures were obtained by DSC (Figure S3 and Table 3.2). The glass transition temperatures of poly(1-PHE-6), poly(1-iPHE-

6)0.24-co-poly(1-PHE-6)0.76, poly(1-iPHE-6)0.44-co-poly(1-PHE-6)0.56 and poly(1- iPHE-6) are 59, 65, 71 and 88 oC, respectively. All four materials have glass transition temperatures far above physiological temperature, which is near 37 oC. It implies that

PEUs are suitable for implantable device manufacturing. The DSC results show that

incorporation of iodine in poly(1-PHE-6) can increase the glass transition temperature,

which is consistent with previously published results.126 This observation is attributed

to two aspects. First, iodine is polarizable, which increases both the inter- and intra- chain interaction and hence, reduces the segmental mobility of polymer chains.

Second, iodine is bulky, which hinders the mobility of polymer chains. Both of these properties become enhanced with increasing iodine content, causing a resultant increase in the glass transition temperature of PEUs with increasing iodine content.

Table 3.2. Characterization summary of PEUs

Td/°C Tg/°C

Sample Mn Mw ĐM (TGA) (DSC)

Poly(1-PHE-6) 87k 148k 1.7 277 59

Poly(1-iPHE-6)0.24-co-poly(1-PHE-6)0.76 181k 307k 1.7 287 65

Poly(1-iPHE-6)0.44-co-poly(1-PHE-6)0.56 117k 238k 2.0 297 71

Poly(1-iPHE-6) 88k 147k 1.7 303 88

54 3.4.3 Mechanical properties of bulk PEU films

Figure 3.3. Stress-strain curves of poly(1-PHE-6), poly(1-iPHE-6)0.24-co-poly(1- PHE- 6)0.76 and poly(1-iPHE-6)0.44-co-poly(1-PHE-6)0.56 measured by dynamic mechanical analysis with a strain rate of 2.5 %/min at room temperature. Three samples were tested for each polymer film. The elastic moduli were obtained in the linear region of the stress-strain curve and the average value of three samples was calculated. Incorporation of iodine in poly(1-PHE-6) makes the normally brittle PEU more ductile. However, with increasing iodine content, the PEUs again become brittle. Poly(1-iPHE-6) homopolymer for example was too brittle to be measured by DMA. The elastic moduli of PEUs decreased following iodine modification.

Figure 3.3 shows the stress-strain curves of poly(1-PHE-6), poly(1-iPHE-

6)0.24-co-poly(1-PHE-6)0.76 and poly(1-iPHE-6)0.44-co-poly(1-PHE-6)0.56 as obtained

by DMA. The poly(1-PHE-6) film was brittle with no observable yield point, while

both poly(1-iPHE-6)0.24-co-poly(1-PHE-6)0.76 and poly(1-iPHE-6)0.44-co-poly(1-PHE-

6)0.56 were ductile with a tensile elongation at break of 205% and 124%, respectively.

Elastic moduli were calculated in the low strain region for all three polymers. As

shown in figure 3.3, elastic moduli for poly(1-PHE-6), poly(1-iPHE-6)0.24-co-poly(1-

55 PHE-6)0.76 and poly(1-iPHE-6)0.44-co-poly(1-PHE-6)0.56 are (2.2±0.12) GPa,

(1.7±0.15) GPa, and (1.5±0.03) GPa, respectively. The incorporation of iodine

decreased the moduli of poly(1-PHE-6). One possible reason is that bulky iodine

atoms interrupt the regular packing of polymer chains. As a result, the hydrogen-

bonding networks of poly(1-PHE-6) are partially broken down since hydrogen

bonding is oriented within a short distance. Less hydrogen bonding interaction leads

to smaller elastic moduli for iodinated poly(1-PHE-6). Modification of poly(1-PHE-6)

with iodine decreases the elastic modulus and also toughens poly(1-PHE-6). The

elongation at break shows a maximum at 205% for poly(1-iPHE-6)0.24-co-poly(1-

PHE-6)0.76. This is possibly due to the competition of two major factors that determine the elongation at break of materials. As mentioned earlier, the incorporation of a small amount of iodine atom decreases hydrogen bonding interactions. The interaction between polymer chains reduces, which makes it easier for chains to slide and results in higher elongation at break for poly(1-iPHE-6)0.24-co-poly(1-PHE-6)0.76 compared

with poly(1-PHE-6). Meanwhile, iodine atoms can be easily polarized, and have

strong interactions with each other, which hinders the sliding of polymer chains. So

with an increase of iodine content in polymers, for example poly(1-iPHE-6)0.44-co-

poly(1-PHE-6)0.56, elongation at break decreases compared with poly(1-iPHE-6)0.24-

co-poly(1-PHE-6)0.76. For poly(1-iPHE-6), The brittleness property is presumed to be

attributed to the strong interactions between polarized iodine atoms.

3.4.4 Radiopacity of PEUs

µ-CT testing (Figure 3.4) shows that incorporation of iodine enhances the

radiopacity of poly(1-PHE-6). In figure 3.4 there are two categories of µ-CT

56 projection images. Figure 3.4 (a-e) are reference aluminum stages with thicknesses of

0.5 mm, 1 mm, 1.5 mm, 2 mm and 2.5 mm, respectively. Since aluminum atoms

hinder X-ray transmission, the radiopacity of reference aluminum stages increases

with larger thickness. By comparing the reference aluminum stages with PEU films

with different iodine content, it is possible to assess the approximate radiopacity of

iodinated PEUs.

Figure 3.4. Micro-CT images of an aluminum wedge 0.5–2.5 mm in 0.5 mm steps (a- e) and PEU films with different iodine content (f-i) with 0.5 mm thickness. Radiopacity of PEUs increases with increasing iodine content. The poly(1-iPHE-6) film (i) has comparable radio contrast to that of the aluminum reference with a thickness of 1 mm (b), the poly(1-iPHE-6)0.44-co-poly(1-PHE-6)0.56 film (h) has comparable radio contrast as that of the aluminum reference with a thickness of 0.5 mm (a), and the radiopacity of poly(1-iPHE-6)0.24-co-poly(1-PHE-6)0.76 film is lower than that of the 0.5 mm thick aluminum reference, but is much higher than that of the poly(1-PHE-6) film (f). Micro-computed tomography (µ-CT) has advantages over traditional x-

radiography as it can be used to reconstruct 3D images of scanned samples showing

internal structures and interconnection.147,148 X-ray attenuation projections from various angles are collected and reconstructed into 2D intersecting slices.149 3D scaffold images can be obtained by stacking the 2D slices together. This method is

57 nondestructive to samples and the 3D images enable us to directly see the inside structure and morphology of the samples and to quantitatively characterize the pore size and porosity.150 Scaffold porosity is related to mechanical properties151-153, degradation,153,154 and cell growth and differentiation.36,150,152,155 Hence, the calculation of porosity is very important. Compared with traditional methods to calculate porosity, such as theoretical methods151,156, mercury porosimetry156,157 and

SEM157,158, µ-CT is able to calculate the volume of closed/open pores, open/total porosity, inter-connectivity and connectivity density.146,159 It is an effective way to obtain quantitative information about scaffold microstructures.

Figure 3.4 (f-i) shows the µ-CT images of poly(1-PHE-6), poly(1-iPHE-6)0.24- co-poly(1-PHE-6)0.76, poly(1-iPHE-6)0.44-co-poly(1-PHE-6)0.56 and poly(1-iPHE-6), respectively. The thickness of all films is 0.5 mm. It is clear that the radiopacity of iodinated PEUs increases with increasing iodine content, as is expected. By comparing the µ-CT results of poly(1-iPHE-6) with the reference, it is obvious that poly(1-iPHE-6) film has similar radiopacity to that of the aluminum reference with a thickness of 1 mm, poly(1-iPHE-6)0.44-co-poly(1-PHE-6)0.56 film has similar radiopacity to that of the aluminum reference with a thickness of 0.5 mm, and the radiopacity of poly(1-iPHE-6)0.24-co-poly(1-PHE-6)0.76 film is lower than the 0.5 mm thickness aluminum reference. Regardless of iodine content, the radiopacity is still much higher than that of the poly(1-PHE-6) film. The contrast of poly(1-PHE-6) is very weak since there is no heavy atom in the material. Hence the radiopacity of polymers can be adjusted by controlling the content of iodine in the copolymer.

Almost any implantable polymeric device needs radio contrast to distinguish it from neighboring tissues and to locate its position within the body. Since different tissues

58 may have different radiopacity, the radiopacity of polymeric implantable devices

should change depending on their implantation location.

Figure 3.5. Reconstruction slices of Micro-CT 3D scanning of porous scaffolds with different iodine content under the same scanning conditions. The images show the cross-section of the scaffold throughout the sample. It is difficult to see the poly(1- PHE-6) scaffold (a) structure because of the poor radiopacity. Poly(1-iPHE-6)0.24-co- poly(1-PHE-6)0.76 (b) and poly(1-iPHE-6)0.44-co-poly(1-PHE-6)0.56) (c) show the internal structure (pore size, pore type and interconnectivity) of the scaffolds. The porosity of poly(1-PHE-6), (poly(1-iPHE-6)0.24-co-poly(1-PHE-6)0.76 and poly(1- iPHE-6)0.44-co-poly(1-PHE-6)0.56) scaffolds were calculated to be (90±1.6)%, (85±0.4)% and (88±0.5)%, respectively.

3.4.5 PEU 3D porous scaffolds analysis

The porosity of scaffolds is an important feature, as it is related to scaffold

mechanical properties and degradation as well as cell attachment, growth and

differentiation.150,151,153 Figure 3.5 shows the µ-CT reconstruction slices from 3D scanning using the same testing conditions between samples. The radiopacity results show the same trend as seen using polymer films. With decreasing iodine content, reduction in contrast is observed. In Figure 3.5, from poly(1-PHE-6) to poly(1-iPHE-

6)0.44-co-poly(1-PHE-6)0.56, the intensity of polymers under µ-CT increases with increasing iodine content. For poly(1-PHE-6) scaffold, it is difficult to see the inside

structures (Figure 5(a)). The copolymer scaffold results show the presence of regular

59 square pores with sizes ranging from 250 µm-400 µm. The porosity of poly(1-PHE-6),

poly(1-iPHE-6)0.24-co-poly(1-PHE-6)0.76 and poly(1-iPHE-6)0.44-co-poly(1-PHE-6)0.56 scaffolds were calculated to be (90±1.6)%, (85±0.4)% and (88±0.5)%, respectively.

While the theory porosity for the scaffolds was 85.6%. The porosity of poly(1-PHE-6) that was calculated to be (90±1.6)% is not reliable since the radio contrast is too low to be accurately calculated, even though exposure time was increased to 70 ms and the voltage was decreased to 40 kV to obtain higher radio contrast for calculation. The fabrication method for the scaffolds was consistent for all three materials, so the porosity difference may be due to some material property, such as brittleness. 3D reconstruction images of scaffolds are shown in the supporting information.

The compression moduli data are summarized in Table 3.3. For all scaffolds, the compression moduli in the wet state are lower than those in the dry state due to water penetration in the scaffolds. In both dry and wet states, the compression modulus has the same trend: poly(1-iPHE-6)0.24-co-poly(1-PHE-6)0.76 scaffold has the

highest compression modulus and poly(1-iPHE-6)0.44-co-poly(1-PHE-6)0.56 has the

lowest. In this test, the compression modulus is mainly related to both inherent

material properties and the physical structure of the scaffolds through this empirical

relationship160:

-bP E=E0e

where E0 is the elastic modulus of the bulk material, P the porosity and b related to the

microstructure. Poly(1-PHE-6) scaffolds easily crumbled following salt-leaching due

to their brittle nature. They do not maintain their original shape after 4 days in PBS,

while iodinated copolymers maintain structural integrity after PBS soaking. Structural

defects in poly(1-PHE-6) scaffolds caused an increase in porosity, hence decrease in

60 the compressive modulus. For iodinated copolymers, poly(1-iPHE-6)0.24-co-poly(1-

PHE-6)0.76 had a superior modulus to poly(1-iPHE-6)0.44-co-poly(1-PHE-6)0.56. This higher compressive modulus is likely related to the higher elastic modulus detected in bulk poly(1-iPHE-6)0.24-co-poly(1-PHE-6)0.76 compared to that of poly(1-iPHE-6)0.44- co-poly(1-PHE-6)0.56.

Table 3.3. Compression modulus of PEU porous scaffolds

Sample Porosity (%) Dry (MPa) Wet (MPa)

Poly(1-PHE-6) (90±1.6)% 2.15±0.17 0.27±0.04

Poly(1-iPHE-6)0.24-co-poly(1- (85±0.4)% 5.03±0.80 0.79±0.08

PHE-6)0.76

Poly(1-iPHE-6)0.44-co-poly(1- (88±0.5)% 1.19±0.04 0.23±0.05

PHE-6)0.56

61 3.4.6 Cell viability and spreading assay

Figure 3.6. MC3T3 cell viability on PEU films (n=3). 10 images were used for calculating cell viability on each sample. (a): poly(1-PHE-6) film; (b): poly(1-iPHE- 6)0.24-co-poly(1-PHE-6)0.76 film; (c): poly(1-iPHE-6)0.44-co-poly(1-PHE-6)0.56) film. (d): Comparison of cell viability on PEU films with different iodine content shows no significant difference in cell viability.

Figure 3.7. MC3T3 cell spreading on PEU films (n=3). 20 images were used for quantification of cell aspect ratio and cell area for each sample. (a) and (d): poly(1- PHE-6) film; (b) and (e): poly(1-iPHE-6)0.24-co-poly(1-PHE-6)0.76 film; (c)and (f): poly(1-iPHE-6)0.44-co-poly(1-PHE-6)0.56) film. (g): Comparison of cell aspect ratio on PEU films with different iodine content. (h): Comparison of cell area on PEU films with different iodine content, which shows that cells are well spread on PEU films, supporting cell viability results. Figure 3.6 shows MC3T3 cells on PEU film with different iodine content.

From the representative pictures, it is evident that living cells predominate and cells are distributed uniformly on the films. The viability of cells on poly(1-PHE-6),

poly(1-iPHE-6)0.24-co-poly(1-PHE-6)0.76, and poly(1-iPHE-6)0.44-co-poly(1-PHE-

6)0.56) is (81.2±11.5)%, (76.8±5.1)% and (81.3±4.6)%, respectively. The observation

of about 20% cell death is presumed due to cell seeding and handling. So PEUs are

non-toxic to cells. This result is supported by cell staining images showing cell actin

63 staining (Figure 3.7). The calculated aspect ratio and cell area are similar for cells seeded on all films, even with different iodine content. This indicates that there is no significant difference in the effect of iodine content on cell activity.

3.5 Conclusion

A compositional series of radiopaque PEUs were synthesized from bis-L- phenylalanine-1,6-hexanediol-diester and bis-4-I-L-phenylalanine-1,6-hexanediol- diester monomers. The polymer compositions were characterized by 1H-NMR and FT-

IR. The data illustrate that iodine can be intrinsically and controllably incorporated into PEUs based on feed ratio with varied content. Iodinated PEUs showed higher glass transition temperatures, thermal stability and radiopacity with a limited decrease in elastic modulus. The radiopacity of 0.5 mm poly(1-iPHE-6) film is comparable to that of a reference aluminum film with a thickness of 1 mm. Importantly, poly(1- iPHE-6)0.24-co-poly(1-PHE-6)0.76 and poly(1-iPHE-6)0.44-co- poly(1-PHE-6)0.56 are ductile. It is unusual for a material to have relatively high modulus and also ductility.

Ductility enables polymer scaffolds to maintain their original shape, which explains why the poly(1-iPHE-6)0.24-co-poly(1-PHE-6)0.76 scaffold has the highest compression modulus. In addition to radiopacity, the modulus and ductility of iodinated PEUs could also be selectively tuned by variation of the iodine incorporation. Cell viability and spreading assays demonstrate iodinated PEUs are non-toxic. We envision that these materials will find widespread application in a number of tissue engineering applications where degradation and contrast are required.

64 3.6 Acknowledgement

The authors are grateful for useful comments from Fei Lin and Yaohua Gao and financial support from the Akron Functional Materials Center.

65 CHAPTER IV

ENHANCED OSTEOGENIC ACTIVITY OF POLY(ESTER UREA) SCAFFOLDS

USING FACILE POST-3D PRINTING PEPTIDE FUNCTIONALIZATION

STRATEGIES

This work has been previously published as

Shan Li, Yanyi Xu, Jiayi Yu and Matthew L. Becker

Biomaterials, 2017, 141, 176-187.

4.1 Abstract

Additive manufacturing has the potential to revolutionize regenerative medicine, but the harsh thermal or photochemical conditions during the 3D printing process limit the inclusion of drugs, growth factors and other biologics within the resulting scaffolds. Functionalization strategies that enable specific placement of bioactive species on the surface of 3D printed structures following the printing process afford a promising approach to sidestep the harsh conditions and incorporate these valuable bioactive molecules with precise control over concentration. Herein, resorbable polymer scaffolds were prepared from propargyl functionalized L- phenylalanine-based poly(ester urea)s (PEUs). Osteogenic growth peptide (OGP) or bone morphogenic protein-2 (BMP-2) peptides was immobilized on PEU scaffolds through surface available propargyl groups via copper-catalyzed azide alkyne cycloaddition (CuAAC) post 3D printing. The presence of either OGP or BMP-2 showed significantly enhanced hMSCs osteogenic differentiation compared to unfunctionalized scaffolds.

66 4.2 Introduction

Advances in additive manufacturing and 3D printing are changing medicine.50,161,162 Collectively, patient specific medical imaging, computer aided design (CAD), and 3D printers with high print fidelity possess many advantages over other lab scale scaffold fabrication methods including salt leaching, phase separation and gas-foaming with respect to processability, reproducibility, and flexibility in architectures and design.163-166 Scaffold design features including pore size, distribution, and interconnectivity are critically important to mechanical properties,152 degradation,167 and bioactivity including cell attachment, proliferation and differentiation.154,155,165,168,169 Using 3D printing techniques, the geometry of scaffolds

170 can be precisely designed to meet the optimized requirements for cell activity.

Synthetic polymers are used widely for 3D printing degradable scaffolds due to their thermal or photochemical processability. However, cells do not have receptors for synthetic materials and uncontrolled protein adsorption often results in poor bioactivity that limits cell attachment, proliferation, and differentiation, leading to slow tissue regeneration or failure of implants.171,172 Modification of scaffolds with growth factors or other bioactive species is necessary and has been shown to promote the formation of extracellular matrix which facilitates tissue regeneration.

Unfortunately, the structure and activity of the bioactive molecules is compromised when exposed to harsh photochemical and thermal printing conditions. Post-3D printing surface modification is thought to be a translationally relevant solution to the problem. Bioactive molecules can be physically173,174 adsorbed or chemically92,175 tethered to the scaffold surface. Since the possible detachment and side effects of bioactive molecules limit the utility of physical adsorption methods, a facile method

67 of chemical tethering is more attractive.176,177 One commonly used surface chemical

attachment strategy involves the use of “click” chemistry such as cycloaddition

reactions, thiol additions, and oxime transformations.178-180 These reactions are

frequently utilized because of their simple reaction conditions, high reaction rate, high

yield with negligible side reactions, and functional group tolerance.181 Zheng et al.

described the post-electrospinning “tri-click” functionalization of nanofibers through

copper-catalyzed azide alkyne cycloaddition (CuAAC), strain-promoted azide alkyne

cycloaddition (SPAAC), and oxime ligation to sequentially attach bioactive Gly-Arg-

Gly-Asp-Ser (GRGDS), BMP-2 peptide, and dopamine molecules on the surface of

nanofibers.182 In another work, GRGDS and Tyr-Ile-Gly-Ser-Arg (YIGSR) peptides were clicked on the surface of nanofibers to enhance Schwann cell attachment and alignment.183 Each had a significant enhancement of bioactivity consistent with the

tethered species. Through rational design, controllable attachment of bioactive

molecules can be achieved and optimized, based on the concentration of surface

182 reactive functional groups on the nanofiber scaffolds.

Osteogenic growth peptide (OGP) and bone morphogenic protein-2 (BMP-2)

are two well-known growth factors that have been shown to regulate cellular activities

for bone regeneration.75,184 OGP is a naturally occurring linear tetradecapeptide

(ALKRQGRTLYGFGG) found in serum.185 Research has shown that OGP surface

functionalized TiO2-coated slides showed enhanced proliferation, osteogenic

differentiation, alkaline phosphatase (ALP) activity and mineralization of seeded

MC3T3-E1 cells in vitro.186 Similarly, it has been reported that OGP upregulates the

osteogenic differentiation and mineralization of hMSCs within OGP-tethered 3D porous scaffolds in vitro. In vivo studies also indicate that OGP promotes collagen

28 deposition and matrix mineralization.

68 Another family of cytokines that promote bone regeneration is bone

morphogenic proteins (BMPs),187 especially BMP-2,188,189

(KIPKASSVPTELSAISTLYL) which has been proved to stimulate stem cell

osteogenic differentiation through recruitment of osteoprogenitor cells.75 Wang and

coworkers reported that BMP-2 boosted the induced pluripotent stem cells (iPSCs)

osteogenic differentiation significantly on ordered nanotopographical BMP-2-phage

matrix characterized by higher ALP activity, osteocalcin (OCN) and osteopontin

(OPN) expression levels than WT-phage matrix (no BMP-2).190 Also, it was shown

that BMP-2 stimulated MC3T3-E1 cell osteogenic differentiation in vitro and new

191 bone formation in vivo in a rabbit calvarial defect model.

As implantable polymeric scaffolds, amino acid-based poly(ester urea)s (PEUs) have shown promise in a number of applications due to their non-toxic properties,

tunable degradation and mechanical properties.23,32,192-194 PEUs can undergo

hydrolytic and enzymatic degradation through the ester and urea bonds. Their

degradation by-products can be metabolized in vivo, which shows that PEUs are non-

toxic and bioresorbable. The limited inflammation observed in vivo suggests that the

degradation by-products are also self-buffering, which is advantageous over

traditional polyesters.29 The PEU properties can be tuned by utilizing different amino

acids and diols.24,27,29,30 In addition, X-ray contrast can also be introduced to PEUs by directly using an iodinated L-phenylalanine derivative.195 Further modification on

PEUs could be designed simply by choosing L-tyrosine as the starting amino acid, in which the functional group or reactive handle is attached directly to the phenol.26,28

The versatility in functionalization of PEUs holds powerful promise for their

application in regenerative medicine. In this paper, we report the copolymerization of

a series of reactive propargyl-functionalized tyrosine monomers, filament generation

69 and 3D printing, and post-printing functionalization of poly(ester urea) scaffolds with

OGP or BMP-2 using a CuAAC “click” reaction. We also demonstrate that peptides

immobilized on the surface of 3D printed PEU scaffolds promote hMSCs osteogenic

differentiation by measuring the expression of osteogenic phenotype markers and

enhanced mineralization.

4.3 Experimental Section

4.3.1 Materials

L-phenylalanine (≥98.5%), Boc-Tyr-OMe (97%), propargyl bromide solution

(80 wt.% in toluene, contains 0.3% magnesium oxide as stabilizer), 1,6-hexanediol

(97%), p-toluene sulfonic acid monohydrate (≥98.5%), activated carbon black, calcium hydride (95%), potassium carbonate (99%), sodium hydroxide (≥98%), sodium carbonate (≥99.5%), triphosgene (98%), di-tert-butyl dicarbonate (99%), hydrogen chloride solution (4 M in dioxane), toluene (GRACS), tetrahydrofuran

(GRACS), dichloromethane (GRACS), chloroform (GRACS), ethanol (200 proof, anhydrous) and N,N-dimethyl formamide (DMF, GRACS) were purchased from

Sigma-Aldrich or Alfa Aesar. Chloroform was dried by calcium hydride and distilled before use. All other chemicals were used as received.

4.3.2 Characterization of chemical structure and thermal properties

1H-NMR (500 MHz) and 13C-NMR (125 MHz) spectra were obtained using

Varian NMR Spectrophotometer. All chemical shifts were reported in ppm (δ).

Abbreviations of s, d and m were used to represent singlet, doublet and multiplet,

70 respectively. The molecular mass of the polymers was obtained from size exclusion chromatography (SEC) analysis (TOSOH HLC-8320 gel permeation chromatograph), using linear polystyrene as standard. Eluent conditions were DMF (with 0.1 M LiBr, flow rate 0.3 mL/min) at 40 °C, and a refractive index detector. UV-visible spectra were recorded using Synergy Mx Microplate Reader (BioTek, Winooski, VT) with scanning wavelength ranging from 230 nm to 325 nm and 2 nm resolution.

Thermogravimetric Analysis (TGA, TA Q500) was used to measure the thermal nd decomposition properties of the polymers at a heating rate of 20 °C/min from room temperature to 600 °C under nitrogen atmosphere. The glass transition temperature,

Tg, was determined using Differential Scanning Calorimetry (DSC, TA Q200) at a

scanning rate of 20 °C/min from 0 °C to 200 °C for 3 cycles. The midpoint of the

transition shown in the second heating cycle was used to determine the Tg.

4.3.3 Synthesis of di-p-toluene sulfonic acid salt of bis-L-phenylalanine-1,6- hexanediol-diester (1-PHE-6 monomer)

The monomer 1-PHE-6 was synthesized according to the previously published

195 methods and is described in the supporting information.

4.3.4 Synthesis of di-hydrochloride acid salt of bis-4-propargyl-L-tyrosine-1,6-

hexanediol-diester (1-pTYR-6 monomer)

The monomer 1-pTYR-6 was prepared as demonstrated in Scheme S4 in the supporting information and characterized by 1H NMR and 13C NMR.

71 4.3.5 Synthesis of homopolymer of 1-PHE-6 monomer (poly(1-PHE-6))

Unfunctionalized PEU was synthesized by interfacial polymerization using 1-

PHE-6 monomer as shown in Scheme 4.1. Briefly, 1-PHE-6 monomer (1.0 equiv.),

sodium carbonate (2.1 equiv.) and distilled water (0.1 M to the monomer) were added

to a 2 L 3-neck round bottom flask. The contents were stirred mechanically in a 35 °C

water bath until the mixture was dissolved. The 35 °C water bath was then replaced

with an ice bath, and upon reaching 0 °C, additional sodium carbonate (1.05 equiv.)

was dissolved in distilled water and added to the flask. Triphosgene (0.35 equiv.)

dissolved in freshly distilled chloroform (0.6 M) was added to the flask quickly. After

30 minutes, an additional aliquot of triphosgene (0.08 equiv.) dissolved in distilled

chloroform (0.6 M) was added to the flask dropwise for 2 h. The crude product was

transferred to a separatory funnel and precipitated into boiling water dropwise to

1 obtain the polymer. (Yield ~92%). H-NMR (500 MHz, DMSO-d6): 1.16 (m, 4H,

−COOCH2CH2CH2−) 1.43 (m, 4H, −COOCH2CH2CH2−) 2.50 (DMSO) 2.83-2.99 (m,

4H, −CHCH2Ar−) 3.29 (s, H2O), 3.95 (m, 4H, −CHCOOCH2CH2−) 4.33-4.40 (m, 2H,

−NHCHCOO−) 6.47 (m, 2 H, −NH−) 7.10-7.33 (m, 10H, aromatic hydrogen). 13C-

NMR (125 MHz, DMSO-d6): 25.32, 28.35, 38.15, 40.03, 54.50, 64.72, 126.97-129.59,

137.33, 157.09, 172.70. Mw = 134 kDa, Mn = 77 kDa, Dm = 1.7. Tg = 59 °C. Td =

294 °C.

4.3.6 Synthesis of copolymer of 1-pTYR-6 monomer and 1-PHE-6 monomer (2:98 molar ratio, poly[(1-pTYR-6)0.02-co-(1-PHE-6)0.98])

Propargyl functionalized PEU (Yield ~94%) was synthesized using the same

procedure as poly(1-PHE-6) described above, except a 2:98 molar feed ratio of 1-

72 1 pTYR-6 and 1-PHE-6 monomer was used. H-NMR (500 MHz, DMSO-d6): 1.15 (m,

4H, −COOCH2CH2CH2−) 1.42 (m, 4H, −COOCH2CH2CH2−) 2.48 (DMSO) 2.81-

3.01 (m, 4H, −CHCH2Ar−) 3.28 (s, H2O), 3.49 (m, 2H, -OCH2C≡CH), 3.93 (m, 4H,

−CHCOOCH2CH2−) 4.32-4.41 (m, 2H, −NHCHCOO−), 4.71 (d, 4H, -OCH2C≡CH),

6.40-6.52 (m, 2 H, −NH−) 7.07-7.31 (m, aromatic hydrogen). 13C-NMR (125 MHz,

DMSO-d6): 25.30, 28.33, 38.14, 40.01, 54.49, 64.71, 126.96-129.58, 137.32, 157.08,

172.69. Mw = 108 kDa, Mn = 67 kDa, Dm = 1.6. Tg = 59 °C. Td = 299 °C.

4.3.7 Synthesis of azide-derivatized peptide

N3-OGP (N3-YGFGG) and N3-BMP-2 (N3-KIPKASSVPTELSAISTLYL)

were synthesized using microwave-assisted solid phase FMOC synthesis (CEM

Microwave Peptide Synthesizer, NC). After dialysis, N3-OGP was characterized using a HCT Ultra II quadrupole ion trap mass spectrometer (Bruker Daltonics, Billerica,

MA) equipped with an electrospray ionization source. (ESI MS, m/z found 638.0

(calculated 638.6), Figure S22). N3-BMP-2 was verified using a Bruker Ultraflex-III

TOF/TOF mass spectrometer (Bruker Daltonics, Inc., Billerica, MA) equipped with a

Nd:YAG laser (355 nm), (MALDI-TOF MS; m/z found 2258.0 (calculated 2257.8),

Figure S22).

4.3.8 3D printed PEU porous scaffold preparation and characterization

To obtain polymer filament devoid of trapped gasses, poly(1-PHE-6) and poly[(1-pTYR-6)0.02-co-(1-PHE-6)0.98] films (100 mm × 80 mm × 1.0 mm) were

vacuum compressed at 160 °C (TMP Technical Machine Products Corp). The

polymer film (3 g) was cut into pieces and fed into a capillary rheometer (Dynisco,

73 LCR 7000 Series) equipped with a 1.5 mm die. The extrusion temperature was 160 °C

and extrusion speed was 16 mm/min. PEU filament (2.4 g) prepared by the capillary

rheometer extrusion was used as feedstock material for 3D printing (CartesioW

equipped with 0.2 mm nozzle). The printing temperature was 160 °C and bed

temperature was 65 °C. Scaffolds with orthogonally knitted porous structure (8.0 mm

in diameter, 1.0 mm in thickness) were printed as designed (Google SketchUp 8) with

2 mm/s printing speed, 20% fill density, rectilinear fill pattern and 0.15 mm layer

height. The structure of the printed porous scaffolds was characterized

nondestructively using X-ray micro-computed tomography (µ-CT, Skyscan 1172). 3D

scanning of scaffolds was carried out using the following parameters: 60 kV voltage,

medium camera, no filter, 30 ms camera exposure preset time and 10.0 µm resolution.

The density of surface reactive propargyl groups on 3D printed scaffolds was

quantified by UV-visible spectroscopy and fluorescence spectroscopy (BioTek,

Winooski, VT) using a fluorescent Chromeo 488 dye surrogate coupled to the surface

via CUAAC reaction. A calibration curve of PEUs was obtained by UV-visible

spectroscopy from polymer solutions at the concentrations of 13.64 μM, 11.37 μM,

9.09 μM, 6.82 μM, 4.55 μM and 2.27 μM in hexafluoro-2-propanol (HFIP). The characteristic UV-visible absorption from the aromatic ring at 257 nm was set as the analytical peak. The calibration curve of Chromeo 488 dye was measured by fluorescence spectroscopy at concentrations of 0.208 nM, 0.166 nM, 0.125 nM, 0.083 nM and 0.042 nM in HFIP using an excitation wavelength (λ485) and measuring the

emission at 511 nm as the analytic peak.

74 4.3.9 Scaffold surface peptide immobilization

N3-OGP and N3-BMP-2 were covalently tethered to the propargyl

functionalized scaffold surface via CuAAC click chemistry. Following ethylene oxide

(EtO) sterilization, poly[(1-pTYR-6)0.02-co-(1-PHE-6)0.98] scaffolds (n=18 for each

group) were immersed in the respective peptide solution (4 mL, 0.5 μM peptide in a

66% v/v solution of ethanol in H2O). 2 mL of CuSO4 solution in H2O (2mg/mL) and 2

mL of sodium ascorbate solution in H2O (4 mg/mL) were then added. The immersed

scaffolds were shaken in a 37 °C incubator for 2 h, then washed using sterilized

distilled H2O (25 mL × 20 times) and 70% v/v ethanol solution (25 mL × 10 times) to

remove any catalyst and non-tethered peptide. Scaffolds were rinsed in Dulbecco's phosphate-buffered saline (DPBS) one hour before cell seeding. All steps were performed in an aseptic environment and solutions used were filtered using a 0.2 μm filter for sterilization.

4.3.10 Human mesenchymal stem cell (hMSC) culture

Female hMSCs (Lonza, Walkersville, MD) were expanded and cultured

following the manufacturer’s protocol using Lonza MSC growth medium kit (Lonza,

Walkersville, MD) at 37 °C in a 5% CO2 humidified atmosphere. Cells (passage 4-5)

were seeded onto 3D printed scaffolds at a density of 2.5 × 105 cells/scaffold and fed

every 3 days for 4 weeks.

75 4.3.11 Immunohistochemical staining and alizarin red staining

All samples were fixed in 3.7% paraformaldehyde (PFA) buffer for 2 hours.

After washing with 1× PBS 3 times, samples were put into the paraffin processor to dehydrate (70% EtOH × 2, 95% EtOH × 2, 100% EtOH × 3, 50% Xylenes/50% EtOH, and 100% Xylenes × 2), embedded in paraffin blocks, sectioned into 5 μm thick sections by microtome and dried at 42 °C overnight before staining. Samples were put into a 60 °C oven for 1h to obtain better attachment between sections and slides, deparaffinized and rehydrated through a series of washes: Xylenes (2 × 2min), 50%

Xylenes/50% EtOH (1 × 2 min), 100% EtOH (2 × 2 min), 95% EtOH (1 × 2 min), 70%

EtOH (1 × 2 min), 50% EtOH (1 × 2 min) and deionized H2O (3 × 2 min).

Immunohistochemistry samples were incubated in pepsin reagent for antigen retrieval (15 min at room temperature) to expose antigenic sites for antibodies to bind.

After washing with 1× TBS buffer (3 × 5 min), samples were incubated in blocking buffer (10% donkey serum in 1×PBS) for 1 h at room temperature to block the non- specific binding. After washing with 1× TBS buffer (3 × 5 min), samples were incubated in runt related transcription factor primary antibody in 1× PBS (RUNX2; v/v 1:100), bone sialoprotein primary antibody in 1× PBS (BSP; v/v 1:100) or osteocalcin primary antibody in 1× PBS (OCN; v/v 1:200) overnight at 4 °C. After washing with 1× TBS buffer (3 × 5 min), samples were incubated in corresponding secondary antibodies conjugated to Alexa Fluor 488 (v/v 1:300 in 1× PBS) or Alexa

Fluor 546 (v/v 1:300 in 1× PBS) for 1 h at room temperature in the dark. After washing with 1× TBS buffer (3 × 5 min), samples were incubated in DAPI solution

(300 nM in 1× PBS) for 15 minutes at room temperature to stain the nuclei. After washing with 1× TBS buffer (3 × 5 min), samples were mounted and imaged using an

76 IX81microscope (Olympus) equipped with Hamamatsu Orca R2 CCD camera and

DAPI, FITC and TRITC fluorescence filters.

After rehydration, histological samples were incubated in freshly made alizarin red solution (2 g in 100 mL dd H2O, pH adjusted to 4.2) for 10 min, then

washed and dehydrated through a series of solvents: 70% EtOH (2 × 2 min), 95%

EtOH (2 × 2 min), 100% EtOH (2 × 2 min), and Xylenes (2 × 2min), mounted by

coverslip and imaged using an IX81microscope (Olympus) equipped with QImaging

Micropublisher 3 camera under bright field.

4.3.12 Alkaline phosphatase (ALP) activity assay

ALP activity was measured by SensoLyte® pNPP Alkaline Phosphatase Assay

Kit (Anaspec, Fremont, CA, AS-72146) following the provided protocol. ALP dilution buffer (1 mL, 1× assay buffer) was added to each scaffold after cell culture.

The system was homogenized by vortex and centrifuged at 2500×g for 10 min at 4 °C to resuspend the cells in the buffer. Following centrifugation, the supernatant was

collected for ALP and total DNA quantification. A standard curve of alkaline

phosphatase was obtained at concentrations of 0, 3.1, 6.2, 12.5, 25, 50, 100, 200

ng/mL to quantify ALP in the sample. A sample/standard solution (50 µL) was mixed

with 50 µL pNPP solution in each well of a 96-well plate. After incubation in the dark

for 1 h at room temperature, the absorbance of ALP was measured at 405 nm. Total

DNA in the samples was measured by Cyquant cell proliferation assay kit (Thermo

Fisher Scientific, C7026). The bacteriophage λ DNA solutions at concentration of 0,

10, 50, 100, 200, 400, 600, 800, and 1000 ng/mL were prepared to obtain a standard

curve to quantify the DNA in the sample. After incubation in the dark for 5 min at

77 room temperature, the fluorescence of the samples was measured by Synergy Mx

Microplate Reader (BioTek, Winooski, VT, excitation at 480 nm and emission at 520

nm).

4.3.13 Quantitative real time reverse transcription polymerase chain reaction (real

time RT-PCR)

RNA for real time RT-PCR characterization was extracted from samples at 2,

3, and 4 week time points by the RNeasy Mini Kit (Qaigen, Valencia, CA) and

quantified by Microplate Reader using a Take3 Multi-Volume Plate. Following the protocol, RNA was reverse transcripted to complementary DNA (cDNA), which was then stored at −20 °C for further use. The real time RT-PCR was performed by 7500

Real Time PCR System (Applied Biosystems) using SYBR Green fluorescence detection system. Human Glyceraldehyde-3-phosphate dehydrogenase (GAPDH) was

used as the housekeeping gene. cDNA derived from hMSCs without any treatment

was set as calibrator sample for data analysis. The value obtained was expressed as

fold changes. PCR primer sets used were as follows: GAPDH forward: 5’-

GACAGTCAGCCGCATCTT-3’; reverse: 5’-CCATGGTGTCTGAGCGATGT-3’;

RUNX2 forward: 5’-GGACGAGGCAAGAGTTTCAC-3’; reverse: 5’-

CAAGCTTCTGTCTGTGCCTTC-3’; BSP forward: 5’-

CCTGGCACAGGGTATACAGG-3’; reverse: 5’-CTGCTTCGCTTTCTTCGTTT-3’;

OCN forward: 5’-CATGAGAGCCCTCACA-3’; reverse: 5’-

AGAGCGACACCCTAGAC-3’.

78 4.3.14 Statistics

Statistical analysis was performed using one-way student’s t-test. * indicates p value < 0.05, and ** indicates p value < 0.01.

4.4 Result

4.4.1 Polymer synthesis

Scheme 4.1. Interfacial polymerization of amino acid-based poly(ester urea)s (PEUs).

1 Figure 4.1. H-NMR spectra (DMSO-d6) of poly(1-PHE-6) and poly[(1-pTYR-6)0.02- * co-(1-PHE-6)0.98]. The resonance b (δ = 4.35-4.39 ppm) is set as the reference peak to calculate the polymer composition. Poly(1-PHE-6): homopolymer of 1-PHE-6 monomer; poly[(1-pTYR-6)0.02-co-(1-PHE-6)0.98]: random copolymer of 1-pTYR-6 monomer and 1-PHE-6 monomer at a molar ratio of 2:98.

80 Figure 4.2. UV-visible Spectra of PEUs. The peak at 257 nm is assigned to phenylalanine absorption due to π−π* transition from aromatic ring. n, π- hyperconjugation from o-propargyl substitution red shifts the π→π* transition from 257 nm to longer wavelength (278 nm). The composition can be easily confirmed by a ratio of the two transitions.

Poly(1-PHE-6) was prepared by the homopolymerization of 1-PHE-6

monomer while PEU containing surface reactive propargyl groups was prepared by a

random copolymerization of the two monomers at a feed ratio of 2:98 (1-pTYR-6

monomer:1-PHE-6 monomer) (Scheme 4.1 and Table 4.1). The chemical structure

and composition were characterized by 1H NMR (Figure 4.1), 13C NMR (Figure S14

and Figure S15) and UV-visible spectroscopy (Figure 4.2). In Figure 4.1, the

+ disappearance of the NH3− protons and the presence of urea protons at 6.46 ppm

demonstrate the successful synthesis of PEUs. In poly[(1-pTYR-6)0.02-co-(1-PHE-

6)0.98] spectrum, the resonance at 4.71 ppm is assigned to the methylene protons in the

propargyl group (-OCH2C≡CH). Using the proton resonance at 4.35-4.39 ppm as a reference, the integration of the proton resonance at 4.71 ppm is calculated to obtain the polymer composition as shown in Table 4.1. The content of each monomer unit in the copolymer is similar to the feed ratio, which demonstrates that the composition of the copolymer can be controlled. In Figure 4.2, the UV-visible absorption at 257 nm corresponds to the π−π* transition in L-phenylalanine. Substitution of the proton in

the para- position of the aromatic ring by o-propargyl group red-shifts the π−π*

transition to longer wavelength (278 nm) since the n, π-hyperconjugation lowers the

energy of π* orbital. 1H NMR, 13C NMR and UV-visible spectroscopy demonstrate the successful polymerization and functionalization of PEUs.

81 Table 4.1. Characterization summary of PEUs

PEU Mw Đm Td Tg

Sample Composition Polymer Film Filament Scaffold Polymer °C °C

A 0:100 134 k 127 k 95 k 84 k 1.7 294 59

B 2:98 * 108 k 101 k 92 k 85 k 1.6 299 59

A: Poly(1-PHE-6) B: Poly[(1-pTYR-6)0.02-co-(1-PHE-6)0.98]

* Precisely, the value from NMR integration calculation is 1.8:98.2.

4.4.2 Thermal properties of PEUs

Since thermal stability of the reactive groups determines if a material can be

3D printed (fused deposition modeling, FDM) processability, the proargyl groups on

PEUs should survive high temperature processing (160 °C). The thermal stability of

PEUs was characterized by TGA (Table 4.1). The temperature at 5% weight loss designates the degradation temperature (Td), which is 294 °C and 299 °C for poly(1-

PHE-6) and poly[(1-pTYR-6)0.02-co-(1-PHE-6)0.98], respectively. Therefore, the processing window, which is the temperature difference between processing temperature and degradation temperature, is fairly wide for phenylalanine and tyrosine-based PEUs. Table 4.1 and Figure S18 show there is only a minimal decrease in molecular mass after three steps of thermal processing. In addition, the 1H NMR

spectra comparing propargyl functionalized PEU polymer and scaffold (Figure S19)

demonstrate that the characteristic peak of methylene protons in the propargyl group

(-OCH2C≡CH) at 4.75 ppm remains unchanged after thermal processing, showing

that the propargyl functional group remains available for future CuAAC click

reactions. For polymers used in bone tissue engineering, the glass transition

82 temperature should be above physiological temperature, which is near 37 °C. The

DSC result shows the glass transition temperatures of poly(1-PHE-6) and poly[(1- pTYR-6)0.02-co-(1-PHE-6)0.98] are 59 °C. Therefore, the thermal stability and glass transition temperature make PEUs a suitable candidate for scaffold fabrication via 3D printing and subsequent implantation in vivo.

Figure 4.3. 3D printed scaffold fabrication procedure. A Dynisco LCR 7000 Capillary Rheometer (a) was used to extrude continuous PEU filament with diameter of 1.75 mm (b). After fused deposition modeling (FDM) 3D printing, porous scaffolds (8 mm in diameter, and 1 mm in thickness) were obtained. (c): Cartesio 3D printer; (d): A schematic of how a FDM 3D printer works; (e): A PEU scaffold being printed. (f): The design of architecture within a scaffold; (g): A photograph of 3D printed PEU scaffold; (h and i): Micro-CT 3D reconstruction images of a 3D printed PEU scaffold. The filament diameter is measured to be around 200 µm and pore size is approximately 400 µm.

83 4.4.3 Filament geometry

The continuous filament prepared using a Dynisco capillary rheometer was measured at 1.75 mm in diameter. The small scale capillary rheometer extrusion method was used successfully to fabricate polymer filament resulting in far less waste compared with traditional extrusion methods (2.4 g filament obtained from 3 g polymer, only 0.6 g polymer wasted).

4.4.4 Printing of scaffolds

The micro-computed tomography (u-CT) results showed that the scaffolds

were printed as designed. The fiber diameter was measured to be 200 µm, which was

consistent with the nozzle size. The orthogonal layer by layer knitted structure

resulted in pore size around 400 µm as shown in the 2D reconstruction image (Figure

S20) and florescent images (Figure S21).

4.4.5 Scaffold surface functional group density

Figure 4.4 shows the quantification of surface functional group concentration

by UV-visible spectroscopy and fluorescence spectroscopy. Poly[(1-pTYR-6)0.02-co-

(1-PHE-6)0.98] (after ethylene oxide sterilization) + dye without Cu catalyst is used as

the control for non-specific physical adsorption quantification. CuAAC click reaction

of poly[(1-pTYR-6)0.02-co-(1-PHE-6)0.98] scaffolds before and after ethylene oxide

sterilization are compared to demonstrate the effect of ethylene oxide on the propargyl

functional groups. Using the calibration curves, the polymer concentration of poly[(1-

pTYR-6)0.02-co-(1-PHE-6)0.98] (after ethylene oxide sterilization) + dye without Cu

84 catalyst, poly[(1-pTYR-6)0.02-co-(1-PHE-6)0.98] (after ethylene oxide sterilization) + dye with Cu catalyst, and poly[(1-pTYR-6)0.02-co-(1-PHE-6)0.98] (before ethylene oxide sterilization) + dye with Cu catalyst are found to be 4.50, 4.31, and 3.53 μM, respectively. According to the 1H NMR integrations, the propargyl functional group in the polymer is 1.8 mol%. The resulting functional group concentration is 81, 78, and 64 nM, respectively. Using the dye calibration curve, the dye concentration attached to these three samples could be calculated as 0.002, 0.014, and 0.011 nM, respectively. Since the reacted dye amount is equal to that of the surface propargyl functional group assuming 100% conversion of the CUAAC reaction, the percentage of surface functionalization (the amount of functional group on the surface relative to that in the bulk) is 0.002%, 0.018%, and 0.017%, respectively. Thus, after subtraction of the physically adsorbed dye, the percentage of surface functionalization group available for click reaction is 0.016% (ethylene oxide sterilized sample) and 0.015%

(Non- ethylene oxide sterilized sample). This result demonstrates that the propargyl functional groups survive the ethylene oxide sterilization protocols. Since the scaffold surface area is 2.36 cm2 and the mass is 14 mg, the surface propargyl functional group density is approximately (75 ± 5) pmol/cm2.

85 Figure 4.4. Quantification of the surface functional groups on the scaffolds by UV- visible spectroscopy (a) and fluorescence spectroscopy (c). (b): Calibration curve for poly[(1-pTYR-6)0.02-co-(1-PHE-6)0.98] (UV-visible absorbance at 257 nm). (d): Calibration curve for Chromeo 488 azide fluorescence emission at 511 nm. Black curve: Propargyl PEU (after EtO) + dye without Cu catalyst; Blue curve: Propargyl PEU (after EtO) + dye with Cu catalyst; Red curve: Propargyl PEU (before EtO) + dye with Cu catalyst. Concentrations of polymer and dye are calculated from calibration curves (insets).

4.4.6 Effects of immobilized peptide on hMSCs osteogenic differentiation

Osteogenic differentiation of hMSCs within 3D printed scaffolds was

monitored by measuring expressions of osteogenic phenotype markers using real time

RT-PCR, ALP activity assay, IHC staining and alizarin red staining. Gene expression was analyzed using real-time RT-PCR study at 2 weeks, 3 weeks, and 4 weeks after cell seeding. RUNX2, as an early stage marker, is a key transcription factor in osteogenic differentiation and bone formation since it triggers osteoblast formation and regulates osteoblast-specific gene expression.196,197 RUNX2 gene expression

(Figure 4.5A) increased by 1.2-fold (p<0.05) and 1.4-fold (p<0.01), respectively, in

OGP and BMP-2 tethered PEU scaffolds compared to the poly(1-PHE-6) control at 2

the week time point. Interestingly, at 3 weeks, despite the increased expression of

RUNX2 in each group, the up-regulation of peptide functionalized group compared

with the control group was no longer observed. The expression of OGP tethered PEU

scaffolds was 0.9-fold (p<0.05) of the control group and there was no significant

difference between the BMP-2 tethered PEU group and control group. The comparison between the peptide functionalized groups and control group at 4 weeks showed similar results; however, the RUNX2 expression level decreased in all groups.

These results suggested that RUNX2 expression reached a peak level before 3 weeks.

86 Overall, for the three scaffolds, RUNX2 expression was enhanced in samples surface functionalized with peptides.

As hMSCs differentiation progressed, the middle stage marker, BSP (Figure

4.5B), was expressed. Bone sialoprotein (BSP) is a non-collagenous protein that

promotes bone development and mineralization. It is expressed by osteoblasts and the

appearance of its expression peak suggests the onset of bone mineralization.189 Thus

BSP expression can be used to determine the differentiation rate. The gene expression

of BSP in OGP and BMP-2 tethered samples was significantly increased by 1.8-fold

(p<0.01) and 3.7-fold (p<0.01), compared with the control at 2 weeks, respectively, which was consistent with the results from RUNX2 expression. The BSP expression

increased and the trend persisted in all three groups at 3 weeks with OGP and BMP-2

tethered samples showing 3.2-fold (p<0.01) and 6.2-fold (p<0.01) higher expression

than control. At 4 weeks, the expression of poly(1-PHE-6) control and OGP tethered

PEU scaffolds increased while BMP-2 tethered PEU scaffolds decreased slightly. The

BSP expression in OGP and BMP-2 tethered samples was 2.5-fold (p<0.01) and 2.2-

fold (p<0.01) higher than the control. It is obvious that BMP-2 tethered samples

passed BSP expression peak before 4 weeks. The increasing rate in OGP tethered

PEU scaffolds dropped (from 3.2-fold at 3 weeks to 2.5-fold at 4 weeks compared

with controls), suggesting there may be a peak expression between 3 weeks and 4

weeks and the hMSCs osteogenic differentiation on BMP-2 tethered samples was

faster than OGP tethered samples.

Another noncollagenous protein in bone is osteocalcin (OCN), a late stage

marker that is secreted by osteoblasts during bone formation. It contains residues of γ- carbossiglutamatic acid that allows the capture of calcium ions, therefore, potentially

87 modulating the growth of hydroxyapatite and regulating the metabolism of bone.198

The gene expression of OCN was consistent with the BSP expression. At 2 weeks, there was no statistical differences between the OGP tethered samples and the control while the expression of BMP-2 tethered samples increased (1.2-fold, p<0.05). The 3 week results showed a similar trend but at higher expression levels. However, at 4 weeks, the expression level in all groups dropped. It is important to note that the drop rate of each group was different, resulting in OCN mRNA expression in BMP-2 tethered samples 1.5-fold (p<0.05) higher than the control. This higher relative expression compared to the 3 week time point indicated the expression peak was reached between 3 weeks and 4 weeks.

IHC staining. Immunohistochemistry staining takes advantage of the principal that antibodies tagged to a fluorophore bind specifically to osteogenic phenotype proteins expressed during differentiation. The fluorescent staining of

RUNX2, BSP and OCN revealed that these three markers were expressed on all samples at 2 weeks and 4 weeks of hMSCs culture on scaffolds (Figure 4.5 D and 4.5

E). RUNX2 (green) was expressed more on scaffolds with OGP or BMP-2 attached at

2 weeks compared with non-functionalized controls, suggesting both peptides upregulated the osteogenic differentiation of hMSCs at the early stage. This confirmed the results from real time RT-PCR. BSP (green) and OCN (red), middle/late stage markers, were expressed more at 4 weeks than 2 weeks as expected.

Figure 4.5. The summary of osteogenic differentiation of hMSCs within 3D printed PEU scaffolds. (A-C): Real time RT-PCR amplification of RUNX2 (early stage marker), BSP (middle stage marker) and OCN (late stage marker) on 2, 3 and 4 weeks in vitro with GAPDH was endogenous control and hMSCs as calibrator sample. Data represent mean ±SEM of 4 determinations. * indicates p value < 0.05, and ** indicates p value < 0.01. (D and E): Immunofluorescence staining for cell nuclei (blue) and BSP (green), OCN (red) and RUNX2 (green) after 2 and 4 weeks in vitro.

ALP activity assay. Alkaline phosphatase (ALP) is an enzymatic catalyst responsible for the phosphate ester hydrolysis into inorganic phosphate, which is critical in the later bone matrix formation.199 Its level and activity is related closely to 89 the mineralization ability. As an early stage marker, at 2 weeks (Figure 4.6), the OGP

functionalized scaffold showed higher ALP expression compared with the poly(1-

PHE-6) control, suggesting the up-regulation of hMSCs osteogenic differentiation,

which is consistent with the results from IHC and real time RT-PCR. There was no

significant difference between the groups at 4 weeks as the hMSCs osteogenic

differentiation had reached the terminal stage.

Figure 4.6. ALP expression (normalized by DNA in sample) of hMSCs within scaffolds after 2 and 4 weeks in vitro. Consistent with the real time RT-PCR results, at 2 weeks, the OGP functionalized scaffolds show higher ALP expression than the poly(1-PHE-6) control scaffold. * indicates p value < 0.05.

Alizarin red staining. To support the results discussed above, mineralization

was also studied using alizarin red staining. Alizarin red chelates with calcium,200

producing a birefringent complex,201 thus calcium deposition is detected. As shown in

Figure 4.7, all samples were stained red, indicating there was calcium deposition on the surface of each scaffold at 2 weeks and 4 weeks. The 4 weeks samples showed more calcium deposition as expected. Furthermore, it is important to note that both the OGP and especially the BMP-2 peptide functionalized scaffolds exhibited

90 significantly more calcium deposition than the poly(1-PHE-6) scaffold at each time point, which is consistent with real time RT-PCR and IHC results.

Figure 4.7. Bright field histological images for calcium deposition by alizarin red staining after 2 and 4 weeks culture in vitro. Fibers are outlined by dotted line. These images clearly demonstrate that the presence of tethered OGP or BMP-2 enhanced hMSCs osteogenic differentiation and mineralization.

4.5 Discussion

In this study, poly(ester urea)s were successfully synthesized with pendant

propargyl group functionality for the post-3D printing attachment of peptide. A combination of FDM 3D printing and click chemistry was used to fabricate bioactive bone tissue engineering scaffolds. These functional scaffolds were designed to enhance the hMSCs osteogenic differentiation. The FDM 3D printing technique is highly flexible in terms of customized design, repeatability, and limited material waste. There is no reporting on the 3D printing of PEUs before this study and this method applies to the whole family of PEUs such as Valine-based PEUs, Leucine-

based PEUs and Phenylalanine-based PEUs. More PEUs 3D printing will be conducted in the Becker lab. With the help of small scale capillary rheometer,

91 filament was fabricated with only 0.6 g polymer waste in each batch. One

disadvantage of FDM printing is that it requires filament, usually manufactured by

extrusion. This is a resource and material intensive process that limits access to small

batches of custom polymer. The capillary rheometer extrusion method solves this

problem by decreasing the required polymer amount from pounds to as little as 3 g or

less. Thus, it opens a door to printing with custom or expensive polymers, which is

particularly meaningful for academic researchers. To the best of our knowledge, no

study using such a small amount of custom polymer in filament fabrication has been

reported.

Identification of scaffold properties that control hMSCs osteogenic differentiation is critical in bone tissue engineering. In our study, the scaffolds were printed as designed (Figure 4.3) with 400 µm interconnected pores to enable cell penetration and nutrition transportation. The concentration of surface available propargyl groups for functionalization was determined to be (75 ± 5) pmol/cm2 using

a combination of UV-visible spectroscopy and fluorescence spectroscopy after

CuAAC with Chromeo 488 dye surrogate. The quantification of 3D printed scaffold

surface is challenging since the filament diameter is usually over 100 µm, especially

in FDM printing. Until now, there have been several reports on the surface

quantification but mostly limited to self-assembled monolayers (SAM) or

nanofibers.182,189 In our scaffold, the fiber diameter is 200 µm. After surface

modification, most of the bulk material remains unreacted. For one propargyl PEU

scaffold (14 mg, 64 µmol repeat unit), there is only 1.16 µmol functional group. The

surface functionalization is (0.015 ± 0.001)%, meaning that the surface reacted

propargyl group is approximately 0.17 nmol. Conventional quantification methods

such as NMR, FTIR and UV-visible spectroscopy alone cannot detect this subtle

92 difference. Surface quantification method, for example, X-ray photoelectron spectroscopy (XPS), can only detect local signal and not the whole scaffold surface.

The combination of UV-visible spectroscopy and fluorescence spectroscopy method provides a universal solution to the quantification of peptide concentration on the 3D scaffold.

OGP and BMP-2 are widely used to promote hMSCs to differentiate into osteoblasts and the subsequent bone mineralization. Policastro et al. has modified

PEUs with OGP by post-polymerization and prepared scaffolds by the salt leaching method,28 resulting in the majority of OGP buried under the scaffold surface. The post

3D printing surface modification method described above requires significantly less

peptides which when coupled with 3D printing afford a translationally relevant and

scalable method for producing bioactive scaffolds.

During the early stage of hMSCs osteoblast lineage differentiation, the hMSCs

first differentiate into RUNX2 and type II collagen producing hMSCs

condensations.202 Then RUNX2 promotes the condensations to differentiate into pre-

osteoblasts. RUNX2, also called core-binding factor 1 (Cbfa1), controls most of the

osteoblast specific genes such as Col I, BSP, OCN and OPN.196,202 Bone maturation

can be hampered with the absence of RUNX2.203 Real time RT-PCR for RUNX2

expression results showed it was up-regulated by OGP or BMP-2 at 2 weeks. Similar

results were observed by Lee et al. that the RUNX2 expression of hMSCs cultured on

PLLA nanofibers in growth medium for 10 days was increased significantly by BMP-

2 derived osteogenic peptide.204 RUNX2 expression depends on the post-translational control by phosphorylation, acetylation, and ubiquitination.205 BMP-2 contributes to the release of RUNX2 from histone deacetylases-7’s (HDAC7) repression by kinase

93 D1 protein activation, which results in HDAC7 phosphorylation and then departure

from the nucleus. BMP-2 also protected RUNX2 from proteolysis by stimulating

205 RUNX2 acetylation. OGP binding to Gi protein cell receptor activates the mitogen-

activated protein kinase signaling pathway. The phosphorylation of kinase 1 and 2

84 protein relates to the regulation of RUNX2.

The elevated expression of RUNX2 in the early stage is essential to induce

osteoblast lineage differentiation. However, for further bone maturation in the late

stage, the expression needs to be suppressed since RUNX2 inhibits the maturation of

osteoblast.206 In our study, all three samples, the poly(1-PHE-6) control, OGP and

BMP-2 tethered samples, showed the trend that RUNX2 expression was first upregulated in immature osteoblast, reached a peak before 3 weeks, and then was downregulated in mature osteoblast. The expression at 2 weeks for all three samples suggested that hMSCs reached preosteoblast stage before 2 weeks, and the expression peak between 2 weeks and 3 weeks indicated hMSCs proceeded to mature osteoblast

before 3 weeks. Lee et al. demonstrated that RUNX2 was strongly expressed at day 7,

and decreased markedly at day 14 in a transfected adipose stem cells osteogenesis

study.207 Similar results were also reported by Mendonca et al. showing the in vivo

RUNX2 mRNA expression of bone tissue surrounding Ru and Zr surface implants

increased by day 14, and then decreased later with all implants showing the lowest

208 RUNX2 expression at day 21.

ALP, a crucial metalloenzyme for bone mineralization, is induced by

RUNX2.209 This explains the up-regulation of ALP activity in peptide tethered scaffolds due to elevated RUNX2 expression at 2 weeks. It was demonstrated that modification of a HA/PLGA substrate with OGP increased ALP activity at both at 7

94 and 14 days.210 The ALP activity stimulation by BMP-2 with the deficiency of

RUNX2 was reported by Komori et al.203 and its mechanism was explored by Kim. et

al.211 It was also shown that OGP could regulate ALP activity through the Rho-

associated protein kinase (ROCK) activation.84 Therefore, both RUNX2 and peptide

regulate the ALP activity. It appears the differences between the 2 week and 4 week

ALP activity in all three samples was not obvious. This is likely due to the fact that

ALP expression increases as the maturation of osteoblast and drops as the maturation

of mineralization.199 According to the expression trend of RUNX2, it is speculated

that there should be an ALP activity peak between 2 weeks and 4 weeks.

As hMSCs osteogenic differentiation precedes from pre-osteoblasts to osteoblasts, BSP, one of the small integrin-binding ligand N-linked glycoprotein

(SIBLING) family212 that regulates mineral formation, was expressed. It is well-

known that BSP is involved in binding to hydroxyapatite through polyglutamic acid

sequences and thus acts as an HA nucleator.213 In our study, BSP was significantly up-

regulated by the OGP or BMP-2 peptide and the expression pattern of BSP might be similar to RUNX2 with a peak between 3 weeks and 4 weeks as described above. We postulate that two factors determine the BSP expression: first, the variation in RUNX2 expression regulates the BSP production positively; second, the RGD (ARG- GLY-

ASP) sequence in BSP promotes cell attachment28 and together with OGP/BMP-2,

79 have a synergistic enhancement on osteogenic differentiation.

In the late stage of hMSCs osteogenic differentiation, osteocalcin, the most

osteoblast-associated marker, is secreted. On our printed scaffolds, OCN expression

peaked between 3 weeks and 4 weeks, which is consistent with the expression of

RUNX2. Run-Tao Gao and coauthors also reported that OCN expression of MSCs

95 cultured in osteogenic-inducing medium showed a peak at 7 days with a RUNX2

expression peak at 3 days.214 As there are many receptor binding sites, the OCN

expression is regulated by various factors.213 The most understood factor is osteoblast-

specific element 2 (OSE2), which has the consensus sequence of CCACA for RUNX2

binding.215 This may explain how RUNX2 works in regulating OCN expression.

4.6 Conclusion

In this manuscript we have demonstrated that 3D printed L-phenylalanine- based poly(ester urea) scaffolds surface functionalized with OGP or BMP-2 peptide are bioavailable and support the up-regulation of hMSCs osteogenic differentiation.

The reactive propargyl handle survives filament generation and the 3D printing process, which will enable the tethering of multiple species following FDM 3D printing. The concentration of the scaffold surface functional groups has been quantified and determined to be (75 ± 5) pmol/cm2 using UV-visible spectroscopy and

fluorescence spectroscopy methods. In vitro hMSCs differentiation analyzed by real

time RT-PCR, IHC staining, ALP activity assay, and alizarin red staining reveal that

both OGP and BMP-2 promote the osteogenic differentiation of hMSCs. Further

studies will be focused on applying these scaffolds to a clinically relevant in vivo rat

cranial critical size defect study.

4.7 Acknowledgement

The authors gratefully acknowledge financial support from The Biomaterials

Division of the National Science Foundation (DMR-1507420) & The Knight

Foundation via the W. Gerald Austen Endowed Chair in Polymer Science and

96 Polymer Engineering. The authors would like to thank Fang Peng for his help in 3D printing.

97 CHAPTER V

CRITICAL SIZED CRANIAL DEFECT REPAIR USING 3D PRINTED

RADIOPAQUE POLY(ESTER UREA) SCAFFOLD MODIFIED WITH

BIOMIMETIC PEPTIDES THAT ENHANCE BONE REGENERATION

5.1 Abstract

Functional repair of critical size bone defects remains a clinical challenge.

While tissue engineering research has yielded significant data on stem cell differentiation, growth factor delivery, and scaffold fabrication methods, few new degradable materials have advanced to pre-clinical evaluation. Herein, the synthesis and characterization of iodine and propargyl functionalized L-phenylalanine-based poly(ester urea)s (PEUs) are reported. Iodine atoms increase the scattering length density and contrast of the materials while the propargyl groups enable post- polymerization and post-3D printing copper-catalyzed azide alkyne cycloaddition

(CuAAC) to immobilize osteogenic growth peptide (OGP) and bone morphogenic protein-2 (BMP-2) to the 3D printed scaffolds. The in vitro human mesenchymal stem cells (hMSCs) cultured on porous scaffolds and in vivo rat cranial defect recovery study using the 3D PEU scaffolds demonstrated that the introduced OGP/BMP-2 promotes in vitro hMSCs osteogenic differentiation, mineralization, and significant new bone formation in vivo. Additionally, the contrast enhanced iodinated PEU scaffolds can be seen clearly after implantation while non-iodinated PEU scaffolds were undetectable under in vivo X-ray imaging.

98 5.2 Introduction

Synthetic bone grafts and scaffolds have emerged with the growing prevalence

of bone tissue engineering therapy for the treatment of critical size bone defects,

which requires the translational in vivo animal study to evaluate the bone remodeling

and possible systemic effects.216 In particular, polymeric materials have attracted immense interest in this field as a consequence of their customizability and cost effectiveness. While polymeric materials have advantages over inorganic counterparts, they are limited in their lack of radiopacity and bioactivity. The absence of radiopacity severely limits imaging of the scaffold once it is implanted and this lack

of diagnostic knowledge prevents polymers from being fully utilized in the clinical

environment. On the other hand, the inert characteristics of polymer-based materials when implanted in the body leave much to be desired.

X-ray based techniques such as X-radiography and computed tomography (CT)

are one of the few diagnostic techniques for implanted materials. The lack of

radiopacity in polymers originates from the elemental composition that is essentially

identical to the surrounding tissue and bone.54 Therefore, approaches to impart

radiopacity in these materials involve chemical or physical introduction of high

atomic mass elements such as iodine, barium, bismuth, bromine, zirconium, strontium

or tungsten.128 Given the drawbacks of physically blending polymers with radio

contrast agents such as inhomogeneous filler dispersion and filler leakage,126,217,218 it

is highly desirable to covalently attach heavy atoms to polymers to generate intrinsic

radiopacity.219,220 Sandker et al. has reported modification of a biodegradable and temperature-responsive PCLA-PEG-PCLA hydrogel with radiopacity through a hydroxyl end group reaction with 2-(2',3',5',-triiodobenzoyl, TIB) moieties followed

99 by the successful assessment of its real time degradation kinetics non-invasively in

221 vitro and in vivo by micro-CT.

Inducing bioactivity in polymeric materials utilizes a similar approach of

physical or chemical introduction of bioactive molecules to the scaffold. In particular,

growth factors are frequently used to regulate cellular activities such as cell

attachment, proliferation, and differentiation. Thus, bone related growth factors have

been studied in applications involving bone regeneration. Among them, osteogenic

growth peptide (OGP) and bone morphogenetic proteins (BMPs) have been explored

extensively.12,71,84,92,222 The amino acid sequence of OGP (ALKRQGRTLYGFGG) is

composed of an active domain (YGFGG) and accessory domain (ALKRQGRTL).84

The active domain promotes proliferation through the activation of mitogen-activated

protein (MAP) kinase signaling pathway and osteogenic differentiation via the RhoA-

ROCK (Rho-associated protein kinase) cellular pathway.84 Xu et al. reported that the

OGP [10-14] functionalized PPF/bioglass composite displayed an increased hMSCs

adhesion, spreading and osteogenic differentiation in vitro compared with controls.223

Besides its enhancement in differentiation, osteogenic differentiation and

mineralization, the availability makes it widely used in the research. The OGP [10-14]

active domain contains 5 amino acids and thus can be obtained easily using chemical

synthetic methods. Bone morphogenic protein-2 (BMP-2) is a particular BMP that has been consistently utilized in bone regeneration. Studies into the knuckle epitope of

BMP-2 (BMP-2 73-92: KIPKASSVPTELSAISTLYL) are of considerable interest as it has shown bioactivity without the negatives of using the native protein.74 Literature

reported that BMP-2 [73-92] promoted the osteogenic differentiation of hMSCs in

vitro characterized by the upregulated osteogenic phenotype marker expression such

as runt-related transcription factor 2 (RUNX2), bone sialoprotein primary (BSP),

100 osteocalcin (OCN) and calcium deposition.224 It has also been demonstrated that bone

recovery was enhanced after BMP-2 [73-92] application in the rat tibial bone defect

model, in which peripheral quantitative computed tomography (pQCT) showed an

77 accelerated and increased new bone formation.

Amino acid-based poly(ester urea)s (PEUs) stand out as a class of bone graft

material due to the unique combination of non-toxic biodegradation, tissue

compatibility, flexible synthesis and tunable mechanical properties.23,32,192,225 PEUs

are hydrolytically and enzymatically degradable due to the ester and urea bonds. The

degradation byproducts can take part in the process of metabolism, which indicates

PEUs are bioresorbable. The neutral degradation byproducts prevent inflammation,

which is advantageous over polyesters.29 By choosing different amino acids and diols according to the material design, PEU properties can be controlled.24,27,29,30 For bone

regeneration applications, L-phenylalanine and 1,6-hexanediol were chosen as the

starting materials because the rigid aromatic side group from L-phenylalanine hinders

the movement of chain and affects chain packing. In addition, 1,6-hexanediol, as a

component of the backbone, contributes limited flexibility. Therefore, the semi-

crystalline structure and the hydrogen bonding in the polymer endow the material

30 strong mechanical properties suitable for bone tissue engineering.

3D printing techniques provide a reproducible and accurate method to

fabricate porous scaffolds in regenerative medicine. With the development of

computer aided design (CAD), 3D printing provides benefits over conventional

scaffold fabrication methods such as salt leaching, phase separation and gas-foaming when it comes to reproducibility and processability.163-166,170,226 Since pore size,

152 distribution, interconnectivity and porosity contribute to mechanical properties,

101 degradation167 and cell bioactivities such as cell penetration, attachment, proliferation and differentiation, nutrition transportation and waste removal,40,154,155,165,168,169,227 it is

significant to fabricate scaffolds in a controlled and repeatable manner. Zhang et al.

has 3D printed hollow-pipe-packed silicate bioceramic scaffolds for vascularized

bone regeneration by a modified core/shell printing nozzle.228 Through their 3D

printing technique, the surface area, morphology and geometry of the scaffold could

be designed to meet the optimized requirements for cell activity accurately and

precisely.170 More importantly, when using tissue engineering methods to repair tissues/organs, patient-custom fabrication is highly preferred. Via CAD and CAM

(computer aided manufacturing) techniques, on-demand and versatile scaffolds could be created as desired. 36,229-231 Hyun-Wook Kang and coworkers utilized an integrated

tissue–organ printer to prepare stable, human-scale tissue constructs of any shape

from cell-laden hydrogels and PCL. The CAD model was created from the clinical

image data and translated into a visualized motion program, which controlled the

232 movement of printer nozzle.

In our previous research,195,224 two modifications have been made separately

to amino acid based poly(1-PHE-6), an excellent bone graft material. First, inherent radiopacity was introduced by the covalent incorporation of iodine and enabled the detection of scaffolds under X-ray imaging. Second, 3D printed scaffolds were functionalized with bioactive peptides, promoting the in vitro osteogenic differentiation and mineralization of hMSCs. This present work combines both the radiopacity and osteogenicity: functionalization of 3D printed PEU porous scaffolds with inherent radiopacity through covalent linked iodine atom to the polymer and osteogenicity by CuAAC to immobilize OGP and BMP-2 peptides on the scaffold

102 surface. The performance of the scaffolds was evaluated in vitro with hMSCs

differentiation assay and in vivo with a rat cranial bone defect recovery study.

5.3 Experimental Section

5.3.1 Materials

All chemicals were purchased from Sigma-Aldrich, Alfa Aesar or Fisher

Scientific and used as received unless otherwise specified. Chloroform was dried by calcium hydride and distilled before use.

5.3.2 Characterization of chemical structure and thermal properties

Chemical structures and thermal properties were characterized with standard

methods. 1H-NMR (500 MHz) and 13C-NMR (125 MHz) spectra were recorded using

a Varian NMR Spectrophotometer. All chemical shifts were reported in ppm (δ).

Singlet, doublet and multiplets are represented by abbreviations of s, d and m,

respectively. Attenuated total reflectance Fourier-transform infrared (ATR-FTIR,

MIRACLE 10, Shimadzu Corp.) spectroscopy was performed using a MIRACLE 10,

Shimadzu Corp. ATR-FTIR Spectrometer with a wavenumber (cm-1) detection

ranging from 4000-400 cm-1 and a resolution of 4 cm-1. The molecular masses and

molecular mass distributions of each of the polymers were determined by size

exclusion chromatography (SEC) using a TOSOH HLC-8320 gel permeation chromatograph that was calibrated with a series of linear polystyrene standards.

Eluograms were collected using DMF containing 0.1 M LiBr as the eluent at a rate of

0.3 mL/min at 40 °C with a refractive index (RI) detector. UV-visible spectra were

103 collected using a Synergy Mx Microplate Reader (BioTek, Winooski, VT) at a

wavelength ranging from 230 nm to 325 nm and resolution of 1 nm. The degradation

temperature of polymers was measured by thermogravimetric analysis (TGA, TA

Q500) at a scanning rate of 20 °C/min from ambient temperature to 600 °C. The 5%

mass loss temperature was set as the degradation temperature (Td). The glass

transition temperature (Tg) was determined using a differential scanning calorimeter

(DSC, TA Q200) over a temperature range of 0 °C - 220 °C at a scanning rate of

20 °C/min for 3 cycles. The midpoint of the second heating cycle was used to

determine the Tg.

5.3.3 Synthesis of di-p-toluene sulfonic acid salt of bis-L-phenylalanine-1,6- hexanediol-diester (1-PHE-6 monomer)

The 1-PHE-6 monomer was synthesized and purified according to the methods

195 previously published.

5.3.4 Synthesis of di-p-toluene sulfonic acid salt of bis-4-I-L-phenylalanine-1,6-

hexanediol-diester (1-iPHE-6 monomer)

The 1-iPHE-6 monomer was synthesized and purified according to the

195 methods previously published.

104 5.3.5 Synthesis of di-hydrochloride acid salt of bis-4-propargyl-L-tyrosine-1,6-

hexanediol-diester (1-pTYR-6 monomer)

The 1-pTYR-6 monomer was synthesized and purified according to the

224 methods previously published.

5.3.6 Synthesis of copolymer of 1-PHE-6 monomer and 1-iPHE-6 monomer (89:11 M ratio, poly[(1-iPHE-6)0.11-co-(1-PHE-6)0.89])

Iodinated PEU was synthesized by interfacial copolymerization as published

using 1-PHE-6 monomer and 1-iPHE-6 monomer at a molar feed ratio of 90:10 as

shown in Scheme 5.1. Unfunctionalized 1-PHE-6 monomer (0.9 equiv.), iodine

functionalized 1-iPHE-6 monomer (0.1 equiv.), sodium carbonate (2.1 equiv.) and distilled water (0.1 M to the monomer) were added to a 2 L 3-neck round bottom flask, followed by mechanical stirring at 35 °C to dissolve the mixture. Ice bath was used to cool the temperature to 0 °C and sodium carbonate (1.05 equiv.) was added to decompose the triphosgene, which was added subsequently to the flask (0.35 equiv. in

0.6 M freshly distilled chloroform). After 30 min of stirring, an additional aliquot of triphosgene (0.08 equiv. in 0.6 M freshly distilled chloroform) was added to the flask dropwise for 2 h. The crude product was transferred to a separatory funnel and the organic layer was precipitated into boiling water dropwise to obtain the polymer.

(Yield ~93%)

1 H-NMR (500 MHz, DMSO-d6): 1.15 (m, 4H, −COOCH2CH2CH2−) 1.42 (m,

4H, −COOCH2CH2CH2−) 2.48 (DMSO) 2.75-2.93 (m, 4H, −CHCH2Ar−) 3.28 (s,

H2O), 3.92 (m, 4H, −CHCOOCH2CH2−) 4.29-4.38 (m, 2H, −NHCHCOO−), 6.46 (m,

105 13 2 H, −NH−) 6.91-7.59 (m, aromatic protons). C-NMR (500 MHz, DMSO-d6): 25.30,

28.33, 38.14, 40.01, 54.49, 64.71, 92.85, 126.96-132.03, 137.32, 157.08, 172.69.

5.3.7 Synthesis of terpolymer of 1-pTYR-6 monomer, 1-iPHE-6 monomer, and 1-

PHE-6 monomer (2:11:87 M ratio, poly[(1-pTYR-6)0.02-co-(1-iPHE-6)0.11-co-(1-PHE-

6)0.87])

Propargyl functionalized iPEU (Yield ~90%) was synthesized by the same interfacial copolymerization method described above but instead using a 2:10:88 M feed ratio of 1-pTYR-6, 1-iPHE-6, and 1-PHE-6 monomers.

1 H-NMR (500 MHz, DMSO-d6): 1.15 (m, 4H, −COOCH2CH2CH2−) 1.42 (m,

4H, −COOCH2CH2CH2−) 2.49 (DMSO) 2.76-2.99 (m, 4H, −CHCH2Ar−) 3.28 (s,

H2O), 3.48 (m, 2H, -OCH2C≡CH), 3.93 (m, 4H, −CHCOOCH2CH2−) 4.31-4.42 (m,

2H, −NHCHCOO−), 4.71 (d, 4H, -OCH2C≡CH), 6.46 (m, 2 H, −NH−) 6.90-7.62 (m,

13 aromatic protons). C-NMR (500 MHz, DMSO-d6): 25.30, 28.33, 38.13, 40.01, 54.58,

64.71, 92.85, 126.96-132.03, 137.31, 157.08, 172.69.

5.3.8 Synthesis of azide-derivatized peptide

N3-OGP [10-14] (N3-YGFGG) and N3-BMP-2 [73-92] (N3-

KIPKASSVPTELSAISTLYL) were obtained using fluorenylmethyloxycarbonyl

(FMOC) microwave-assisted solid phase peptide synthesis using a Liberty 1 peptide synthesizer (CEM Microwave Peptide Synthesizer). The peptide was cleaved from the

Wang resin using a cocktail containing trifluoroacetic acid (TFA): H2O: triisopropylsilane (TIPS) at 95: 2.5: 2.5 v/v and then dialyzed against water for 2 days.

106 Electrospray ionization (ESI) mass spectroscopy (HCT Ultra II quadrupole ion trap

mass spectrometer equipped with an electrospray ionization source, Bruker Daltonics,

Billerica, MA) was used to characterize N3-OGP and matrix-assisted laser

desorption/ionization-time of flight (MALDI-TOF) mass spectroscopy (Bruker

Ultraflex-III TOF/TOF equipped with a Nd:YAG laser (355 nm), Bruker Daltonics,

Inc., Billerica, MA) was used for N3-BMP-2 analysis.

5.3.9 PEU porous scaffold 3D printing and surface functionalization

The PEUs were vacuum compressed into a film at 150 °C (TMP Technical

Machine Products Corp), fractured and then fed into a capillary rheometer (Dynisco,

LCR 7000 Series) equipped with a 1.5 mm die to extrude filament at a speed of 16

mm/min at a temperature of 150 °C. The resulting PEU filament was then printed

using a CartesioW 3D printer equipped with a 0.2 mm nozzle. A 3D printed PEU

scaffold with 8.0/7.8 mm diameter and 1.0 mm thickness was designed by Google

SketchUp 8 and imported into a Slic3r program to set up the printing parameters with

a printing temperature of 150 °C, bed temperature of 65 °C, printing speed of 2 mm/s,

rectilinear fill pattern, 20% fill density, and 0.15 mm layer height. The diameter for

the in vitro cell study was 8 mm. To adjust for the in vivo 8 mm critical size defect,

the scaffold design was adjusted to 7.8 mm in diameter. The radiopacity and structure

of the 3D printed scaffolds were measured using X-ray micro-computed tomography

(µ-CT, Skyscan 1172) and reconstructed using NRecon program.

Azide-derivatized peptides were attached to the scaffold surface using a

CuAAC Huisgen 1,3 cycloaddition click reaction. After ethylene oxide (EtO)

sterilization, propargyl functionalized iPEU scaffolds were immersed in a 4 mL azide-

107 derivatized peptide solution (0.5 µM peptide in a 66% v/v mixture solution of ethanol

in H2O). 2 mL CuSO4 solution (2 mg/mL in H2O) and 2 mL sodium ascorbate

solution (4 mg/mL in H2O) were added and the system was shaken in a 37 °C incubator for 2 h. Distilled H2O (25 mL × 20 times) and 70% v/v ethanol solution in

H2O (25 mL × 10 times) were used to remove the catalyst and remaining peptide. For

in vitro cell seeding, scaffolds were rinsed in Dulbecco's phosphate-buffered saline

(DPBS) for one hour before cell seeding. For the in vivo animal surgery, scaffolds

were rinsed in sterile saline before implantation. All the solutions used here were

sterilized by filtering using the 0.2 µm filters. (Corning, NY14831, Germany)

5.3.10 Human mesenchymal stem cell (hMSC) culture

Female hMSCs (Lonza, Walkersville, MD) were expanded and cultured

following the manufacturer’s protocol using Lonza MSC growth medium kit (Lonza,

Walkersville, MD) at 37 °C in a 5% CO2 humidified atmosphere. hMSCs (passage 4-

5) were seeded onto 3D printed porous scaffolds at a density of 2.5×105 cells/scaffold

and fed every 3 days for up to 4 weeks.

5.3.11 In vitro hMSC osteogenic differentiation on 3D printed peptide functionalized

scaffolds

Alkaline Phosphatase (ALP) activity assay: ALP activity after 2 and 4

weeks culture was measured using a SensoLyte® pNPP Alkaline Phosphatase Assay

Kit (Anaspec, Fremont, CA, AS-72146) following the provided protocol. After cell culture, each scaffold was immersed into 1 mL lysis buffer (20 µL Triton X-100 in 10

mL 1× Assay Buffer) and stored at -80 °C until use. After thawing under room

108 temperature, the sample was homogenized by vortexing and centrifuged at 2500 ×g

for 10 min at 4 °C to completely release ALP from the sample, and the supernatant

was then collected for ALP and total DNA quantification. To quantify the ALP, a

calibration curve of ALP was obtained at concentrations of 0, 3.1, 6.2, 12.5, 25, 50,

100, 200 ng/mL. A 50 µL sample of each solution was mixed with 50 µL pNPP

solution in each well of a 96-well plate and incubated in the dark for 1 h at room temperature. ALP absorbance at 405 nm was measured via a plate reader. Each

experiment was replicated three times. The standard curve was fit using a linear

equation by plotting absorbance vs ALP concentration. The ALP activity was

normalized by the total DNA of the corresponding samples, which was measured by a

CYQUANT® cell proliferation assay kit (Thermo Fisher Scientific, C7026). A

standard calibration curve of λ DNA was obtained at concentrations of 0, 10, 50, 100,

200, 400, 600, 800, and 1000 ng/mL to quantify the DNA in the sample. The

fluorescence intensity of each solution (Ex480 nm, Em520 nm) was measured via a plate

reader. The standard curve was fit using a linear equation by plotting fluorescence

intensity vs DNA concentration and the fluorescence intensity of samples was used to

calculate the total DNA.

Immunohistochemical staining: Protein expression was also evaluated after

2 and 4 weeks of cell culture. Three osteogenic markers were used in the present

study: RUNX2, BSP and OCN. Harvested samples were fixed in 3.7%

paraformaldehyde (PFA) buffer for 2 hours at room temperature and washed with 1×

PBS buffer 3 times. Samples were then processed by a paraffin processor with

different solvents (70% EtOH × 2, 95% EtOH × 2, 100% EtOH × 3, 50% Xylenes/50%

EtOH, and 100% Xylenes × 2), and embedded in paraffin blocks. Then, 5 μM thick

sections of the blocks were cut by a microtome (Leica RM2255, Leica Biosystems),

109 put into a 60 °C oven for 1 h for better attachment between sections and slides, and

finally deparaffinized and rehydrated through a series of washes: Xylenes (2 × 2min),

50% Xylenes/50% EtOH (1 × 2 min), 100% EtOH (2 × 2 min), 95% EtOH (1 × 2

min), 70% EtOH (1 × 2 min), 50% EtOH (1 × 2 min) and deionized H2O (3 × 2 min).

Before labeling, samples were incubated in 0.5% pepsin reagent for antigen

retrieval for 15 min at room temperature. After washing with 1× TBS buffer (3 × 5 min), slides were incubated in blocking buffer (10% donkey serum in 1× PBS) for 1 h at room temperature to block the non-specific antibody binding. After 1× TBS buffer washing, slides were incubated in RUNX2 primary antibody in 1× PBS (v/v 1:100),

BSP primary antibody in 1× PBS (v/v 1:100) or OCN primary antibody in 1× PBS

(v/v 1:200) overnight at 4 °C. After 1× TBS buffer washing, samples were incubated in corresponding secondary antibodies conjugated to Alexa Fluor 488 (v/v 1:300 in

1× PBS) or Alexa Fluor 546 (v/v 1:300 in 1× PBS) for 1 h at room temperature in the dark. After 1× TBS buffer washing, samples were incubated in DAPI solution (300 nM in 1× PBS) for 15 min at room temperature to label the nuclei. After 1× TBS buffer washing, samples were mounted and imaged using an IX81 microscope

(Olympus) equipped with a Hamamatsu Orca R2 CCD camera and DAPI, FITC and

TRITC fluorescence filters.

Alizarin red S. staining: After paraffin removal and rehydration washes,

samples were incubated in freshly made alizarin red solution (2 g in 100 mL dd H2O, pH adjusted to 4.2) for 10 min at room temperature, then washed and dehydrated through a series of solvents: 70% EtOH (2 × 2 min), 95% EtOH (2 × 2 min), 100%

EtOH (2 × 2 min), and Xylenes (2 × 2min), mounted by coverslip and imaged using

110 an IX81 microscope (Olympus) equipped with a QImaging Micropublisher 3 camera

under bright field.

5.3.12 In vivo rat critical size defect recovery study using 3D printed scaffolds

5.3.12.1 Animal surgeries

The rat critical size defect model was approved (17-01-01 BRD) by the

Institutional Animal Care and Use Committee (IACUC) at The University of Akron.

Procedures were followed according to the protocol.233 After acclimation for at least 6

days, sixty 12-week old male Sprague-Dawley rats were randomly assigned to 5

experimental groups for implantation of polymer cranial scaffolds made of the

following materials: PLLA control; PEU; PEU + BMP-2; iPEU; and iPEU + BMP-2.

Rats were initially anesthetized with 3% isoflurane/O2 gas in a chamber after which

the top of cranium was shaved. The head was fixed into a rat stereotaxic instrument

equipped with an anesthesia mask to maintain isoflurane/O2 gas during the surgery. A

subcutaneous injection of 0.5% lidocaine (5mg/mL, 7mg/kg dosage) and

buprenorphineSR (slow release) (1.0mg/kg) were given at the incision site and

shoulder for analgesia, respectively. The incision site was then sterilized using aseptic

technique with an alcohol wipe and povidone-iodine swab. A mid-cranial incision of approximately 2 cm was gently made starting from behind the orbital to the posterior nuchal line to expose the periosteal layer. After the separation of the skin layer from periosteum layer, a blunt incision was made to the periosteum, which was then separated from the underneath cranial bone. After identifying the key cranial markers of bregma and lambda, an 8 mm defect was created using a trephine bur with care to

avoid damage to the underlying dura and superior sagittal sinus vein. Constant saline 111 solution was used to wash away the bone debris to prevent overheating. The craniotomy segment was then removed and replaced by the 3D printed scaffolds. A hemostatic gauze called SURGICEL® (Ethicon, Somerville, New Jersey) was placed over the defect to stop profuse bleeding. After implantation, the periosteal incision was closed using a running 4-0 PGA resorbable suture and the skin was closed using a simple interrupted 3-0 plain gut suture. Lastly, a subcutaneous injection of 4 mL normal saline was given for possible fluid loss during the operative period. Two implantation time points were tested: 4 and 12 weeks. At each time point, six rats from each of the five groups (n=6) were euthanized with an intraperitoneal injection of 1 mL Fatal-Plus containing sodium pentobarbital. The dorsal portion of the skull with the embedded implants was retrieved and fixed in 10% phosphate buffered formalin for 48 h at 4 °C. Sample type was kept blind throughout the entire experiment by utilizing a numbering system. The survival rate of the animal surgery was 85%.

5.3.12.2 Micro-CT evaluation

Following fixation, samples were wrapped in parafilm and scanned with a µ-

CT scanner (Skyscan 1172) vertically to the coronal aspect of the cranial bone.

Acquisition parameters were: 10 µm resolution, medium camera, 0.5 mm aluminum filter, 40 kV, 250 µA, and 0.4° rotation step. The resulting 2D micro-radiographic images in different angles were reconstructed by NRecon software to generate a series of 2D cross sectional images of the sample, which were further analyzed by a CT analyzer program to quantify the new bone growth. The thresholding range was set from -727 to 11709 Hounsfield units (Hu) to remove the effect of contrast from

112 iodinated scaffold. After thresholding, processing and calculation, the new bone volume was obtained.

5.3.12.3 Histology analysis

After µ-CT scanning, the specimens were dehydrated using a series of acetones, 50%, 75%, and 100% × 2, each for 24-48 hours followed by immersion in

PMMA monomer 2× for 24-48 h. Each specimen was individually embedded in freshly made PMMA polymer, allowed to infiltrate in the refrigerator for 48 h and then allowed to polymerize at room temperature until hard. The block was removed from the glass embedding vial and trimmed to the region of interest using an Isomet low speed saw (Buehler Ltd., Evanston, IL), then cut at 5 μm using a Leica RM2255 microtome outfitted with the appropriate block holder and a carbide-tungsten blade.

Sections were placed on microscope slides that had been coated with Haupt’s adhesive (Newcomer Supply, Middleton, WI), flattened with plastic wrap, clamped, and allowed to dry for 48 h in a 60 °C oven. The slides were separated, plastic wrap removed, and the samples were de-plasticized in methyl acetate for 30 min, allowed to briefly air dry, then re-hydrated in graded alcohols (100% × 2, 95% × 2 and 70%) each for 2 min, before rinsing in distilled water. H&E and Goldner’s trichrome staining were done using routine staining protocols. All the sections were imaged using an IX81 microscope (Olympus) and analyzed by Olympus cellSens Dimension software. Quantification of new bone growth was calculated by color thresholding.

113 5.3.13 Statistics

Statistical analysis was performed using two-way analysis of variance

(ANOVA). All the experiments in vitro and in vivo were blinded.

5.4 Result

5.4.1 Polymer synthesis

Scheme 5.1. Interfacial polymerization of amino acid-based poly(ester urea)s (PEUs).

114

1 Figure 5.1. H-NMR spectra (DMSO-d6) of poly[(1-iPHE-6)0.11-co-(1-PHE-6)0.89] and * poly[(1-pTYR-6)0.02-co-(1-iPHE-6)0.11-co-(1-PHE-6)0.87]. The resonance b (δ = 4.29- 4.42 ppm) is set as the reference peak to calculate the polymer composition. Poly[(1- iPHE-6)0.11-co-(1-PHE-6)0.89]: random copolymer of 1-PHE-6 monomer and 1-iPHE- 6 monomer at a molar ratio of 10:90; poly[(1-pTYR-6)0.02-co-(1-iPHE-6)0.11-co-(1- PHE-6)0.87]: random terpolymer of 1-pTYR-6 monomer, 1-iPHE-6 monomer and 1- PHE-6 monomer at a molar ratio of 2:10:88.

Figure 5.2. UV-visible spectra of PEUs. The peak at 257 nm corresponds to L- phenylalanine absorption due to π−π* transition from aromatic ring (B band), which in the propargyl functionalized PEU red shifts to 278 nm due to n, π-hyperconjugation with o-propargyl substitution. The strong absorbance at 238 nm results from the red shift of π−π* transition (K band) due to n, π-hyperconjugation with iodine atom.

115 We have previously demonstrated the synthesis and characterization of L-

phenylalanine-based poly(ester urea)s homopolymers. Herein, we show that the

properties and functionality can be tuned widely with a copolymerization method

using monomers with different functionality. In this study, three monomers were

prepared: an unfunctionalized 1-PHE-6 monomer, an iodinated monomer (1-iPHE-6)

to introduce radiopacity, and a propargyl derivatized tyrosine monomer (1-pTYR-6)

as a reactive handle for peptide attachment. The random copolymerization resulted in

a number of polymers with different derivatization chemistries. Poly[(1-iPHE-6)0.11-

co-(1-PHE-6)0.89] was prepared by the random copolymerization of 1-PHE-6

monomer and 1-iPHE-6 monomer at a molar ratio of 10:90 and poly[(1-pTYR-6)0.02-

co-(1-iPHE-6)0.11-co-(1-PHE-6)0.87] was prepared by random copolymerization of 1-

pTYR-6 monomer, 1-iPHE-6 monomer and 1-PHE-6 monomer at a molar ratio of

2:10:88. Their chemical structures and composition were confirmed by 1H NMR

(Figure 5.1), 13C NMR (Figure S23 and S24), FT-IR spectroscopy (Figure S25) and

UV-visible spectroscopy (Figure 5.2).

In Figure 5.1, the chemical shift at 6.46 ppm corresponds to the urea resonance in the polymers, demonstrating the successful synthesis of PEUs. As previously published, both polymers possess characteristic resonances of poly(1-PHE-6)

(aromatic resonance at 7.1 to 7.3 ppm) and poly(1-iPHE-6) (aromatic resonance at

6.95 and 7.6 ppm), showing that the para-substituted iodinated copolymer was successfully synthesized. These distinct integrations were used to determine the amount of iodinated monomer in the polymer, with the proton resonance at 4.29-4.42 ppm as reference. For the poly[(1-pTYR-6)0.02-co-(1-iPHE-6)0.11-co-(1-PHE-6)0.87]

spectrum, the chemical shift of the methylene protons in the propargyl group (-

OCH2C≡CH) appears at 4.71 ppm, showing the incorporation of clickable reactive

116 handles. The integration based on the reference peak was used to calculate the amount

of propargyl groups in the polymer chain as shown in Table 5.1. In Figure 5.2, the

UV-visible peak at 257 nm is assigned to the L-phenylalanine absorption due to π−π*

transition from aromatic ring (B band), which in the propargyl functionalized PEU red

shifts to 278 nm due to n, π-hyperconjugation with O-propargyl substitution. Both

PEUs have strong characteristic absorbance at a wavelength shorter than 230 nm due to π−π* transition (K band), which is beyond the spectrometer limitation. However,

the iodine substitution of the aromatic proton red-shifts this π−π* transition band to

longer wavelength due to n, π-hyperconjugation. Both polymers have strong absorbance at 238 nm, which corresponds to the K band of the 11% iodinated component. From the result of 1H NMR and UV-visible spectroscopy, it is

demonstrated that PEUs were successfully polymerized at the desired mole ratio of

functional species as designed.

Table 5.1. Characterization summary of PEUs

PEU Composition Mw Đm Td/°C Tg/°C

Sample PHE:iPHE:pTYR Polymer Film Filament Scaffold Polymer

A 89:11:0 99 k 88 k 65 k 57 k 1.6 276 61

B 87:11:2 144 k 110 k 102 k 97 k 1.6 299 60

A: Poly[(1-iPHE-6)0.11-co-(1-PHE-6)0.89]

B: Poly[(1-pTYR-6)0.02-co-(1-iPHE-6)0.11-co-(1-PHE-6)0.87]

117 5.4.2 Thermal properties of poly(ester urea)s

To utilize the 3D printing technique of fused deposition modeling (FDM), a polymer must be formed into a filament which can then be extruded and printed in the melt state. Thus, the polymers are required to withstand high temperature processing which may be an issue with degradable polymers where mass loss due to hydrolysis can be a significant problem. From TGA analysis, the degradation temperatures (Td) of poly[(1-iPHE-6)0.11-co-(1-PHE-6)0.89] and poly[(1-pTYR-6)0.02-co-(1-iPHE-6)0.11- co-(1-PHE-6)0.87] were 276 °C and 299 °C, respectively, which demonstrates that the

PEUs can be melt processed at 150 °C. In addition to TGA analysis, the resulting products from each step of thermal processing were characterized by SEC to assess thermal degradation. Table 5.1 and Figure S28 (SEC eluograms) show that for poly[(1-iPHE-6)0.11-co-(1-PHE-6)0.89], the molecular mass was reduced from 99 kDa to 57 kDa over three processing steps and for poly[(1-iPHE-6)0.11-co-(1-PHE-6)0.89] and for poly[(1-pTYR-6)0.02-co-(1-iPHE-6)0.11-co-(1-PHE-6)0.87], the molecular mass dropped from 144 kDa to 97 kDa using the same procedures. While there is an appreciable decrease in the molecular mass in each step, which is expected, the resulting scaffold molecular mass is still high enough to maintain the bulk thermal and mechanical properties that are inherent in PEUs. In addition, for poly[(1-pTYR-

6)0.02-co-(1-iPHE-6)0.11-co-(1-PHE-6)0.87], it is critical for the propargyl groups to survive high temperature processing for future peptide attachment. This is characterized by comparing the 1H NMR spectra between each of the thermal processing steps between the polymer synthesis and scaffold generation (Figure S29).

The relative intensity of the methylene protons in the propargyl group (-OCH2C≡CH) observed at 4.75 ppm remains constant after three steps of thermal processing,

118 proving that the propargyl groups are not lost during thermal processing up to the

final fabrication step.

The glass transition temperature (Tg) of polymers was characterized using

DSC (Table 5.1), which is 61 °C for poly[(1-iPHE-6)0.11-co-(1-PHE-6)0.89] and 60 °C

for poly[(1-iPHE-6)0.11-co-(1-PHE-6)0.89] and poly[(1-pTYR-6)0.02-co-(1-iPHE-6)0.11-

co-(1-PHE-6)0.87]. The glass transition temperature is important in that it delineates

the 3D printing bed and annealing temperature of the printed product. For FDM

printing, one disadvantage is the potential for residual stress in the product, which

could lead to the scaffold deformation via curving and shape instability. The residual

stress in minimized by choosing a suitable bed temperature and annealing temperature,

usually near the glass transition temperature to facilitate short range chain segment

movement. In addition, for bone graft applications, polymers should have glass transition temperature/melting temperature above 37 °C to provide structural stability and mechanical support to the implantation site.

119 5.4.3 3D printed scaffold characterization

Figure 5.3. 3D porous scaffold printing and characterization. (A): a schematic of the working principle of FDM printing; (B): 3D printed PEU scaffold; (C-E): µ-CT 3D scanning results of scaffolds: poly(1-PHE-6) control (C), poly[(1-iPHE-6)0.11-co-(1- PHE-6)0.89] scaffold (D), and poly[(1-pTYR-6)0.02-co-(1-iPHE-6)0.11-co-(1-PHE-6)0.87] scaffold (E). From top to bottom: shadow projection of scaffolds under X-ray, 2D reconstructed cross-section of scaffolds; and 3D reconstructed images. The iodinated PEU scaffolds show higher radiocontrast than the poly(1-PHE-6) control.

The radiopacity of PEUs and morphology of scaffolds were characterized by

µ-CT. Compared with poly(1-PHE-6), the incorporation of iodine increases the

scattering length density of the polymer which serves to enhance the X-ray contrast.

Figure 5.3 illustrates that the reconstructed image of the non-iodinated PEU, poly(1-

PHE-6), displays minimal contrast between the polymer and the surrounding air,

leading to poorly refined images under X-ray. Since the scattering length density is

similar to the body tissue due to the compositional similarity of hydrogen, nitrogen,

120 carbon and oxygen atoms, tracking of the scaffold after implantation using

fluoroscopic imaging would be difficult. The iodinated PEU scaffolds render distinct

images while also exhibiting a strut distance of 400 µm and a strut size of 200 µm as

expected from the FDM nozzle size. The orthogonal knitted struts are clearly seen in

the scaffold with interconnected channels around 400 µm.

5.4.4 In vitro hMSCs osteogenic differentiation on peptide functionalized scaffolds

Figure 5.4. Effect of OGP or BMP-2 peptide on the protein expression of osteogenic markers, RUNX2 (early stage marker), BSP (middle stage marker) and OCN (late stage marker) in hMSCs cultured in 3D printed scaffolds in vitro. Immunofluorescence staining after 2 weeks and 4 weeks with cell nuclei stained with blue color, BSP with green color, OCN with red color and RUNX2 with green color. The introduction of peptide increased the expression of osteoblast protein markers.

121

To examine the osteogenic potential of peptide functionalized scaffolds, hMSCs were seeded into the 3D scaffolds and cultured in growth medium for 2 and 4 weeks. Osteogenic differentiation and mineralization were analyzed using osteoblast related markers using an ALP activity assay, immunohistochemical staining and alizarin red staining.

On the protein level, up-regulation of the osteoblast specific markers was observed using immunohistochemistry staining after 2 and 4 weeks of culture in vitro

(Figure 5.4). Represented by strong fluorescence (green color), the RUNX2 protein production was higher in the peptide attached scaffolds at both 2 and 4 week and in each group, the 4 weeks samples showed more RUNX2 secretion than 2 week samples. For BSP (green) and OCN (red) protein, similar expression profiles were observed where 4 week scaffolds produced more protein than 2 week scaffolds. BSP and OCN were expressed more on the peptide functionalized scaffolds than the control at 4 weeks. Therefore, the protein production by IHC staining demonstrated the peptide promotion effect on the osteogenic differentiation.

To further analyze the osteogenic differentiation, an ALP activity assay was conducted. Figure 5.5 shows that at 2 weeks, the ALP activity of OGP or BMP-2 attached samples increased by 2.1-fold and 1.9-fold, respectively, compared with the iPEU control. There results demonstrated an up-regulation of hMSCs osteogenic differentiation within the peptide-functionalized scaffolds, which corresponds with the results from IHC. No significant differences were found between the groups at 4 weeks as hMSCs differentiation entered later stages of differentiation.

122

Figure 5.5. Effect of OGP or BMP-2 peptide on the ALP activity (normalized by total DNA) of hMSCs cultured in 3D printed scaffolds after 2 and 4 weeks in vitro. Significant difference was detected at early stage of osteoblast differentiation (2 weeks) on the peptide attached samples. (N=3, each sample were tested three times). Groups that do not share a letter are significantly different.

123 Figure 5.6. Effect of OGP or BMP-2 peptide on the mineralization of hMSCs seeded in the 3D printed scaffolds after 2 and 4 weeks culture in vitro analyzed by alizarin red staining, in which matrix calcification is shown with red deposition and black mineral nodules. The dotted line outlines the 3D printed polymer struts. The peptide functionalized samples revealed significantly higher calcification deposition than the control at both 2 and 4 weeks.

To detect the matrix mineralization, the final stage of osteoblast differentiation,

calcium deposition was stained with alizarin red. As shown in Figure 5.6, the

mineralization was strongly stimulated by peptide at both time points, especially 4

weeks, as seen on the polymer strut surface that is covered by an extensive amount of

black mineral nodules and red deposition of the mineralized extracellular matrix. In

comparison, there is a minimal amount of black mineral nodules and red deposition

on the scaffold surface of the control, even after 4 weeks of culture. In addition, the

iPEU + BMP-2 group showed more mineral nodules than the iPEU + OGP group at 2 and 4

weeks, which suggested that BMP-2 promoted the hMSCs differentiation and mineralization

further than OGP.

5.4.5 In vivo rat cranial critical size defect recovery

An 8 mm rat cranial critical size defect model was used to demonstrate the

clinical translational potential of radiopaque and BMP-2 modified 3D porous PEU scaffolds. The new bone formation was systematically investigated at 4 and 12 weeks using µ-CT 3D scanning, H&E staining and Goldner’s trichrome staining (Figure 5.7 and 5.8).

The µ-CT images (Figure 5.7a) demonstrated that the new bone grew from the defect margins and dural side into the pores. As anticipated, non-iodine functionalized scaffolds showed no contrast with the tissue, so they were indiscernible using X-ray

124 imaging after implantation. For the iodinated PEU groups, the scaffold contrast was enhanced so they could be easily distinguished from the surrounding tissues and bones in vivo under X-ray. The µ-CT quantification result (Figure 5.7b) illustrated that BMP-2 promoted new bone formation, which was consistent with the in vitro results. At 4 weeks, PEU + BMP-2 scaffolds stimulated bone repair (9.48 ± 1.07 mm3) as compared to PEU control (6.54 ± 0.87 mm3) and the iPEU + BMP-2 scaffolds

(9.24 ± 0.51 mm3) showed 1.22-fold increase in bone volume compared with the iPEU scaffolds (7.60 ± 0.72 mm3). At 12 weeks, all groups showed increased bone volume compared to the earlier time point. Significantly more bone formation was found in the PEU + BMP-2 scaffolds (19.0 ± 1.41 mm3) than the PEU control (14.21

± 1.34 mm3).

Undecalcified transverse PMMA sections were stained with Goldner’s trichrome and decalcified transverse PMMA sections were stained with H&E to evaluate the bone tissue such as osteoid and mineralized new bone within the defect site (Figure 5.7a, 5.7c and 5.8). PEU + BMP-2 sample at 12 weeks was used to show the detailed information of new bone ossification. From the H&E images in Figure

5.8, the scaffold strut, calcified collagen (new bone), uncalcified collagen (fibrous connective tissue), newly formed blood vessels, and osteoblasts in the defect site were clearly identified. As early as 4 weeks, the scaffold pores and gap between the scaffold and host bones were filled with abundant uncalcified fibrous connective tissue and some calcified new bone near the dural side and defect margins. At 12 weeks, more of the uncalcified fibrous connective tissue matured into calcified new bone from the defect margins toward the center. The scaffold struts were directly surrounded by uncalcified fibrous connective tissue, calcified newly formed bone or adjacent to bone with a thin layer of a fiber capsule between, as shown in Figure 5.8,

125 which suggested minimal to no chronic inflammation in the defect. It is interesting to

note that in the BMP-2 attached scaffold, two varying new bone configurations were

formed: lamellar bone under the scaffold and woven bone in the pores of scaffold

(Figure 5.8a, 5.8b and 5.8d). The woven bone in the pores was similar in morphology

to the fibrous connective tissue and the lamellar bone under the scaffold was similar

to the host bone.

Similar observations were found in the Goldner’s trichrome labeled samples in

which the uncalcified fibrous connective tissue was stained red and mineralized bone

was stained green. A third new bone structure was found in Figure 5.8(e-h), showing

blue mineralized fibrous connective tissue interwoven with the unmineralized fibrous

connective tissue. The quantification of mineralized bone in the defect is shown in

Figure 5.7c. The PEU control group demonstrated (9.10 ± 2.13)% and (22.90 ± 5.27)%

bone recovery at 4 and 12 weeks, respectively. For the PEU + BMP-2 scaffolds, bone

recovery was (9.42 ± 2.19)% at 4 weeks and (30.91 ± 4.05)% at 12 weeks. The bone

recovery of PEU + BMP-2 group was 1.35-fold more than PEU control at 12 weeks.

By comparing the iPEU group (11.90 ± 1.68)% with the iPEU + BMP-2 group (17.04

± 1.77)% at 4 weeks, the BMP-2 improved the bone recovery by 1.43-fold.

126

Figure 5.7. Bone formation of each group after 4 and 12 weeks. Visible in the µ-CT 3D results, new bone grew inside the pores. Non-iodine functionalized scaffolds

127 showed no contrast with the tissue, so they were invisible under X-ray after implantation. However, from the iodinated PEU groups, the scaffold can be distinguished from the surrounding tissue and bone. In the µ-CT 2D cross-section and histology images stained by H&E and Goldner’s trichrome, new bone grew from the periphery and the dural side. Goldner’s trichrome stained uncalcified osteoid red and mineralized bone green. Quantification results from µ-CT (b) and Goldner’s trichrome (c) demonstrate that BMP-2 enhances new bone growth in the cranial defect. Groups that do not share a letter are significantly different.

Figure 5.8 The histological images of PEU + BMP-2 scaffold at 12 weeks post- surgery stained with H&E (a-d) and Goldner’s trichrome (e-h), showing detailed new bone growth information. (a): H&E image of the defect site. New bone binds with the host bone. (b-d): scaffold pores are filled with new bone and fibrous connective tissue. New bone is formed in two configurations: lamellar bone under the scaffold and woven bone in the pores of scaffold (b and d). (e): Goldner’s trichrome image of the defect site. (f-h) shows that the fibrous connective tissue is composed of unmineralized collagen (osteoid) and mineralized collagen. S: scaffold; HB: host bone; NB: new bone; FT: fibrous connective tissue; BV: blood vessel; and Ob: osteoblast.

128 5.5 Discussion

Bone defects of critical size remain a serious challenge for surgeons. Previous

studies of amino acid-based PEUs resulted in PEUs exhibiting strong mechanical

properties, non-toxic biodegradation, and bioavailable functionality. In this study, two modifications were successfully imparted to the PEUs through a copolymerization strategy designed to advance these materials towards clinical applications. The

radiopacity imparted by iodine incorporation enables the in vivo detection by X-ray imaging and the enhanced bioactivity by the post-printing attachment of bioactive peptides through the propargyl reactive site promotes bone regeneration. The

combination of both enhanced radiopacity and bioactivity overcomes critical hurdles

for the clinical adaptation of polymer-based implants. The radiolucent property of conventional polymers significantly limits their application in a number of in vivo procedures and devices as most polymers cannot be detected by a postoperative imaging method such as µ-CT. There have been a number of studies on the incorporation of iodine into polymers to impart radiopacity.98,234,235 Previously, iodine functionalized L-phenylalanine-based poly(ester urea)s (iPEUs) with different iodine content were synthesized and their radiopacity increased with higher iodine contents.195 In our current study, only 11% iodine functionalization was sufficient to

ensure contrast with surrounding tissues. For polymeric grafts, bioactive species such

as growth factors and bioactive glass are usually incorporated to modify polymers. In

our study, we attached BMP-2/OGP to the radiopaque PEU scaffold surface to

investigate the osteogenic differentiation of hMSCs in vitro and new bone formation

in vivo using a rat cranial bone defect model.

129 During bone defect healing, bone forms in a process called intramembranous ossification. This process is mediated by osteoblasts, which secrete the organic bone extracellular matrix proteins (osteoid), including type I collagen, ALP, BSP, and OCN and are responsible for the mineralization of osteoid.236 In the defect site, a significant

recruitment of osteoblasts is required for bone healing. This recruitment is facilitated

by the differentiation of mesenchymal stem cells, which can differentiate into four

lineages by the expression of specific transcriptional factors that regulate

differentiation.236 RUNX2 is the corresponding key regulator for osteoblast

differentiation. Its activation is used as a marker for the hMSCs differentiation into

preosteoblasts.237 As the first transcription factor expressed, it triggers the sequential

expression of major bone matrix related genes in the early stage of osteoblast

differentiation, directing the formation of immature osteoblasts.196 In our current

study, it was observed that RUNX2 expression was activated in all three groups

before the 2 week time point, which represented preosteoblast differentiation. More

expression in the peptide tethered groups than the control at 2 weeks demonstrates the

promotional effect of the OGP and BMP-2 peptides. Similar results were observed in previous studies.224 The pathways for BMP-2 and OGP in the promotion of osteoblast differentiation are well studied in which BMP-2 releases RUNX2 from histone deacetylases-7’s (HDAC7) repression and protects RUNX2 from proteolysis205 and

OGP binds to Gi protein cell receptor to activate the mitogen-activated protein kinase

(MAPK) signaling pathway for the regulation of RUNX2.84 The RUNX2 regulation

of the middle stage marker BSP and late stage marker OCN was also demonstrated in

Figure 5.4. BSP and OCN expression were more in the peptide attached scaffolds than

the control at 4 weeks, corresponding with the RUNX2 expression profiles at 2 weeks,

confirming the promotional effect of both peptides.

130 The hMSCs mineralization in the peptide functionalized samples was

significantly higher than the control at both 2 and 4 weeks, which is in accordance

with the hMSCs differentiation results. By comparing between the peptide attached

groups, BMP-2 promoted the hMSCs mineralization further than OGP, demonstrated

by more mineral nodules at 2 and 4 weeks, which is consistent with our previously

224 published paper.

Overall, our in vitro findings suggested that both OGP and BMP-2 promoted and enhanced the hMSCs osteogenic differentiation compared to controls. Based on the effects of OGP and BMP-2, the later was chosen for the further in vivo rat cranial critical size defect recovery study, which was characterized by the new bone formation using µ-CT 3D scanning, H&E staining and Goldner’s trichrome staining.

The scaffolds showed excellent bone integration with the surrounding host

calvarium, as evidenced by ample osteoid/bone growth through the pores and gaps

between the scaffolds and host bone, especially with the BMP-2 attached sample showed in Figure 8. The new bone growth started from the dural side and the scaffold margins. The observance of new bone formation in different configurations (loose mineralized fibrous tissue in the scaffold pores, woven bone in the scaffold pores and lamellar bone under the scaffold) can be explained by the “four-stage-model”, in which the fracture healing involves four overlapping stages: inflammation, soft callus formation, hard callus formation, and remodeling. In the remodeling stage, the woven bone in the hard calluses remodel into the original bone configuration.8 This explains

why bone formed under the scaffold (dural side) was similar to the host bone. Ma et al.

have reported that in a rabbit segmental long bone defect using OGP incorporated

PLGA scaffolds, radiological images at 8 weeks demonstrated the segmental defect

131 was healed with newly formed bone, which further developed into cortical bone at 12

weeks, and finally remodeled as host bone at 16 weeks.94 A similar result was also

reported that in a rabbit calvarial critical size defect, where fibrous connective tissue

grew into the scaffold, new bone formation started from the defect margins, and

woven bone islets were formed in the middle of the defect at 4 weeks and part of the

238 new bone matured to lamellar-structure at 12 weeks.

The µ-CT results demonstrated that at 4 weeks, BMP-2 promoted bone

healing as supported by the increased new bone formation of 1.45-fold and 1.22-fold

compared to the PEU control and iPEU control respectively. Previously, rhBMP-2-

loaded mPCL–TCP/collagen scaffolds increased bone 1.69-fold as compared to

scaffolds alone at 4 weeks.71 The new bone formation confirmed the in vitro result

that BMP-2 promoted osteogenic differentiation of hMSCs, in which more osteoid

was deposited and the subsequent mineralization was enhanced. We observed a trend

of new bone growth of the five groups that is slightly different than the histological

evaluations, most likely because µ-CT calculates the bone volume in the 3D defect

site while histological analysis only takes into account the 2D sections, which can

potentially vary at different sectioned positions. This also explains the larger error bar of histological analysis than µ-CT. Another factor that contributes to the difference in

trends is that the µ-CT calculates the dense new bone with strong radiocontrast while

the histology analysis calculates the mineralized bone tissue colorimetrically, which

includes not only the newly formed dense bone but also the loosely mineralized

fibrous connective tissue as demonstrated in Figure 5.8h.

The 11% iodine incorporation enhances the radiocontrast between the scaffold

and its surrounding tissues. After implantation, the radiopacity of the non-iodinatd

132 PEU is similar to the soft tissue in the rats, and there is no contrast between scaffolds

and the unmineralized tissue. However, the radio contrast enhanced-iodinated

scaffolds render clear 3D scaffold morphology and can be easily distinguished from

unmineralized tissue and mineralized bone. Therefore, radio contrast enhanced by

iodine is promising and provides valuable diagnostic data in the application of

implanted biomedical devices.

5.6 Conclusion

In this study, we have demonstrated a novel amino acid based biomaterial with

dual modifications for surface tethered growth factor (BMP-2) activity and increased

radiopacity by iodine incorporation. FDM printing has been used to fabricate

customized scaffolds with a defined pore structure. The detection of iodinated

functionalized scaffolds under X-ray after implantation facilitates in vivo noninvasive

evaluation of polymeric devices. The in vitro enhancement of hMSCs osteogenic

differentiation and in vivo promotion of new bone formation in the rat cranial critical

size defect strongly indicate this as a promising strategy for critical size bone defect

repair in clinical applications.

5.7 Acknowledgement

The authors gratefully acknowledge financial support from The Biomaterials

Division of the National Science Foundation (DMR-1507420) & The John S. And

James L. Knight Foundation via the W. Gerald Austen Endowed Chair in Polymer

Science and Polymer Engineering. Support for RKW was through the Margaret F.

Donovan Endowed Chair for Women in Engineering.

133

CHAPTER VI

SUMMARY

In this dissertation, L-phenylalanine-based poly(ester urea)s (PEUs) were used as scaffold material since research has shown that they are mechanically strong, biodegradable, and non-toxic. Modification was based on the copolymerization of monomers with functional groups at different feed ratios. Heavy atom, iodine, was introduced to the L-phenylalanine-based monomer (1-iPHE-6) to impart radiopacity.

The resultant copolymerized iPEUs with different iodine content were synthesized and their radiopacity were compared, which was directly proportional to the iodine content. The iodinated PEUs showed increased thermal properties with higher degradation temperature and glass transition temperature. The iodine incorporation into PEUs also tuned the mechanical properties by decreasing the elastic modulus and toughening the polymer. The decrease in elastic modulus is not favorable in their application as bone grafts, but the ductile property of polymer is good for scaffold fabrication. The brittle poly(1-PHE-6) scaffolds crumbled easily during the porous scaffold fabrication process. They cannot maintain their original shape in wet state while the iodinated PEU scaffolds can maintain structural integrity. Therefore, by tuning the iodine content in the copolymer, a balance should be kept between the two competing properties: radiopacity and mechanical properties. The in vitro MC3T3 cell viability and spreading tests demonstrated iodinated PEUs are non-toxic to cells.

By using the L-tyrosine amino acid instead of L-phenylalanine, further

osteogenic modification of polymers could be achieved via direct attachment of

134 bioactive molecules to the phenol. Propargyl functionalized 1-TYR-6 monomer (1- pTYR-6) was copolymerized with the 1-PHE-6 monomer to prepare PEUs with reactive handles for

OGP/BMP-2 attachment to the scaffold surface using CuCCA. The 3D printing condition of

PEUs was explored and the propargyl functional group density on the 3D printed scaffold surface was quantified to be (75 ± 5) pmol/cm2 using a combination method of UV-visible spectroscopy and fluorescence spectroscopy. The functional group survived the high temperature 3D printing process, so the peptide could be attached post-printing to the scaffold surface. The in vitro hMSCs seeded on the peptide attached 3D printing scaffolds for osteogenic differentiation analysis. ALP activity of peptide functionalized scaffold showed an increase at early stage of differentiation. The oesteoblast related gene expression and protein expression of

RUNX2 (early stage marker), BSP (middle stage marker), and OCN (late stage marker) demonstrated both OGP and BMP-2 promoted the hMSCs osteogenic differentiation. Calcium deposition also confirmed this result.

Lastly, the iodine and/or BMP-2 functionalized scaffolds were implanted in rats with 8 mm cranial critical size defect. The iodinated PEU scaffolds could be distinguished in vivo from the surrounding bones and tissues since their radiopacity was different. The ingrowth of new bones in the scaffold pores were observed directly under X-ray. While for the non-iodine functionalized groups, since their radiopacity was similar to the surrounding soft tissues, they were invisible via X-ray analysis.

This dissertation is a comprehensive study from polymer synthesis and functionalization, scaffold fabrication, in vitro cell study, and in vivo animal study to verify the dual functionalization of PEUs. The results are positive, but there is still room for improvement to optimize the peptide amount on the scaffold surface for the best bone defect repair. This work can be applied to other system for biomedical applications.

135 REFERENCES

(1) Wiese, A.; Pape, H. C.: Bone Defects Caused by High-energy Injuries, Bone

Loss, Infected Nonunions, and Nonunions. Orthop Clin N Am 2010, 41, 1-4.

(2) Raggatt, L. J.; Partridge, N. C.: Cellular and Molecular Mechanisms of Bone

Remodeling. Journal of Biological Chemistry 2010, 285, 25103-25108.

(3) Eriksen, E. F.: Cellular mechanisms of bone remodeling. Rev Endocr Metab Dis

2010, 11, 219-227.

(4) Clarke, B.: Normal Bone Anatomy and Physiology. Clin J Am Soc Nephro

2008, 3, S131-S139.

(5) Hadjidakis, D. J.; Androulakis, I. I.: Bone remodeling. Women's Health and

Disease: Gynecologic, Endocrine, and Reproductive Issues 2006, 1092, 385-

396.

(6) Lian, J. B.; Stein, G. S.; van Wijnen, A. J.; Stein, J. L.; Hassan, M. Q.; Gaur, T.;

Zhang, Y.: MicroRNA control of bone formation and homeostasis. Nat Rev

Endocrinol 2012, 8, 212-227.

(7) Matsumoto, M.; Biguetti, C.; Fonseca, A.; Saraiva, P.: Bone Tissue Healing

Dynamics: Fram damage to reconstruction, 2016.

(8) Schindeler, A.; McDonald, M. M.; Bokko, P.; Little, D. G.: Bone remodeling

during fracture repair: The cellular picture. Semin Cell Dev Biol 2008, 19, 459-

466.

(9) Oryan, A.; Alidadi, S.; Moshiri, A.; Maffulli, N.: Bone regenerative medicine:

classic options, novel strategies, and future directions. J Orthop Surg Res 2014,

136 (10) DeCamp, C. E.; Johnston, S. A.; D©*jardin, L. c. M.; Schaefer, S. L.: Brinker,

Piermattei and Flo's handbook of small animal orthopedics and fracture repair;

Fifth edition ed.

(11) Calori, G. M.; Mazza, E.; Colombo, M.; Ripamonti, C.: The use of bone-graft

substitutes in large bone defects: Any specific needs? Injury 2011, 42, S56-

S63.

(12) Luvizuto, E. R.; Tangl, S.; Zanoni, G.; Okamoto, T.; Sonoda, C. K.; Gruber, R.;

Okamoto, R.: The effect of BMP-2 on the osteoconductive properties of beta-

tricalcium phosphate in rat calvaria defects. Biomaterials 2011, 32, 3855-3861.

(13) Hutmacher, D. W.: Scaffolds in tissue engineering bone and cartilage.

Biomaterials 2000, 21, 2529-2543.

(14) Albrektsson, T.; Johansson, C.: Osteoinduction, osteoconduction and

osseointegration. Eur Spine J 2001, 10, S96-S101.

(15) O'Brien, F. J.: Biomaterials & scaffolds for tissue engineering. Mater Today

2011, 14, 88-95.

(16) Zimmermann, G.; Moghaddam, A.: Allograft bone matrix versus synthetic bone

graft substitutes. Injury 2011, 42, S16-S21.

(17) Crane, G. M.; Ishaug, S. L.; Mikos, A. G.: Bone Tissue Engineering.

Nat Med 1995, 1, 1322-1324.

(18) Chan, B. P.; Leong, K. W.: Scaffolding in tissue engineering: general

approaches and tissue-specific considerations. Eur Spine J 2008, 17, S467-

S479.

(19) Dhandayuthapani, B.; Yoshida, Y.; Maekawa, T.; Kumar, D. S.: Polymeric

Scaffolds in Tissue Engineering Application: A Review. Int J Polym Sci 2011.

137 (20) Pang, X.; Chu, C. C.: Synthesis, characterization and biodegradation of

functionalized amino acid-based poly(ester amide)s. Biomaterials 2010, 31,

3745- 3754.

(21) Sun, H. L.; Meng, F. H.; Dias, A. A.; Hendriks, M.; Feijen, J.; Zhong,

Z. Y.: alpha-Amino Acid Containing Degradable Polymers as Functional

Biomaterials: Rational Design, Synthetic Pathway, and Biomedical

Applications. Biomacromolecules 2011, 12, 1937-1955.

(22) 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(alpha-amino acid) alpha,omega-alkylene

diesters, and aliphatic dicarboxylic acids. J Polym Sci Pol Chem 1999, 37, 391-

407.

(23) 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.

Macromolecular Chemistry and Physics 1997, 198, 1921-1932.

(24) Gao, Y.; Childers, E. P.; Becker, M. L.: l-Leucine-Based Poly (ester urea) s for

Vascular Tissue Engineering. ACS Biomaterials Science & Engineering 2015,

1, 795-804.

(25) Peterson, G. I.; Childers, E. P.; Li, H.; Dobrynin, A. V.; Becker, M. L.: Tunable

Shape Memory Polymers from alpha-Amino Acid-Based Poly(ester urea)s.

Macromolecules 2017, 50, 4300-4308.

(26) Lin, F.; Yu, J.; Tang, W.; Zheng, J.; Xie, S.; Becker, M. L.: Postelectrospinning

“click” modification of degradable amino acid-based poly (ester urea)

nanofibers. Macromolecules 2013, 46, 9515-9525.

138 (27) 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 2014, 16, 266-274.

(28) Policastro, G. M.; Lin, F.; Smith Callahan, L. A.; Esterle, A.; Graham, M.;

Sloan Stakleff, K.; Becker, M. L.: OGP functionalized phenylalanine-based

poly (ester urea) for enhancing osteoinductive potential of human mesenchymal

stem cells. Biomacromolecules 2015, 16, 1358-1371.

(29) Stakleff, K. S.; Lin, F.; Callahan, L. A. S.; 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 Biomaterialia 2013, 9, 5132-5142.

(30) Yu, J.; Lin, F.; Lin, P.; Gao, Y.; Becker, M. L.: Phenylalanine-based poly (ester

urea): synthesis, characterization, and in vitro degradation. Macromolecules

2013, 47, 121-129.

(31) Gao, Y. H.; Childers, E. P.; Becker, M. L.: L-Leucine-Based Poly(ester urea)s

for Vascular Tissue Engineering. ACS Biomaterials Science & Engineering

2015, 1, 795-804.

(32) Yu, J.; Lin, F.; Becker, M. L.: Branched Amino Acid Based Poly (ester urea) s

with Tunable Thermal and Water Uptake Properties. Macromolecules 2015, 48,

2916-2924.

(33) Middleton, J. C.; Tipton, A. J.: Synthetic biodegradable polymers as orthopedic

devices. Biomaterials 2000, 21, 2335-2346.

140 (34) Athanasiou, K. A.; Agrawal, C. M.; Barber, F. A.; Burkhart, S. S.: Orthopaedic

applications for PLA-PGA biodegradable polymers. Arthroscopy 1998, 14, 726-

737.

(35) von Burkersroda, F.; Schedl, L.; Gopferich, A.: Why degradable polymers

undergo surface erosion or bulk erosion. Biomaterials 2002, 23, 4221-31.

(36) Hollister, S. J.: Porous scaffold design for tissue engineering. Nat Mater 2005,

4, 518-524.

(37) An, J.; Teoh, J. E. M.; Suntornnond, R.; Chua, C. K.: Design and 3D Printing of

Scaffolds and Tissues. Engineering-Prc 2015, 1, 261-268.

(38) Chen, G. P.; Ushida, T.; Tateishi, T.: Scaffold design for tissue engineering.

Macromol Biosci 2002, 2, 67-77.

(39) Lu, L. C.; Mikos, A. G.: The importance of new processing techniques in tissue

engineering. Mrs Bull 1996, 21, 28-32.

(40) Bose, S.; Vahabzadeh, S.; Bandyopadhyay, A.: Bone tissue engineering using

3D printing. Mater Today 2013, 16, 496-504.

(41) Lo, H.; Ponticiello, M. S.; Leong, K. W.: Fabrication of controlled release

biodegradable foams by phase separation. Tissue Engineering 1995, 1, 15-28.

(42) Mooney, D. J.; Baldwin, D. F.; Suh, N. P.; Vacanti, L. P.; Langer, R.: Novel

approach to fabricate porous sponges of poly(D,L-lactic-co-glycolic acid)

without the use of organic solvents. Biomaterials 1996, 17, 1417-1422.

(43) Ozbolat, I. T.; Hospodiuk, M.: Current advances and future perspectives in

extrusion-based bioprinting. Biomaterials 2016, 76, 321-343.

(44) Sachs, E. M.; Haggerty, J. S.; Cima, M. J.; Williams, P. A.: Three- dimensional

printing techniques. Google Patents, 1993.

141 (45) Zhu, W.; Ma, X. Y.; Gou, M. L.; Mei, D. Q.; Zhang, K.; Chen, S. C.: 3D

printing of functional biomaterials for tissue engineering. Curr Opin Biotech

2016, 40, 103-112.

(46) Nyberg, E. L.; Farris, A. L.; Hung, B.; Dias, M.; Garcia, J. R.; Dorafshar, A. H.;

Grayson, W. L.: 3D-Printing Technologies for Craniofacial Rehabilitation,

Reconstruction, and Regeneration. Ann Biomed Eng 2017, 45, 45-57.

(47) Munaz, A.; Vadivelu, R. K.; St John, J.; Barton, M.; Kamble, H.; Nguyen, N.

T.: Three-dimensional printing of biological matters. J Sci 2016, 1, 1-17.

(48) Kang, H. W.; Lee, S. J.; Ko, I. K.; Kengla, C.; Yoo, J. J.; Atala, A.: A 3D

bioprinting system to produce human-scale tissue constructs with structural

integrity. Nat Biotechnol 2016, 34, 312-+.

(49) Rengier, F.; Mehndiratta, A.; von Tengg-Kobligk, H.; Zechmann, C. M.;

Unterhinninghofen, R.; Kauczor, H. U.; Giesel, F. L.: 3D printing based on

imaging data: review of medical applications. Int J Comput Ass Rad 2010, 5,

335-341.

(50) Gross, B. C.; Erkal, J. L.; Lockwood, S. Y.; Chen, C. P.; Spence, D. M.:

Evaluation of 3D Printing and Its Potential Impact on Biotechnology and the

Chemical Sciences. Anal Chem 2014, 86, 3240-3253.

(51) Jin, Y.-a.; Li, H.; He, Y.; Fu, J.-z.: Quantitative analysis of surface profile in

fused deposition modelling. Additive Manufacturing 2015, 8, 142-148.

(52) Varella, E.; Ohio Library and Information Network: Conservation science for

the cultural heritage applications of instrumental analysis. Springer: Berlin ;

London, 2012.

(53) Kohn, J. B.; Bolikal, D.; Pendharkar, S. M.: Radio-opaque polymer

biomaterials. Google Patents, 2002.

142 (54) Wang, T.; Zhang, Z. H.; Song, R.; Tuo, X. L.; Wu, J. G.: Radiopaque

polyurethanes with flexible iodine-containing chain extender. J Appl Polym Sci

2015, 132.

(55) Chang, W. J.; Pan, Y. H.; Tzeng, J. J.; Wu, T. L.; Fong, T. H.; Feng, S. W.;

Huang, H. M.: Development and Testing of X-Ray Imaging-Enhanced Poly-L-

Lactide Bone Screws. Plos One 2015, 10.

(56) Hong, S.; Carlson, J.; Lee, H.; Weissleder, R.: Bioorthogonal Radiopaque

Hydrogel for Endoscopic Delivery and Universal Tissue Marking. Adv Healthc

Mater 2016, 5, 421-426.

(57) Bourke, S. L.; Kohn, J.: Polymers derived from the amino acid L- tyrosine:

polycarbonates, polyarylates and copolymers with poly(ethylene glycol).

Advanced drug delivery reviews 2003, 55, 447-466.

(58) Solheim, E.: Growth factors in bone. Int Orthop 1998, 22, 410-416.

(59) Polo-Corrales, L.; Latorre-Esteves, M.; Ramirez-Vick, J. E.: Scaffold Design

for Bone Regeneration. J Nanosci Nanotechno 2014, 14, 15-56.

(60) Urist, M. R.: Bone: formation by autoinduction. Science 1965, 150, 893-9.

(61) Luyten, F. P.; Cunningham, N. S.; Ma, S.; Muthukumaran, N.; Hammonds, R.

G.; Nevins, W. B.; Woods, W. I.; Reddi, A. H.: Purification and partial amino

acid sequence of osteogenin, a protein initiating bone differentiation. J Biol

Chem 1989, 264, 13377-80.

(62) Grgurevic, L.; Pecina, M.; Vukicevic, S.: Marshall R. Urist and the discovery of

bone morphogenetic proteins. Int Orthop 2017, 41, 1065-1069.

(63) Rao, S. M.; Ugale, G. M.; Warad, S. B.: Bone morphogenetic proteins:

periodontal regeneration. N Am J Med Sci 2013, 5, 161-8.

143 (64) Burr, D. B.; Allen, M. R.: Basic and applied bone biology.

(65) Rengachary, S. S.: Bone morphogenetic proteins: basic concepts.

Neurosurg Focus 2002, 13, e2.

(66) Goodman, S. R.: Medical cell biology; 3rd ed ed.; Academic: Oxford, 2007.

(67) Cai, J.; Pardali, E.; Sanchez-Duffhues, G.; ten Dijke, P.: BMP signaling in

vascular diseases. FEBS Lett 2012, 586, 1993-2002.

(68) Guzman, A.; Zelman-Femiak, M.; Boergermann, J. H.; Paschkowsky, S.;

Kreuzaler, P. A.; Fratzl, P.; Harms, G. S.; Knaus, P.: SMAD versus Non-

SMAD Signaling Is Determined by Lateral Mobility of Bone Morphogenetic

Protein (BMP) Receptors. Journal of Biological Chemistry 2012, 287.

(69) Zhang, Y. E.: Non-Smad pathways in TGF-beta signaling. Cell Res

2009, 19, 128-39.

(70) Quinlan, E.; Thompson, E. M.; Matsiko, A.; O'Brien, F. J.; Loez- Noriega, A.:

Long-term controlled delivery of rhBMP-2 from collagen-hydroxyapatite

scaffolds for superior bone tissue regeneration. J Control Release 2015, 207,

112-119.

(71) Sawyer, A. A.; Song, S. J.; Susanto, E.; Chuan, P.; Lam, C. X. F.; Woodruff, M.

A.; Hutmacher, D. W.; Cool, S. M.: The stimulation of healing within a rat

calvarial defect by mPCL-TCP/collagen scaffolds loaded with rhBMP-2.

Biomaterials 2009, 30, 2479-2488.

(72) Madl, C. M.; Mehta, M.; Duda, G. N.; Heilshorn, S. C.; Mooney, D. J.:

Presentation of BMP-2 Mimicking Peptides in 3D Hydrogels Directs Cell Fate

Commitment in Osteoblasts and Mesenchymal Stem Cells. Biomacromolecules

2014, 15, 445-455.

144 (73) Lauzon, M. A.; Drevelle, O.; Faucheux, N.: Peptides derived from the knuckle

epitope of BMP-9 induce the cholinergic differentiation and inactivate

GSk3beta in human SH-SY5Y neuroblastoma cells. Sci Rep-Uk 2017, 7.

(74) Saito, A.; Suzuki, Y.; Ogata, S.; Ohtsuki, C.; Tanihara, M.: Activation of osteo-

progenitor cells by a novel synthetic peptide derived from the bone

morphogenetic protein-2 knuckle epitope. Bba-Proteins Proteom 2003, 1651,

60-67.

(75) Amso, Z.; Cornish, J.; Brimble, M. A.: Short Anabolic Peptides for Bone

Growth. Med Res Rev 2016, 36, 579-640.

(76) Saito, A.; Suzuki, Y.; Ogata, S.; Ohtsuki, C.; Tanihara, M.: Prolonged ectopic

calcification induced by BMP-2-derived synthetic peptide. Journal of

Biomedical Materials Research Part A 2004, 70A, 115-121.

(77) Saito, A.; Suzuki, Y.; Ogata, S.; Ohtsuki, C.; Tanihara, M.: Accelerated bone

repair with the use of a synthetic BMP-2-derived peptide and bone- marrow

stromal cells. Journal of Biomedical Materials Research Part A 2005, 72A, 77-

82.

(78) Zhou, X. J.; Feng, W.; Qiu, K. X.; Chen, L.; Wang, W. Z.; Nie, W.; Mo, X. M.;

He, C. L.: BMP-2 Derived Peptide and Dexamethasone Incorporated

Mesoporous Silica Nanoparticles for Enhanced Osteogenic Differentiation of

Bone Mesenchymal Stem Cells. Acs Appl Mater Inter 2015, 7, 15777-15789.

(79) Moore, N. M.; Lin, N. J.; Gallant, N. D.; Becker, M. L.: Synergistic

enhancement of human bone marrow stromal cell proliferation and osteogenic

differentiation on BMP-2-derived and RGD peptide concentration gradients.

Acta Biomaterialia 2011, 7, 2091-2100.

145 (80) Xu, Y. Y.; Luong, D.; Walker, J. M.; Dean, D.; Becker, M. L.: Modification of

Poly(propylene fumarate)-Bioglass Composites with Peptide Conjugates to

Enhance Bioactivity. Biomacromolecules 2017, 18, 3168-3177.

(81) Bab, I.; Gazit, D.; Muhlrad, A.; Shteyer, A.: Regenerating Bone- Marrow

Produces a Potent Growth-Promoting Activity to Osteogenic Cells.

Endocrinology 1988, 123, 345-352.

(82) Gazit, D.; Shteyer, A.; Bab, I.: Further Characterization of Osteogenic- Cell

Growth-Promoting Activity Derived from Healing Bone-Marrow. Connect

Tissue Res 1989, 23, 153-161.

(83) Bab, I.; Gazit, D.; Chorev, M.; Muhlrad, A.; Shteyer, A.; Greenberg, Z.;

Namdar, M.; Kahn, A.: Histone H4-Related Osteogenic Growth Peptide (Ogp) -

a Novel Circulating Stimulator of Osteoblastic Activity. Embo J 1992, 11,

1867-1873.

(84) Policastro, G. M.; Becker, M. L.: Osteogenic growth peptide and its use as a

bio-conjugate in regenerative medicine applications. Wires Nanomed Nanobi

2016, 8, 449-464.

(85) Gabarin, N.; Gavish, H.; Muhlrad, A.; Chen, Y. C.; Namdar-Attar, M.;

Nissenson, R. A.; Chorev, M.; Bab, I.: Mitogenic G(i) protein-MAP kinase

signaling cascade in MC3T3-E1 osteogenic cells: Activation by C-terminal

pentapeptide of osteogenic growth peptide [OGP(10-14)] and attenuation of

activation by cAMP. J Cell Biochem 2001, 81, 594-603.

(86) Zhu, H. B.; Cao, B. R.; Zhen, Z. P.; Laxmi, A. A.; Li, D.; Liu, S. R.; Mao, C.

B.: Controlled growth and differentiation of MSCs on grooved films assembled

from monodisperse biological nanofibers with genetically tunable surface

chemistries. Biomaterials 2011, 32, 4744-4752.

146 (87) Chen, Z. X.; Wang, X. F.; Shao, Y. C.; Shi, D. Y.; Chen, T. Y.; Cui, D. F.;

Jiang, X. X.: Synthetic osteogenic growth peptide promotes differentiation of

human bone marrow mesenchymal stem cells to osteoblasts via RhoA/ROCK

pathway. Mol Cell Biochem 2011, 358, 221-227.

(88) Rolfe, B. E.; Worth, N. F.; World, C. J.; Campbell, J. H.; Campbell, G. R.: Rho

and vascular disease. Atherosclerosis 2005, 183, 1-16.

(89) Pigossi, S. C.; Medeiros, M. C.; Saska, S.; Cirelli, J. A.; Scarel- Caminaga, R.

M.: Role of Osteogenic Growth Peptide (OGP) and OGP(10-14) in Bone

Regeneration: A Review. International journal of molecular sciences 2016, 17.

(90) Chen, Y. C.; Bab, I.; Mansur, N.; Muhlrad, A.; Shteyer, A.; Namdar- Attar, M.;

Gavish, H.; Vidson, M.; Chorev, M.: Structure-bioactivity of C-terminal

pentapeptide of osteogenic growth peptide [OGP(10-14)]. J Pept Res 2000, 56,

147- 156.

(91) Chen, Z. X.; Chang, M.; Peng, Y. L.; Zhao, L.; Zhan, Y. R.; Wang, L. J.; Wang,

R.: Osteogenic growth peptide C-terminal pentapeptide [OGP(10-14)] acts on

rat bone marrow mesenchymal stem cells to promote differentiation to

osteoblasts and to inhibit differentiation to adipocytes. Regul Peptides 2007,

142, 16-23.

(92) Moore, N. M.; Lin, N. J.; Gallant, N. D.; Becker, M. L.: The use of immobilized

osteogenic growth peptide on gradient substrates synthesized via click

chemistry to enhance MC3T3-E1 osteoblast proliferation. Biomaterials 2010,

31, 1604-1611.

(93) Fei, Q. M.; Guo, C. J.; Xu, X. Y.; Gao, J. J.; Zhang, J. A.; Chen, T. Y.; Cui, D.

F.: Osteogenic growth peptide enhances the proliferation of bone marrow

mesenchymal stem cells from osteoprotegerin-deficient mice by CDK2/cyclin

147 A. Acta Bioch Bioph Sin 2010, 42, 801-806.

(94) Ma, S. Q.; Wang, K. Z.; Dang, X. Q.; Wang, W.; Zhang, M. Y.; Wu, D. C.:

Osteogenic growth peptide incorporated into PLGA scaffolds accelerates

healing of segmental long bone defects in rabbits. Journal of Plastic

Reconstructive and Aesthetic Surgery 2008, 61, 1558-1560.

(95) Kiran, S.; James, N. R.; Joseph, R.; Jayakrishnan, A.: Synthesis and

characterization of iodinated polyurethane with inherent radiopacity.

Biomaterials 2009, 30, 5552-5559.

(96) James, N. R.; Philip, J.; Jayakrishnan, A.: Polyurethanes with radiopaque

properties. Biomaterials 2006, 27, 160-166.

(97) Yang, Y. Y.; Dorsey, S. M.; Becker, M. L.; Lin-Gibson, S.; Schumacher, G. E.;

Flaim, G. A.; Kohn, J. C.; Simon, C. G.: X-ray imaging optimization of 3D

tissue engineering scaffolds via combinatorial fabrication methods. Biomaterials

2008, 29, 1901-1911.

(98) Aamer, K. A.; Genson, K. L.; Kohn, J.; Becker, M. L.: Impact of Polymer-

Bound Iodine on Fibronectin Adsorption and Osteoblast Cell Morphology in

Radiopaque Medical Polymers: Tyrosine-Derived Polycarbonate Blends as a

Model System. Biomacromolecules 2009, 10, 2418-2426.

(99) Boelen, E. J. H.; van Hooy-Corstjens, C. S. J.; Bulstra, S. K.; van Ooij, A.; van

Rhijn, L. W.; Koole, L. H.: Intrinsically radiopaque hydrogels for nucleus

pulposus replacement. Biomaterials 2005, 26, 6674-6683.

(100) Davy, K. W. M.; Anseau, M. R.; Berry, C.: Iodinated methacrylate copolymers

as X-ray opaque denture base acrylics. J. Dent. 1997, 25, 499-505.

(101) Dawlee, S.; Jayabalan, M.: Development of segmented polyurethane elastomers

with low iodine content exhibiting radiopacity and blood compatibility. Biomed.

148 Mater. 2011, 6.

(102) Dawlee, S.; Jayabalan, M.: Iodinated glycidyl methacrylate copolymer as a

radiopaque material for biomedical applications. J. Biomater. Appl. 2013, 28,

28- 37.

(103) Vivan, R. R.; Ordinola-Zapata, R.; Bramante, C. M.; Bernardineli, N.; Garcia,

R. B.; Duarte, M. A. H.; de Moraes, I. G.: Evaluation of the radiopacity of some

commercial and experimental root-end filling materials. Oral Surg. Oral Med.

Oral Pathol. Oral Radiol. Endod. 2009, 108, E35-E38.

(104) Dawlee, S.; Jayakrishnan, A.; Jayabalan, M.: Studies on novel radiopaque

methyl methacrylate: glycidyl methacrylate based polymer for biomedical

applications. J. Mater. Sci.-Mater. Med. 2009, 20, 243-250.

(105) Mottu, F.; Rufenacht, D. A.; Laurent, A.; Doelker, E.: Iodine- containing

cellulose mixed esters as radiopaque polymers for direct embolization of

cerebral aneurysms and arteriovenous malformations. Biomaterials 2002, 23,

121-131.

(106) Pekkan, G.; Aktas, A.; Pekkan, K.: Comparative radiopacity of bone graft

materials. Journal of cranio-maxillo-facial surgery : official publication of the

European Association for Cranio-Maxillo-Facial Surgery 2012, 40, e1-4.

(107) Kohn, J. B.; Bolikal, D.; Pendharkar, S. M.: Radio-opaque polymer

biomaterials. Rutgers, The State University, 2002.

(108) Aberg, J.; Pankotai, E.; Hulsart Billstrom, G.; Weszl, M.; Larsson, S.; Forster-

Horvath, C.; Lacza, Z.; Engqvist, H.: In vivo evaluation of an injectable

premixed radiopaque calcium phosphate cement. International journal of

biomaterials 2011, 2011, 232574.

(109) He, J. W.; Soderling, E.; Lassila, L. V. J.; Vallittu, P. K.: Incorporation of an

149 antibacterial and radiopaque monomer in to dental resin system. Dent. Mater.

2012, 28, E110-E117.

(110) Aldenhoff, Y. B. J.; Kruft, M. A. B.; Pijpers, A. P.; van der Veen, F. H.;

Bulstra, S. K.; Kuijer, R.; Koole, L. H.: Stability of radiopaque iodine-

containing biomaterials. Biomaterials 2002, 23, 881-886.

(111) Aguilai, F. G.; Garcia, L. D. R.; Rossetto, H. L.; Pardini, L. C.; Pires- de-Souza,

F. D. P.: Radiopacity Evaluation of Calcium Aluminate Cement Containing

Different Radiopacifying Agents. J. Endod. 2011, 37, 67-71.

(112) Duarte, M. A. H.; El Kadre, G. D. D.; Vivan, R. R.; Tanomaru, J. M. G.;

Tanomaru, M.; de Moraes, I. G.: Radiopacity of Portland Cement Associated

With Different Radiopacifying Agents. J. Endod. 2009, 35, 737-740.

(113) Watts, D. C.: RADIOPACITY VS COMPOSITION OF SOME BARIUM

AND STRONTIUM GLASS COMPOSITES. J. Dent. 1987, 15, 38-43.

(114) Deb, S.; Vazquez, B.: The effect of cross-linking agents on acrylic bone

cements containing radiopacifiers. Biomaterials 2001, 22, 2177-2181.

(115) Kim, H. C.; Chung, Y. S.: Preparation and radiopaque properties of

chitosan/BaSO4 composite fibers. Fiber. Polym. 2013, 14, 292-297.

(116) Kurtz, S. M.; Villarraga, M. L.; Zhao, K.; Edidin, A. A.: Static and fatigue

mechanical behavior of bone cement with elevated barium sulfate content for

treatment of vertebral compression fractures. Biomaterials 2005, 26, 3699-

3712.

(117) Shasteen, C.; Kwon, S. M.; Park, K. Y.; Jung, S. Y.; Lee, S. H.; Park,

C. G.; Kim, M. H.; Kim, S.; Son, W. C.; Choi, T. H.; Choy, Y. B.: Biodegradable

internal fixation plates enabled with X-ray visibility by a radiopaque layer of

beta- tricalcium phosphate and poly (lactic-co-glycolic acid). J. Biomed. Mater.

150 Res. Part B 2013, 101B, 320-329.

(118) Jin, Y. S.; Li, Y. Y.; Ma, X. B.; Zha, Z. B.; Shi, L. L.; Tian, J.; Dai, Z. F.:

Encapsulating tantalum oxide into polypyrrole nanoparticles for X-ray

CT/photoacoustic bimodal imaging-guided photothermal ablation of cancer.

Biomaterials 2014, 35, 5795-5804.

(119) Hainfeld, J. F.; Smilowitz, H. M.; O'Connor, M. J.; Dilmanian, F. A.; Slatkin,

D. N.: Gold nanoparticle imaging and radiotherapy of brain tumors in mice.

Nanomedicine-Uk 2013, 8, 1601-1609.

(120) Pepiol, A.; Teixidor, F.; Saralidze, K.; van der Marel, C.; Willems, P.; Voss, L.;

Knetsch, M. L. W.; Vinas, C.; Koole, L. H.: A highly radiopaque vertebroplasty

cement using tetraiodinated o-carborane additive. Biomaterials 2011, 32, 6389-

6398.

(121) Deb, S.; Abdulghani, S.; Behiri, J. C.: Radiopacity in bone cements using an

organo-bismuth compound. Biomaterials 2002, 23, 3387-3393.

(122) Hernandez, L.; Fernandez, M.; Collia, F.; Gurruchaga, M.; Goni, I.: Preparation

of acrylic bone cements for vertebroplasty with bismuth salicylate as

radiopaque agent. Biomaterials 2006, 27, 100-107.

(123) Mottu, F.; Rufenacht, D. A.; Doelker, E.: Radiopaque polymeric materials for

medical applications - Current aspects of biomaterial research. Investigative

Radiology 1999, 34, 323-335.

(124) Galperin, A.; Margel, D.; Baniel, J.; Dank, G.; Biton, H.; Margel, S.:

Radiopaque iodinated polymeric nanoparticles for X-ray imaging applications.

Biomaterials 2007, 28, 4461-4468.

(125) Aviv, H.; Bartling, S.; Kieslling, F.; Margel, S.: Radiopaque iodinated

copolymeric nanoparticles for X-ray imaging applications. Biomaterials 2009,

151 30, 5610-5616.

(126) Carbone, A. L.; Song, M.; Uhrich, K. E.: Iodinated salicylate-based

poly(anhydride-esters) as radiopaque Biomaterials. Biomacromolecules 2008, 9,

1604-1612.

(127) van Hooy-Corstjens, C. S. J.; Govaert, L. E.; Spoelstra, A. B.; Bulstra,

G. M. R.; Wetzels, G. M. R.; Koole, L. H.: Mechanical behaviour of a new

acrylic radiopaque iodine-containing bone cement. Biomaterials 2004, 25,

2657-2667.

(128) James, N. R.; Jayakrishnan, A.: On imparting radiopacity to a poly(urethane

urea). Biomaterials 2007, 28, 3182-3187.

(129) Mawad, D.; Poole-Warren, L. A.; Martens, P.; Koole, L. H.; Slots, T. L. B.; van

Hoy-Corstjens, C. S. J.: Synthesis and characterization of radiopaque iodine-

containing degradable PVA hydrogels. Biomacromolecules 2008, 9, 263-268.

(130) Jin, E.; Lu, Z. R.: Biodegradable iodinated polydisulfides as contrast agents for

CT angiography. Biomaterials 2014, 35, 5822-5829.

(131) Yin, Q.; Yap, F. Y.; Yin, L. C.; Ma, L.; Zhou, Q.; Dobrucki, L. W.; Fan, T. M.;

Gaba, R. C.; Cheng, J. J.: Poly(iohexol) Nanoparticles As Contrast Agents for

in Vivo X-ray Computed Tomography Imaging. J Am Chem Soc 2013, 135,

13620-13623.

(132) Ghosh, P.; Das, M.; Rameshbabu, A. P.; Das, D.; Datta, S.; Pal, S.; Panda, A.

B.; Dhara, S.: Chitosan Derivatives Cross-Linked with Iodinated 2,5-

Dimethoxy-2,5-dihydrofuran for Non-Invasive Imaging. Acs Appl Mater Inter

2014, 6, 17926-17936.

(133) Benzina, A.; Kruft, M. A. B.; Bar, F.; Vanderveen, F. H.; Bastiaansen,

C. W.; Heijnen, V.; Reutelingsperger, C.; Koole, L. H.: STUDIES ON A NEW

152 RADIOPAQUE POLYMERIC BIOMATERIAL. Biomaterials 1994, 15, 1122-

1128.

(134) Qu, W. Q.; Xia, W. J.; Feng, C.; Tuo, X. L.; Qiu, T.: Synthesis and

Characterization of Radiopaque Poly(ether urethane) with Iodine-Containing

Diol as Chain Extender. J. Polym. Sci. Pol. Chem. 2011, 49, 2191-2198.

(135) Kohn, J. B.; Bolikal, D.; Pendharkar, S. M.: Radio-opaque polymer

biomaterials. RutgersThe State University, 2006.

(136) Lin, F.; Yu, J. Y.; Tang, W.; Zheng, J. K.; Xie, S. B.; Becker, M. L.:

Postelectrospinning "Click" Modification of Degradable Amino Acid-Based

Poly(ester urea) Nanofibers. Macromolecules 2013, 46, 9515-9525.

(137) Huang, S. J.; Bansleben, D. A.; Knox, J. R.: Biodegradable Polymers -

Chymotrypsin Degradation of a Low-Molecular Weight Poly(Ester-Urea)

Containing Phenylalanine. J Appl Polym Sci 1979, 23, 429-437.

(138) 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) alpha,omega-alkylene

diesters. Macromolecular Chemistry and Physics 1997, 198, 1921-1932.

(139) Yu, J. Y.; Lin, F.; Lin, P. P.; Gao, Y. H.; Becker, M. L.: Phenylalanine- Based

Poly(ester urea): Synthesis, Characterization, and in vitro Degradation.

Macromolecules 2014, 47, 121-129.

(140) Stakleff, K. S.; Lin, F.; Callahan, L. A. S.; 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.

(141) Strable, E.; Prasuhn, D. E.; Udit, A. K.; Brown, S.; Link, A. J.; Ngo, J. T.;

153 Lander, G.; Quispe, J.; Potter, C. S.; Carragher, B.; Tirrell, D. A.; Finn, M.

G.: Unnatural amino acid incorporation into virus-like particles. Bioconjugate

Chem 2008, 19, 866-875.

(142) Kiick, K. L.; Saxon, E.; Tirrell, D. A.; Bertozzi, C. R.: Incorporation of azides

into recombinant proteins for chemoselective modification by the Staudinger

ligation. P Natl Acad Sci USA 2002, 99, 19-24.

(143) Kiick, K. L.; Tirrell, D. A.: Protein engineering by in vivo incorporation of non-

natural amino acids: Control of incorporation of methionine analogues by

methionyl-tRNA synthetase. Tetrahedron 2000, 56, 9487-9493.

(144) Watts, D. C.; McCabe, J. F.: Aluminium radiopacity standards for dentistry: an

international survey. J. Dent. 1999, 27, 73-78.

(145) Mikos, A. G.; Thorsen, A. J.; Czerwonka, L. A.; Bao, Y.; Langer, R.; Winslow,

D. N.; Vacanti, J. P.: Preparation and Characterization of Poly(L-Lactic Acid)

Foams. Polymer 1994, 35, 1068-1077.

(146) Ho, S. T.; Hutmacher, D. W.: A comparison of micro CT with other techniques

used in the characterization of scaffolds. Biomaterials 2006, 27, 1362- 1376.

(147) Gonzalez-Garcia, R.; Monje, F.: The reliability of cone-beam computed

tomography to assess bone density at dental implant recipient sites: a

histomorphometric analysis by micro-CT. Clin. Oral Implant. Res. 2013, 24,

871-879.

(148) Schambach, S. J.; Bag, S.; Schilling, L.; Groden, C.; Brockmann, M. A.:

Application of micro-CT in small animal imaging. Methods 2010, 50, 2-13.

(149) Hallouard, F.; Anton, N.; Choquet, P.; Constantinesco, A.; Vandamme, T.:

Iodinated blood pool contrast media for preclinical X-ray imaging applications -

A review. Biomaterials 2010, 31, 6249-6268.

154 (150) Otsuki, B.; Takemoto, M.; Fujibayashi, S.; Neo, M.; Kokubo, T.; Nakamura, T.:

Pore throat size and connectivity determine bone and tissue ingrowth into

porous implants: Three-dimensional micro-CT based structural analyses of

porous bioactive titanium implants. Biomaterials 2006, 27, 5892-5900.

(151) Shimko, D. A.; Shimko, V. F.; Sander, E. A.; Dickson, K. F.; Nauman,

E. A.: Effect of porosity on the fluid flow characteristics and mechanical

properties of tantalum scaffolds. J. Biomed. Mater. Res. Part B 2005, 73B, 315-

324.

(152) Karageorgiou, V.; Kaplan, D.: Porosity of 3D biomaterial scaffolds and

osteogenesis. Biomaterials 2005, 26, 5474-5491.

(153) Nasri-Nasrabadi, B.; Mehrasa, M.; Rafienia, M.; Bonakdar, S.; Behzad, T.;

Gavanji, S.: Porous starch/cellulose nanofibers composite prepared by salt

leaching technique for tissue engineering. Carbohydr. Polym. 2014, 108, 232-

238.

(154) Rezwan, K.; Chen, Q. Z.; Blaker, J. J.; Boccaccini, A. R.: Biodegradable and

bioactive porous polymer/inorganic composite scaffolds for bone tissue

engineering. Biomaterials 2006, 27, 3413-3431.

(155) Mastrogiacomo, M.; Scaglione, S.; Martinetti, R.; Dolcini, L.; Beltrame, F.;

Cancedda, R.; Quarto, R.: Role of scaffold internal structure on in vivo bone

formation in macroporous calcium phosphate bioceramics. Biomaterials 2006,

27, 3230-3237.

(156) Shi, X. F.; Sitharaman, B.; Pham, Q. P.; Liang, F.; Wu, K.; Billups, W. E.;

Wilson, L. J.; Mikos, A. G.: Fabrication of porous ultra-short single-walled

carbon nanotube nanocomposite scaffolds for bone tissue engineering.

Biomaterials 2007, 28, 4078-4090.

155 (157) Ghasemi-Mobarakeh, L.; Semnani, D.; Morshed, M.: A novel method for

porosity measurement of various surface layers of nanofibers mat using image

analysis for tissue engineering applications. J Appl Polym Sci 2007, 106, 2536-

2542.

(158) Park, S. N.; Park, J. C.; Kim, H. O.; Song, M. J.; Suh, H.: Characterization of

porous collagen/hyaluronic acid scaffold modified by 1-ethyl-3- (3-

dimethylaminopropyl)carbodiimide cross-linking. Biomaterials 2002, 23, 1205-

1212.

(159) Jones, A. C.; Arns, C. H.; Sheppard, A. P.; Hutmacher, D. W.; Milthorpe, B. K.;

Knackstedt, M. A.: Assessment of bone ingrowth into porous biomaterials using

MICRO-CT. Biomaterials 2007, 28, 2491-2504.

(160) Rice, R. W.: Comparison of Stress-Concentration Versus Minimum Solid Area

Based Mechanical Property Porosity Relations. J Mater Sci 1993, 28, 2187-

2190.

(161) Wang, M. O.; Vorwald, C. E.; Dreher, M. L.; Mott, E. J.; Cheng, M. H.; Cinar,

A.; Mehdizadeh, H.; Somo, S.; Dean, D.; Brey, E. M.; Fisher, J. P.: Evaluating

3D-Printed Biomaterials as Scaffolds for Vascularized Bone Tissue

Engineering. Adv Mater 2015, 27, 138-144.

(162) Childers, E. P.; Wang, M. O.; Becker, M. L.; Fisher, J. P.; Dean, D.: 3D

printing of resorbable poly(propylene fumarate) tissue engineering scaffolds.

Mrs Bull 2015, 40, 119-126.

(163) Bajaj, P.; Schweller, R. M.; Khademhosseini, A.; West, J. L.; Bashir, R.: 3D

Biofabrication Strategies for Tissue Engineering and Regenerative Medicine.

Annu Rev Biomed Eng 2014, 16, 247-276.

(164) Derby, B.: Printing and Prototyping of Tissues and Scaffolds. Science

156 2012, 338, 921-926.

(165) Melchels, F. P. W.; Feijen, J.; Grijpma, D. W.: A poly(D,L-lactide) resin for the

preparation of tissue engineering scaffolds by stereolithography. Biomaterials

2009, 30, 3801-3809.

(166) Inzana, J. A.; Olvera, D.; Fuller, S. M.; Kelly, J. P.; Graeve, O. A.; Schwarz, E.

M.; Kates, S. L.; Awad, H. A.: 3D printing of composite calcium phosphate and

collagen scaffolds for bone regeneration. Biomaterials 2014, 35, 4026- 4034.

(167) Agrawal, C. M.; McKinney, J. S.; Lanctot, D.; Athanasiou, K. A.: Effects of

fluid flow on the in vitro degradation kinetics of biodegradable scaffolds for

tissue engineering. Biomaterials 2000, 21, 2443-2452.

(168) Woodfield, T. B. F.; Malda, J.; de Wijn, J.; Peters, F.; Riesle, J.; van

Blitterswijk, C. A.: Design of porous scaffolds for cartilage tissue engineering

using a three-dimensional fiber-deposition technique. Biomaterials 2004, 25,

4149-4161.

(169) Hutmacher, D. W.; Sittinger, M.; Risbud, M. V.: Scaffold-based tissue

engineering: rationale for computer-aided design and solid free-form fabrication

systems. Trends Biotechnol 2004, 22, 354-362.

(170) Kao, C. T.; Lin, C. C.; Chen, Y. W.; Yeh, C. H.; Fang, H. Y.; Shie, M. Y.:

Poly(dopamine) coating of 3D printed poly(lactic acid) scaffolds for bone tissue

engineering. Mat Sci Eng C-Mater 2015, 56, 165-173.

(171) Domingos, M.; Intranuovo, F.; Gloria, A.; Gristina, R.; Ambrosio, L.; Bártolo,

P.; Favia, P.: Improved osteoblast cell affinity on plasma-modified 3-D

extruded PCL scaffolds. Acta Biomaterialia 2013, 9, 5997-6005.

(172) Prabhakaran, M. P.; Venugopal, J.; Chan, C. K.; Ramakrishna, S.: Surface

modified electrospun nanofibrous scaffolds for nerve tissue engineering.

157 Nanotechnology 2008, 19, 455102.

(173) Wu, Y.-C.; Shaw, S.-Y.; Lin, H.-R.; Lee, T.-M.; Yang, C.-Y.: Bone tissue

engineering evaluation based on rat calvaria stromal cells cultured on modified

PLGA scaffolds. Biomaterials 2006, 27, 896-904.

(174) Cai, K.; Yao, K.; Yang, Z.; Li, X.: Surface modification of three- dimensional

poly (d, l-lactic acid) scaffolds with baicalin: a histological study. Acta

Biomaterialia 2007, 3, 597-605.

(175) Ho, M.-H.; Wang, D.-M.; Hsieh, H.-J.; Liu, H.-C.; Hsien, T.-Y.; Lai, J.-Y.;

Hou, L.-T.: Preparation and characterization of RGD-immobilized chitosan

scaffolds. Biomaterials 2005, 26, 3197-3206.

(176) Goddard, J. M.; Hotchkiss, J.: Polymer surface modification for the attachment

of bioactive compounds. Progress in polymer science 2007, 32, 698-725.

(177) Yoo, H. S.; Kim, T. G.; Park, T. G.: Surface-functionalized electrospun

nanofibers for tissue engineering and drug delivery. Advanced drug delivery

reviews 2009, 61, 1033-1042.

(178) Zheng, J.; Liu, K.; Reneker, D. H.; Becker, M. L.: Post-assembly derivatization

of electrospun nanofibers via strain-promoted azide alkyne cycloaddition. J Am

Chem Soc 2012, 134, 17274-17277.

(179) Goldmann, A. S.; Glassner, M.; Inglis, A. J.; Barner‐Kowollik, C.:

Post‐Functionalization of Polymers via Orthogonal Ligation Chemistry.

Macromolecular rapid communications 2013, 34, 810-849.

(180) Kim, S.-E.; Wang, J.; Jordan, A. M.; Korley, L. T.; Baer, E.; Pokorski,

J. K.: Surface modification of melt extruded poly (ε-caprolactone) nanofibers:

toward a new scalable biomaterial scaffold. ACS Macro Letters 2014, 3, 585-

589.

158 (181) Kolb, H. C.; Finn, M.; Sharpless, K. B.: Click chemistry: diverse chemical

function from a few good reactions. Angewandte Chemie International Edition

2001, 40, 2004-2021.

(182) Zheng, J.; Hua, G.; Yu, J.; Lin, F.; Wade, M. B.; Reneker, D. H.; Becker, M. L.:

Post-electrospinning “triclick” functionalization of degradable polymer

nanofibers. ACS Macro Letters 2015, 4, 207-213.

(183) Zheng, J.; Kontoveros, D.; Lin, F.; Hua, G.; Reneker, D. H.; Becker, M. L.;

Willits, R. K.: Enhanced Schwann cell attachment and alignment using one-pot

“Dual Click” GRGDS and YIGSR derivatized nanofibers. Biomacromolecules

2014, 16, 357-363.

(184) Zouani, O. F.; Kalisky, J.; Ibarboure, E.; Durrieu, M. C.: Effect of BMP-2 from

matrices of different stiffnesses for the modulation of stem cell fate.

Biomaterials 2013, 34, 2157-2166.

(185) Ma, Y. R.; Zheng, J. K.; Amond, E. F.; Stafford, C. M.; Becker, M. L.: Facile

Fabrication of "Dual Click" One- and Two-Dimensional Orthogonal Peptide

Concentration Gradients. Biomacromolecules 2013, 14, 665-671.

(186) Tang, W.; Policastro, G. M.; Hua, G.; Guo, K.; Zhou, J. J.; Wesdemiotis, C.;

Doll, G. L.; Becker, M. L.: Bioactive Surface Modification of Metal Oxides via

Catechol-Bearing Modular Peptides: Multivalent-Binding, Surface Retention,

and Peptide Bioactivity. J Am Chem Soc 2014, 136, 16357-16367.

(187) Park, J. S.; Yang, H. N.; Jeon, S. Y.; Woo, D. G.; Na, K.; Park, K. H.:

Osteogenic differentiation of human mesenchymal stem cells using RGD-

modified BMP-2 coated microspheres. Biomaterials 2010, 31, 6239-6248.

(188) Lin, Z.-Y.; Duan, Z.-X.; Guo, X.-D.; Li, J.-F.; Lu, H.-W.; Zheng, Q.- X.; Quan,

D.-P.; Yang, S.-H.: Bone induction by biomimetic PLGA-(PEG-ASP) n

159 copolymer loaded with a novel synthetic BMP-2-related peptide in vitro and in

vivo. Journal of Controlled Release 2010, 144, 190-195.

(189) Ma, Y.; Policastro, G. M.; Li, Q.; Zheng, J.; Jacquet, R.; Landis, W. J.; Becker,

M. L.: Concentration-Dependent h MSC Differentiation on Orthogonal

Concentration Gradients of GRGDS and BMP-2 Peptides. Biomacromolecules

2016, 17, 1486-1495.

(190) Wang, J.; Wang, L.; Yang, M.; Zhu, Y.; Tomsia, A.; Mao, C.: Untangling the

effects of peptide sequences and nanotopographies in a biomimetic niche for

directed differentiation of iPSCs by assemblies of genetically engineered viral

nanofibers. Nano letters 2014, 14, 6850-6856.

(191) Jun, S.-H.; Lee, E.-J.; Jang, T.-S.; Kim, H.-E.; Jang, J.-H.; Koh, Y.-H.: Bone

morphogenic protein-2 (BMP-2) loaded hybrid coating on porous

hydroxyapatite scaffolds for bone tissue engineering. Journal of Materials

Science: Materials in Medicine 2013, 24, 773-782.

(192) Huang, S. J.; Bansleben, D. A.; Knox, J. R.: Biodegradable polymers:

Chymotrypsin degradation of a low molecular weight poly (ester‐urea)

containing phenylalanine. J Appl Polym Sci 1979, 23, 429-437.

(193) 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.

(194) Peterson, G. I.; Dobrynin, A. V.; Becker, M. L.: α-Amino Acid-Based Poly

(Ester urea) s as Multishape Memory Polymers for Biomedical Applications.

ACS Macro Letters 2016, 5, 1176-1179.

160 (195) Li, S.; Yu, J.; Wade, M. B.; Policastro, G. M.; Becker, M. L.: Radiopaque,

iodine functionalized, phenylalanine-based poly (ester urea) s.

Biomacromolecules 2015, 16, 615-624.

(196) Komori, T.: Regulation of osteoblast differentiation by Runx2. In

Osteoimmunology; Springer, 2009; pp 43-49.

(197) Liu, Z.; Yao, X.; Yan, G.; Xu, Y.; Yan, J.; Zou, W.; Wang, G.: Mediator

MED23 cooperates with RUNX2 to drive osteoblast differentiation and bone

development. Nature communications 2016, 7, 11149.

(198) Neve, A.; Corrado, A.; Cantatore, F. P.: Osteocalcin: Skeletal and extra‐skeletal

effects. Journal of cellular physiology 2013, 228, 1149-1153.

(199) Golub, E. E.; Boesze-Battaglia, K.: The role of alkaline phosphatase in

mineralization. Current Opinion in Orthopaedics 2007, 18, 444-448.

(200) Tang, W.; Policastro, G. M.; Hua, G.; Guo, K.; Zhou, J.; Wesdemiotis, C.; Doll,

G. L.; Becker, M. L.: Bioactive surface modification of metal oxides via

catechol-bearing modular peptides: Multivalent-binding, surface retention, and

peptide bioactivity. J Am Chem Soc 2014, 136, 16357-16367.

(201) Porter, J. R.; Henson, A.; Popat, K. C.: Biodegradable poly (ɛ- caprolactone)

nanowires for bone tissue engineering applications. Biomaterials 2009, 30, 780-

788.

(202) Ducy, P.; Zhang, R.; Geoffroy, V.; Ridall, A. L.; Karsenty, G.: Osf2/Cbfa1: a

transcriptional activator of osteoblast differentiation. Cell 1997, 89, 747-754.

(203) Komori, T.; Yagi, H.; Nomura, S.; Yamaguchi, A.; Sasaki, K.; Deguchi, K.;

Shimizu, Y.; Bronson, R.; Gao, Y.-H.; Inada, M.: Targeted disruption of

161 Cbfa1results in a complete lack of bone formation owing to maturational arrest

of osteoblasts. Cell 1997, 89, 755-764.

(204) Lee, J.; Madhurakkat Perikamana, S. K.; Ahmad, T.; Lee, M. S.; Yang,

H. S.; Kim, D. G.; Kim, K.; Kwon, B.; Shin, H.: Controlled Retention of BMP-2-

Derived Peptide on Nanofibers Based on Mussel-Inspired Adhesion for Bone

Formation. Tissue Eng Part A 2017.

(205) Bab, I. A.; Sela, J. J.: Cellular and molecular aspects of bone repair. In

Principles of bone regeneration; Springer, 2012; pp 11-41.

(206) Komori, T.: Regulation of osteoblast differentiation by transcription factors. J

Cell Biochem 2006, 99, 1233-1239.

(207) Lee, J.-S.; Lee, J.-M.; Im, G.-I.: Electroporation-mediated transfer of Runx2

and Osterix genes to enhance osteogenesis of adipose stem cells. Biomaterials

2011, 32, 760-768.

(208) Mendonca, G.; Mendonca, D. B. S.; Simoes, L. G. P.; Araujo, A. L.; Leite, E.

R.; Golin, A. L.; Aragao, F. J. L.; Cooper, L. F.: Nanostructured implant surface

effect on osteoblast gene expression and bone-to-implant contact in vivo. Mat

Sci Eng C-Mater 2011, 31, 1809-1818.

(209) Fujita, T.; Azuma, Y.; Fukuyama, R.; Hattori, Y.; Yoshida, C.; Koida, M.;

Ogita, K.; Komori, T.: Runx2 induces osteoblast and chondrocyte

differentiation and enhances their migration by coupling with PI3K-Akt

signaling. J Cell Biol 2004, 166, 85-95.

(210) Wang, Z.; Chen, L.; Wang, Y.; Chen, X.; Zhang, P.: Improved Cell Adhesion

and Osteogenesis of op-HA/PLGA Composite by Poly (dopamine)-Assisted

Immobilization of Collagen Mimetic Peptide and Osteogenic Growth Peptide.

Acs Appl Mater Inter 2016, 8, 26559-26569.

162 (211) Kim, Y.-J.; Lee, M.-H.; Wozney, J. M.; Cho, J.-Y.; Ryoo, H.-M.: Bone

morphogenetic protein-2-induced alkaline phosphatase expression is stimulated

by Dlx5 and repressed by Msx2. Journal of Biological Chemistry 2004, 279,

50773- 50780.

(212) Staines, K. A.; MacRae, V. E.; Farquharson, C.: The importance of the

SIBLING family of proteins on skeletal mineralisation and bone remodelling.

Journal of Endocrinology 2012, 214, 241-255.

(213) Kirkham, G.; Cartmell, S.: Genes and proteins involved in the regulation of

osteogenesis. Ashammakhi N, Reis R, Chiellini E, editores. Topics in Tissue

Engineering 2007, 3.

(214) Gao, R.-T.; Zhan, L.-P.; Meng, C.; Zhang, N.; Chang, S.-M.; Yao, R.; Li, C.:

Homeobox B7 promotes the osteogenic differentiation potential of

mesenchymal stem cells by activating RUNX2 and transcript of BSP.

International journal of clinical and experimental medicine 2014, 8, 10459-

10470.

(215) Porte, D.; Tuckermann, J.; Becker, M.; Baumann, B.; Teurich, S.; Higgins, T.;

Owen, M.; Schorpp-Kistner, M.; Angel, P.: Both AP-1 and Cbfa1-like factors

are required for the induction of interstitial collagenase by parathyroid

hormone. Oncogene 1999, 18, 667-678.

(216) Martin, J. T.; Milby, A. H.; Ikuta, K.; Poudel, S.; Pfeifer, C. G.; Elliott,

D. M.; Smith, H. E.; Mauck, R. L.: A radiopaque electrospun scaffold for

engineering fibrous musculoskeletal tissues: Scaffold characterization and in

vivo applications. Acta Biomaterialia 2015, 26, 97-104.

(217) Kiran, S.; Joseph, R.: Synthesis and characterization of a noncytotoxic, X-ray

opaque polyurethane containing iodinated hydroquinone bis(2-hydroxyethyl)

163 ether as chain extender for biomedical applications. Journal of Biomedical

Materials Research Part A 2014, 102, 3207-3215.

(218) Lu, X. J.; Zhang, Y. A.; Cui, D. C.; Meng, W. J.; Du, L. R.; Guan, H. T.;

Zheng, Z. Z.; Fu, N. Q.; Lv, T. S.; Song, L.; Zou, Y. H.; Lu, W. L.; Fan, T. Y.:

Research of novel biocompatible radiopaque microcapsules for arterial

embolization. Int J Pharmaceut 2013, 452, 211-219.

(219) Kiran, S.; James, N. R.; Jayakrishnan, A.; Joseph, R.: Polyurethane

thermoplastic elastomers with inherent radiopacity for biomedical applications.

Journal of Biomedical Materials Research Part A 2012, 100A, 3472-3479.

(220) Dawlee, S.; Jayabalan, M.: Intrinsically radiopaque polyurethanes with chain

extender 4, 4′-isopropylidenebis [2-(2, 6-diiodophenoxy) ethanol] for

biomedical applications. Journal of biomaterials applications 2015, 29, 1329-

1342.

(221) Sandker, M. J.; Petit, A.; Redout, E. M.; Siebelt, M.; Muller, B.; Bruin, P.;

Meyboom, R.; Vermonden, T.; Hennink, W. E.; Weinans, H.: In situ forming

acyl- capped PCLA-PEG-PCLA triblock copolymer based hydrogels.

Biomaterials 2013, 34, 8002-11.

(222) Kim, S. J.; Shin, H. S.; Shin, S. W.: Effect of bone block graft with rhBMP-2

on vertical bone augmentation. Int J Oral Max Surg 2010, 39, 883-888.

(223) Xu, Y.; Luong, D.; Walker, J. M.; Dean, D.; Becker, M. L.: Modification of

Poly (propylene fumarate)–Bioglass Composites with Peptide Conjugates to

Enhance Bioactivity. Biomacromolecules 2017.

(224) Li, S.; Xu, Y. Y.; Yu, J. Y.; Becker, M. L.: Enhanced osteogenic activity of

poly(ester urea) scaffolds using facile post-3D printing peptide

functionalization strategies. Biomaterials 2017, 141, 176-187.

164 (225) Yu, J.; Xu, Y.; Li, S.; Seifert, G. V.; Becker, M. L.: Three-Dimensional Printing

of Nano Hydroxyapatite/Poly(ester urea) Composite Scaffolds with Enhanced

Bioactivity. Biomacromolecules 2017, 18, 4171-4183.

(226) Li, X. M.; Cui, R. R.; Sun, L. W.; Aifantis, K. E.; Fan, Y. B.; Feng, Q. L.; Cui,

F. Z.; Watari, F.: 3D-Printed Biopolymers for Tissue Engineering Application.

Int J Polym Sci 2014.

(227) Zhu, W.; Ma, X.; Gou, M.; Mei, D.; Zhang, K.; Chen, S.: 3D printing of

functional biomaterials for tissue engineering. Curr Opin Biotechnol 2016, 40,

103- 112.

(228) Zhang, W.; Feng, C.; Yang, G.; Li, G.; Ding, X.; Wang, S.; Dou, Y.; Zhang, Z.;

Chang, J.; Wu, C.: 3D-printed scaffolds with synergistic effect of hollow- pipe

structure and bioactive ions for vascularized bone regeneration. Biomaterials

2017, 135, 85-95.

(229) Pfister, A.; Landers, R.; Laib, A.; Hubner, U.; Schmelzeisen, R.; Mulhaupt, R.:

Biofunctional rapid prototyping for tissue-engineering applications: 3D

bioplotting versus 3D printing. J Polym Sci Pol Chem 2004, 42, 624-638.

(230) Berman, B.: 3-D printing: The new industrial revolution. Bus Horizons

2012, 55, 155-162.

(231) Melchels, F. P. W.; Feijen, J.; Grijpma, D. W.: A review on stereolithography

and its applications in biomedical engineering. Biomaterials 2010, 31, 6121-

6130.

(232) Kang, H.-W.; Lee, S. J.; Ko, I. K.; Kengla, C.; Yoo, J. J.; Atala, A.: A 3D

bioprinting system to produce human-scale tissue constructs with structural

integrity. Nat Biotech 2016, 34, 312-319.

165 (233) Spicer, P. P.; Kretlow, J. D.; Young, S.; Jansen, J. A.; Kasper, F. K.; Mikos, A.

G.: Evaluation of bone regeneration using the rat critical size calvarial defect.

Nat Protoc 2012, 7, 1918-1929.

(234) Yang, Y.; Dorsey, S. M.; Becker, M. L.; Lin-Gibson, S.; Schumacher,

G. E.; Flaim, G. M.; Kohn, J.; Simon, C. G.: X-ray imaging optimization of 3D

tissue engineering scaffolds via combinatorial fabrication methods.

Biomaterials 2008, 29, 1901-1911.

(235) Johnson, P. A.; Luk, A.; Demtchouk, A.; Pate, H.; Sung, H. J.; Treiser,

M. D.; Gordonov, S.; Sheihet, L.; Bolikal, D.; Kohn, J.; Moghe, P. V.: Interplay

of anionic charge, poly(ethylene glycol), and iodinated tyrosine incorporation

within tyrosine-derived polycarbonates: Effects on vascular smooth muscle cell

adhesion, proliferation, and motility. Journal of Biomedical Materials Research

Part A 2010, 93A, 505-514.

(236) Rutkovskiy, A.; Stenslokken, K. O.; Vaage, I. J.: Osteoblast Differentiation at a

Glance. Med Sci Monit Basic 2016, 22, 95-106.

(237) Zhang, Y.; Khan, D.; Delling, J.; Tobiasch, E.: Mechanisms Underlying the

Osteo-and Adipo-Differentiation of Human Mesenchymal Stem Cells. Sci

World J 2012.

(238) Tuusa, S. M. R.; Peltola, M. J.; Tirri, T.; Puska, M. A.; Roytta, M.; Aho, H.;

Sandholm, J.; Lassila, L. V. J.; Vallittu, P. K.: Reconstruction of critical size

calvarial bone defects in rabbits with glass-fiber-reinforced composite with

bioactive glass granule coating. Journal of Biomedical Materials Research Part

B-Applied Biomaterials 2008, 84B, 510-519.

166

APPENDIX

1 Figure S1. HNMR (DMSO-d6) of 1-PHE-6 monomer and 1-iPHE-6 monomer

167

13 Figure S2. CNMR (DMSO-d6) of PEUs. (a) iodinated phenylalanine-based poly(1- iPHE-6); (b) copolymer of 44% poly(1-iPHE-6) and 56% poly(1-PHE-6); (c) copolymer of 24% poly(1-iPHE-6) and 76% poly(1-PHE-6); (d) phenylalanine-based poly(1-PHE-6). In the benzyl ring, substitution of one hydrogen atom with an iodine atom results in a shift from around 130 ppm to 93 ppm. With the increase in iodine content, the intensity of the characteristic C-I peak at 93 ppm increases.

168

Figure S3. DSC of PEUs at a scanning rate of 20 oC/min. The second cycle was used to determine Tg after the removal of any thermal history in the first cycle.

169 Synthesis of di-p-toluene sulfonic acid salt of bis-L-phenylalanine-1,6-hexanediol- diester (1-PHE-6 monomer): 1-PHE-6 monomer was synthesized as described previously, and shown in Scheme S4 (a). 1,6-hexanediol (20.00 g, 1.0 equiv., 0.17 mol), L-phenylalanine (64.32 g, 2.3 equiv., 0.39 mol), p-toluene sulfonic acid monohydrate (77.29 g, 2.4 equiv., 0.41 mol) and toluene (500 mL) were mixed in a

1L one-neck round-bottomed flask using a magnetic stir bar with a dean stark trap.

The system was refluxed at 110 °C for 21 h. The crude product was vacuum filtered to remove toluene, decolorized by activated carbon black (4.00 g) and recrystallized from boiling water 4 times to yield 109 g (yield 85%). 1H-NMR (500 MHz, DMSO- d6): 1.08 (m, 4H, −COOCH2CH2CH2−), 1.40 (m, 4H, −COOCH2CH2CH2−), 2.29 (s,

6H, CH3Ar−), 2.50 (m, DMSO), 2.99-3.17 (m, 4H, −CHCH2−Ar−), 3.31 (s, H2O),

+ 3.98-4.03 (m, 4H, −COOCH2CH2−), 4.27-4.30 (m, 2H, NH3CHCOO−), 7.11-7.13 (d,

4 H, aromatic H), 7.22-7.32 (m, 10H, aromatic H), 7.42-7.51 (d, 8H, aromatic H),

+ 13 8.38 (s, 6H, NH3−). C-NMR (125 MHz, DMSO-d6): 21.23, 25.12, 28.06, 36.64,

39.98, 53.75, 65.91, 125.96, 127.68-129.75, 135.12, 138.41, 145.65, 169.49.

Synthesis of di-hydrochloride acid salt of bis-4-propargyl-L-tyrosine-1,6- hexanediol-diester (1-pTYR-6 monomer): To a 100 mL round-bottom flask, added sequentially Boc-Tyr-OMe (5.0 g, 1.0 equiv., 0.017 mol), K2CO3 (4.7 g, 2.0 equiv.,

0.034 mol), and anhydrous DMF 20 mL. Propargyl bromide solution (3.8 mL, 2.0 equiv., 0.034 mol) was added dropwisely. The resulting mixture was stirred at room temperature for 24 h, then diluted with ethyl acetate (150 mL), and washed with water

(200 mL×3) three times. The organic layer was collected, dried with anhydrous

MgSO4 and concentrated under reduced pressure. The resulting yellow oil product

(5.2 g, 92% yield) was pure enough for the next step. This product was then dissolved in 25 mL THF, and 25 mL NaOH aqueous (1.0 M) was added dropwise. The mixture

170 was stirred at room temperature for 8 h, then diluted with 50 mL water, and extracted

with 100 mL ethyl acetate. The aqueous layer was collected and adjusted to PH=2~3

with HCl (1.0 M in water). The resulting mixture was extracted with ethyl acetate (50

mL×2). The combined organic layer was dried with anhydrous MgSO4 and

concentrated under reduced pressure. The resulting orange oil product (4.5 g, 90%

yield) was pure enough for the next step. Product from above step (4.5 g, 2.4 equiv.,

14 mmol), 1,6-hexanediol (0.69 g, 1.0 equiv., 5.8 mmol) and 4-(N,Ndimethylamino)

puridinium 4-toluenesulfonate (DPTS, 0.69 g, 0.4 equiv., 2.3 mmol) were dissolved in

30 mL CH2Cl2. Then this mixture was cooled down to 0 °C by immersing into an ice

bath. 1,3-diisopropyl cabodiimide (DPIC, 2.7 mL, 18 mmol, 3 equiv) was added with a syringe. The ice bath was removed and this reaction was carried out at room temperature for 24 h. The insoluble solid was removed by filtration. The collected

organic solution was concentrated for chromatography purification on silica gel (ethyl

acetate/hexane 1:2 v/v). 3.74 g yellow gel like product (yield: 88%) was obtained

under reduced pressure. This boc-protected precursor was dissolved in HCl/dioxane

(4 M) and stirred under nitrogen at room temperature for 12 h. 2.95 light yellow solid

(96%) was obtained by freeze drying to remove the solvent. 1H-NMR (500 MHz,

DMSO-d6): 1.13 (m, 4H, −COOCH2CH2CH2−), 1.45 (m, 4H, −COOCH2CH2CH2−),

2.50 (m, DMSO), 2.99-3.15 (m, 4H, −CHCH2−Ar−), 3.31 (s, H2O), 3.55 (m, 2H, -

+ OCH2C≡CH), 4.00 (m, 4H, −COOCH2CH2−), 4.16 (m, 2H, NH3CHCOO−), 4.75 (d,

4H, -OCH2C≡CH), 6.89-7.00 (d, 4 H, aromatic H), 7.13-7.21 (d, 4H, aromatic H),

+ 13 8.67 (s, 6H, NH3−). C-NMR (125 MHz, DMSO-d6): 25.19, 28.11, 35.59, 40.01,

53.74, 55.84, 65.86, 66.80, 78.62-79.70, 115.30, 127.77, 130.91, 156.91, 169.48.

171 Scaffold surface propargyl functional group density calculation

Calibration curve for poly[(1-pTYR-6)0.02-co-(1-PHE-6)0.98]: y=0.1004x, where y is

UV-visible spectra intensity and x is polymer concentration.

Polymer concentration= 𝑈𝑈𝑈𝑈−𝑉𝑉𝑉𝑉𝑉𝑉𝑉𝑉𝑉𝑉𝑉𝑉𝑉𝑉0.1004𝐼𝐼𝐼𝐼𝐼𝐼 𝐼𝐼𝐼𝐼𝐼𝐼𝐼𝐼𝐼𝐼𝐼𝐼

Functional group concentration in the polymer=Polymer concentration×1.8%

(determined from NMR integration)

Calibration curve for Chromeo 488 azide dye: y=100699x, where y is fluorescence spectra intensity and x is dye concentration.

Dye concentration= = Surface functional group concentration 100699 𝐹𝐹𝐹𝐹𝐹𝐹𝐹𝐹𝐹𝐹𝐹𝐹𝐹𝐹𝐹𝐹𝐹𝐹𝐹𝐹𝐹𝐹𝐹𝐹 𝐼𝐼𝐼𝐼𝐼𝐼𝐼𝐼𝐼𝐼𝐼𝐼𝐼𝐼𝐼𝐼𝐼𝐼

Surface functional group concentration Percentage of surface functionalization= ×100% Functional group concentration in the polymer

Thus, the surface propargyl functional group density

×1.8%× Percentage of surface functionalization = ℎ Repeat unit molecular weight× 𝑆𝑆𝑆𝑆𝑆𝑆𝑆𝑆𝑆𝑆𝑆𝑆𝑆𝑆𝑆𝑆 𝑤𝑤𝑤𝑤𝑤𝑤𝑤𝑤 𝑡𝑡 𝑆𝑆𝑆𝑆𝑆𝑆𝑆𝑆𝑆𝑆𝑆𝑆𝑆𝑆𝑆𝑆 𝑠𝑠𝑠𝑠𝑠𝑠𝑠𝑠𝑠𝑠𝑠𝑠𝑠𝑠 𝑎𝑎𝑎𝑎𝑎𝑎𝑎𝑎

172 (a)

(b)

Scheme S4. (a) 1-PHE-6 monomer and (b) 1-pTYR-6 monomer synthesis. (a) 1,6- hexanediol is condensed with 2.3 equivalents L-phenylalanine. (b) Synthesis of 1- pTYR-6 monomer. A propargyl functional group is easily introduced to the monomer.

173

1 Figure S5. H-NMR (DMSO-d6) of compound 1 in the synthesis of 1-pTYR-6 monomer. Solvent residuals are marked as asterisks.

174

1 Figure S6. H-NMR (DMSO-d6) of compound 2 in the synthesis of 1-pTYR-6 monomer. Solvent residuals are marked as asterisks.

175

1 Figure S7. H-NMR (DMSO-d6) of compound 3 in the synthesis of 1-pTYR-6 monomer. Solvent residuals are marked as asterisks.

176

1 Figure S8. H-NMR (DMSO-d6) of compound 4 in the synthesis of 1-pTYR-6 monomer. Solvent residuals are marked as asterisks.

177

1 Figure S9. H-NMR (DMSO-d6) of the synthesis of 1-pTYR-6 monomer. Solvent residuals are marked as asterisks.

178

1 Figure S10. H-NMR (DMSO-d6) of 1-PHE-6 monomer.

179

1 Figure S11. H-NMR (DMSO-d6) of 1-pTYR-6 monomer. Solvent residuals are marked as asterisks.

180

13 Figure S12. C NMR (DMSO-d6) of 1-PHE-6 monomer.

181

13 Figure S13. C NMR (DMSO-d6) of 1-pTYR-6 monomer, which shows characteristic peaks of o-propargyl substitution at 84 (-OCH2C≡CH), 66.80 (-

OCH2C≡CH), and 78.62-79.70 ppm (-OCH2C≡CH).

182

13 Figure S14. C NMR (DMSO-d6) of poly(1-PHE-6).

183

13 Figure S15. C NMR (DMSO-d6) of poly[(1-pTYR-6)0.02-co-(1-PHE-6)0.98]. Because of the resolution of 13C NMR, the 2% propargyl group in polymer is impossible to be detected.

184

Figure S16. SEC traces of PEUs to determine the molecular mass of polymers

185

Figure S17. DSC of PEUs at a scanning rate of 20 °C/min. The second cycle was used to determine Tg after the removal of any thermal history in the first cycle.

186

Figure S18. SEC traces of PEUs after each step of thermal processing in scaffold fabrication.

187

1 Figure S19. Comparison of H-NMR (DMSO-d6) between polymers and scaffolds.

The characteristic peak intensity of methylene hydrogens in the propargyl group (-

OCH2C≡CH) at 4.75 ppm remains the same after thermal processing.

188

Figure S20. (a) Micro-CT 3D reconstruction slice of PEU scaffolds.

189

Figure S21. (a) Fluorescence images (IX81microscope (Olympus)) of printed poly(1-

PHE-6) porous scaffold after CuAAC click reaction with dye. Since there is no functional group in the polymer, there is no dye clicked on the surface. The mean intensity is 181. (b) Fluorescence images of printed poly[(1-pTYR-6)0.02-co-(1-PHE-

6)0.98] porous scaffold after CuAAC click reaction with dye. The mean intensity is 328.

By comparing these two images, it is obvious that the dye is clicked on the surface.

The scale bar is 200 μm.

190 (a)

6000 2257.981

M+ + + 5000 M +K

4000

M+ +Na+ 3000

[a.u.] Intens.

2000

1000

0 1000 1200 1400 1600 1800 2000 2200 2400 2600 2800 m/z

(b)

Intens. -MS, 0.1-1.0min #(1-43) x105

637.0 637.0

579.9 282.9

497.9

440.9 751.1

0 500 1000 1500 2000 2500 m/z

Figure S22. Mass spectra of peptides. (a): N3-BMP2. (b): N3-OGP.

191 Table S1. Primers used in real time RT-PCR

Gene Primers

GAPDH Forward: 5’- GACAGTCAGCCGCATCTT-3’

Reverse: 5’-CCATGGTGTCTGAGCGATGT-3’ RUNX2

Forward: 5’-GGACGAGGCAAGAGTTTCAC-3’

Reverse: 5’-CAAGCTTCTGTCTGTGCCTTC-3’ BSP Forward: 5’-CCTGGCACAGGGTATACAGG-3’

Reverse: 5’-CTGCTTCGCTTTCTTCGTTT-3’

OCN Forward: 5’-CATGAGAGCCCTCACA-3’

Reverse: 5’-AGAGCGACACCCTAGAC-3’

192

13 Figure S23. C NMR (DMSO-d6) of poly[(1-iPHE-6)0.11-co-(1-PHE-6)0.89].

193

13 Figure S24. C NMR (DMSO-d6) of poly[(1-pTYR-6)0.02-co-(1-iPHE-6)0.11-co-(1-

PHE-6)0.87]. The resolution is not enough to detect the 2% propargyl group.

194

Figure S25. FT-IR of PEUs. It shows the C-I stretching at 1007 cm-1. 2% propargyl group can not be detected.

195

Figure S26. SEC traces of PEUs to determine the molecular mass of polymers

196

Figure S27. DSC of PEUs at a scanning rate of 20 °C/min to determine glass transition temperature of polymers.

197

Figure S28. SEC traces of PEUs during scaffold fabrication to evaluate the molecular mass decrease due to thermal degradation.

198

1 Figure S29. Comparison of H-NMR (DMSO-d6) between polymer and scaffold. The propargyl functional group (-OCH2C≡CH) survives the high temperature processing.

199

Figure S30. Animal surgery and implantation of PEU scaffolds in an 8 mm cranial defect.

200