Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Tandem modular protein-based hydrogels as extracellular matrix mimetic biomaterials Lv, Shanshan 2013

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata

Download

Media
24-ubc_2013_spring_lv_shanshan.pdf [ 3.95MB ]
Metadata
JSON: 24-1.0073719.json
JSON-LD: 24-1.0073719-ld.json
RDF/XML (Pretty): 24-1.0073719-rdf.xml
RDF/JSON: 24-1.0073719-rdf.json
Turtle: 24-1.0073719-turtle.txt
N-Triples: 24-1.0073719-rdf-ntriples.txt
Original Record: 24-1.0073719-source.json
Full Text
24-1.0073719-fulltext.txt
Citation
24-1.0073719.ris

Full Text

Tandem Modular Protein-based Hydrogels as Extracellular Matrix Mimetic Biomaterials  by SHANSHAN LV  B.Sc., University of Science and Technology of China, 2008  A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF Doctor of Philosophy in THE FACULTY OF GRADUATE STUDIES (Chemistry) THE UNIVERSITY OF BRITISH COLUMBIA (Vancouver)  April 2013 ©Shanshan Lv, 2013  ABSTRACT New generations of protein-based hydrogels are being developed rapidly over the last several decades for potential applications in biomedicine as well as basic biological studies. Protein-based hydrogels have been explored as synthetic extracellular matrices (ECM) for applications in cell culture and tissue regeneration. A variety of proteins, including silk protein, fibrin, collagen and elastin have been studied. Most of these proteins are non-globular proteins. However, a large number of ECM proteins are tandem modular proteins that consist of many individually folded domains. We hypothesize that tandem modular proteins may also be used in constructing novel hydrogels that can mimic the physical and biochemical characteristics of natural extracellular matrices. In this dissertation, we explored the feasibility of using tandem modular proteins for constructing protein-based biomaterials. First, through self-assembly of two complementary leucine zipper sequences, artificial tandem modular proteins were engineered to form physically cross-linked hydrogels mimicking ECM. Going a step further, a photochemical cross-linking strategy is employed to covalently cross-link engineered artificial elastomeric protein to biomaterials  ii  that exhibit mechanical properties mimicking the passive elasticity of muscles. To optimize the biocompatibility of the tandem modular protein-based hydrogels, a protein domain which contains cell-binding sequences is used to construct ECM-mimetic hydrogels. The hydrogels can support cell adhension. Our result also suggests a possible method to design functional hydrogels. To prove the possibility of designing functional hydrogels, an xylanase was used to design enzymatic hydrogels. Our result shows that the enzymes remain active after being cross-linked into hydrogels. The possibility was further proved by fluorescent hydrogels designed from tandem modular protein based on Cyan Fluorescent Protein (CFP) and Yellow Fluorescent Protein (YFP). The fluorescent hydrogels can be applied as force sensors in cells with picoNewton (pN) sensitivity.  iii  PREFACE Portions of this thesis benefited from intellectual and experimental contributions of others. Chapter 2, a version of this chapter has been published as “Tandem Modular Protein-Based Hydrogels Constructed Using a Novel Two-Component Approach Shanshan Lv, Yi Cao, and Hongbin Li Langmuir 2012 28 (4), 2269-2274”, adapted with permission, copyright (2012) American Chemical Society. Dr. Hongbin Li and Dr. Yi Cao conceived the experiments. Dr. Yi Cao constructed the polyprotein genes. Dawei Zhou and Dr. Yi Cao performed the AFM imaging. I am responsible for the remaining experiments. I wrote the manuscript together with Dr. Hongbin Li. Chapter 3, a version of this chapter has been published as “Designed biomaterials to mimic the mechanical properties of muscles Shanshan Lv, Daniel M. Dudek,Yi Cao, M. M. Balamurali, John Gosline, Hongbin Li, Nature (2010) 465, 69-73", adapted with permission, Macmillan Publishers Limited, all rights reserved. Dr. Hongbin Li conceived the project. Dr. Hongbin Li and Dr. John Gosline designed the overall experiments. Dr. M. M. Balamurali constructed the polyprotein genes. Dr. Yi Cao and Dr. John Gosline measure the birefringence of the biomaterials. Dr. Daniel M. Dudek, Dr. Yi Cao and I carried out the tensile tests. Dr. Daniel M. Dudek and I iv  measured the Poisson’s ratio. I performed all other experiments and analyzed data with the help of Dr. Yi Cao. Dr. Hongbin Li wrote the manuscript. Chapter 4 has been published on line as “Shanshan Lv, Tianjia Bu, Jona Kayser, Andreas Bausch, Hongbin Li, Towards constructing extracellular matrix-mimetic hydrogels: An elastic hydrogel constructed from tandem modular proteins containing tenascin FnIII domains, Acta Biomaterialia, Available online 5 January 2013, ISSN 1742-7061, 10.1016/j.actbio.2013.01.002. (http://www.sciencedirect.com/science/article/pii/S1742706113000044)". Dr. Hongbin Li and I conceived the experiments. Tianjia Bu constructed the polyprotein gene. Jona Kayser and Dr. Andreas Bausch carried out the rheology tests. I performed all the other experiments and analyzed the data. In Chapter 5, Dr. Hongbin Li and I conceived the experiments. Dr. Stephen Withers suggested the active site titration. I performed all the experiments. I wrote the manuscript together with Dr. Hongbin Li. In Chapter 6, Dr. Hongbin Li and I conceived the experiments. I performed all the the experiments and I wrote the chapter.  v  TABLE OF CONTENTS ABSTRACT ......................................................................................................... ii PREFACE ........................................................................................................... iv TABLE OF CONTENTS ................................................................................... vi LIST OF TABLES ............................................................................................. ix LIST OF FIGURES ............................................................................................ x LIST OF SCHEMES......................................................................................... xv LIST OF ABBREVIATIONS .......................................................................... xvi ACKNOWLEDGEMENTS ........................................................................... xviii DEDICATION ................................................................................................... xx CHAPTER 1 Introduction.................................................................................. 1 1.1 Protein-based hydrogels as extracellular matrix mimetic hydrogels .....................1 1.2 Globular proteins used in hydrogel ........................................................................22 1.3 Biomaterials designed to mimic the passive elasticity of muscles using tandem modular titin-mimetic proteins .......................................................................35 1.4 Natural tandem modular ECM protein-fibronectin type III (FnIII) domain .....44 1.5 Modular structured enzyme protein-BCX .............................................................47 1.6 Fluorescent protein based hydrogel ........................................................................50  CHAPTER 2 Tandem Modular Protein-Based Hydrogels Constructed Using a Novel Two-Component Approach ..................................................... 55 vi  2.1 Introduction ..............................................................................................................55 2.2 Materials and Methods ............................................................................................58 2.3 Results .......................................................................................................................63 2.4 Discussions and Conclusions ...................................................................................77  CHAPTER 3 Biomaterials Designed to Mimic the Passive Elasticity of Muscles based on Tandem Modular Titin-mimetic Proteins ........................ 80 3.1 Introduction ..............................................................................................................80 3.2 Materials and Methods ............................................................................................83 3.3 Results .......................................................................................................................95 3.4 Discussions ..............................................................................................................118 3.5 Conclusions .............................................................................................................120  CHAPTER 4 Towards Constructing Extracellular Matrix-Mimetic Hydrogels: An Elastic Hydrogel Constructed from Tandem Modular Proteins Containing Tenascin FnIII Domains .............................................. 121 CHAPTER 5 Functional Hydrogels with Enzymatic Activity Based on Tandem Modular Protein Recombinant with Modular Structured Enzyme Protein-BCX...................................................................................... 122 5.1 Introduction ............................................................................................................122 5.2 Materials and Methods ..........................................................................................125 5.3 Results and Discussions ..........................................................................................131  vii  5.4 Conclusions .............................................................................................................152  CHAPTER 6 Fluorescent Hydrogels Constructed from Tandem Modular Proteins based on CFP/YFP FRET Pair as Force Sensors Capable of Estimating Swelling Force on Single Peptide Chain ................. 154 6.1 Introduction ............................................................................................................154 6.2 Materials and Methods ..........................................................................................158 6.3 Results and Discussions ..........................................................................................165 6.4 Conclusions .............................................................................................................183  CHAPTER 7 Conclusions and Future Directions ........................................ 184 7.1 Conclusions .............................................................................................................184 7.2 Future Directions ....................................................................................................188  REFERENCES ................................................................................................ 193  viii  LIST OF TABLES Table 6.1 The intensity ratio values.................................................................. 170  ix  LIST OF FIGURES Figure 1.1 Tensile testing result on resilin. ....................................................................13 Figure 1.2 Coiled-coil protein domains. ........................................................................19 Figure 1.3 Polyprotein (GB1)8 has significant mechanical properties. ..........................28 Figure 1.4 Design of artificial tandem modular protein based reversible hydrogel. ......30 Figure 1.5 The schematics of the heterodimeric CCE/CCK coiled-coil. .......................32 Figure 1.6 Molecular structure of muscle and titin. .......................................................37 Figure 1.7 Force- Sarcomere length curve of a cardiac myocyte. .................................39 Figure 1.8 Ig domains unfold/refold contributing to the elasticity of the sarcomere. ....40 Figure 1.9 Design of GB1-resilin based protein as titin mimics ....................................43 Figure 1.10 The structure of the third fibronectin type III domain of tenascin-C..........45 Figure 1.11 Mechanical unfolding behaviors of TNCfn3. .............................................46 Figure 1.12 Cartoon representation of the protein backbone structure of globular BCX domain. ....................................................................................................................48  Figure 1.13 The force-extension curves of polyprotein containing BCX flanked by (GB1)4 handles. ................................................................................................................49  Figure 1.14 Structure of GFP (PDB:1GFL). .................................................................51 Figure 1.15 Structure of CFP (PDB: 2Q57) and YFP (PDB:1HUY). ...........................54  x  Figure 2.1 Schematics of the two tandem modular proteins AG4A and CG5CG5C used to construct the protein hydrogel. .............................................................................66  Figure 2.2 CD spectra. ...................................................................................................68 Figure 2.3 Thermal melting behaviours of a GB1-CCE/GB1-CCK mixture. ................69 Figure 2.4 7 % aqueous solution of AG4A and CG5CG5C mixture forms protein hydrogel at neutral pH. .....................................................................................................72  Figure 2.5 Schematic drawing of the proposed gel network of the AG4A and CG5CG5C hydrogel. .........................................................................................................73  Figure 2.6 Rheology measurement. ...............................................................................74 Figure 2.7 Surface morphology of the freeze-dried hydrogel made of 7 % AG4A and CG5CG5C aqueous solution (the molar ratio of A: C is 1:1). ....................................76  Figure 2.8 Erosion profile of 100 mg 7 % AG4A and CG5CG5C hydrogel. ..................77  Figure 3.1 Coomassie blue stained PAGE gel for polyproteins (GR)4, GRG5RG4R, GRG5R, GRG9R and G8 (left to right). .............................................................................95  Figure 3.2 GB1-resilin-based polyproteins exhibit mechanical properties that are similar to those of titin at the single molecule level. ........................................................97  Figure 3.3 Mechanical properties of GB1-resilin-based polyproteins at the single-molecule level. .......................................................................................................98  xi  Figure 3.4 Far ultraviolet circular dichroism (CD) spectrum of R12 indicates that R12 is largely unstructured. ...............................................................................................99  Figure 3.5 Photographs of GB1-resilin-based biomaterials. ........................................100 Figure 3.6 Photocross-linking scheme and the schematic structure of hydrogel based on GRG5RG4R. .....................................................................................................101  Figure 3.7 Mechanical properties of (GR)4 and GRG5RG4R-based biomaterials. .......103 Figure 3.8 Consecutive stress-strain curves of GRG5RG4R during cyclic experiments. ...................................................................................................................105  Figure 3.9 Consecutive stretching-relaxation curves of GRG5RG4R at a pulling speed of 200 mm/min with waiting time between consecutive cycles of zero. ..............106  Figure 3.10 GRG5RG4R-based biomaterials can recover hysteresis under residual stress. ..............................................................................................................................107  Figure 3.11 GB1-resilin-based biomaterials exhibit pronounced stress relaxation behaviours. .....................................................................................................................108  Figure 3.12 Monte Carlo simulation on the force-relaxation of GRG5RG4R at constant extensions. ........................................................................................................110  Figure 3.13 Modulate the mechanical properties of macroscopic materials by affecting the folded state of GB1 domains using urea. ...................................................113  Figure 3.14 Mechanical properties of different types of GB1-resiline-based cross-linked biomaterials. ...............................................................................................115  xii  Figure 3.15 Swelling ratio of different types of GB1-resiline-based cross-linked biomaterials samples in PBS. .........................................................................................116  Figure 3.16 Mechanical properties of GRG5RG4R biomaterials with different protein concentration. .....................................................................................................117  Figure 3.17 Mechanical properties of GRG5RG4R biomaterials cross-linked at different APS concentration. ..........................................................................................118  Figure 5.1 Construction of recombinant protein containing BCX, G-R-G-BCX-G4-R. .........................................................................................................133  Figure 5.2 The activity of BCX in polyprotein G-R-G-BCX-G4-R. ............................134 Figure 5.3 Formation of hydrogels containing BCX. ..................................................137 Figure 5.4 The activity of G-R-G-BCX-G4-R-based hydrogels. .................................138 Figure 5.5 The activity of hydrogels with different G-R-G-BCX-G4-R concentrations.................................................................................................................140  Figure 5.6 Determination of active enzyme concentration by active site titration.......143 Figure 5.7 Characterization of CCE-G-G-BCX-G-CCE protein and CCE-G-G-BCX-G-CCE -based hydrogels. ....................................................................146  Figure 5.8 The enzyme activity of the hydrogels embedded with enzymes in repeated reactions. ..........................................................................................................150  xiii  Figure 6.1 Schematic illustration of how the FRET efficiency depends on the length of CCK linker sequences. ....................................................................................166  Figure 6.2 Structure of CFP and YFP and Coomassie blue stained SDS-PAGE picture for the constructed proteins. ...............................................................................168  Figure 6.3 Fluorescence spectroscopy of (GR)2-CFP-xtz3-YFP-(GR)2, (GR)2-CFP-CCK-YFP-(GR)2,(GR)2-CFP-CCK-YFP-(GR)2 after coiling with CCE, (GR)2-CFP-(GR)2. ..........................................................................................................169  Figure 6.4 The Iratio of (GR)2-CFP-(GR)2 remains the same after addition of CCE, indicating CCE did not have any effect on (GR)2-CFP-(GR)2. ......................................171  Figure 6.5 Fluorescence spectroscopy of (GR)2-CFP-CCK-YFP-(GR)2 before and after adding cross-linking matrix....................................................................................173  Figure 6.6 Photographs of hydrogels under fluorescence. ...........................................175 Figure 6.7 Relation of Iratio of (GR)2-CFP-xtz3-YFP-(GR)2, (GR)2-CFP-CCK-YFP-(GR)2, (GR)2-CFP-CCK-YFP-(GR)2 after coiling with CCE, (GR)2-CFP-(GR)2 with linker lengths. ...........................................................................178  Figure 6.8 Force-distance relation of CCK peptide sequence derived from WLC model. .............................................................................................................................180  xiv  LIST OF SCHEMES Scheme 3.1 The expected mechanism of Ru(II)-mediated photo-initiated protein cross-linking reaction. ......................................................................................................86  Scheme 3.2 Photos of a custom-made plexiglass mold for ring shaped hydrogel fabrication.........................................................................................................................88  Scheme 3.3 Schematic of tensile test. ............................................................................93  Scheme 5.1 Reaction mechanism of BCX with 2, 5-DNPX2, where 2, 5-DNP represents 2, 5-dinitrophenolate. ....................................................................................134  Scheme 5.2 Reaction mechanism of BCX with 2F-DNPX2, where 2F corresponds to 2-fluoro-β-xylobioside, and 2, 4-DNP to 2, 4-dinitrophenolate. ................................142  Scheme 6.1 DNA and amino acid sequences of CCE and CCK. .................................161  xv  LIST OF ABBREVIATIONS 2,5-DNPX2  2,5-dinitrophenol β-xylobioside  2,5-DNP  2,5-dinitrophenolate  AFM  atomic force microscopy  APS  ammonium persulfate  BCX  an xylanase from Bacillus circulans  BFP  blue fluorescent protein  CD  circular dichroism  CFP  cyan fluorescent protein  ECM  extracellular matrix  ELP  elastin-like peptide  E. coli  Escherichia coli  FnIII  third fibronectin type III domain  FRET  Förster Resonance Energy Transfer  GB1  streptococcal B1 immunoglobulin-binding domain of protein G  GFP  green fluorescent protein  Ig  Immunoglobulin  IPTG  isopropyl-1-β-D-thiogalactoside  LB  lysogeny broth  xvi  PBS  phosphate-buffered saline  PEVK  proline, glutamic acid, valine, and lysine  RFP  red fluorescent protein  RGD  Arg-Gly-Asp  RLP  resilin-like peptide  [Ru(bpy)3]2+  tris-bipyridylruthenium(II) cation  SEM  scanning electron microscopy  TN-C  tenascin-C  TNCfn3  the third fibronectin type III domain of tenascin-C  WLC  Worm Like Chain  YFP  yellow fluorescent protein  xvii  ACKNOWLEDGEMENTS The author thanks Dr. Hongbin Li for his patience and guidance over the course of this work. The passion, ambition and creativity that he brings to his work are perceptible. I am deeply impressed by his effort and energy that he consistently shared with those around him, and I have learned an immeasurable amount during my time in his lab. I greatly appreciate the opportunities that Dr. Li has given. I would also like to thank all the members of the Li lab for their patience and willingness to help with any problem encountered on a day to day basis. I wish them the best in their future endeavours, and hope that they achieve the success they deserve. The author would also like to thank the various collaborators and members of the UBC community that made the research presented here possible. Dr. Robert Campbell graciously provided CFP and YFP sequences. The author would also like to thank Dr. John Gosline, Dr. Margo Lillie and Dr. Robert Shadwick for discussions and their help in Instron test. The author would also like to thank Dawei Zou for his generous help in AFM imaging of the hydrogel. The author would also like to thank Drs. E. Polishchuk and G. Lamour for their general technical assistance in cell culture and cell imaging. The author would also like to thank Ms. A. Jollymore for critical comments xviii  on the manuscript of Chapter 4. The author would also like to thank Dr. Stephen Withers, Dr. Hongmin Chen, Dr. Ethan Goddard-Borger and Dr. Jamie Rich for their help in enzymatic activity tests. The author would also like to thank Dr. Matthew Roberts and Dr. Michael Wolf for their help in fluorescence measurement. The author would also like to thank those associated with the Biological Services Facility for their help and for maintaining a facility critical to the research presented here. I would also like to thank Dr. Leslie Burtnick, Dr. Stephen Withers and Dr. Mark Thachuk for providing guidance as my supervision committee. I especially thank Dr. Leslie Burtnick for critical evaluation of this manuscript. Last but not least, I would like to thank my parents for always being there, for their deep love and greatest support through all these years and my entire life.  xix  DEDICATION For my parents, my brother, my uncles and aunts, and LF.  xx  CHAPTER 1 Introduction 1.1 Protein-based hydrogels as Extracellular Matrix (ECM) mimetic hydrogels New generations of protein-based biomaterials are being developed rapidly over the last two decades for potential applications in biomedicine as well as basic biological studies including tissue engineering, drug delivery and biocatalysis [1-9]. Proteins display interesting properties in biomaterials, especially their excellent biocompatibility and biodegradability, which make them promising candidates for biomedical applications. Moreover, as proteins comprise the structural basis for extracellular matrices, biomaterials formed from proteins are particularly suited for tissue engineering [10]. Among these protein-based biomaterials, protein-based hydrogels are of particular interests, widely used in biomedical applications such as the controlled delivery of drugs and cells and in tissue engineering [11, 12]. Hydrogels are water-swollen, three-dimensional networks of cross-linked polymeric structures. Protein-based hydrogels not only retain the biocompatibility, biodegradability from the protein component, but also gain unique physical properties from the hydrogel network structure, such as mechanical and structural properties, which can be made to resemble those of many natural tissues. As a consequence of these unique properties, a 1  variety of protein-based hydrogels have been developed to be used as scaffolds capable of supporting cell viability and function [11] for tissue engineering applications. An important template in the design of hydrogels for tissue engineering applications is extracellular matrix (ECM). The ECM is material secreted by cells. Its major components are macromolecules of glycoproteins, proteoglycans, polysaccharides, and proteins, such as collagen, elastin, fibronectin, and laminins. The ECM fills the space between cells in tissues, providing structural support for cells. It also provides an environment for new tissue formation. Moreover, it plays an essential role in regulating cell dynamics such as cell proliferation, differentiation and migration [13-17]. Cells can interact with their ECM through binding their cellular receptors (for example, integrin) to the ECM components, (for example, fibronectin, laminins and collagens) [18]. These cell binding partners can then elicit corresponding cues to regulate cell dynamics through mechanotransduction on a molecular level. For instance, the protein fibronectin has been reported to undergo force-induced unfolding of protein structure, converting mechanical signals into biochemical signals to modulate cellular processes [19-21]. As described above, the ECM controls cell adhesion, proliferation, migration and differentiation. Hence, the ECM has been an important guide 2  in developing biomaterials that promote cell/tissue growth and mimicking the ECM has been a good tool in the design of hydrogels in tissue engineering. Protein-based hydrogels have been explored as synthetic ECM to provide artificial extracellular microenvironments that will mimic the physical and biochemical characteristics of natural ECM for applications in cell studies and tissue regeneration. For such purposes, a variety of proteins have been studied extensively. Among these different protein-based hydrogels, those based on proteins, which are of natural origin, such as natural ECM proteins elastin, collagen/gelatin, fibrin and silk protein, have been exploited most frequently as ECM mimics for tissue engineering and are described in detail in this section.  1.1.1 Elastin and Elastin-like-polypeptides (ELPs) Elastin is a major structural protein of the ECM found in connective tissues, lungs, blood vessels, skin, elastic ligaments and cartilage. Elastin is known to provide tensile strength and elasticity to these tissues, allowing these tissues to recover their shapes after deformation. Elastin is also known to regulate cell-cell interactions and cell-ECM interactions [22]. The precursor of elastin is tropoelastin. Tropoelastin has a molecular weight ~ 60 kDa. It 3  contains variable amino acid repeats of hydrophobic domains Val-Pro-Gly-X-Gly (X is other hydrophobic amino acids such as alanine, leucine, and isoleucine) alternated with alanine-rich domains containing lysine [22]. The amino acid sequence gives rise to a random coil structure of tropoelastin. Upon an increase in temperature, tropoelastin can self-assemble through a coacervation (inverse transition) process [23-26]. Meanwhile, the lysine in the alanine-rich domains can form cross-links, leading to an insoluble network, that is, elastin. Due to its major content in ECM and excellent mechanical properties, elastin has been suggested for many potential applications [27-31]. However, because of difficulty in purification and other problems [28, 32-34], elastin could not be used very often in tissue engineering. Therefore, a great deal of research has been conducted on developing elastin-like peptides (ELP). Initial studies on chemically synthesized ELP were carried out by Urry et al. In their study, a penta peptide repeat, poly(Val-Pro-Gly-Val-Gly), was cross-linked by γ-irradiation [35, 36]. Later studies expanded the ELPs to Val-Pro-Gly-X-Gly, in which X can be any natural amino acid, except proline [36-38]. These ELPs self-assembled to hydrogels for various applications [36-38]. Tirrell and coworkers also produced ELPs for tissue engineering applications [39]. ELPs alternated with fibronectin-derived 4  Arg-Gly-Asp (RGD) cell-binding domains were chemically cross-linked to form hydrogels. The resultant ELP-based hydrogels not only exhibited mechanical properties similar to those of natural elastin, but also promoted cell spreading and adhesion, making them suitable candidates for use as vascular graft replacement materials [40-43]. Tirrell and coworkers also synthesized an ELP containing non-natural amino acids to explore other cross-linking methods, providing opportunities for photopatterning of ELPs [44]. Many other functional hydrogels based on elastin-like recombinant protein have also been explored for drug delivery applications [45-48], tissue engineering and many other biomedical applications [45, 49-54].  1.1.2 Collagen and Gelatin-Based Hydrogels. Collagen is another major component of the ECM and is the most abundant protein found in mammals, particularly found in connective tissues, such as articular/cartilage/cornea, bone tissues and blood vessels, as well as in fibrous tissues such as tendon and skin. Collagens have great tensile strength and are essential for many important mechanical functions. Collagens support most tissues and play an important role in tissue development. Collagens can also support cells and interact with cells through binding to cells via integrin receptors, playing an important role in signal transduction 5  [55, 56]. Native collagen presents as a triple helix, consisting of amino acid repeats as Gly-Pro-X and Gly-X-hydroxyproline (Hyp), where X can be any amino acid other than Gly, Pro or Hyp [57]. There are various types of collagen, the most common ones are collagen type I to V. Different types of collagen have been used as many tissue engineering biomaterials [58]. Of all the collagens, collagen type I is the most abundant, mostly found in tissues like skin, tendon and bone, as well as in damaged articular tissues. Collagen type I is considered to play an important role in tissue regeneration and repair. Furthermore, collagen type I can spontaneously polymerized into gels at physiologic pH/temperatures [56]. As a result, collagen type I hydrogels have been widely used for biomedical and tissue engineering applications [56]. Studies revealed that articular chondrocytes remain viable and functional upon attachment to collagen type I hydrogels [59-69] and collagen type I gels containing chondrocytes were used to repair articular defects in animals [70]. Due to its major content in bone, collagen type I has also been used to mimic natural ECM of bones. Collagen type I from various sources, including mammalian sources such as bovine [71-73] and marine organisms such as marine sponges [74, 75] and fish skin [76, 77], has been used as material for bone replacement. In addition to collagen type I, collagen type II has also been 6  reported to be used as artificial ECM materials, such as articular cartilage matrix [78]. Despite of the wide applications, there are some disadvantages of collagen-based gels including low mechanical strength, limited availability and fast degradation rate [10, 79]. Therefore, extensive research has explored the design of collagen-based synthetic hydrogels with controllable properties for tissue engineering applications. Collagen type I based hydrogels containing different concentrations of fibronectin, laminin, and collagen type IV were fabricated and used for stem cell differentiation study [80]. Collagen has also been used to form triblock thermoreversible gels with predictable viscoelastic properties [81, 82]. In another study, the repeating sequence Gly-X-Hyp of collagen was modified and used to construct gels. The gels were further tested with in vitro cell culture experiments and it was reported that the gels were biocompatible/suitable for Mouse embryonic fibroblast cells (3T3) [83]. Beside of the usage as a biomaterial in its native form, collagen is widely used after denaturation. When the triple-helix of collagen is denatured into single-strands, collagen becomes gelatin. In other words, gelatin is the denatured form of collagen. There are two types of gelatin, gelatin type A and B. Gelatin type A is obtained through acidic pretreatment of collagen 7  followed by thermal breakdown, while gelatin type B is obtained through thermal breakdown after alkaline treatment. Similar to its native form, gelatin can form gels upon a change in temperature. Gelatin gels also exhibit biodegradable and biocompatible properties that are suitable for biomedical applications. Hence, gelatin-based materials have also been explored for tissue engineering. Lots of effort has been devoted to produce gelatin gels with improved thermal and mechanical stability. Cross-linked gelatin has been reported [84, 85]. For example, genipin cross-linked gelatin was used in development of a tissue resembling native cartilage [86].  1.1.3 Silk Silks are naturally occurring protein fibres extruded by insects such as silk worms, cocoons and spiders [52, 58]. Silk derived from the silk worm called Bombyx mori is silk fibroin. Silk fibroin is a fibrous protein, mainly consisting of glycine (45.9 %), alanine (30.3%) and serine (12.1 %). It has two components: fibroin as structural center and serecin as glue-like sticky coating surrounding the fibroin. Fibroin contains amino acid repeat sequences such as Gly-Ala-Gly-Ala-Gly-Ser and Gly-Ser-Gly-Ala-Gly-Ala, which form β sheets. Upon assembly to the β-sheets, the protein becomes insoluble in water, leading to the formation of hydrogels [87]. Silk fibroin 8  hydrogels exhibit unique mechanical properties and slow degradation rates. Moreover, silk fibroin is biocompatible, being reported to support stem cell adhesion, proliferation, and differentiation in vitro and to promote tissue repair in vivo [88, 89]. All of these advantages make it suitable for biomedical applications. For example, it has been shown that silk fibroin hydrogels stimulated cell proliferation and permitted healing of bone defects in rabbits [90]. Hydrogels based on silk fibroin have also been used for in vitro cartilage tissue engineering [91-93]. In addition to silk fibroin, silks from spiders provide alternative options to develop silk-based biomaterials for tissue-engineering applications. Among all the different types of silks produced by spiders, dragline silk and flagelliform silk are the most well studied. The dragline silk has a high strength and is used as the frame of the orb web, while the flagelliform silk exhibits high extensibility and is used as the core fibers of the spiral of the web [94]. Like silk fibroin, flagelliform silk also consists of repetitive amino acid sequence of Gly-Pro-Gly-Gly-X [95]. Spider silk has also been explored as biomaterial for tissue engineering applications. For example, a recombinant polypeptide which mimics the repeating sequence Gly-Pro-Gly-Gly-X of flagelliform silk has been synthesized [96].  9  1.1.4 Fibrin-Based Hydrogels. Fibrin, another fibrous protein found in mammals, is the major structural component of blood clots and plays an important role in tissue regeneration and repair. Fibrin is formed from enzymatic cleavage of the protein fibrinogen by the serine protease thrombin. The cleavage of fibrinogen initially gives rise to molecules called fibrin monomer. These fibrin monomers can bind to each other, forming oligomers. These oligomers further aggregate to fibrin gels [97]. In vivo, this polymerization process of fibrin monomers to fibrin gels forms a plug at a wound site, part of the blood clotting process. In vitro, the fibrin gels show excellent biocompatible and mechanical properties. Moreover, the mechanical properties, stability and solubility of fibrin gels can be tuned by varying the formation conditions [98-100]. As a result, fibrin has been used extensively in biomedical applications, ranging from tissue engineering to wound healing [10, 56, 101, 102]. For instance, a series of studies demonstrated that fibrin gel was suitable for cartilage tissue engineering [103, 104].  1.1.5 Resilin Resilin is an elastic protein mainly found in the cuticle in most insects. The protein was first discovered in the 1960s by Weis-Fogh in the flight systems 10  of locusts and dragonflies [105, 106]. Resilin was shown to have remarkable mechanical properties such as extraordinary elasticity, high strain, low stiffness, efficient energy storage and a long lifetime. Since its first description, resilin has been found in specialized regions of the cuticle of almost all insects where it affects mechanical properties of the cuticle. It is also found to play important roles in many biomechanics including locomotion (flight and jump), feeding, respiration and sensory among a wide variety of insect, such as fleas, spittle bugs [106-110], dragonfly, locust [24, 26, 111-114], cicadas [115], and moths [116]. A gene from Drosophila melanogaster (CG15290) was identified as the precursor for Drosophila resilin [117], which is called pro-resilin [25] . The N-terminus of the gene is dominated by glycine- and proline-rich repeats of a 15-residue motif sequence GGRPSDSYGAPGGGN [118]. These repeats are reported to be connected by di-and tri-tyrosine cross-links into a three-dimensional network with disordered polypeptide chains. Upon stretching, a loss in conformational entropy of the disordered chains will give rise to a restoring force. Since the force is caused by entropy change, it is called entropic force. The force will cause the polypeptide chains to return to the unstretched state, giving rise to entropic elasticity of  resilin [112]. In  addition to the entropic elasticity, the high content of hydrophilic residues in 11  resilin suggests that there could be enthalpic elasticity, which involves energy exchange in contrast to entropic elasticity [119-121]. Interests in the outstanding properties of resilin have motivated exploration of resilin-like polypeptides (RLPs). Elvin and coworkers successfully cloned and expressed a recombinant form of the Drosophila melanogaster CG15290 gene, rec-resilin (R). The R were cross-linked via Ru(II)-mediated photochemical methods to obtain di-tyrosine linkages and cast into rubber-like biomaterials. Moreover, the cross-linked hydrated RLP was highly elastic and resilient (Figure 1.1) [122-124]. Their studies also demonstrated that the synthetic resilin was mostly unstructured [120]. Elvin and co-workers also expanded the resilin-like protein to three different proteins named Rec1-resilin (the first exon of the Drosophila CG15920 gene), Dros16 ((GGRPSDSYGAPGGGN)16) and An16 ((AQTPSSQYGAP)16). All three proteins exhibited similar mechanical properties [123, 125].  12  Figure 1.1 Tensile testing results on resilin. Purple line is a typical stress-strain plot for a strip of rec-resilin tested in a stretch-release cycle, where stress corresponds to force per unit area and strain equals to  lengthextended − lengthinitial . During stretching, as the strip is extended, the lengthinitial strain increases and the force/stress also increase. Upon releasing, both the strain and stress decrease. Sample cycled to 225%, showing resilience of 97%. Sample later tested to failure showing extension at break of 313% (blue curve). Also shown for comparison is the theoretical curve (green curve). Reprinted from reference [122] by permission from Macmillan Publishers Ltd: Nature, copyright (2005).  The unique mechanical properties of resilin and RLPs have motivated and provided opportunities in developing materials for tissue regeneration. For 13  example, a design of RLPs sequences has been reported that combines the resilin-like polypeptide with cell-binding domains RGDSP to promote cell adhesion [126]. A heparin-binding domain has also been reported to be incorporated into the resilin-like polypeptide sequence [124, 127-129]. These reports together with other studies [52, 130-133] show the potential of RLPs as matrix in tissue regeneration.  1.1.6 Abductin Abductin is another natural elastomeric protein that is found in hinge region between shell junctions of bivalve mollusc shells. It contributes to shell opening and closing, which are involved in scallop swimming [52, 134]. Abductin is predominantly made of three amino acids, glycine, methionine, and phenylalanine. Due to its high glycine content, the abductin adopts a random coil conformation [134, 135]. Like other elastomeric proteins, abductin contains repetitive sequences linked through cross-links. The C-terminal domain of abductin contain glycine- and methionine-rich penta-/deca-peptide sequences [134]. These repeating peptide sequences are reported to be connected through lysine and tyrosine cross-linking sites [136].  14  Abductin is so far the only elastomeric protein in nature that exhibits compressible elasticity for functioning [134]. Because of this, it draws lots of attention in biomaterials research. Abductin-like proteins have been designed and synthesized. For example, a peptide mimetic of abductin was reported to self-assemble into a biphasic aggregate with compressive elasticity like abductin [137].  1.1.7 Wheat gluten Wheat gluten is an elastomeric protein found in plants. It accounts for about half of the total protein in some dry grain, serving as storage of nutrients (such as carbon, nitrogen and sulfur) for seedlings. The elastic properties of wheat gluten have no known biological role, but give elastic properties including elasticity, viscosity and extensibility to dough, which allow the dough to be made into foods often with a chewy texture. Wheat gluten is composed of gliadin and glutenin. Upon mixing with water, glutenin cross-links leading to a continuous network attached to gliadin, which forms wheat gluten. The gliadin and glutenin proteins are considered to be responsible for the elastic properties of wheat gluten. Similar to other known elastomeric proteins, the proteins in wheat gluten contain repetitive domains located in between of N- and C-terminal domains. The repetitive 15  domains are glycine and proline-rich, including hexa-amino-acid motifs Pro-Gly-Gln-Gly-Gln-Gln and nona-amino-acid motifs Gly-Try-Try-Pro-Thr-Ser-Pro-Gln-Gln [138]. The repetitive domains can undergo deformation and this deformation process is considered to contribute to entropic elasticity of wheat gluten. In contrast to other known elastomeric proteins, the remaining amino acid residues of the repetitive domains are hydrophilic [138, 139]. These peptides are reported to form repetitive β-turns between the repeats, which are considered to contribute to enthalpic elasticity of wheat gluten [139-141]. The N- and C-terminal domains of the proteins are globular, non-repetitive amino acid sequences and contain cysteine residues for inter-chain covalent cross-linking to other proteins [140]. There were also recombinant polypeptides based on subunit of wheat gluten synthesized and cross-linked by γ-irradiation, resulting in hydrated materials that were elastic [142].  1.1.8 Hydrogels Based on Self-Assembled Peptides Self-assembled peptides have also been a useful tool in hydrogel design because the formation of α-helical coiled coil is a natural way of self-assembly of biological systems [143-147]. Naturally occurring α-helical coiled coil is found in transcription factors [148-152]. The x-ray crystal 16  structures of α-helical coiled coil were first solved by Kim and Alber and co-workers. The study was on a homodimer formed by the transcription factor GCN4 coiled coil domain. The structure was a left-handed superhelix with two parallel strands, showing a typical core packing with hydrogen bonds formed in the core and salt bridges formed between charged residues of the two strands [148, 150]. As concluded from previous studies, structurally, the α-helical coiled coil domain is a supercoil of two or more α-helical strands. The repeat peptides of α-helical coiled coil form a helix. The repeat is mostly a heptad amino acid peptide labled as abcdefg (Figure 1.2 [8]) according to the amino acid location in the helix. The helices contain two turns per heptad, thus 3.5 residues per turn. At positions a and d are hydrophobic amino acids like leucine, while at positions e and g are usually charged amino acids. The hydrophobic residues can form a hydrophobic plane of the a and d positions. This hydrophobic effect leads to hydrophobic inter-helical interfaces and drive the helix aggregation to coiled coil dimers, which is the formation of a coiled coil structure [6]. The coiled coil structure can be stabilized/de-stabilized by electrostatic interaction between charged residues found at positions e and g and other residues at positions b, c and f [149, 152]. The coiled coil is also known as α-helical leucine zipper domain 17  because of the leucine residues commonly found at the a and d positions and that two such domains can form zipper-like dimers. In addition to dimers, Alber, Kim and co-workers also designed trimeric and tetrameric coiled coils configurations [151]. Engineering of five [153] and even seven [154] stranded coiled coils has also been demonstrated. The formation of the coiled coil structures can be used as physical cross-linking to drive the formation of hydrogels. Hydrogels formed by self-assembled peptides have been suggested as candidate biomaterials in drug delivery and tissue engineering applications [143].  18  Figure 1.2 Coiled coil protein domains. A) Schematic of the interaction of two heptads. The interactions of amino acid residues at positions a and d form a hydrophobic core. B) Helical wheel representation of dimerization of two-stranded, antiparallel α-helical coiled coils. C) Schematic of the parallel or antiparallel orientation of two coiled coil strands. Reprinted from reference [8], Copyright (2007), with permission from Elsevier.  Based on naturally occurring coiled coil proteins, the Tirrell group designed and studied a protein containing six heptad peptide coiled coil repeats [6]. A triblock copolymer adopting an ABA triblock architecture was designed with the coiled coil domains (A) flanking a water-soluble central domain (B). Through the self-assembly/dissembly of the coiled coil domains, the triblock 19  copolymer formed a reversible hydrogel. The hydrogel formation was thermal-/pH-dependent. This approach not only allows control of the coiled coil domains, but also allows the control of the central domain through which other functions/properties can be introduced into the hydrogels [6]. Xu and co-workers designed a series of triblock polypeptides based on the coiled coil domain. In one of their studies, they designed two terminal coiled coil motifs, named A and C, and a central block represented by B. A triblock copolymer with an ABC architecture was designed and used to form hydrogels. These hydrogels are responsive to stimuli and reversible and, therefore, have potential applications in the biomedical field [155]. The Woolfson group designed two 28-residue peptides as coiled coil sequences. In their design, the effect of the amino acids at position b, c, and f were investigated. With three alanines on one peptide and three glutamines on another at position b, c, and f, the designed sequences were shown to form gels. The hydrogels were shown to support rat adrenal pheochromocytoma cell differentiation and cell growth, providing a good platform as substrates for cell growth [156, 157]. There are also a wide variety of other self-assembled peptides developed for hydrogels design and fabrication [147, 158]. Stupp and collaborators [158-162] have developed a class of self-assembled peptides that can 20  self-assemble into rodlike shapes [161]. Hydrogels have been produced with these self-assembled rodlike peptides. These self-assembled peptides were attached with other functional motifs to promote cellular adhesion and other biological functions [163]. Zhang and collaborators [164-167] fabricated another kind of self-assembled peptides that self-assembled into β-sheets and then form hydrogels. The hydrogels made with this kind of self-assembled peptides were used to encapsulate cells as well as to generate three-dimensional environments for a variety of cells, such as stem cells [164-166, 168].  All of these protein-based hydrogels described above have a number of unique features, which makes them suitable candidates as ECM mimetic biomaterials to provide artificial extracellular microenvironments that will mimic the physical and biochemical characteristics of natural ECM for applications in cell culture and tissue engineering. However, most of these proteins are non-globular proteins that behave like entropic springs upon stretching. As described previously, in natural extracellular matrices, a large number of ECM proteins are tandem modular proteins that consist of many individually folded functional domains to confer the required biological functionalities for cell-ECM interactions. 21  Many of these tandem modular proteins are subject to stretching forces under physiological conditions and can undergo force-induced conformational changes to modulate biological processes via mechano-signal transduction [19-21]. These tandem modular ECM proteins are promising building blocks for constructing novel biomaterials for a variety of biomedical applications. However, they have not been explored extensively for such purposes. We hypothesize that these tandem modular proteins may also be used in constructing novel hydrogels that can mimic the physical and biochemical characteristics of natural extracellular matrices.  In this thesis, we explored the feasibility of using such tandem modular proteins for constructing protein-based biomaterials with five model systems, including an artificial elastomeric protein (GB1 domain), a natural ECM protein (FnIII domain), a modular structured enzyme (BCX), as well as fluorescent proteins (CFP/YFP).  1.2 Globular proteins used in hydrogel 1.2.1 Overview of globular proteins used in hydrogel Many globular proteins have been reported to form gels, such as β -lactoglobulin, lysozyme, bovine serum albumin, and ovalbumin [58, 169]. 22  β-lactoglobulin is a protein widely used in food and pharmaceutical industry, and thus well studied. Studies on the gelation processes of this globular protein show that upon an increase in temperature, the protein will denature and form some small aggregates and some large linear fibrils. The small aggregates can grow larger and these larger aggregates can form gels above certain concentration; while the large linear fibrils can undergo phase separation, then aggregate to even longer fibrils and finally self-assemble into gels [170-173]. Lysozyme is a globular protein which can be isolated from hen egg white. This protein contains β-sheets and α-helices, as well as disulfide bridges which help stabilize the globular structure of the protein. Upon addition of reducing agent, the disulfide bridges will be disrupted. The protein structure will become more flexible, leading to the self-assembly and formation of a hydrogel. Lysozyme has been investigated for applications in biomaterials. For example, a transparent hydrogel was formed through heating a dithiothreitol (DTT)-containing lysozyme solution to 85 °C followed by cooling down to room temperature. This lysozyme gel was used to investigate cell-hydrogel interactions. Results showed that the lysozyme hydrogels were biocompatible, promoting 3T3 fibroblasts cell proliferation and supporting cell spreading. These results demonstrates the potential of 23  lysozyme hydrogels as biomaterials in tissue engineering applications [174, 175]. Other progress has also allowed for incorporating a single globular domain into hydrogels using recombinant protein technology. These hydrogels allow for many potential applications, including tissue engineering scaffolds and heterogeneous biocatalysis [176].  Although the progress described above has allowed for the utilization of globular protein domains in hydrogels, the random coil nature of the used globular domains remains the key design principle (since most of the hydrogels form after proteins denature). However, many naturally occurring extracellular matrix proteins are tandem modular elastomeric proteins, which are composed of individually folded domains [177]. The tandem modular construction is one of the common features of natural elastomeric proteins, making it possible for them to unfold sequentially when subject to stretching forces [21, 178-180]. This unfolding process is considered to convey high toughness to the elastomeric proteins and makes them perfect shock-absorbers, giving rise to the elastic characteristics of a wide variety of natural materials [181-185]. Upon removal of stretching force, the unfolded proteins can refold back to their original folded modular structure. This 24  ability to refold enables the elastomeric proteins to maintain their structure and mechanical properties after constantly repeated stretching-relaxation cycles [186], demonstrating remarkable consistency and reliability in their mechanical functions in the elasticity of natural adhesives, cell adhesion proteins and muscle proteins [187-189] . Thus, incorporating tandem modular folded proteins into hydrogels is very important in creating a synthetic extracellular matrix that closely mimics naturally occurring ones.  Inspired by naturally occurring elastomeric proteins, researchers have started to explore naturally non-mechanical proteins to construct artificial tandem modular elastomeric proteins. Previous studies demonstrated that such modular artificial elastomeric proteins also display significant mechanical stability [190-198]. Our group reported the engineering of the first artificial tandem modular protein-based reversible hydrogel through direct incorporation of such tandem modular proteins into hydrogels starting with recombinant proteins. Results show that the engineered hydrogel exhibits unique properties combining a slow erosion rate and a fast and reversible sol-gel transition [199].  25  1.2.2 Artificial elastic protein domain-GB1 The model modular protein our group reported is an artificial protein domain GB1 (Figure 1.3). GB1 is the streptococcal B1 immunoglobulin-binding domain of protein G (PDB: 1PGA). It is a small globular, α+β protein with only 56 amino acid residues. Our group and others have identified that GB1 has significant mechanical stability. The average unfolding force is 184±41 pN revealed by Atomic Force Microscopy (AFM) studies [191, 193], which is comparable to the mechanical stability of the I27 domain from the natural elastomeric protein titin [200]. Besides, similar to naturally occurring elastomeric proteins [179, 189, 201-203], there is significant energy dissipation as heat in the process of mechanical unfolding of GB1 domains, which entails high toughness for GB1 and makes it an ideal shock-absorber [179, 189, 204]. In addition, GB1 can fold fast to efficiently recover its mechanical stability and avoid mechanical fatigue [205]. These features when combined together enable GB1 to fulfill one of the prerequisites for a non-mechanical protein to function as an artificial elastomeric protein. To mimic the tandem modular architecture of natural elastomeric proteins, our group engineered polyprotein (GB1)8, consisting of eight tandem repeats of GB1 domains (Figure 1.3). Our previous single molecule AFM experiments have demonstrated that polyprotein (GB1)8 shows excellent 26  mechanical properties, combining low mechanical fatigue during repeated stretching-relaxation cycles, ability to fold against residual forces, fast folding kinetics and high folding fidelity, all of which are comparable or superior to those of known naturally occurring elastomeric proteins [191, 206]. These mechanical features make GB1 polyprotein an ideal artificial protein-based molecular spring for bottom-up construction of biomaterials [207].  27  Figure 1.3 Polyprotein (GB1)8 has significant mechanical properties. A) A cartoon representation of the three-dimensional structure of the non-mechanical protein GB1. B) A schematic diagram of polyprotein (GB1)8. C) Typical force-extension curves of (GB1)8 polyproteins. D) A pair of typical stretching (black) and relaxation (red) curves of a polyprotein GB1. The hysteresis (shaded area) reflects the energy dissipated during the mechanical unfolding of GB1 domains. E) The refolding kinetics of GB1 are very fast. Plot of the refolding probability, Nrefolded/Ntotal, versus t. The polyprotein is first stretched to unfold all the GB1 domains and record the 28  total number of domains, then the protein is quickly relaxed to its original length. After a relaxation time t, the protein is stretched again to record the number of domains that have refolded. F) GB1 can refold in the presence of residual forces. The protein is stretched to unfold all the GB1 domains. Then, it is rapidly relaxed to a shorter length and held for certain time. The protein is stretched again to record the number of domains refolded during the waiting time under the force. Plot of Nrefolded/Ntotal versus residual force. G) GB1 does not show significant mechanical fatigue during a repeated stretching-relaxation experiment. Reprinted from reference [206] by permission from Macmillan Publishers Ltd: [Nature Materials], copyright (2007).  As a step toward engineering extracelluar mimetic hydrogels, previous work of our group developed the first artificial tandem modular protein-based hydrogel. The formation of the hydrogel was mediated by self-assembly of leucine zipper domains. As described in the previous section, pioneering work has demonstrated that leucine zipper domains are excellent building blocks with which to construct self-assembled protein hydrogels and have been widely used to engineer protein-based hydrogels for biomedical applications. Such proteins typically adopt an ABA/ABC triblock 29  architecture with leucine zipper based coiled coil motifs at both ends and a center block [6, 7, 155, 176, 198, 208-213]. To construct tandem modular (GB1)8 (G8)-based hydrogel, our group previously adopted the standard triblock protein design for hydrogel: G8 was used as the center block and the well-characterized leucine zipper domains A was used to flank the center block at its N- and C-termini (Figure 1.4). The leucine zipper A used was designed by Tirrell and co-workers [6] and has been reported to be able to self-associate into oligomers and mediate formation of hydrogels [6, 209, 210]. Our results demonstrated the engineering of the first artificial tandem modular protein-based reversible hydrogel with unique properties, such as improved erosion properties and ability to bind Immunoglobulin G (IgG) antibodies [199].  Figure 1.4 The figure has been published in reference [199]. According to the copyright policy of Royal Society of Chemistry (RSC), the reuse of RSC figures in a thesis/dissertation is not permissible. Therefore, Figure 1.4 has been removed when the thesis is published. As an alternative, the link to the reference's DOI is provided as following: http://pubs.rsc.org/en/content/articlelanding/2008/cc/b806684a  30  Our approach, using tandem modular protein G8, which is made of eight identical tandem repeats of a small protein GB1, as the center block in a triblock protein ABA to construct a novel protein hydrogel [199], provides a new approach to tune the topology and physical properties of the protein hydrogels via genetic engineering, and opens the possibility to systematically explore the use of large native extracellular proteins to engineer extracellular matrix-mimetic hydrogels. However, there are some limitations of the triblock protein approach, such as relatively fast erosion rate. Therefore, new design strategies must be conceived to engineer protein hydrogels containing tandem modular proteins.  To obviate this problem, in one project of this thesis, we explore a novel two-component approach to engineer tandem modular protein-based hydrogels. This methodology makes use of two complementary leucine zipper sequences (CCE and CCK), which do not self-associate but self-assemble into heterodimeric coiled coils at neutral pH, as functional groups to drive the self-assembly of hydrogels. The resultant hydrogels show improvements in various aspects of the hydrogel properties over the first-generation tandem modular protein-based hydrogels [199]. 31  Figure 1.5 The schematics of the heterodimeric CCE/CCK coiled coil and schematics of the two tandem modular proteins CCE-G4- CCE and CCK-G5-CCK-G5-CCK used to construct the protein hydrogel. Green and blue coils represent the A (CCE) and C (CCK) domains under neutral pH, respectively; red globules represent the folded GB1 domains. Reprinted with permission from reference [214] (Tandem Modular Protein-Based Hydrogels Constructed Using a Novel Two-Component Approach Shanshan Lv, Yi Cao, and Hongbin Li Langmuir 2012 28 (4), 2269-2274) Copyright (2012) American Chemical Society.  Employing the CCE/CCK coiled coil sequences, two tandem modular proteins were engineered to form hydrogel mimicking extracellular matrix. This approach overcomes the problems we encountered in constructing hydrogels based on tandem modular proteins carrying coiled coil sequences, which will be described in detail later (Chapter 2).  32  This method provides a new approach to use large native extracellular proteins to engineer extracellular matix-mimetic hydrogels. However, the hydrogels formed in this method are typically mechanically weak due to the weak interactions through physical cross-linking that arise from the association of the coiled coil sequences. Therefore, the hydrogels cannot be used for tissue engineering applications where high mechanical stability is required. Previous studies suggest that chemical/covalent cross-links are potentially a powerful approach to generate stronger interactions and thus more mechanically stable structures [215-220]. Recently, it has been demonstrated that a ruthenium(II) (Ru(II))-mediated photochemical cross-linking strategy [221], allows the cross-linking of two tyrosine residues in close proximity into di-tyrosine adducts. This method was used successfully to cross-link recombinant resilins into solid biomaterials [122]. Further studies on hydrogels formed using the same approach showed that various fibrins/gelatins could also be rapidly (within seconds) cross-linked to form highly elastic hydrogels with mechanical properties comparable to some of those of native elastomeric proteins. In addition, the ability of this biomaterial to rapidly form hydrogels enables its potential use in situ for  33  many biomedical applications and for generating tissue engineered structures with strong mechanical stability [222].  Therefore, going a step further, in one project of this thesis, we employed this well-developed photochemical cross-linking strategy to engineer protein-based biomaterials that exhibit mechanical properties that closely mimic the passive elastic properties of muscles. We designed and engineered an artificial elastomeric protein, which combines folded tandem modular protein building block with random coil-like sequences, to closely mimic the architecture and domain arrangement of the giant muscle protein titin. By cross-linking these mini-titin mimetic proteins photochemically, we successfully engineered novel protein-based biomaterials that exhibit mechanical properties that closely mimic the passive elastic properties of muscles. With unique mechanical properties, this new type of biomaterials has potential applications in tissue engineering and material sciences and will be described in detail in later sections.  34  1.3 Biomaterials designed to mimic the passive elasticity of muscles using tandem modular titin-mimetic proteins 1.3.1 Introduction to titin and its role in muscles Titin is the third most abundant filamentous protein in muscle, only after the thin and thick filaments which are composed of actin and myosin, respectively. Titin is a giant protein with molecule weight of 3-4 MDa, encoded by a single gene (TTN) which locates to a 294-kb region on the long arm of chromosome 2 in both human and mouse. A single titin molecule spans half a sarcomere, connecting Z-disk and M-band [187]. It plays an important role in organizing sarcomeres by serving as a scaffold that coordinates assembly of many proteins with a variety of functions in the sarcomere. Aside from its role in organizing sarcomeres, titin provides an elastic link between Z-disc and A-band. Through serving as an elastic “molecular spring”, titin is mainly responsible for the passive elasticity of muscles, which combines strength, toughness and elasticity [223].  1.3.2 Molecular structure and elastic properties of the I-band region of titin The molecular structure of titin is the key to its function in giving muscle its distinct mechanical properties and integrity during contraction, relaxation, 35  and stretch [179]. Although a titin molecule is a giant protein, only a fraction of the molecule is extensible [224-226] and involved in the passive elasticity [227, 228], that is, the I-band region of titin beginning around 100 nm away from the center of the Z-disk. Upon stretching, the I-band region of titin will gradually extend, developing passive forces, while other regions of titin appear to be inextensible. This function contributes to the development of restoring forces that maintain sarcomere integrity during muscle stretch and contraction [187]. The titin’s amino acid sequence, determined by Labeit and Kolmerer [187], showed that the elastic I-band of titin has a complex structure, consisting of distinct elements (Figure 1.6), including repeated tandem modular proximal and distal immunoglobulin (Ig) domain regions of variable lengths with multiple, unique, non-repetitive sequences in between that are mostly unstructured, such as PEVK (proline (P), glutamic acid (E), valine (V), and lysine (K)) sequences, an N2A element in all skeletal and some cardiac isoforms, and an N2B element expressed exclusively in cardiac variants [187, 204, 229].  36  Figure 1.6 Molecular structures of muscle and titin. a) Electron micrograph of a cardiac muscle sarcomere. Titin is a giant protein connecting Z-line to M-line in a muscle sarcomere. b) modular structure of the elastic I-band section of human cardiac N2B-titin isoform. Titin is composed of folded domains, such as Ig, and unique sequences, such as PEVK and N2B. Reprinted from reference [204] by permission from Macmillan Publishers Ltd: [Nature], copyright (2002).  Studies show that the I-band titin elasticity is a combination of the elasticity of these structural elements: as stretch is initiated, at a low sarcomere length range where the passive force is very small, the Ig domain regions will extend first by straightening of linker sequences. As extension increases where the passive force increases, unstructured unique sequences such as the PEVK segment will be extended. At even higher forces the extension of 37  random coil sequences in the N2B element will extended. If the extension increases even higher, at extreme sarcomere extensions, some of individual Ig domains will unfold [230, 231]. Specific studies on stretching titin-specific antibody labelled sarcomeres have been performed focusing in particularly on stress relaxation/force decay and force hysteresis, which are two typical features of the viscoelasticity of nonactivated skeletal-muscle sarcomeres. Force hysteresis at higher sarcomere length during a stretch-release cycle (Figure 1.7) means that force during releasing is smaller than the force during stretching. It was shown that force decay exhibited by myofibrils subjected to stretch-release cycles can be reproduced by a simulation, taking into account both the Ig-domain unfolding and the PEVK entropic elasticity, indicating the force decay is explainable by the unfolding of only a very small number of Ig domains per titin molecule [183, 213]. The rate constants of hysteresis recovery of sarcomere compares well with the refolding rates of Ig-like domains. Another feature is force relaxation at a constant sarcomere length (Figure 1.7), which refers to force decreasing at a constant sarcomere length. Monte Carlo simulations showed that by taking into account the entropic elasticity regions and the unfolding characteristics of Ig domains measured in single-molecule experiments, it would be possible to reproduce the force 38  decay in stress relaxation of myofibrils. These findings suggest that the passive elasticity of myofibrils in a physiological (length) range arises from unfolding and refolding of a small number of Ig domains [183].  Figure 1.7 Force-Sarcomere length curve of a cardiac myocyte. A) Force hysteresis (force during releasing is smaller than force during stretching) is observed at higher sarcomere length. Reprinted from reference [213] swith permission from Wolters Kluwer, promotional and commercial use of the material in print, digital or mobile device format is prohibited without the permission from the publisher Lippincott Williams & Wilkins. B) Stretch protocol (bottom) and force response (top) of skeletal myofibrils. At a constant sarcomere length of 2.9 µm, force relaxation is observed. Reprinted from reference [183] with permission from Elsevier.  Single-molecule mechanical studies on isolated single titin molecules also demonstrated that Ig domains could unfold and refold in response to external 39  forces upon stretch-release and thereby contribute to the elasticity of the sarcomere [179, 229]. Single molecule AFM experiments showed that stretching a single titin molecule resulted in saw-tooth like force-extension curves [179]. It has been demonstrated that the passive elasticity of muscles can be explained by scaling up the single molecule AFM result of titin molecules (Figure 1.8) [204].  Figure 1.8 Ig domains unfold/refold contributing to the elasticity of the sarcomere. A) Force extension curves obtained by stretching titin proteins show periodic saw-tooth like features. The saw-tooth pattern is consistent with the sequential unfolding of individual titin domains. Figure from [179]. Reprinted with permission from AAAS. B) Single-molecule data explains the force-extension curve of cardiac muscle. The red line shows the calculated end-to-end length of the I-band titin versus a stretching force. The black line shows the calculated end-to-end length of the I-band titin versus a stretching force without considering unfolding of Ig regions. The black spots plot force-extension measurements from nonactivated rabbit cardiac 40  myofibrils. Reprinted from reference [204] by permission from Macmillan Publishers Ltd: [Nature], copyright (2002).  1.3.3 Titin-mimetic materials The I band region of the giant muscle protein titin largely determines the passive elasticity of muscles, with combinations of excellent mechanical properties, such as high modulus, toughness and resilience [223, 232-234]. This phenomenon of titin conveying the excellent mechanical properties of muscles has been the basis for design of biomimetic materials. For example, Guan et al. reported a new synthetic polymer inspired by the protein titin. Several biomimetic modules have been designed. These modules reversibly unfolded, mimicking the Ig domains in titin, and enhanced mechanical properties after been incorporated into linear polymers. The resulting synthetic polymer showed very interesting mechanical properties with Young’s modulus of around 200 MPa and a maximal strain greater than 100 %. The material could undergo large plastic deformations with small increase in stress. However, even though the plastic deformation was not permanent, it took a long time to recover (>18 h at room temperature) [235, 236].  41  1.3.4 The design of biomaterials to mimic the passive elasticity of muscles using tandem modular titin-mimetic proteins Since the nanomechanical properties of titin plays an important role in determining the macroscopic passive elasticity of muscles, we hypothesize that if we could design a protein to mimick titin, then we combine the proteins together, we would be able to produce a material that can mimick the passive elasticity of muscles. Therefore, in one project of this thesis, using artificial titin mimetic elastomeric proteins as building blocks, we constructed a biomaterial which exhibits mechanical properties mimicking the passive elasticity of muscle at the macroscopic level. We first engineered artificial elastomeric proteins that mimic titin at the molecular level, using well-characterized GB1 domains [206] to mimic the folded Ig domains and the random coil-like protein resilin [122] to mimic the unstructured unique sequences (Figure 1.9). Single molecule AFM results showed that the overall mechanical properties of a single (GB1-Resilin)4 polyprotein largely mimicked those of individual titin molecules. We then cross-linked these titin-like proteins into solid biomaterials using a well established Ru(II)-mediated photochemical cross-linking method [221] . Our results showed that GB1-resilin based biomaterials were elastic and could be stretched to a strain as high as 135 %. The Young’s modulus is ~70 kPa for 42  (GB1-Resilin)4. And the resilience of these materials decreases with the increase in strain. Such mechanical behaviours indicate that these materials behave like shock absorbers at higher strains, mimicking those of muscles very well [182, 234]. In addition, the mechanical properties of these materials could be tuned both by adjusting mechanical properties of individual domains and by adjusting the composition of the proteins. The engineered biomaterials represent a new type of muscle-mimic and we anticipate that they will have applications in material sciences as well as in tissue engineering and other fields in medicine and biology [2]. More detailed results will be described in a later section (Chapter 3).  Figure 1.9 Design of GB1-resilin based protein as titin mimics. A) The protein structure of globular GB1 (PDB: 1PGA) domains and the amino acid sequence of resilin random coil. B) Schematic of one of the GB1-resilin based protein GRG5RG4R. Adapted from reference [237] with permission Macmillan Publishers Limited. All rights reserved. 43  1.4 Natural tandem modular ECM protein-fibronectin type III (FnIII) domain Using a well-developed photochemical cross-linking strategy, we have engineered protein-based biomaterials that exhibit mechanical properties that closely mimic the passive elastic properties of muscles [237]. This new technique also offers the possibility to make use of natural tandem modular ECM protein as building blocks to construct novel hydrogels and explore their use in biological studies and biomedical engineering. Natural ECM proteins have an advantage that a large number of natural ECM protein domains contain cell-binding sequences which can interact with cells. We hypothesize that through incorporating natural ECM protein domains that contain cell-binding sequences, we could design biocompatible hydrogels suitable for cell studies. Therefore, in one project of this study, we endeavor to use the third fibronectin type III (FnIII) domain (TNfn3, PDB code: 1TEN)) from Homo sapiens tenascin-C (TN-C, TNCfn3) as a model system to construct tandem modular protein-based hydrogels. TN-C is an ECM tandem modular protein [238]. TN-C interacts with a variety of ligands and regulates cell-matrix interactions [239]. TNCfn3 is an all-β protein of 90 amino acid residues with a typical immunoglobulin-like  44  β- sandwich structure. As shown in Figure 1.10, two β sheets of TNCfn3 pack against each other with two force- bearing β strands.  Figure 1.10 The structure of the third fibronectin type III domain of tenascin-C (TNCfn3)-a typical β- sandwich structure (PDB: 1TEN). RGD motif is indicated in green.  Single-molecule AFM was previously employed to characterize the mechanical unfolding behavior of TNCfn3. Stretching polyprotein (TNCfn3)8 resulted in force-extension curves with characteristic saw-tooth patterns (Figure 1.11). Each individual peak resulted from the unfolding of each TNCfn3 domain, with an average unfolding force of 125±14 pN (Figure 1.11), at a pulling speed of 400 nm/s.  45  Figure 1.11 Mechanical unfolding behaviors of TNCfn3. A) Force-extension curves of polyprotein (TNfn3)8 with a characteristic saw-tooth pattern and a histogram of unfolding forces of TNfn3. Worm Like Chain Model fitting (red lines) of the consecutive unfolding force peaks measure an average  contour length increment ∆Lc of ∼29.0±0.8 nm, consistent with the  mechanical unfolding of the TNCfn3 domains in the polyprotein chain. B) The unfolding force histogram with an average value of 125± 14 pN (n=4198) at a pulling speed is 400 nm/s. Reprinted from reference [240], Copyright (2009), with permission from Elsevier.  Beside having the capability of force-induced unfolding/refolding reactions, the TNCfn3 domain was chosen because this domain contains the cell-adhesive RGD sequence, and is known to interact with integrins [239, 241]. Our results demonstrated that tandem modular proteins containing 46  TNCfn3 could be readily photocross-linked into elastic hydrogels with Young’s modulus in the range of 20 kPa. In vitro studies showed that none of the components of the photochemical cross-linking reaction were cytotoxic at the level tested, and the hydrogel supported cell spreading of human lung fibroblast cells. More detailed results will be described in a later section (Chapter 4).  1.5 Modular structured Enzyme protein-BCX The cell spreading result indicates that the RGD motif remains functionally active after being cross-linked in hydrogels. This suggests that the ability to make use of tandem modular protein domains as building blocks to construct novel hydrogels also permits design of hydrogels with multiple functionalities, like fluorescence, enzymatic activity and electron conduction and exploration of their applications in biocatalysis [242-244], bioelectrocatalysis [245-247] and biosensing [248-253]. In the case of enzymatic activity, as enzymes are mostly active in an aqueous environment, protein-based hydrogels have attracted attention in enzyme immobilization studies thanks to their high water content. Various methods have been explored for protein-based enzymatic hydrogels [44, 243, 245, 254, 255]. Here we propose a design of a photochemically cross-linked 47  enzymatic hydrogel with xylanase activity based on tandem modular protein to demonstrate the possibility to engineer enzymatic hydrogels. The xylanase we used is one originally from Bacillus circulans, called BCX [256]. BCX has been the subject of much previous study using a variety of techniques including X-ray crystallography, which shows a well-defined modular structure [257-259].  Figure 1.12 Cartoon representation of the protein backbone structure of globular BCX domain. (PDB: 1HV0) [260]  BCX has also been studied by NMR spectroscopy, and mutational analysis, leading to an understanding of enzymatic mechanism, and development of enzymatic activity assaying methods [258, 260-264]. Previous studies in our group also report the characterization of the unfolding/folding kinetics of the BCX protein by single-molecule AFM, with an average ∆Lc of 64±8 nm  48  (average±standard deviation) upon unfolding, and an average mechanical unfolding force of 55 ±30 pN 400 nm/s [265].  Figure 1.13 The force-extension curves of polyprotein containing BCX flanked by (GB1)4 handles. The unfolding of GB1 is indicated in red. The peaks with a ∆Lc of around 64 nm correspond to unfolding of BCX domains. A schematic of the polyprotein (GB1)4-BCX-(GB1)4 is also shown at the top Reprinted from reference [265], Copyright (2010), with permission from Elsevier.  The well established knowledge of BCX has made it a good model enzyme system for examining enzymatic hydrogels. Here, we engineered a recombinant protein fused with folded tandem modular protein domain GB1, random coil-like sequences from resilin and BCX protein domains. The 49  protein retained partial enzymatic activity of non-fusion BCX. Then the protein was cross-linked to hydrogels using a Ru(II)-mediated photocross-linking method. The resultant hydrogel exhibited BCX enzymatic activity. (See details in Chapter 5)  1.6 Fluorescent protein The possibility to design functional hydrogels through incorporation of modular protein domains was also demonstrated with hydrogels constructed from fluorescent protein pairs to be used as force sensors capable of measuring swelling force of a single peptide chain with pN sensitivity. Mechanical force is one of the most influential physical factors in cells [266-268]. However, there are limited biosensors that can measure force in cells with pN sensitivity without causing damage to the cells [269]. Therefore, in one project of this study, the feasibility of design fluorescent hydrogels that can be applied as force sensors was explored. Green fluorescent protein (GFP) is a spontaneously fluorescent protein originally isolated from the bioluminescent jellyfish, Aequorea Victoria, by Shimomura et al [270]. It has a molecular weight of 27 kDa with 238 amino acid residues. High-resolution crystallography (PDB: 1GFL) [271, 272] has shown that structure of GFP is an 11-stranded β-barrel with an α-helix 50  attached with a chromophore buried in the center [271, 273-276]. Wild-type GFP exists as dimers (Figure 1.14) [271].  Figure 1.14 Structure of GFP (PDB:1GFL) [271] shows an 11-stranded β-barrel with an α-helix bearing a chromophore.  Utilization of GFP has become well established. Mutagenesis and engineering of GFP into chimeric proteins provide new tools in research fields such as physiological indicators and photochemical biosensors [277]. Ever since the first cloning and expression of GFP in Escherichia coli and Caenorhabditis elegans by Chalfie et al. in 1994 [278], GFP has been widely studied and explored in biochemistry and cell biology as a marker of gene expression and protein targeting in cells and organisms thanks to its highly visible intrinsic fluorophore. GFP can be genetically concatenated to many other proteins without affecting the biochemical functions of these proteins, 51  meanwhile, the fluorescence from GFP can be used to monitor the localization and fate the proteins of interest [279]. Due to its wide applications, a number of derivatives of GFP have been developed, such as Blue Fluorescent Protein (BFP), Cyan Fluorescent Protein (CFP), Yellow Fluorescent Protein (YFP) and Red Fluorescent Protein (RFP). The most common applications of GFP and its derivatives involve Förster Resonance Energy Transfer (FRET) between fluorescent proteins to show the localization of specific proteins and detect their interactions in living cells. FRET is energy transfer from a donor fluorophore to an acceptor fluorophore when the two fluorophores are in molecular proximity (<10 nm). FRET efficiency (E), the degree of FRET, is distanceand orientation-dependent. The E of singlet energy transfer between chromophores varies with the inverse sixth power of the inter-chromophore distance r, following E =  1 (R0 is the Förster distance, the r 6 1+ ( ) R0  distance at which E=50%.) [280]. Thus, the longer the distance between the donor and the acceptor fluorophores, the lower the E, and thus the lower the FRET generated fluorescence signal of the acceptor fluorophore. Since the E is distance dependent, the FRET strategy can be used to study biological phenomenon involving changes in the distance between the fluorophores 52  [281-283], such as distance change resulting from protease activity [284-286], distance change upon association of two distinct moieties within a polypeptide chain or as a result of dimerization [287], and a conformational changes upon binding small molecule ligands like Ca²+ [288]. A basic design of a FRET experiment involves use of genetically fused fluorescent proteins with a flexible linker in between [289]. Among the FRET pairs, two of the GFP mutants, CFP and YFP, have appropriate fluorescence excitation and emission properties for measurement of short molecular distances, serving as the donor and acceptor fluorophores [289], which have been used in many studies [290, 291]. For one project of this thesis, the FRET pair CFP/YFP was chosen as a model system to test the efficacy of the hydrogel for functional protein immobilization. The CFP/YFP pairs we used incorporated an improved cyan fluorescent protein variant (ECFP/S72A/Y145A/H148D) named Cerulean [292] and a new YFP (GFP/ Thr203Tyr/ Q69M) named Citrine [293].  53  Figure 1.15 Structure of A) CFP (PDB: 2Q57) [292] and B) YFP (PDB:1HUY) [293].  Through the well developed Ru(II) photochemical cross-linking method, a hydrogel was constructed using a tandem modular protein containing the CFP/YFP FRET pair. In this study of the FRET pair embedded hydrogels, swelling effects on single polymer chain length and swelling force on single polymer chain were also investigated. (See details in Chapter 6)  54  CHAPTER 2 Tandem Modular Protein-Based Hydrogels Constructed Using a Novel Two-Component Approach 2.1 Introduction Hydrogels based on engineered proteins have attracted great interest over the last two decades due to their great potential in a wide range of biomedical applications, ranging from drug delivery carriers, synthetic extracellular matrices, tissue engineering scaffolds, as well as biocatalysts [1-7, 9]. Although elastin-mimetic protein and silk-elastin-like proteins have been widely used to engineer protein-based hydrogels [294-297], proteins containing coiled coil motifs, as pioneered by Tirrell and co-workers [6], have proven a versatile building block for constructing protein-based hydrogels [3, 9]. Such proteins typically adopt an ABA/ABC triblock architecture with leucine zipper based coiled coil motifs at both ends and a sequence at the center [6, 155, 176, 208, 209, 298]. Coiled coil is one of the basic folding motifs found in native proteins and consists of two or more α-helices winding together to form a superhelix [299-301]. The gelation of the aqueous solution of block protein ABA/ABC is mediated by the self-assembly/aggregation of the terminal leucine zipper sequences, which form oligomers to serve as the physical cross-linking points [6]. Using recombinant DNA technology, it has become possible to precisely tailor 55  coiled coil-containing triblock proteins at the gene level (i.e., defined amino acid sequence, composition and molecular weight) in order to tune the physical and functional properties of the resultant hydrogels. Since the center block is largely responsible for retaining the water in the hydrogel, various functional motifs have been readily incorporated into the center domain to confer new feats and functionalities to the resultant hydrogels [6, 155, 176, 209, 298]. Inspired by the tandem modular structure of many extracellular matrix proteins [238], we recently have successfully used tandem modular protein (GB1)8, which is made of eight identical tandem repeats of a small protein GB1 [302], as the center block in a triblock protein ABA to construct a novel protein hydrogel [199]. These efforts are not only important for creating a synthetic extracellular matrix that closely mimics naturally occurring ones, but also demonstrate the unique physical properties brought up by the incorporation of tandem modular proteins. However, two limitations of the triblock protein approach have made it difficult to further our efforts in engineering protein hydrogels containing tandem modular proteins for mimicking synthetic extracellular matrix. One is the relatively fast erosion rate due to the formation of intramolecular loops from the same protein chain. The other limitation is the high viscosity of the protein solution, which makes the purification of the protein under native conditions 56  very difficult. Using a triblock protein ABC containing two dissimilar coiled coil sequences in the same gelator protein has overcome the first limitation [209]. This approach makes the purification of proteins under native conditions even more difficult (sometimes even impossible), as tandem modular proteins carrying two dissimilar terminal coiled coil sequences could more easily form intermolecular aggregates inside bacterial cells and the solubility of the protein is very poor. On the other hand, because it is difficult to refold tandem modular proteins after denaturation, purifications of these gelator proteins from denaturing conditions generally gives rise to proteins with compromised functionalities. Therefore, new design strategies must be conceived to overcome these two hurdles to engineer protein hydrogels containing tandem modular proteins. Here we report a novel two-component approach as an alternative method to engineer tandem modular protein-based hydrogels that show improvements in various aspects of the hydrogel properties over the first-generation tandem modular protein-based hydrogels [199].  57  2.2 Materials and Methods 2.2.1 Protein Engineering. The gene encoding protein GB1 was a generous gift from David Baker of the University of Washington. The gene that encodes (GB1)4 and (GB1)5 was constructed as previously reported [199]. The DNA sequences of coiled coil domains CCE and CCK [303], flanked with a 5' BamHI restriction site and 3' BglII and KpnI restriction sites, were synthesized by PCR (polymerase chain reaction)-based oligonucleotide assembly. The resulting sequences of the CCE and CCK peptides are shown in Figure 2.1A. The expression vector pQE80L-CCE-(GB1)4-CCE was constructed by iterative cloning CCE, (GB1)4, and CCE genes into an empty pQE80L vector (with His-tag built in), on the basis of the identity of the sticky ends generated by BamHI and BglII restriction enzymes. The expression vectors of pQE80L-CCK-(GB1)5-CCK-(GB1)5- CCK was constructed in the same way. The expression vector was transformed into Escherichia coli (E. coli) strain DH5α. Cultures were grown at 37 ˚C in 2.5% lysogeny broth (LB) containing 100 mg/L ampicillin, and induced with 1 mM  isopropyl-1-β-D-thiogalactoside (IPTG) when its optical density reached ∼1.  Protein expression continued for 5 h. The cells were harvested by 58  centrifugation at 4000 rpm for 12 min and lysed using lysozyme. The soluble fraction was purified using Co²+ affinity chromatography. The yield of the proteins CCE-(GB1)4-CCE (designated as AG4A) and CCK-(GB1)5-CCK-(GB1)5-CCK (designated as CG5CG5C) was around 60 mg and 30 mg per liter of culture, respectively. The purity of the purified protein is above 90 %, as estimated from sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS-PAGE) using AlphaEaseFC software (Version 4.0.0, Alpha Innotech Corporation, San Leandro, CA).  2.2.2 Circular Dichroism (CD) Spectroscopy. CD spectra were recorded on a Jasco-J815 spectropolarimeter flushed with nitrogen gas. The spectra were recorded in a 0.1 cm path length cuvette at a scan rate of 50 nm/min. The protein samples were dissolved in 0.2×phosphate-buffered saline (PBS, 1×PBS contains 137 mM NaCl, 2.7 mM KCl, 10 mM Na2HPO4 and 2 mM KH2PO4) at pH 7.2 to a final concentration of 0.1 mg/ml. Data have been corrected for buffer contributions. For each protein sample, the CD signal was converted into mean residue ellipticity (MRE) using the following equation: θMRE = (100×θobs)/[dC (n-1)], where θobs is the observed ellipticity (in deg), d is the path length (in cm), C is the concentration of protein samples (in M), and n 59  is total number of amino acids in the protein. For thermal melting measurements, the θobs at 222 nm was monitored when the temperature was increased from 15 to 90 ˚C with a rate of 3 ˚C/min. The θMRE was calculated for graphical presentation.  2.2.3 Hydrogel Preparation. The purified protein was dialyzed against deionized water for 2 days to remove all the salt from elution buffer. During dialysis, the water was changed every 5 h. The protein was then lyophilized after dialysis. The hydrogel was made by redissolving the protein sample into PBS buffer. Vigorously mixing helps dissolution of proteins. The trapped air bubbles can be removed by fast spinning at 2500 rpm for 30 min. The fast spinning also helps flatten the hydrogel surface.  2.2.4 Rheology Rheological measurements were carried out on a Haake RheoStrss 6000 rheometer at 20 ℃. The samples were oil-sealed to prevent evaporation. The measuring system consisted of a cone and plate sensor with a diameter of 20 mm and cone angle of 1°. Frequency sweep tests were performed over a  60  range of frequencies from 1 to 100 rad sˉ¹ at 1% constant strain. All reported G’ and G” values represent an average of at least three independent samples.  2.2.5 Atomic Force Microscopy (AFM). About 5 µL of the protein hydrogel was placed on the surface of a newly cleaved mica surface and allowed to adsorb for 10 min. Then the mica plate with the hydrogel sample was shock-frozen using liquid nitrogen. Then, the surface of the mica plate was flushed with nitrogen gas to remove unattached proteins before analysis. The images were taken at room temperature using a NanoWizard II AFM (JPK, Germany) operating in intermittent contact mode (conditions: scan rate, 1Hz; cantilever, AC160 from Olympus, Japan; number of pixels, 512×512).  2.2.6 Erosion Rate Measurement. The erosion rate of hydrogel was measured using a method similar to that previous reported [199]. AG4A and CG5CG5C proteins were dissolved in PBS buffer to a concentration of 7 %. The two protein solutions are mixed so that the molar ratio of CCE and CCK is 1:1. One hundred milligrams of hydrogel was transferred into a cylindrical glass tube with a flat bottom (1.05 cm diameter). The glass tube with the hydrogel was then centrifuged at 2500 61  rpm for 30 min to completely flatten the hydrogel sample to the bottom and smooth the surface of the hydrogel. The hydrogel was allowed to stand overnight. Then the thin gel film together with the glass tube was soaked in 5 ml of PBS, pH 7.2, in a scintillation vial. The whole setup was placed on a compact rocker (FINEPCR) tilting at 50 rpm with an amplitude of 9˚ at room temperature. The erosion profiles were determined by measuring the protein absorbance at 280 nm of the supernatant at successive time points using a Nanodrop ultraviolet-visible spectrophotometer. Two different samples were measured, and the average value was reported.  2.2.7 Measurement of the Degree of Hydration. The hydrogel (weight ranging from 60 mg to 200 mg) was soaked in 1.0 ml of PBS (pH= 7.2) for 30 min at room temperature, and was allowed to sit on a filter paper to drain the excess PBS. The hydrogel was weighed to obtain the swollen weight (Ws). Then the hydrogel was lyophilized and weighed again to obtain the dry weight (Wd). The degree of hydration of the hydrogel was calculated as [(Ws-Wd)/ Ws]×100%.  62  2.3 Results 2.3.1 The Two-Component Methodology to Construct Protein Hydrogels. The new method we developed is a two-component approach, which uses two different triblock proteins to form the hydrogel. The two triblock proteins carry two different and complementary leucine zipper coiled coil sequences CCE and CCK to drive the gelation process. CCE and CCK are engineered coiled coil motifs, which were designed by Kopecek and co-workers [303], and have been used to engineer polymer/protein hybrid hydrogels [303-306]. The sequences of CCE and CCK are shown in Figure 2.1A. CCE and CCK have opposite charges at neutral pH: CCE is negatively charged, while CCK is positively charged under the same condition. Due to electrostatic repulsion, CCE or CCK cannot form homo coiled coils; instead, they can only form antiparallel heterodimers between CCE and CCK (Figure 2.1A). The two components we engineered to form the hydrogel are CCE-(GB1)4-CCE (designated as AG4A) and CCK-(GB1)5-CCK-(GB1)5-CCK (designated as CG5CG5C), where G4 and G5 are tandem modular proteins containing four and five tandem repeats of GB1 and are to mimic tandem modular extracellular matrix proteins; A represents CCE, and C represents CCK (Figure 2.1B). Both CCE and CCK 63  are designed to provide intermolecular cross-linking through the formation of antiparallel heterodimers between CCE and CCK. Tandem modular proteins made of GB1 have superior solubility in aqueous solution and are used to prevent precipitation of the tandem modular protein chain and retain water in the resultant hydrogel. AG4A is a bifunctional protein, while CG5CG5C is a trifunctional one. Thus, a mixture of AG4A and CG5CG5C with equal-molar functional groups A and C should be able to cross-link both proteins into a physical network through the self-association of coiled coil sequences A and C. There are a few considerations in the design to improve the properties of the resultant hydrogels against the first-generation gelator protein A-(GB1)8-A we constructed before [199]. First, since CCK and CCE are positively (or negatively) charged, they cannot aggregate on their own, making it possible to express and purify AG4A and CG5CG5C with high yield under native conditions. Second, the two proteins are bifunctional and trifunctional, respectively, which facilitate the formation of physically cross-linked inter-molecular networks. Third, the number of repeats of GB1 between coiled coil sequences in the two proteins is different, preventing the formation of protein dimers between AG4A and CG5CG5C so that the coiled coil sequences can be fully utilized for cross-linking. Using standard molecular biology techniques, we expressed and purified the 64  bifunctional AG4A and trifunctional protein CG5CG5C in E. coli under  native conditions. The yield for AG4A and CG5CG5C are ∼60 mg and 30 mg  per liter of LB culture, respectively. The viscosity of both protein aqueous solutions was low and did not pose any adverse effect on the purification of both proteins under native conditions. Figure 2.1C shows the SDS-PAGE  picture of both purified proteins. The apparent molecular weight is ∼34 kDa  for AG4A and ∼78 kDa for CG5CG5C, in close agreement with the  theoretical molecular weights.  65  Figure 2.1. A) The amino acid sequences of the CCE and CCK peptides and the schematics of the heterodimeric CCE/CCK coiled coil. B) Schematics of the two tandem modular proteins AG4A and CG5CG5C used to construct the protein hydrogel. Green and blue coils represent the A (CCE) and C (CCK) domains under neutral pH, respectively; red globules represent the folded GB1 domains. C) 12% SDS-PAGE picture of purified AG4A and CG5CG5C. Lane 1: protein molecular weight marker; Lane 2: AG4A; Lane 3: CG5CG5C.  2.3.2 Far-UV CD Spectroscopy. To confirm that CCE and CCK can indeed form a coiled coil in the presence of GB1 domains, we carried out far-UV CD spectroscopy experiments. Since 66  the CD signal of AG4A and CG5CG5C is dominated by that of GB1, we constructed GB1-CCE and GB1-CCK to confirm that CCE and CCK can form a coiled coil in the presence of folded GB1 domains. The CD spectra of GB1-CCE and GB1-CCK show two broad negative minima at 208 and 222 nm, which resulted from the α+β structure of the folded GB1 domains (Figure 2.2) [199, 307]. The CD spectrum of 1:1 mixture of GB1-CCE and GB1-CCK shows a clear increase in the MRE of the two bands at 208 nm and 222 nm, suggesting that GB1-CCE and GB1-CCK indeed form a coiled coil structure, leading to the increase in CD bands (208 and 222 nm) characterized by α-helix structures. Furthermore, a dilute aqueous solution of GB1-CCE/ GB1-CCK mixture (1:1 molar ratio) exhibits two distinct thermal unfolding transitions (Figure 2.3A), as probed by the ellipticity at 222 nm. The first transition occurs at Tm (temperature of the transition midpoint) of 38 ˚C and corresponds to the thermal dissociation of the coiled coil heterodimer formed by CCE and CCK [303]; the second transition occurs at Tm of 76 ˚C and corresponds to the thermal denaturation of the folded GB1 domains [199, 307]. Having established that CCE and CCK can indeed associate to form heterocoiled coils in the presence of folded GB1 domains, we then characterized the dilute solution of AG4A/CG5CG5C (1:1 molar ratio of the leucine zipper sequences A and C) using temperature-dependent CD 67  spectroscopy (Figure 2.3). Two thermal unfolding transitions were observed: a sharp one at 76 ˚C, which corresponds to the unfolding of folded GB1  domains, and a broad transition at ∼38 ˚C, which can be attributed to the  thermal unfolding of CCE/CCK coiled coils. This broad melting transition of CCE/CCK is consistent with the behaviour observed for polymers cross-linked by CCE/CCK sequences [303], suggesting that the cooperativity of the thermal melting of CCE/CCK is reduced upon formation of cross-links between AG4A and CG5CG5C.  Figure 2.2 CD spectra of GB1-CCK (red curve), GB1-CCE (blue curve), and a 1:1 mixture of GB1-CCK and GB1-CCE (black curve) in 0.2×PBS, pH 7.2.  68  Figure 2.3 Thermal melting behaviours of a GB1-CCE/GB1-CCK mixture A) and AG4A/CG5CG5C B) in 0.2×PBS at pH7.2 monitored by using the ellipticity at 222 nm. In both mixtures, the molar ratio of the functional groups A (CCE) and C (CCK) is 1:1. Two transitions were observed in both  mixtures at similar melting temperatures: one is at ∼38 ˚C, corresponding to  the thermal dissociation of CCE/CCK coiled coil, and the other one is at ∼76  ˚C, which corresponds to the thermal melting of GB1. It is evident that the cooperativity of the thermal melting transition of CCE/CCK in the mixture of AG4A/CG5CG5C is reduced as compared with that of the GB1-CCE/GB1-CCK mixture. In panel A, red and blue lines correspond to sigmoid fits to the first and second transitions, respectively.  69  2.3.3 Protein Hydrogels Self-Assembled from CCE/CCK Coiled Coils. The CD results suggest that the CCE/CCK sequence can self-associate to form coiled coils at neutral pH, giving rise to the possibility of constructing protein hydrogels using AG4A and CG5CG5C. Indeed, a 7 % (w/w) aqueous solution of AG4A and CG5CG5C mixture in PBS buffer (pH 7.2) readily forms a transparent hydrogel (Figure 2.4). The resultant gel can hang at the bottom of the vial without flowing down for days. In comparison, a 7 % aqueous solution of pure AG4A or CG5CG5C in PBS (pH 7.2) is a clear transparent solution (Figure 2.4). This result indicates that the AG4A or CG5CG5C does not self-associate, and the gelation of the AG4A and CG5CG5C mixture is due to the formation of intermolecular coiled coils by the CCE and CCK sequences. In addition, since the association and dissociation of CCE/CCK coiled coils are temperature dependent, we anticipate that the two-component hydrogel formation is also dependent upon temperature. Indeed, the formation of such two-component hydrogels is fully reversible as a function of temperature. When the temperature is increased to 40 ˚C, the hydrogel readily dissolves into viscous solution (Figure 2.4B). These results clearly confirm that this two-component hydrogel formation is mediated by the formation of physical cross-links between CCE/CCK coiled coil sequences. By varying the concentration of 70  proteins at pH 7.2, we also estimated the gelation point (Figure 2.4C). We found that the two-component protein solution can form a hydrogel at a concentration as low as 3.5 %. Given the high molecular weight of the two proteins, such a gelation concentration is quite low, indicating the efficiency of the two proteins to form hydrogels. These results clearly demonstrate the feasibility of engineering protein-based hydrogels using this novel two-component approach. A schematic of the formation of AG4A and CG5CG5C hydrogel network at the molecular level is shown in Figure 2.5, where the physical cross-linking is formed via the formation of coiled coils by the CCE and CCK sequences.  71  Figure 2.4 (Top) 7 % aqueous solution of AG4A and CG5CG5C mixture forms protein hydrogel at neutral pH, while 7 % AG4A or CG5CG5C solution alone does not form hydrogels. Photographs of each sample are shown. (Middle) Hydrogel formation is temperature-dependent. Photographs of the hydrogel at room temperature and 40 ˚C are shown. (Bottom) Mixture of AG4A and CG5CG5C aqueous solution can form hydrogel at a concentration as low as 3.5 %. Photographs of AG4A/CG5CG5C mixture at different concentrations are shown.  72  Figure 2.5 Schematic drawing of the proposed gel network of the AG4A and CG5CG5C hydrogel. Hydrogels are formed through the self-association of the formation of a coiled coil from CCE and CCK sequences (green and blue helices).  2.3.4 Rheology After confirmation of hydrogel formation, it is critical to study the process of self-assembly of hydrogels. An important phenomenon to study is the kinetics of self-assembly and viscoelastic properties of the resultant hydrogels. Rheology can be a good start and may provide preliminary insight into the details of hydrogel structure. Therefore, rheology measurements were performed to evaluate the self-assembly of AG4A and CG5CG5C hydrogel.  73  1000 G' G''  G', G'' (Pa)  100 10 1 0.1 2  0.1  4  6 8  2  4  6 8  1 10 Frequency (rad/s)  2  4  6 8  100  Figure 2.6. Rheology measurement. Dynamic modulus in a frequency range of 0.1-100 rad sˉ¹ for 7% w/v AG4A and CG5CG5C hydrogel in PBS buffer, pH 7.2, at room temperature. Diamonds represent storage moduli G' (Pa); squares represent loss moduli G'' (Pa).  As shown in Figure 2.6, the moduli of the hydrogel are a little low compared with those of previously reported AC10A hydrogel [208]. This might be caused by the intrinsic differences between C10 and G proteins. However, the storage modulus G' increases as the frequency increases in the range of 0.1-100 rad sˉ¹. This behavior is different from that of previous reported AC10A hydrogel of which the storage modulus G' reaches a plateau after a frequency of 0.1 rad sˉ¹[208]. This different response to frequence may potentially be used in applications where storage modulus is required to  74  increase at high frequency range, such as for injectable hydrogel [103, 164, 255, 308-312].  2.3.5 Morphology of Shock-Frozen Two-Component Protein Hydrogels. The morphology of the freeze-dried two-component protein hydrogel self-assembled from 7 % AG4A and CG5CG5C solution was characterized using AFM. As shown in Figure 2.7, the hydrogel shows an interconnected porous network structure, a sponge-like morphology. The pore size shows a big variation, ranging from tens of nanometers to a few hundred nanometers. These results suggest that the physical cross-links mediated by CCE/CCK coiled coils lead to the formation of an interconnected network structure. Due to the fast association/dissociation rate of CCE/CCK, the physical cross-linking in the gel is not permanent but in fast equilibrium. Although this feature is beneficial for the prompt formation of hydrogel upon mixing two components, this may also make the resultant hydrogels prone to erosion.  75  Figure 2.7 Surface morphology of the shock-frozen hydrogel made of 7 % AG4A and CG5CG5C aqueous solution (the molar ratio of A:C is 1:1).  2.3.5 Swelling and Erosion Properties. The degree of hydration for the 7 % two-component hydrogel is 93.0 % ± 0.2 % (n =6) after being equilibrated in PBS for 30 min. To measure the erosion profile of the two-component hydrogel formed by AG4A and CG5CG5C, we monitored the mass loss as a function of time in an open aqueous environment (Figure 2.8). The erosion of this two-component hydrogel shows a linear mass loss versus time profile (Figure 2.8), indicating that the erosion occurs at the surface of the hydrogel [199, 208]. The erosion of this hydrogel is very slow: it takes about 100 h for the hydrogel to completely dissolve in PBS buffer, giving rise to an erosion rate of  76  1.34×10ˉ³mg cmˉ²minˉ¹, which is about a 3-fold improvement over that for AG8A hydrogel we reported before [199].  Figure 2.8 Erosion profile of 100 mg 7 % AG4A and CG5CG5C hydrogel with a surface area of 0.865 cm² at room temperature in PBS buffer pH 7.2. A linear regression (solid line) measures an erosion rate of 1.34× 10ˉ³mg cmˉ²minˉ¹.  2.4 Discussions and Conclusions Designing protein hydrogels with new functions and improved properties is of imperative importance for realizing the great potential of protein hydrogels in the field of biomaterials and biomedical engineering. The use of coiled coil sequences to drive the assembly of block protein-based hydrogels has become an important avenue in this area and has enabled the construction of protein hydrogels with various functions and physical properties. Triblock protein, ABA or ABC (where A and C are coiled coil 77  sequences and can self-associate to drive the gelation process), has been the standard approach [6, 155, 176, 208, 209, 298]. In this approach, the use of triblock proteins to form hydrogels relies on the efficient oligomerization of the coiled coils. In our previous work, we have incorporated tandem modular proteins (GB1)8 into the triblock protein as the center block to construct tandem modular protein-based hydrogels [199]. However, efficient oligomerization and low gelation point of our triblock tandem modular proteins resulted in significant increase of the viscosity of the protein solution, making the purification of proteins under native conditions very difficult. In this work, we have developed a novel two-component methodology to engineer tandem modular protein-based hydrogels. This methodology utilizes two dissimilar yet complementary coiled coil sequences as the gelation motif, and the gelation is accomplished by the mixing of a bifunctional protein AG4A and a trifunctional block protein CG5CG5C. This new method effectively obviates the problem we encountered in expression and purification of high molecular weight tandem modular proteins carrying coiled coil sequences for constructing hydrogels. Since CCE and CCK sequences do not self-associate, two proteins AG4A and CG5CG5C were readily expressed and purified at high yield under native conditions. The 78  resultant protein hydrogel shows much improved erosion properties relative to previously engineered tandem modular protein based-hydrogels. Therefore, this new method will open the possibility to systematically explore large fragments of native extracellular matrix proteins, such as fibronectin and tenascin, for the construction of extracellular matrix mimicking hydrogels. Furthermore, this method allows for the use of abundant heterodimeric coiled coil sequences, both naturally occurring and de novo designed ones, for the engineering of protein-based hydrogels, thus significantly expanding the toolbox of protein building blocks for hydrogel formation. Therefore, this novel approach reported here complements the standard triblock approach and will offer some new possibilities to fine-tune the topology and physical properties of the protein hydrogels via genetic engineering.  79  CHAPTER 3 Biomaterials Designed to Mimic the Passive Elasticity of Muscles based on Tandem Modular Titin-mimetic Proteins 3.1 Introduction Titin is a giant muscle protein, connecting the Z-disk to the M-line in a muscle sarcomere. The I-band part of titin is composed of a series of individually folded immunoglobulin-like (Ig) domains and largely unstructured unique sequences, like N2B and PEVK sequences [181, 204]. These different domains have different mechanical properties. The folded immunoglobulin domains have higher persistence length than the unstructured sequences and will extend first during stretching. Upon further stretching, the unstructured sequences will be straightened. When the stretching force reaches even higher, a small number of folded immunoglobulin domains will unfold, dissipating energy to prevent damage due to overstretching [183]. When these features are combined together, they will provide the desired passive elastic properties of muscle at the macroscopic level [181, 232, 233, 313], which are a unique combination of strength with a Young’s modulus close to 100 kPa, extensibility, resilience, increasing energy dissipation at higher sarcomere length, and stress relaxation at a constant strain [182, 183, 234]. Single-molecule atomic force 80  microscopy (AFM) studies have demonstrated that the macroscopic mechanical properties of muscles are directly determined by the nanomechanical properties of titin, and suggest a potential method to obtain mechanical features at the macroscopic level by engineering features at the molecular level. This possibility inspired us to design biomaterials mimicking the passive elasticity of muscle based on artificial titin-like elastomeric proteins. Towards this goal, we first engineered artificial elastomeric proteins to mimic the molecular architecture and mechanical properties of titin at the molecular level. Considering the structure of titin, folded globular proteins and unstructured sequences are necessary in order to engineer an artificial titin-like elastomeric protein. Single molecule AFM experiments demonstrated that the protein GB1 (PDB: 1PGA), a small globular protein, has mechanical properties that are comparable to those of Ig domains from the natural elastomeric protein titin [206]. And it was shown that a resilin like protein- rec-resilin (R), a largely unstructured protein [122], can be cast into rubber-like biomaterials by rapid photochemical cross-linking. Moreover, previous tensile tests show that R is highly elastic and resilient. Therefore, we used the well-characterized GB1 domain (G) to mimic the folded Ig domains, and R sequence to mimic unstructured sequences. Based 81  on these two building blocks, we constructed artificial elastomeric proteins. Our single molecule AFM results showed that these engineered artificial elastomeric proteins show nanomechanical properties similar to those of titin. Then we used a [Ru(bpy)3]2+-mediated photochemical cross-linking strategy [122], which has been used to cross-link rec-resilin to rubber-like materials, to construct GB1-resilin-based biomaterials. Our results show that the designed materials serve as high resilience rubber-like materials at low strain and shock absorber-like materials at high strain. These properties are comparable to those of muscle within the physiological range of sarcomere length [182] and thus these materials represent a novel type of muscle-mimetic materials. An advantage of our study is that we incorporated mechanically resistant globular domains in artificial elastomeric proteins which makes it possible to investigate how nanomechanical properties engineered into individual protein molecules can be translated into macroscopic properties in materials by tuning the mechanical strength of proteins. Here we investigated this issue in several ways, including affecting the folded state of GB1 domains using urea, adjusting the composition of elastomeric polyproteins and varying the protein concentration and cross-linker APS concentration.  82  Through this study, we demonstrated that we successfully built materials mimicking the passive elasticity of muscles using titin-mimetic proteins and the mechanical properties of these biomaterials can be tuned to mimick different types of muscles. We anticipate that these biomaterials will find applications in tissue engineering [2].  3.2 Methods and materials 3.2.1 AFM Single molecule AFM experiments were performed on a custom-designed AFM following procedures described elsewhere [206]. All of the force-extension measurements were carried out in PBS buffer, unless noted otherwise. In a typical experiment, ~1µl polyprotein sample was deposited onto a clean glass cover slip covered by PBS buffer (50 µl) and was allowed to adsorb for approximately 5 min before force-extension measurements were taken. The spring constant of each individual cantilever (Si3N4 cantilevers from Vecco, with a typical spring constant of 40 pN/nm) was calibrated in solution using the equipartition theorem before each experiment. And the experiments were performed at a typical pulling speed of 400 nm/s unless otherwise noted.  83  3.2.2 Protein Engineering GB1-resilin-based polyprotein genes were constructed using standard molecular biology techniques following a stepwise construction scheme [314]. The gene encoding protein GB1 was a generous gift from David Baker of University of Washington. And one 15 amino acid consensus resilin repetitive sequence (GGRPSDSYGAPGGGN) from the first exon of the Drosophila melanogaster gene CG15920 was used. The DNA sequence of R, flanked with a 5’ BamHI restriction site and 3’ BglII and KpnI restriction sites, was synthesized by polymerase chain reaction (PCR) based oligonucleotide assembly. The expression vector of pQE80L-(GR)4 was constructed by iterative cloning of G and R genes into empty pQE80L vector (with His-tag built in), on the basis of the identity of the sticky ends generated by BamHI and BglII restriction enzymes. GRG5RG4R, GRG5R, GRG9R and G8 were constructed in the same way. The resultant polyproteins carry two additional cysteines at their C-terminus. The expression of polyproteins was carried out in Escherichia coli strain DH5α. Cultures were grown at 37 °C in 2.5 % LB containing 100 mg/L ampicillin, induced with 1.0 mM isopropyl-1-β-D-thiogalactoside (IPTG) when their optical density reached ~1, and allowed to express for 5 hours. The cells were pelleted by centrifugation at 4000 rpm for 10 min and cell lysis was done using 84  lysozyme from egg white (100 mM, SigmaAldrich). The soluble fraction was purified using Co²+ affinity chromatography. The yield of the polyproteins is between 40 mg to 50 mg per liter of culture. The purified proteins were then used in single-molecule atomic force microscopy (AFM) experiments. To prepare GB1-resilin-based biomaterials, the purified proteins were dialyzed against deionized water for 3 days (water was changed every 5 hours). Then, the protein was lyophilized.  3.2.3 Far ultraviolet circular dichroism (CD) spectrum The far-UV CD measurements were carried out on a Jasco-J810 spectropolarimeter flushed with nitrogen gas, with a scan rate of 20 nm/ min. 0.2-cm path length cuvettes were used as the sample containers. The data are expressed as mean residue ellipticity (MRE), an average of three scans. Before carrying out the experiments, the hybrid proteins were dialyzed against deionized water to get rid of imidazole. The protein samples were prepared in PBS at pH 7.4. And the data are corrected for buffer contributions. The CD signal was converted into mean MRE using the following equation: θMRE = (100·θobs)/ [dC (n-1)], where θobs is the observed ellipticity (in deg), d is path length (in cm), C is concentration of protein samples (in M), and n is total number of amino acids in the protein. 85  3.2.4 Cross-linking method The photo cross-linking method is a method that uses a photo-generated oxidant to mediate cross-linking of proteins. The Ru(II) containing complex, tris-bipyridylruthenium(II) cation [Ru(bpy)3]2+, is a photo sensitive molecule, with a maximum absorbance at 452 nm. In the Ru(II)-mediated photo cross-linking reaction, [Ru(bpy)3]2+ is exposed under visible light in the presence of both ammonium persulfate and the protein. Photolysis of the Ru(II) containing complex will oxidize the Ru(II) to Ru(III) and generate an electron. The electron will be donated to an electron acceptor, persulfate, producing sulfate radical, and sulfate anion, followed by cross-linking of proteins through coupling of the tyrosyl radical present in the proteins or other possible mechanisms [221].  Scheme 3.1 The expected mechanism of Ru(II)-mediated photo-initiated protein cross-linking reaction. [Ru(bpy)3]²+ is photolyzed in the presence of a persulfate and the reaction would proceed to form cross-linked products between two tyrosine residues, that is, di-tyrosine [221].  86  Due to its fast reaction rate (less than 1 second) and high yields of cross-linked products, this method has been widely used for protein cross-linking [221]. Elvin et al. utilized the Ru(II)-mediated photo cross-linking method for resilin studies, and successfully cross-linked recombinant resilins that contain tyrosine residues into solid materials. In this study, resilin-like proteins are introduced to use the Ru(II)-mediated photo cross-linking method for hydrogel fabrication. In our study, hydrogels were cast into ring-shaped samples for tensile testing to minimize difficulties that arise from gripping soft dogbone-shaped specimens in ASTM (American Society for Testing and Materials) standard tensile tests. Because gripped material would thin substantially upon stretching, the material would fail at the grips where it was tightly clamped. The dogbone shapes will cause problems and limit our ability to adequately describe the mechanical properties of the materials studied here. Therefore, we followed previously published methods for testing arterial elastin rings to avoid these problems [315]. Unless otherwise noted, hydrogels were cast in a custom-made plexiglass mold with inner diameter 8 mm, outer diameter 10 mm and height 3 mm (Scheme 3.2).  87  Scheme 3.2 Photos of a custom-made plexiglass mold for ring shaped hydrogel fabrication.  To prepare GB1-resilin-based biomaterials, proteins were dissolved in PBS to 200 mg/ml. Ammonium persulfate (APS) and [Ru(bpy)3]²+ were then added to a final solution containing ~200 µM [Ru(bpy)3]²+ and 50 mM APS in PBS buffer. The solution was cast into the custom-made plexiglass mold described above. The sample was then irradiated for 10 minutes using a 200 W fiber optical white light source. The source of irradiation was 10 cm away from the mold. The ring was then taken out from the mold and stored in PBS buffer with 0.05 % (w/v) sodium azide at 4 ˚C.  88  3.2.5 Fluorescence (di-tyrosine) Di-tyrosine emits blue fluorescence under ultraviolet (UV). The absorption and emission maxima are 315 and 409 nm, respectively. Pictures of rings were taken under both white light and UV light.  3.2.6 Anisotropy measurement To determine the anisotropy of the hydrogel, polarized light microscopy is used to measure the birefringence of the biomaterials. Cubic samples of the biomaterials with dimension of 5mm×5mm×1mm (thickness) were viewed between crossed polarizers on a Leitz Orthoplan polarizing microscope with a 10× achromaticpol lens. The retardation was determined at 546 nm using a Leitz Senarmont compensator and a 546 nm interference filter. The rotational angle between the maximum brightness and darkness was determined and the retardation caused by the sample is calculated as 3.03 nm per degree of the analyzer angle. Birefringence was then calculated by dividing the retardation of the cube by its thickness. The smallest birefringence that can be resolved is ~3×10ˉ6.  89  3.2.7 Tensile test Tensile tests were performed using an Instron-5500R tensometer with a custom-made force gage described elsewhere [316]. Unless otherwise noted, these tests were done at a constant temperature of 22 ˚C in PBS buffer and strain rate at 25 mm/min. Tensile tests were conducted and results were analyzed following the principles described below. Tensile test is a type of test to characterize mechanical properties of materials, including Young's modulus, Poisson's ratio, tensile strength and maximum elongation. During a typical test, a material is extended/compressed by uni-axial force, the material will response to the applied force through generating a restoring force. The extension is recorded against the generated force. The recorded extension is used to calculate the tensile strain, ε, which is defined as  ε=  L − Linitial ∆L = (Equation 3.1), Linitial Linitial  where ∆L is the change in material length, Linitial is the initial length, and L is the final length. The recorded force is used to calculate the tensile stress, σ, which is defined as  σ=  F (Equation 3.2), A 90  where F is the force and A is the cross-section area of the material under force. Plotting ε and σ will give rise to a stress-strain curve as typically reported from a tensile test. The slope of the elastic (initial, linear) portion of a stress-strain curve in which range Hooke's Law holds is defined as Young's modulus (E), which is used to quantify the stiffness of an elastic material. It can be calculated by dividing the tensile stress by the tensile strain in the elastic portion of the stress-strain curve.  E=  stress σ = (Equation 3.3) strain ε  The testing machine we used is an Instron-5500R tensometer with a custom-made force gage described elsewhere [316]. This machine has two main parts. One part, the strain gauge, is adjusted to accurately and precisely control the length of the test specimen, the other part, the force transducer, is driven to apply force to the specimen at required speed. In a typical stretch-release test, the test sample is placed in the testing machine, first extended, and then released at a certain constant speed. In a stress relaxation test, the sample is extended at very fast speed to certain stress-strain, held at that particular stress-strain for a period of time, and then released back to initial length or with certain residual stress.  91  In our stress-strain testing, the sample is hung with two bars (Scheme 3.3). One bar is fixed and connected to force transducer, the other one is connected to strain gauge controlling the extension (distance between two bars, Ex). We start testing from zero extension, which means the sample has the circular shape. The stress (σ) equals to force (F) divided by the cross section area (A) of the ring sample as shown in Equation 3.2 σ =  F . In A  order to calculate the cross section area, we measured the width and the thickness of the ring. It is of note that the force is generated by both sides of the ring when the tensile test is performed in this way. Thus, the cross section area equals two times the product of the width (w) and the thickness (t), that is,  A = 2 wt (Equation 3.4). Therefore, stress can be calculated in the equation,  σ=  F F = (Equation 3.5). A 2 wt  In order to calculate strain, we need to define a parameter, dL, which is the extension at which we started to deform the sample. The dL is measured from force-extension curves and defined as x-intercept of regression through first linear region of force-extension curves. After we measured dL, the change in length is calculated using change in length,  ∆L = 2( E x − dL) (Equation 3.6). 92  In addition, we need to measure the thickness of the ring (t) and the radius of the bars (r) which are used to hold the sample. Then we calculated the initial length at which we started to deform the sample using,  t Linitial = 2π (r + ) + 2dL (Equation 3.7). 2 Finally, strain is calculated using the following equation,  ε=  ∆L = Linitial  2( E x − dL) (Equation 3.8). t 2π (r + ) + 2dL 2  Scheme 3.3 Schematic of tensile test. The sample is hanging with two bars (radius=1.5 mm). One bar is fixed and connected to force transducer, the other one is connected to strain gauge controlling the extension (distance between two bars). We start testing from zero extension, which means the sample has the circular shape.  93  3.2.8 Monte Carlo simulations on the force-relaxation of GRG5RG4R Monte Carlo simulations on the force-relaxation of GRG5RG4R protein were carried out according to published protocols [317]. (GRG5RG4R)6 was used to mimic the polyprotein in a hydrogel network. During the simulations, the polyprotein was quickly stretched to a given extension and then kept constant. The stretching force was recorded as a function of time. The unfolding of GB1 domains was described using the Bell-Evans model [318, 319]  α ( F ) = α 0 ⋅ exp(  F∆xu ) (Equation 3.9), k BT  where α(F) is the unfolding rate constant at a stretching force F, α0 is the intrinsic unfolding rate constant at zero force, ∆xu is the unfolding distance and kBT is the thermal energy. The force-extension relationship of polyproteins was described using the WLC model of polymer elasticity, with a persistence length of resilin as 0.5 nm and a persistence length for the folded GB1 domains as 10 nm. We simulated the force relaxation of GRG5RG4R at constant extensions and measured the extension-dependence of the relaxation rate with an unfolding rate constant of 1x10ˉ4 sˉ¹, an unfolding distance of 0.2 nm and a folding rate constant of 300 sˉ¹ for GB1.  94  3.3 Results 3.3.1 The design of titin-mimic elastomeric proteins Based on the two building blocks-GB1 and resilin-like proteins, we constructed artificial proteins (GR)4 and GRG5RG4R to serve as miniature titin-like elastomeric proteins. Figure 3.1 shows Coomassie blue stained SDS-PAGE picture for the constructed proteins (GR)4 (lane 2) and GRG5RG4R (lane 3).  Figure 3.1 Coomassie blue stained PAGE gel for polyproteins (GR)4, GRG5RG4R, GRG5R, GRG9R and G8 (left to right). The first lane is the broad range molecular weight marker (New England Biolabs).  We first characterized the nanomechanical properties of both polyproteins at the single-molecule level using the AFM techniques [179, 314]. Stretching 95  (GR)4 and GRG5RG4R yields force-extension relationships with a characteristic saw-tooth pattern (Figure 3.2). Due to the dimerization of (GR)4 and GRG5RG4R via the oxidation of C terminal cysteine residues, the force-extension curves may show as many as 8 GB1 unfolding peaks in (GR)4 and 20 in GRG5RG4R. These force peaks are characterized by a contour length increment of ~18 nm and unfolding force of ~180 pN. According to previous single molecule AFM studies [191, 206], this character is a classical “fingerprint” of the GB1 protein. Thus we attributed these peaks to the mechanical unfolding of GB1 domains. However, there were no other characteristic peaks, indicating that R sequence does not show measurable mechanical stability, and that R sequence behaves largely as an entropic spring. One thing to note is that there are spacers before GB1 unfolding peaks, which corresponds to stretching GB1 domain and R sequence. By fitting the force extension curves leading up to unfolding peaks with a Worm-like-chain model (Figure 3.2 grey lines), we measured the unfolding force for GB1 191±42 pN in (GR)4 and 180±41 pN in GRG5RG4R and a persistence length (p) of 0.49±0.09 nm in both polyproteins (Figure 3.3). Since GB1 domains are much more rigid than the resilin sequence, the p we measured is largely due to the resilin sequence, and is comparable to that of random coil-like sequences PEVK in titin [204, 320, 321]. 96  Figure 3.2 GB1-resilin-based polyproteins exhibit mechanical properties that are similar to those of titin at the single molecule level. Stretching (GR)4 and GRG5RG4R yields force-extension relationships with a characteristic saw-tooth pattern. This saw-tooth pattern is consistent with consecutive unfolding of GB1 domains. Grey lines are worm-like chain model fits to the experimental data.  97  Figure 3.3 Mechanical properties of GB1-resilin-based polyproteins at the single-molecule level. A, B) Histogram of spacer length L0 of (GR)4 and GRG5RG4R. C,D) Histogram of persistence length of resilins measured in both polyproteins. E, F) histogram of unfolding force of GB1 domains in both proteins.  To confirm that R is largely unstructured, we also constructed a polyprotein with 12 repeats of the sequence (R12) and performed CD measurements on it. Our CD data (Figure 3.4) showed a minimum in ellipticity at ~200 nm, which indicates that the R12 is largely unstructured and which is consistent with our observation in AFM experiments.  98  -2000  2  [θ]MRE(deg•cm •dmol  -1  )  0  -4000  -6000  -8000 200  210  220  230  240  250  260  Wavelength (nm)  Figure 3.4 Far ultraviolet circular dichroism (CD) spectrum of R12 indicates that R12 is largely unstructured.  3.3.2 Biomaterials constructed from titin-mimetic proteins As shown above, the single molecule AFM experiments on (GR)4 result in saw-tooth-patterned unfolding peaks of GB1 domains, which is similar to that of Ig domains in titin. And CD measurement indicates that the R12 polyprotein is largely unstructured, which is similar to the PEVK sequences in titin. All of these results demonstrate that we have successfully designed artificial elastomeric proteins that exhibit mechanical properties similar to those of titin. We then used these titin-like polyproteins to construct biomaterials to mimic the passive elasticity of muscles. In muscles, individual titin molecules are well aligned [223]. But for now, we cannot produce such ordered structures. Instead, we created a GB1-resilin network 99  using the well-developed [Ru(bpy)3]2+-mediated photochemical cross-linking method. This method has been reported to successfully cross-link recombinant resilins into solid materials [122]. And that is one of the reasons why we introduce R sequence into our construct. We found that GB1-resilin polyproteins solutions (with protein concentrations >150 mg/ml) can be cross-linked into solid materials after exposed under white light (Figure 3.5).  Figure 3.5 Photographs of GB1-resilin-based biomaterials. Photographs of molded rings built from (GR)4 (left, intact) and GRG5RG4R (right, after being loaded to failure in tensile test) under white light (middle panel) and UV illumination (top panel).  The isotropic nature of the GB1-resilin-biomaterials was proved by a birefringence of 5×10ˉ6 to 1×10ˉ5. This birefringence is about 20 to 50 times smaller than that of optically isotropic elastin samples (2×10ˉ4) [322], and 100  vanishingly smaller than that of anisotropic major ampullate silks from A.diadematus (2.5×10ˉ² in a dry state and 6.1×10ˉ³ in a hydrated state) [323]. Since the molecules are cross-linked via tyrosine residues, the cross-linking sites were mainly located in the R sequences, which contain one tyrosine residue in each repeat. However, since one of the tyrosines in the folded GB1 domains is solvent exposed, they may also contribute to the photocross-linking. Figure 3.6 shows the photocross-linking reaction equation, our proposed schematic structure of the materials.  Figure 3.6 Photocross-linking scheme and the schematic structure of hydrogel based on GRG5RG4R.  3.3.3 Tensile test on resultant materials We tested the mechanical properties of the resultant materials using tensile measurements. Figure 3.7 shows typical stress-strain curves of (GR)4 and 101  GRG5RG4R-based materials. Rings made from GRG5RG4R can be stretched to more than 100% strain. From the stress-strain curves, the Young's modulus (stress/strain) is measured to be ~70 kPa for (GR)4-based materials and ~50 kPa GRG5RG4R-based materials, which are comparable to those of myofibrils [182, 234]. As mentioned above, one important feature of muscle is that at small sarcomere length, it behaves like rubber, while at long sarcomere length, it behaves like a shock-absorber. To demonstrate that, we also performed tensile tests at different strains. As shown in Figure 3.7A, B, no hysteresis between stretching and releasing was observed at low strains, indicating that the resultant materials are highly elastic, like rubber. But as strain increased, hysteresis began to develop, indicating that there was energy dissipation. This energy storing and dissipating property can be described by resilience, which is the ratio of the energy recovered upon unloading over the energy required to deform the material. In this manner, we compared the behaviours of the resultant materials at different strains. As shown in Figure 3.7C, the resilience of the materials decreased as the strain increased. Again this behaviour mimics muscle well, behaving like rubber at low strain and like a shock-absorber at higher strain [182, 183, 213]. However, the behaviour of GB1-resilin-based biomaterials is significantly different from resilin-based 102  biomaterials, which do not show hysteresis even at 250 % strain (Figure 3.7C) [122, 324].  Figure 3.7 Mechanical properties of (GR)4 and GRG5RG4R-based 103  biomaterials. Representative stress-strain curves of (GR)4 A) and GRG5RG4R B) measured in PBS. For clarity, stress-strain curves in A) and B) are offset relative to one another. Insets show the superposition of the stress-strain curves at different strains. C) Resilience of GB1-resilin-based biomaterials decreases with the increase of strain. For comparison, biomaterial constructed from resilin does not show any appreciable hysteresis (data taken from Ref[122]).  During the tensile tests, we also noticed that the stress-strain curves during consecutive cycles are superimposable. Figure 3.8 shows typical stress-strain curves after consecutive stretch-release cycles of GRG5RG4R at a pulling speed of 100 mm/min with a waiting time between consecutive cycles of 1 s. This result suggests that the force hysteresis observed is reversible, which is similar to the reversible hysteresis behaviour of myofibrils and myocytes [183, 213].  104  Figure 3.8 Consecutive stress-strain curves of GRG5RG4R during cyclic experiments.  The observed hysteresis indicates that stretching GB1-resilin-based biomaterials to high strains involved the breakage of weak non-covalent bonds [325] in the cross-linked network. And because the hysteresis observed in GB1-resilin-based biomaterials can be fully recovered upon relaxation, the breaking of bonds must be reversible. Moreover, the recovery of hysteresis is very fast, even when there was no waiting time between consecutive cycles (Figure 3.9). This behavior suggests that the recovery of hysteresis occurs in less than 1s (the dead time of Instron), which is consistent with the fast folding rate of GB1 domains measured in single-molecule AFM experiments [206].  105  Figure 3.9 Consecutive stretching-relaxation curves of GRG5RG4R at a pulling speed of 200 mm/min with waiting time between consecutive cycles of zero.  The recovery of hysteresis can also occur even under residual stress (Figure 3.10). When the biomaterial was partially relaxed to a strain above 35 %, no recovery of hysteresis was observed, while partial recovery started to occur at strain below 35 %. And the degree of recovery is residual stress dependent: the lower the residual stress, the more the recovery.  106  Figure 3.10 GRG5RG4R-based biomaterials can recover hysteresis under residual stress. The blue lines stand for the initial stretching trace. The inset shows the protocol of the experiments.  3.3.4 Measuring stress-relaxation of resultant materials To further compare the mechanical properties of the designed biomaterials with those of myofibrils/myocytes, we also carried out stress-relaxation experiments at constant strains. As shown in Figure 3.11, when GRG5RG4R was rapidly stretched to a certain strain and held at that particular strain afterwards, it is clear that there is stress relaxation, indicating energy dissipation. Also we observed a strain dependent relation, the larger the initial strain, the greater the amplitude of stress relaxation. The stress-relaxation behaviours can be well described by double-exponential fits 107  and the relaxation rate constants (k1 and k2) increase with the increase of strain. (Figure 3.11b)  Figure 3.11 GB1-resilin-based biomaterials exhibit pronounced stress relaxation behaviours. Representative stress-relaxation curves of GRG5RG4R at varying strains. It is evident that the stress relaxation is most pronounced at the highest initial strain. The relaxation rate constants k1 (filled squares) and k2 (open triangles) were obtained by a double-exponential fitting of the stress-relaxation. Error bars indicate fitting errors. 108  The stress-relaxation behaviours of GB1-resilin-based biomaterials are also similar to those of myofibrils [183, 313]. The unfolding of a few immunoglobulin domains has been proposed as a possible molecular mechanism to explain the stress-relaxation behaviours of myofibrils. We did Monte Carlo simulations (Figure 3.12) on force-relaxation behaviour of GRG5RG4R at constant extension. The result revealed that the unfolding of some GB1 domains can lead to force-relaxation behaviours similar to those seen in our experiments. The fast-phase relaxation rate depends on extension, which is similar to our experimental data. However, the simulated slow-phase relaxation rate differed from the experimental data. It is clear that GB1-resilin molecules are not well-aligned in the cross-linked network and the force on individual molecules cannot be measured directly. Also other microscopic processes such as possible chain friction during stretching may contribute to the stress-relaxation behaviours of GB1-resilin-based biomaterials. Therefore, a more detailed model combining the possibility of GB1 unfolding with a three-dimensional network is required to better describe the stress-relaxation behaviour of the GB1-resilin-based biomaterials.  109  Figure 3.12 Monte Carlo simulation on the force-relaxation of GRG5RG4R at constant extensions. A) Monte Carlo simulated force-relaxation curves of GRG5RG4R at different extension. B) Relaxation rate constant fitted from Monte Carlo simulation increases as a function of the initial force, which can be determined by the constant extension applied to the polyprotein in the force-relaxation experiments. The bars indicate standard deviation of the rate constants. 110  3.3.5 Tuning the properties of muscle-mimetic materials Tuning the mechanical strength is important because it tailors the hydrogels for specific applications, such as controlling the release of drugs and tissue morphogenesis. Moreover, the need for a hydrogel system with tunable mechanical strength is highlighted by research that studies the impact of matrix stiffness on cell functions, tissue morphogenesis and stem cell differentiation. We have demonstrated that we successfully built materials mimicking the passive elasticity of muscles using titin-mimetic proteins. An advantage of this study is that we incorporated folded, mechanically resistant globular domains into artificial elastomeric proteins. This makes it possible to investigate how nanomechanical properties engineered into individual polyprotein can be translated into macroscopic properties in materials by tuning the mechanical strength of proteins.  3.3.5.1 Fine-tuning the nanomechanical properties of molecular building blocks at the single-molecule level As mentioned before, the GB1-resilin-based materials behave differently from resilin-based materials. This indicates that the folded domains played an important role in determining the mechanical properties of GB1-resilin 111  based materials. We therefore tried to modulate the mechanical properties of macroscopic materials by affecting the folded state of GB1 domains using urea, a widely used chemical denaturant. First, we characterized mechanical properties of elastomeric polyproteins in the presence of urea at the single molecule level. From the AFM results (Figure 3.13), we can see that there were fewer unfolding events and the spacers prior to the unfolding force peaks were longer, which we attribute to stretching of denatured GB1 domains in urea. Correspondingly, at the macroscopic level, the Young's modulus decreased as urea concentration increased, which is consistant with the single molecule level results.  112  Figure 3.13 Modulation of the mechanical properties of macroscopic materials by affecting the folded state of GB1 domains using urea. a. Force-extension curves of single GRG5RG4R molecules in PBS and in the presence of 4 M urea. b. Young’s modulus of GRG5RG4R-based biomaterial as a function of concentration of urea.  113  3.3.5.2 Adjusting the composition of elastomeric polyproteins Another way to modulate the mechanical properties of such GB1-resilin-based biomaterials could be to adjust the composition of elastomeric polyproteins based on the fact that different types of muscles have different molecular compositions. For example, in the human heart, PEVK motifs of titin have 183 residues, while in the human soleus muscle, they have 2,174 residues [187]. Also we have showed that Young's modulus of (GR)4-based materials is higher than that of GRG5RG4R-based materials. To verify this possibility, we constructed three more elastomeric proteins, GRG5R, GRG9R and G8, and their corresponding biomaterials. As expected, change in the composition of the polyprotein changes the macroscopic properties of these biomaterials (Figure 3.14). In addition, the resultant rings show different initial dimensions due to different degrees of swelling in PBS buffer (Figure 3.15).  114  Figure 3.14 Mechanical properties of different types of GB1-resilin-based cross-linked biomaterials.  115  Figure 3.15 Swelling ratio of different types of GB1-resilin-based cross-linked biomaterials samples in PBS. Error bar indicates standard deviation (n≥5).  These preliminary results indicate that GB1-resilin based materials behave differently as the composition changes. We conducted AFM experiments to compare these different constructs at the single molecule level (Figure 3.3). However, stretching polyprotein GRG5RG4R resulted in force-extension curves with similar appearance to (GR)4. The only significant difference is shorter spacers due to fewer R sequences in GRG5RG4R (Figure 3.3). Thus, detailed work is still needed to fully understand these relationships. Nevertheless, these studies prove the possibility of modulating the macroscopic properties of these biomaterials by tuning their composition, and thus the structure of individual proteins, in the same way that different 116  isoforms of titin provide the passive elastic properties of different muscles [187].  3.3.5.3 Adjusting protein concentration and APS concentration Control of most chemically cross-linked systems is limited to varying the gel precursor and cross-linker concentration [310, 311], which changes the mechanical strength of the hydrogel. A third way to tune the mechanical properties of the muscle mimetic materials is to modulate the protein and APS concentration.  3  Young's Modulus (Pa)  60x10  50 40 30 20 10 0 80  100  120  140  160  180  200  Protein Concentration (mg/ml)  Figure 3.16 Mechanical properties of GRG5RG4R biomaterials with different protein concentration.  117  3  Young's Modulus (Pa)  40x10  30  20  10  0 0  100  200  300  400  APS Concentration (nM)  Figure 3.17 Mechanical properties of GRG5RG4R biomaterials cross-linked at different APS concentration.  3.4 Discussion Our results show that the behaviour of GB1-resilin-based biomaterials is similar to that of myofibrils/myocytes in force hysteresis and stress-relaxation. Thus, these biomaterials represent a new type of muscle-mimetic materials. However, how these two kinds of elements, GB1 and resilin-like sequence, work collectively to provide the mechanical properties of GB1-resilin-based biomaterials is not clear. On the one hand, our study manifests the role played by folded GB1 domain in the mechanical properties of designed biomaterials. Our results suggest that the unfolding of GB1 domains may explain the observed hysteresis. For example, the behaviour of GB1-resilin-based biomaterials is significantly 118  different from resilin-based biomaterials, which do not show hysteresis even at 250 % strain (Figure 3.7 C) [122, 324], indicating that folded GB1 domains play an important role in the resultant materials. By affecting the folded state of GB1 domains using urea, we tuned the mechanical properties of macroscopic materials, which also provides insight into the role played by folded GB1 domains. The hysteresis observed during stretch-release cycles indicates there is non-covalent bond breaking involved during stretching at higher strain. However, we also notice that the stress-strain curves during consecutive cycles are superimposable (Figure 3.8). This indicates that GB1-resilin-based biomaterials can fully recover their hysteresis within 1s, suggesting the bond breaking/formation is reversible. In the case of muscle, as mentioned before, a small number of folded Ig domains can unfold to dissipate energy when the stretching force is high enough. It has been reported that, in single-molecule AFM experiments, upon stretching, force-induced rupture of non-covalent bonds can lead to the unfolding of GB1 domains and dissipation of energy. And upon relaxation, the unfolded GB1 domains can refold very quickly [206]. Thus, for the engineered GB1-resilin-based biomaterials, it is possible that the observed hysteresis resulted from the unfolding of folded GB1 domains. On the other hand, it is important to note that the force hysteresis and 119  stress-relaxation are considered as viscoelastic properties at the macroscopic level. Similar macroscopic behavior may be caused by different microscopic mechanisms. Although unfolding of GB1 domains can lead to stress-relaxation behavior of GB1-resilin-based biomaterials, there is no direct demonstration of domain unfolding in macroscopic materials for now. Therefore, it is possible that other microscopic processes may also contribute to the stress-relaxation behaviors of GB1-resilin-based biomaterials.  3.5 Conclusions In summary, we have successfully built materials that exhibit mechanical properties mimicking the passive elastic properties of muscles using designed artificial elastomeric protein as building blocks. Incorporating folded globular protein enabled us to modulate the mechanical properties of these materials by tuning nanomechanical properties of proteins at the single molecule level. Our results represent a new example of "bottom-up" design, obtaining mechanical features at the macroscopic level by engineering features at the molecular level. And these biomaterials represent a new type of muscle-mimetic material. We anticipate that they will find applications in material sciences as well as in tissue engineering.  120  CHAPTER 4 Towards Constructing Extracellular Matrix-Mimetic Hydrogels: An Elastic Hydrogel Constructed from Tandem Modular Proteins Containing Tenascin FnIII Domains A version of Chapter 4 has been published on line as “Shanshan Lv, Tianjia Bu, Jona Kayser, Andreas Bausch, Hongbin Li, Towards constructing extracellular matrix-mimetic hydrogels: An elastic hydrogel constructed from tandem modular proteins containing tenascin FnIII domains, Acta Biomaterialia, Available online 5 January 2013, ISSN 1742-7061, 10.1016/j.actbio.2013.01.002. According to the copyright policy of Elsevier, the republishing of Elsevier full articles in a thesis/dissertation at the present time is not permissable. Therefore, Chapter 4 has been removed when the thesis is published on-line. As an alternative, the link to the article's DOI is provided as following: http://www.sciencedirect.com/science/article/pii/S1742706113000044  121  CHAPTER 5 Functional Hydrogels with Enzymatic Activity Based on Recombinant Tandem Modular Protein with Modular Structured Enzyme Protein-BCX 5.1 Introduction Protein-based hydrogels have attracted extensive attention as engineered materials with multiple functionalities, such as fluorescence, enzymatic activity and electron conduction, for applications in biocatalysis, bioelectrocatalysis and biosensing [1, 4, 326-328].In the design of hydrogels with enzymatic activities, a variety of strategies have been developed, either through physically encapsulating/entrapping [176, 243, 245, 298, 329] or covalent cross-linking [330] enzymes into hydrogels. For example, recombinant fusion proteins bearing affinity tags were engineered to create artificial polypeptide scaffold that immobilize glutathione-S-transferase [44]. Recombinant fusion proteins of organophosphate hydrolase bearing affinity tags were also engineered to form self-assembling hydrogels through physical cross-linking of affinity tags. The resultant enzyme-containing hydrogel had dual purposes of cell sequestration and metabolite detection [243]. Many other physically cross-linked protein based hydrogel were designed to immobilize enzymes for applications in the development of bioelectrocatalysis for biosensing and biofuel cells [245]. Alternative 122  covalent cross-linking methods have also been explored. For example, photocross-linked serum albumin hydrogels have also been designed as enzyme immobilization materials [330]. However, there are still some limitations of these strategies. For the physically cross-linked hydrogels, the bioactive species may be gradually released due to erosion properties of the hydrogel. Even though the present covalent cross-linking methods can help solve the erosion limitation, complex chemical modifications are usually required either on the hydrogel matrix or on the enzyme [330, 331]. Here we propose a novel design of a photochemically cross-linked enzymatic hydrogel based on tandem modular protein to demonstrate the possibility to design enzymatic hydrogels through incorporation of modular structured functional protein domains. As described in previous chapters, using the well-developed photochemical cross-linking strategy, we have engineered tandem modular protein-based biomaterials that exhibit mechanical properties that closely mimic the passive elastic properties of muscles, as well as biomaterials that are able to closely mimic the function and behavior of naturally-occuring large tandem modular proteins in the ECM. The ability to make use of tandem modular protein domain as building blocks to construct novel hydrogels also provides the tool set to design hydrogels with multiple functionalities, such as 123  fluorescence, enzymatic activity and electron conduction. To prove this feasibility, here we report an enzymatic functionalized hydrogel with xylanase (BCX) as a model system. Xylans are the most widely distributed polysaccharides, second only in mass to cellulose in plant biomass. The backbone of xylan is comprised of endo β (1→4) linked β xylopyranose residues [332]. The most important enzymes for xylan degradation are endo-xylanases, which cleave the backbone. Because xylanases can diminish the need for chemicals in the bleaching of pulp [333], there is increasing interest in these enzymes, both academically as well as commercially, particularly in the pulp and paper industry, which is involved in the decomposition of plant materials [254, 333, 334]. Among these endo-β-1, 4-xylanases, there is one originally from Bacillus Circulans, called BCX. BCX is a member of the family 11 xylanases [256]. It has a small molecular weight (20 kDa) and has been the subject of many previous studies using a variety of techniques including X-ray crystallography, which shows a well-defined modular structure [257, 258, 260, 261]. BCX has also been studied by NMR spectroscopy, and mutational analysis, leading to an understanding of enzymatic mechanism, and to well developed enzymatic activity assay methods [258, 260-264, 335]. The well established knowledge  124  of the BCX has made it a good model enzyme system for examining enzymatic hydrogels. In this study, we engineered a recombinant protein fused with folded tandem modular protein domain GB1 (for mechanical integrity), random coil-like sequences from resilin (for cross-linking) and BCX protein domains (for enzymatic function). The protein retained partial enzymatic activity of non-fusion BCX. Then the protein was cross-linked to hydrogel using a Ru(II)-mediated photochemical cross-linking method. Enzymatic activity assays demonstrated that the resultant hydrogel exhibits enzymatic activity of BCX. Furthermore, an active site titration method was employed to quantify the active enzymes immobilized in hydrogels. Moreover, the hydrogels embedded with enzymes retained activity testing for several times, indicating the reusability of in the enzymatic hydrogels which would be important in industrial applications.  5.2 Materials and methods 5.2.1 Protein engineering The gene encoding the tandem modular protein GB1-Resilin-GB1-BCX-GB14- Resilin (G-R-G-BCX-G4-R) was constructed using standard molecular biology techniques following published protocols 125  [314]. Tandem modular protein domain GB1(G) and random coil sequences from resilin (R) were employed to provide mechanical integrity and sites for cross-linking, respectively. The gene encoding protein GB1 was a generous gift from David Baker of the University of Washington. The gene that encodes(GB1)4 was constructed as previously reported [199]. A 15 amino acid consensus repetitive sequence (GGRPSDSYGAPGGGN) of resilin [122] from the first exon of the Drosophila melanogaster gene CG15920 was used during a polymerase chain reaction (PCR). The plasmid containing a semi-synthetic gene encoding BCX was described previously [336]. The BCX gene used here was amplified and engineered as previously reported [265]. The gene encoding G-R-G-BCX-G4-R was cloned into an empty vector pQE80L (with His-tag built in). The expression of the polyprotein G-R-G-BCX-G4-R was carried out in the Escherichia coli strain DH5α. Cell cultures were grown at 37°C in 2.5 % LB medium containing 100mg/L ampicillin until the optical density of the cell culture reached ~1. Protein over-expression was induced with 1.0 mM isopropyl-1-β-D-thiogalactoside (IPTG), and allowed to express for 5 hours. Cells were harvested by centrifugation at 12,000 g for 10 min and lysed using lysozyme from egg white (100 mM, SigmaAldrich). The polyprotein was purified from supernatant using Co2+ affinity chromatography with a yield of ~40 mg per 126  liter of culture. The purified G-R-G-BCX-G4-R was dialyzed against deionized water for 3 days and then lyophilized before enzymatic activity assays and constructing hydrogel. Another polyprotein GRG5RG4R was constructed in the same way and used as bulk matrix of the G-R-G-BCX-G4-R-based hydrogel.  5.2.2 Enzymatic activity assays To test the enzymatic activity of the engineered polyprotein G-R-G-BCX-G4-R, a substrate depletion assay for hydrolysis of 2,5-dinitrophenol β-xylobioside (2,5-DNPX2) was performed. The rates of enzymatic hydrolysis for the substrate were determined using a continuous assay. 0.05 mM substrate in 20 mM MES buffer, containing 50 mM NaCl and 0.1 % BSA, pH 6.0 was pre-warmed to 37 °C. Reaction was initiated by the addition of enzyme to a final concentration of 0.001 mM. Substrate hydrolysis was monitored using UV/Vis spectrometer by measuring the rate of phenolate release at 440 nm, ∆ε=3.57 mMˉ¹cmˉ¹ [337].  5.2.3 Hydrogel formation To prepare G-R-G-BCX-G4-R-based hydrogels, different amount of G-R-G-BCX-G4-R protein was mixed with GRG5RG4R and was dissolved in 127  PBS to a total protein concentration of 200 mg/ml. The final concentrations of G-R-G-BCX-G4-R in the protein mixtures were 0.2 mg/ml, 2 mg/ml and 20 mg/ml. Ammonium persulfate (APS) and [Ru(bpy)3]2+ were then added to the protein solution to a final concentration of ~260 µM for [Ru(bpy)3]2+ and ~50 mM for APS. For each hydrogel sample, 20 µl of the protein solution was cast into a custom-made cylinder mold (with a diameter of 5 mm), and irradiated for 10 minutes using a 150 W fiber optical white light source placed 10 cm away from the mold. The resultant hydrogel was removed from the mold and stored in MES buffer at 4˚C. The hydrogel samples were disk-like with a diameter of 5 mm. It is of note that hydrogels made from 20 µl 2 mg/ml G-R-G-BCX-G4-R protein solution contain the same amount of enzyme as that in enzymatic activity assay of the non-cross-linked G-R-G-BCX-G4-R. Hydrogels were also fabricated with an alternative physical cross-linking method in order to investigate the effect of chemical cross-linking on the activity of the enzymes. The physical cross-linking method is, as described in Chapter 2, a two-component approach based on CCE-G4-CCE and CCK-G5-CCK-G5-CCK [214]. Hydrogels form through self-assembly of coiled coil domain CCE and CCK [303]. Following the previous design of CCE-G4-CCE, a protein fused with BCX, CCE-G-G-BCX-G-CCE, was 128  constructed. The proteins CCE-G-G-BCX-G-CCE and CCK-G5-CCK-G5-CCK were mixed at a 1:1 molar ratio of CCE and CCK and redissolved in MES buffer after lyophilisation, leading to formation of hydrogels. The resultant physical cross-linked hydrogels were tested for enzymatic activity in a similar way to the chemical cross-linked hydrogels.  5.2.4 Hydrogel enzymatic activity assay Enzymatic activity of hydrogel samples was determined by the substrate depletion assay for hydrolysis of 2, 5-dinitrophenol β-xylobioside (2,5-DNPX2). 0.05 mM substrate in 20 mM MES, containing 50 mM NaCl buffer and 0.1 % BSA, pH 6.0 was pre-warmed to 37 °C. Reaction was initiated by the addition of hydrogel samples. The reaction mixture was stirred constantly during the whole measurement for better mixing. At different times during the reaction, the substrate hydrolysis was monitored using UV/Vis spectrometer by measuring the rate of phenolate release at 440 nm, using ∆ε=3.57 mMˉ¹cmˉ¹ [337]. Two different samples were measured, and the average value was reported.  129  5.2.5 Active site titration using 2F-DNPX2 To determine the active enzyme concentration, the non-cross-linked enzyme (20mg/ml G-R-G-BCX-G4-R protein solution) and enzymes cross-linked in hydrogels (made from protein solution containing 20 mg/ml G-R-G-BCX-G4-R) were inactivated using 2,4-dinitrophenyl 2-deoxy-2-fluoro-β-xylobioside (2F-DNPX2) in MES buffer at 40 °C as described elsewhere [338]. The release of 2,4-dinitrophenol was monitored by UV/Vis spectrophotometer at 400 nm until a linear rate of hydrolysis of the inactivator was observed. The titration of non-cross-linked enzymes was determined using a continuous assay, while that of cross-linked enzymes was in a discontinuous way at different times along the reaction course due to the need for a stirring system. The y-intercept of the linear region was determined by linear regression. The obtained values and the extinction coefficient of 2,4-dinitrophenol (∆ε=11.40 mMˉ¹cmˉ¹) were then used to calculate the concentration of 2,4-dinitrophenolate released and thereby the concentration of active enzyme. Two different hydrogel samples were measured, and the average value was reported.  130  5.2.6 Enzymatic activity through multiple test cycles Enzymatic activity of hydrogel samples was followed through multiple test cycles by monitoring the hydrolysis of 2,5-DNPX2. 0.05 mM substrate in 20 mM MES, containing 50 mM NaCl buffer and 0.1 % BSA, pH 6.0 was pre-warmed to 37 °C. Reaction was initiated by the addition of hydrogel samples and continued for 3 hrs. The total amount of phenolate released was recorded using UV/Vis spectrometer by measuring the absorbance at 440 nm, using ∆ε=3.57 mMˉ¹cmˉ¹ [337]. Hydrogel enzymatic activity is normalized to the activity of the same amount of non-cross-linked enzyme after reaction under the same conditions. Hydrogel samples were tested several times. After each cycle of testing, hydrogels were washed with MES buffer and then used for the next cycle. Recyclable enzymatic activity is expressed as % retained activity relative to the activity of the first test cycle. Three different hydrogel samples were measured, and the average value was reported.  5.3 Results and Discussion 5.3.1 Design considerations In our previous study, we designed and engineered an artificial elastomeric protein GRG5RG4R [237], which combined folded tandem modular protein building block GB1 (Figure 5.1A) with random coil-like sequences from 131  resilin (Figure 5.1A). By cross-linking these designed proteins using the Ru(II)-mediated photochemical cross-linking strategy, we successfully engineered protein-based biomaterials that exhibited unique mechanical properties [237]. Following this strategy of constructing tandem modular protein based hydrogel, here we design a recombinant protein fused with folded tandem modular protein domain GB1(G), random coil-like sequences from resilin (R) and enzymatic protein BCX domain (Figure 5.1A), G-R-G-BCX-G4-R (Figure 5.1). Tandem modular protein domain GB1 and random coil sequences from resilin are used to provide mechanical integrity and sites for cross-linking, respectively. The GRG5RG4R is also constructed and used as bulk matrix of the G-R-G-BCX-G4-R-based hydrogel. The use of GRG5RG4R as bulk matrix provides possibility of tuning enzyme concentration in hydrogels without significantly changing structural network of the hydrogels, which can provide the same environment for enzymes at different concentrations. It can also be used to reduce the amount of enzymes needed for hydrogel fabrication. In addition, the use of GRG5RG4R can maintain the unique mechanical properties of the GB1-resilin-based biomaterials.  132  Figure 5.1. Construction of recombinant protein containing BCX, G-R-G-BCX-G4-R. A) The protein structure of globular BCX [265] (PDB: 1HV0). B) Schematic of the artificial protein G-R-G-BCX-G4-R and GRG5RG4R [237]. C) 12 % denaturing SDS-PAGE picture of G-R-G-BCX-G4-R. The first lane is a broad-range molecular weight marker (New England Biolabs).  5.3.2 Activity of the designed tandem modular protein To confirm that the fusion with GB1 domains and resilin sequences is not detrimental to the enzymatic activity of BCX, we carried out a substrate depletion assay using 2, 5-dinitrophenol β-xylobioside (2,5-DNPX2) as substrate. The hydrolysis of 2,5-DNPX2 will release 2,5-dinitrophenolate characterized by UV/Vis absorbance at 440nm [337]. 133  Scheme 5.1. Reaction mechanism of BCX with 2,5-DNPX2, where 2,5-DNP represents 2,5-dinitrophenolate [337].  Figure 5.2. The activity of BCX in polyprotein G-R-G-BCX-G4-R. A) Enzymatic activity is demonstrated by an obvious colour change after a reaction time of 20 minutes, demonstrating the presence of the substrate cleavage product of 2,5-dinitrophenolate that is yellow in solution. B) The formation of the product 2,5-dinitrophenolate was monitored using UV-Vis spectroscopy at 440 nm. The substrate depletion curve was fitted to a first-order rate equation to determine the rate constant of 0.006 sˉ¹. The kcat/Km was found to be 6.0 sˉ¹mMˉ¹.  Clear color change was observed upon addition of G-R-G-BCX-G4-R (Figure 5.2A). The formation of product was also monitored using UV-Vis 134  at 440 nm (Figure 5.2B). The substrate depletion curve was fitted to a first-order rate equation to determine the rate constant of 0.006±0.002 sˉ¹. kcat/Km was calculated to be ~6.0 sˉ¹mMˉ¹. The final absorbance was measured and calculated to give a total amount of 0.03 µmol of 2,5-dinitrophenolate, which suggests complete hydrolysis of all the added substrates. The kcat/Km of G-R-G-BCX-G4-R is smaller than previously reported kcat/Km of wild type BCX, which is around 35 sˉ¹mMˉ¹ [337]. This decrease might be caused by the fusion with GB1 domains and resilin sequence. Previously constructed fusion protein combining GB1 and BCX showed kcat/Km around 29 sˉ¹mMˉ¹ [265]. Although the kcat/Km is still comparable to that of wild type BCX, the decrease is obvious. In the G-R-G-BCX-G4-R, not only the GB1, but also the resilin will interact with the BCX domain to some extent, which might cause a further decrease in the kcat/Km. There are also other possible reasons for the decrease in the kcat/Km, such as the lyophilization step in protein preparation. More studies are needed to address this issue. Nevertheless, the polyprotein G-R-G-BCX-G4-R retains partial enzymatic activity of non-fusion BCX.  135  5.3.3 Hydrogel Formation After determining the activity of the designed protein G-R-G-BCX-G4-R, we used the Ru(II)-mediated photochemical cross-linking method to cross-link the protein to hydrogel. The Ru(II)-mediated cross-linking method allows the cross-linking of two tyrosine residues in close proximity into di-tyrosine adducts. The use of resilin repeats, which provide the majority of cross-linking sites in the polyproteins, enables an efficient approach with which to prepare GB1-resilin-based biomaterials. We found that under illumination with white light, the polyproteins G-R-G-BCX-G4-R and GRG5RG4R can readily be cross-linked into solid and transparent hydrogels at room temperature from their concentrated solutions. Figure 5.3 shows a photograph of the moulded disk-like hydrogel of the polyproteins. The formation of di-tyrosine cross-links was indicated by their characteristic blue fluorescence upon ultraviolet irradiation[122] (Figure 5.3).  136  Figure 5.3. Formation of hydrogels containing BCX. Photograph of a moulded disk like hydrogel built from G-R-G-BCX-G4-R and GRG5RG4R. The bottom panel is taken under white light, while the upper panel was taken under ultraviolet illumination.  5.3.4 Activity of hydrogels embedded with enzymes Enzymatic activity of hydrogels embedded with enzymes was tested using an assay similar to that used for enzymatic activity assay of the polyprotein G-R-G-BCX-G4-R. Hydrogels containing the same amount of G-R-G-BCX-G4-R as that used for enzymatic activity assay of the polyprotein G-R-G-BCX-G4-R was added to substrate solution. The reaction  137  rate was monitored by measuring the UV/Vis absorbance of 2,5-dinitrophenolate at 440nm.  Figure 5.4. The activity of G-R-G-BCX-G4-R-based hydrogels. A) Enzymatic activity of the hydrogel is demonstrated by an obvious colour change after reaction with hydrogel for 2 hours. B) Reaction rate was measured to be k=0.0005±0.0002 sˉ¹.  The reaction rate for hydrolysis of 2,5-DNPX2 was measured to be 0.0005±0.0002 sˉ¹, which is around 10 times slower than that of non-cross-linked polyprotein G-R-G-BCX-G4-R. In the case of hydrogels embedded with enzymes, because the enzymes are localized in the hydrogels, and thus cannot distribute homogeneously in the reaction mixture, the concentration of enzymes cannot be determined. Therefore, kcat/Km value cannot be calculated using the same method as that used for enzymatic activity assay of the polyprotein G-R-G-BCX-G4-R. The final absorbance 138  was measured and calculated to give a total amount of cleaved 2,5-dinitrophenolate around 0.03 µmol, which also suggests complete hydrolysis of all the added substrates (in ~2 hours). The enzymatic activity assay result shows that the enzyme reaction rate of BCX was apparently decreased, with a smaller rate constant value, after being cross-linked to hydrogels. Similar behaviour was observed on other other immobilized enzymes [339, 340]. This decrease could be caused by diffusional limitations of the network structure of hydrogels to either the substrates’ diffusion to the active site of cross-linked enzymes or the products’ diffusion out from the hydrogels [327, 341]. The diffusion effect may also result in changes in optimum conditions for the enzymatic reaction, such as pH and temperature [342]. The decreased reaction rate might also be caused by specific interactions of substrate, product and enzyme within the matrix which can potentially change the kinetic properties of enzymes [343]. Another possible reason for the decreased reaction rate is the changes in enzyme structure induced during the cross-linking preparation of hydrogels and the hydration/swelling process afterwards. More studies are needed to address this issue.  139  5.3.5 Varying the enzyme concentration in the hydrogel As considered in the hydrogel design, the use of GRG5RG4R as bulk matrix provides the possibility of tuning the enzyme concentration in hydrogel. We then varied the concentration of enzymes in the hydrogels. Despite of the change in enzyme concentration, there is no significant change in reaction rate (Figure 5.5).  Figure 5.5. The activity of hydrogels with different G-R-G-BCX-G4-R concentrations at 0.2 mg/ml and 20 mg/ml. The reaction rate of hydrogel containing 0.2 mg/ml G-R-G-BCX-G4-R was fitted to be 0.0003±0.0002 sˉ¹ and 0.0004±0.0001 sˉ¹ for 20 mg/ml. 140  5.3.6 Active site titration Enzymatic activity assay results demonstrate that the hydrogels embedded with enzymes are enzymatically active. However, as mentioned before, in the case of enzymes cross-linked in hydrogels, kcat/Km value cannot be calculated using the same method as that used for enzymatic activity assay of the polyprotein G-R-G-BCX-G4-R, because the concentration of the cross-linked enzymes in the reaction mixture cannot be determined as homogeneous catalysis reaction. Thus we lack a well defined parameter for activity comparison. As a matter of fact, it has been difficult to make comparisons of enzyme activity, especially of the different methodologies from the thousands of papers published on enzyme immobilization. Most researchers compare the activity of the immobilized enzyme with that of the free enzyme, but in some cases, researchers have different definitions of the parameter for activity comparison. Moreover, there are rarely ways to compare different methods of immobilization [326]. In this study, we propose the use of an active site titration method to make comparisons of immobilized enzymes as a supplement to the present publications on enzyme immobilization. We quantified the amount of active enzymes in polyprotein G-R-G-BCX-G4-R solution and hydrogels by an active site titration method 141  employing the inhibitor 2,4-dinitrophenyl 2-deoxy-2-fluoro-β-xylobioside (2F-DNPX2) [263, 344, 345].  Scheme 5.2. Reaction mechanism of BCX with 2F-DNPX2, where 2F corresponds to 2-fluoro-β-xylobioside, and 2,4-DNP to 2,4-dinitrophenolate.  As shown in Scheme 5.2, the 2F-DNPX2 rapidly forms a glycosyl-enzyme intermediate with stoicheometric release of 2,4-dinitrophenol characterized by UV/Vis absorbance at 400 nm. Because the turnover of the glycosyl-enzyme intermediate is much slower than its formation, monitoring the release of 2,4-dinitrophenol will lead to a curve with a burst increase followed by a steady linear increasing. From these curves, the y-intercept of the extrapolation of the linear region gives the absorbance of all the 2,4-dinitrophenol released. Through its extinction coefficient, the 2,4-dinitrophenol concentration can be calculated. Since the release of 2,4-dinitrophenol is stoicheometric, enzyme concentration can be determined [338].  142  A) 0.30  Abs_400nm  0.25 0.20 0.15 0.10 0.05  0  500  1000  1500  Time (s) B)  Abs_400nm  0.3  0.2  0.1  0.0 0  500  1000  1500  2000  2500  3000  Time (s)  Figure 5.6. Determination of active enzyme concentration by active site titration. The active site titration curves for polyprotein G-R-G-BCX-G4-R solution A) and G-R-G-BCX-G4-R-based hydrogels B) are shown with a linear fit for each. The y-intercept and the extinction coefficient of 2,4-dinitrophenol were used to calculate the active site concentration.  The active site titration curves for polyprotein G-R-G-BCX-G4-R solution and G-R-G-BCX-G4-R-based hydrogels are shown in Figure 5.6. It is of note that a total number of 10 nanomol (nmol) of BCX was used for each active site titration. By fitting the gradually increasing trend of the curve and 143  converting the intercept value, the number of active site was calculated to be 5.94 nmol for the tested non-cross-linked G-R-G-BCX-G4-R enzyme. The number is a smaller than the total number of enzymes (10 nmol). This might be caused by the lyophilization during the protein preparation. The number of active site for same amount of enzymes cross-linked in hydrogel was calculated to be 5.87±0.19 nmol, which is comparable to that of non-cross-linked enzyme. The similarity of number of active site in polyprotein G-R-G-BCX-G4-R solution and G-R-G-BCX-G4-R-based hydrogels can explain and is consist with the enzymatic activity assay result that hydrogels containing the same amount of G-R-G-BCX-G4-R as that used for enzymatic activity assay of the polyprotein G-R-G-BCX-G4-R catalyzed the same amount of substrate conversion.  5.3.7 Effect of chemical cross-linking on enzymatic activity Both the activity assay and active site titration results suggest that the chemical cross-linking method used here does not abolish activity of the enzyme. In order to confirm this, we employed an alternative physical cross-linking method to produce enzymatic hydrogels and compared the activities of the chemical and physical cross-linked hydrogels.  144  As described in Chapter 2, we developed a two-component approach as a method to engineer tandem modular protein-based hydrogels [214] using coiled coil domains CCE and CCK [303]. This new method provides the possibility to use large globular proteins in the construction of functional hydrogels. Therefore, we constructed a BCX fused protein, CCE-G-G-BCX-G-CCE, following the previous design of CCE-G4-CCE. CCK-G5-CCK-G5-CCK was also constructed to form hydrogels with CCE-G-G-BCX-G-CCE. Enzymatic activity assays based on hydrolysis of 2,5-DNPX2 were performed on both CCE-G-G-BCX-G-CCE protein solutions and hydrogels formed through self-assembly of CCE-G-G-BCX-G-CCE and CCK-G5-CCK-G5-CCK at a 1:1 molar ratio of CCE and CCK.  145  Figure 5.7. Characterization of CCE-G-G-BCX-G-CCE protein and CCE-G-G-BCX-G-CCE -based hydrogels. A) Activity of uncross-linked CCE-G-G-BCX-G-CCE protein. The substrate depletion assay was initiated by adding the protein to 0.6ml of 0.05mM 2,5-DNPX2 substrate to a final BCX concentration of 0.0016mM. The curves are fitted with an average reaction rate of 0.0026± 0.0009 sˉ¹. kcat/Km was calculated to be 1.6 sˉ¹mMˉ¹. B) 7 % aqueous solution of CCE-G-G-BCX-G-CCE and CCK-G5-CCK-G5-CCK mixture forms protein hydrogel. C) Enzymatic 146  activity of the hydrogel is demonstrated by an obvious colour change after reaction with hydrogel for 2 hours. D) Enzymatic activity assay of CCE-G-G-BCX-G-CCE-based hydrogels. Reaction rate was measured to be ~0.0006 sˉ¹. It is of note that the amount of BCX in hydrogel activity tests is around 5 times the amount of uncross-linked proteins.  Figure 5.7A shows the substrate depletion assay result on CCE-G-G-BCX-G-CCE protein. The kcat/Km of CCE-G-G-BCX-G-CCE was calculated to be 1.6 sˉ¹mMˉ¹. Similar to that of G-R-G-BCX-G4-R, this kcat/Km value was smaller than previously reported ones [265]. Nevertheless, the final absorbance gave the total amount of 0.03 µmol cleaved 2,5-dinitrophenolate, which also suggests complete hydrolysis of all the added substrates. As shown in Figure 5.7B, the designed proteins CCE-G-G-BCX-G-CCE and CCK-G5-CCK-G5-CCK indeed form hydrogels. Enzymatic activity is demonstrated by an obvious colour change (Figure 5.7C). The enzymatic activity of the physical hydrogels was also tested with UV spectrometer. Due to the high concentration required for self-assembly of hydrogels and difficulty in handling the hydrogels, the amount of BCX in hydrogel activity tests are around 5 times that in uncross-linked proteins activity tests. However, even though the amount of enzyme is 5 times that in 147  protein solution, the reaction rate for hydrolysis of 2,5-DNPX2 was ~0.0006 sˉ¹. This reaction rate is much slower than that of non-cross-linked protein (~0.0026 sˉ¹), considering the amount of enzymes used for tests. This slowing down effect upon physical hydrogel formation is also similar to that of chemically cross-linked hydrogels, which proves that the Ru(II) chemical cross-linking method does not cause any more significant change on enzyme activity than physical cross-linking method. Compared with physically cross-linked self-assembling hydrogels [214], the Ru(II) mediated cross-linked hydrogel is chemically cross-linked with a resultant erosion rate that is extremely low, which is significantly improved over physically cross-linked hydrogels [243, 245].  5.3.8 Repeated use of enzymatic hydrogels As it can be seen, the hydrogels are enzymatically active, indicating the enzymes cross-linked in the hydrogels remain functional and active. This result suggests that the design of tandem modular protein based hydrogels not only provides a method to design enzymatic hydrogels, but also provides a simple and general method to immobilize enzymes that remain active. Enzyme immobilization has been a well-known useful strategy both for fundamental research on the reaction mechanisms of enzymes that are 148  naturally associated with insoluble structures, such as enzymes across membranes [342, 343, 346], and for industrial-scale applications in biocatalysis [242-244], bioelectrocatalysis [245-247] and biosensing [248-253] thanks to its many advantages that can help reduce costs in industry [347]. For example, immobilization allows easy separation of enzyme after reaction and thus easy purification of desired products. Immobilization can even improve catalytic properties, such as stability and activity, through protecting the activity of the enzyme from harsh environment conditions, including temperature and different denaturing agents [348-350]. Immobilization also allows collection, recovery and reuse of the enzymes [327, 330, 351, 352], which reduces the enzyme cost in their production and purification [353]. In the present study, enzymes were cross-linked/immobilized into the tandem modular protein based hydrogels through covalently cross-linking fusion protein containing the modular structured enzyme. Enzymes cross-linked in hydrogels can be easily collected and recovered from the reaction system after usage and can repeatedly react with the substrate. We investigated the enzyme activity of the hydrogel after repeated reactions. Hydrogel enzyme activity is normalized to the activity of same amount of free enzyme after reaction under the same conditions. Recycled enzymatic 149  activity is expressed as % retained activity relative to the activity of the first cycle of test. As shown in Figure 5.8, the activity of enzymatic hydrogel was ~90 % retained after up to 3 reactions. Even though the activity drops on the 4th use, it still can recover activity to about ~60 %, slight decreasing to ~30% as the number of reactions increased up to 9.  Figure 5.8. The enzyme activity of the hydrogels embedded with enzymes in repeated reactions. The activity of enzymatic hydrogel was ~90 % retained after up to 3 reactions, with further decreases as the number of reactions increased up to 9. Three different hydrogel samples were measured, and the average value was reported.  Even though the fusion and cross-linking process affect the enzymatic activity of the enzyme BCX (based on the comparison of catalytic efficiency kcat/Km and reaction rate), fusion and cross-linking provides reusability to enzymes, which can compensate for the loss of the enzymatic activity. This 150  reusability of the designed enzymatic hydrogels demonstrates the potential of the enzyme-containing hydrogels in industrial applications.  As it can be concluded from the above results, using the Ru(II)-mediated photocross-linking method, the designed polyprotein G-R-G-BCX-G4-R can easily be cross-linked to hydrogel with enzymatic activity. This proves that Ru(II)-mediated covalent cross-linking not only provides a method to design functional, in this case, enzymatic hydrogels, but also provides an enzyme immobilization method solving the limitations encountered in the use of protein-based hydrogels for enzyme immobilization applications. Compared with physically cross-linked self-assembling hydrogels [243, 245], the hydrogel is chemically cross-linked with a resultant erosion rate that is extremely low, which is significantly improved over physically cross-linked hydrogels [243, 245]. Besides, the covalent photochemically cross-linking method used here is relatively easy and efficient without any need for complex modification on matrix or enzymes. G-R-G-BCX-G4-R-based hydrogels also exhibit some other unique properties and advantages for enzyme immobilization. Firstly, as the protein building block is a tandem modular protein, it is possible to incorporate additional functional protein domain(s) into hydrogels using a modular 151  approach to endow the hydrogel with multiple functionalites. Secondly, as the hydrogels are designed following hydrogels mimicking the passive elasticity of muscles, which has potential applications as extra cellular matrix or tissue engineering scaffolds, upon replacing the BCX with other enzymes, physiological identified in single/multicellular organisms, the hydrogels can be used for biomedical use and in reaction mechanism and kinetics study of natural enzymes. In addition to reaction mechanism and kinetics study, due to the unique mechanical properties of the hydrogel, it can be used to study enzymes that may subject to stretching forces under physiological conditions and investigate the force effect on enzymatic activity. Lastly, the retained enzymatic activity and reusability of the designed enzymatic hydrogels demonstrates their potential in industrial applications.  5.4 Conclusions In summary, through fusions of GB1 domains that are tandem modular mechanically stable and random coil sequence from resilin that can be covalently cross-linked, to the BCX, we designed a protein building block that can form hydrogel with both unique mechanical properties and enzymatic activity. The designed tandem modular protein fused with BCX 152  retained partial enzymatic activity of non-fusion BCX. The protein were easily cross-linked to hydrogel with BCX built in using a Ru(II)-mediated photocross-linking method. Enzymatic activity assay demonstrated that the hydrogels were enzymatically active, even though the apparent reaction rate is slower. Active site titration showed that the number of active enzyme in the hydrogels was the same as that in non-cross-linked enzyme. In addtion, the hydrogels embedded with enzymes retained activity after repeatedly tests, indicating the reusability of the enzymatic hydrogels and their potential in industrial applications. Moreover, the utilization of the active site titration provides an easy and direct parameter for quantification of immobilized enzymes. It also provides a general method to make comparisons of enzyme activity, especially of the different methods of immobilization, which is a significant supplement to the present publications on enzyme immobilization. Furthermore, the design of enzymatic hydrogel with xylanase activity demonstrates the possibility to immobilize functional proteins through incorporation of modular protein domains. This not only provides a novel method to covalently immobilize enzymes into a hydrogel system, but also provides a general method for functional and multi-functional hydrogel design through incorporation of functional modular protein domains. 153  CHAPTER 6 Fluorescent Hydrogels Constructed from Tandem Modular Proteins based on CFP/YFP FRET Pair as Force Sensors Capable of Estimating Swelling Force on Single Peptide Chain 6.1 Introduction Using a well-developed photochemical cross-linking strategy, we have designed a new tandem modular protein-based hydrogel with a natural ECM protein TNfn3 domain as a building block [354]. The TNfn3-containing hydrogels can support cell adhesion and cell spreading, indicating that TNfn3 RGD ligands are effectively presented on the hydrogel surface. This suggests that the new technique making use of tandem modular protein domains as building blocks to construct novel hydrogels offers a possible method to design hydrogels with multiple functionalities, such as enzymatic activity and fluorescence. Through incorporating an enzyme domain-BCX, an enzymatic functionalized hydrogel has been constructed. The enzyme-containing hydrogels exhibit enzymatic activity, providing good evidence that this tandem modular protein-based technique provides a tool set to design novel hydrogels with multiple functionalities and explore their applications in biocatalysis, bioelectrocatalysis and biosensing. This feasibility can be further demonstrated with hydrogels constructed from  154  fluorescent protein pairs as force sensors capable of estimating swelling forces on single peptide chains. As described in Chapter 1, FRET between fluorescent protein pairs has been used to study changes in distance between the fluorophores [281-288]. Among the FRET pairs (FPs), genetically fused CFP and YFP, with a flexible linker inbetween, has been used in many studies for measurement of short molecular diastances [290, 291]. For the project described here, the CFP/YFP FP was chosen as a model system to test the efficacy of designing functional hydrogels. The CFP and YFP we used are an improved cyan fluorescent protein variant (ECFP/S72A/Y145A/H148D) named Cerulean [292] and a new YFP (GFP/ Thr203Tyr/ Q69M) named Citrine [293]. A hydrogel embedded with the CFP/YFP FP separated by a linker sequence was constructed. In order to explore application of the designed hydrogels as force sensors, we investigated hydrogel swelling force as a case of study. Hydrogels are highly hydrated cross-linked polymer networks that can exchange solution with their environment, undergoing a volume change. For instance, dehydrated hydrogels can swell by absorbing solvents. The swelling and drying behaviors of hydrogels are very important in many applications, such as swelling-controlled drug delivery [355, 356]. In addition, the kinetics of 155  hydrogel swelling can be tuned in response to some environmental stimuli. These stimuli can be classified as either physical (temperature, electrical voltage, magnetic fields, mechanical stress, etc.) or chemical (pH, ionic strength, concentration of organic compounds in water, etc.). This stimulus-responsiveness of hydrogels can be utilized in applications as sensors, such as chemical sensors, pH sensors and in vivo glucose sensors [357-359]. Hydrogels with appropriate mechanical properties have many other applications in tissue-engineering [2]. There have also been reports on the effects of swelling on mechanical properties of hydrogels [360-363] . Previous studies have developed many kinds of apparatus to measure the swelling force of different polymer gels at the macroscopic level [364]. It is widely accepted that during swelling, polymer chains will be extended. However, there is limited experimental information available regarding the effects of swelling on single polymer chain length and single polymer chain force [269]. Motivated by the interest in developing a force sensor based on a hydrogel system, especially one sensor that can detect the swelling force on single polymer chain, we used a FP embedded hydrogel system to study the force of "swelling" materials. The hydrogel is constructed from a CFP/YFP FP separated by a sequence named CCK [303]. The CCK sequence is capable of existing at different 156  polymer chain lengths. During swelling, the polymer chain will be extended by the swelling stress, including the CCK sequence. As CCK is stretched, the distance between the CFP/YFP FP will increase, thus resulting in decreased FRET efficiency (E). An alternative method to decrease E is to let CCK peptide sequence form coiled coil with CCE, because the length of the coiled coil is longer than the expected end-to-end distance of the CCK sequence [303]. This is also one reason that CCK is used as the linker between CFP/YFP FP (Figure 6.1). The following principles for the single polymer chain length swelling force detection are used in the hydrogel system based on the swelling behavior of the polymer chain: from the E, given certain assumptions, the length of CCK peptide sequence can be estimated [290]. Using a Worm Like Chain (WLC) Model, forces generated by the CCK peptide sequence at corresponding lengths can be estimated. Since the CCK peptide sequence is part of the cross-linked polymer chain, the force on the single polymer chain should be the same as the force of the CCK peptide sequence. In this way, swelling force on single polymer chain can be detected. Our results showed that the resultant hydrogels are fluorescent. And the extent of FRET depends on differences in distance between the FRET pair component in different protein structural states. The distance between FRET 157  pairs was estimated to change from ~2nm in protein solution to ~6nm in swollen hydrogel. Also, forces generated by the CCK peptide sequence at corresponding distances were estimated according to the E. This “proof of concept” study, for the first time, provides direct evidence of a swelling effect on single polymer chain length. The present study, to our knowledge, reports a potential force sensor of swelling force on single polymer chain for the first time. Our results also suggest the capability of proper functioning of fluorescent proteins in hydrogels, providing further evidence of the feasibility of using tandem modular protein in construction of novel hydrogels with multiple functions.  6.2 Materials and Methods 6.2.1 Protein engineering GB1-resilin-based polyprotein genes were constructed using standard molecular biology techniques following a well-established stepwise construction. The gene encoding protein GB1 was a generous gift from David Baker of University of Washington. The gene encoding the DNA sequence of GB1 was in the pUC19 plasmid. In pUC19-GB1, GB1 was flanked by a 5' BamHI site and a 3' BglII site followed by a KpnI site. One 15 amino acid consensus resilin repetitive sequence 158  (GGRPSDSYGAPGGGN) from the first exon of the Drosophila melanogaster gene CG15920 [122] was used in this study to construct GB1-resilin based polyproteins. The DNA sequence of resilin, flanked with a 5’ BamHI restriction site and 3’ BglII and KpnI restriction sites, was synthesized by PCR (polymerase chain reaction) based oligonucleotide assembly. The genes encoding CFP and YFP were generous gifts from Robert Campbell of University of Alberta. One 39 amino acid peptide sequence, CCK (LGKVSALKEKVSALKEEVSANKEKVSALKEKVSALKELG) [303], was used as a linker sequence between the CFP and YFP in the FP. A 20 amino acid peptide sequence, xtz3, was also used as a linker [290]. The gene encoding xtz3 (KKGTGTGNPATGKGTGTGQE) was also a gift from Robert Campbell of University of Alberta. Also another peptide sequence, CCE (LGEVSALEKEVSALEKKNSALEKEVSALEKEVSALEKLG), which has been reported to form coiled-coil with CCK, was used (Scheme 6.1). The expression vector of pQE80L- GRG5RG4R (G stands for GB1 and R stands for resilin) was constructed by iterative cloning of G and R genes into empty pQE80L vector (with Hig-tag built in), on the basis of the identity of the sticky ends generated by BamHI and BglII restriction enzymes. (GR)2-CFP-CCK-YFP-(GR)2 was constructed in the same way. 159  To improve the expression of CCE, we insert the sequence of CCE into the sequence of wild type (wt) GB1. In order to facilitate the insertion, we first inserted a nonpalindromic AvaI restriction site (CTC GGG) between codons for the 39th and 40th amino acid residues of wt GB1. The insertion of the AvaI site was carried out using standard protocols of site-directed mutagenesis, resulting in pUC19-GB1(AvaI). The presence of the AvaI restriction site in purified clones was confirmed by restriction digestion followed by DNA sequencing. To construct the mutants GB1(AvaI)-CCE, the strands for DNA encoding the sequence of CCE flanked with the AvaI restriction site at both ends were synthesized separately by oligosynthesis (Integrated DNA Technologies). The DNA strands were mixed to a final concentration of 50 µM in 1mM Tris-HCl buffer, pH 8.0, and incubated at 50 °C for 10 min. The product was digested with restriction enzyme AvaI and ligated into pUC19-GB1(AvaI) digested with the same enzyme. The ligation mixture was transformed into Escherichia coli strain DH5α. The resulting GB1(AvaI)-CCE (CCE) contains the 39-residue insertion LG-CCE-LG between the 39th and 40th amino acid residues of wt GB1, respectively. The extra inserted LG are resulted from the AvaI restriction sites. A gene encoding the sequence of CCK flanked with the AvaI  160  restriction site at both ends was also synthesized by oligosynthesis and used in the construction of (GR)2-CFP-CCK-YFP-(GR)2.  Scheme 6.1. DNA and amino acid sequences of CCE and CCK, in which RS results from ligation of restriction site BglII and BamHI, LG results from handle sequence of primers for PCR, and C is added to the sequences for historical reansons.  6.2.2 Protein Expression and Purification The expression of polyproteins was carried out in Escherichia coli strain DH5α. Cultures were grown at 37 °C in 2.5% LB containing 100mg/L ampicillin, and induced with 0.8mM isopropyl-1-β-D-thiogalactoside (IPTG) when the optical density reached ~1. Protein expression continued for 5 hours. The cells were harvested by centrifugation at 4000rpm for 10min and cell lysis was done using lysozyme from egg white (100mM, SigmaAldrich). 161  The soluble fraction was purified using Ni²+ affinity chromatography. The yield of the polyprotein GRG5RG4R was in the range of 40mg to 50mg per liter of culture, while the yield of (GR)2-CFP-CCK-YFP-(GR)2 was in the range of 10mg to 20mg per liter of culture. The purity of the purified polyproteins was around 90%, as estimated by SDS-PAGE using AlphaEaseFC software (Version 4.0.0, Alpha Innotech Corporation, San Leandro, CA). The 10% “impurity” was likely to be truncated fragments of polyproteins. To prepare GB1-resilin-based biomaterials, the purified proteins were then dialyzed against deionized water for 3 days to remove all the salt from the elution buffer. During dialysis, the water was changed every 5 hours. The protein was lyophilized following dialysis.  6.2.3 Preparation of Biomaterials We used a well-developed [Ru(bpy)3]²+-mediated photochemical cross-linking strategy [365] to prepare GB1-resilin-based biomaterials. This photochemical strategy allows the cross-linking of two tyrosine residues that are in close proximity into a di-tyrosine adduct, and leads to rapid and quantitative formation of di-tyrosine cross-links between soluble proteins. To prepare hydrogels, lyophilized proteins were redissolved in PBS. In a typical experiment, 18 mg of the protein was weighted using analytical 162  balance and added to a microcentrifuge tube containing 84.4µl of PBS ( pH 7.4) and 4.5µl of ammonium persulfate (APS) (1M). The trapped air bubbles were removed by centrifugation at 1000 g for 5 minutes. 0.9µl of [Ru(bpy)3]²+ (20mM) solution was then added to the microcentrifuge tube and quickly mixed with the protein solution by tapping the bottom of the tube. The final solution contained 200mg/ml of protein, ~200µM [Ru(bpy)3]2+ and 50 mM APS in PBS buffer. The solution was cast into a custom-made plexiglass mold described in Chapter 3. The sample was then irradiated for 30 seconds using a 200 W fiber optical white light source. The irradiation source was 10 cm away from the mold. The ring was then taken out from the mold and stored in PBS buffer (with 0.05% (w/v) sodium azide). For samples containing 0.5% (GR)2-CFP-CCK-YFP-(GR)2, 0.09mg (GR)2-CFP-CCK-YFP-(GR)2 and 17.91mg GRG5RG4R was weighted and then cross-linked in the same way. The resultant rings showed a degree of swelling in PBS buffer. But, these chemically cross-linked biomaterials showed superior long term stability, and no noticeable erosion was observed in PBS buffer (in the presence of 0.5% azide) over a period of one year.  163  6.2.4 Fluorescence spectrum Fluorescence for (GR)2-CFP-CCK-YFP-(GR)2 was recorded on a Cary Eclipse Fluorescence spectrophotometer (Varian Inc.). The excitation wavelength was set as 430nm (YFP has no significant emission upon excitation at 430nm). The ratio of fluorescence intensity at 530nm over that at 480nm was calculated as an indication of the E [290].  6.2.5 Worm Like Chain Model of CCK peptide sequence The force generated by the CCK peptide sequence at corresponding distances were estimated using Worm Like Chain (WLC) Model [179, 366] shown as  F ( x) =  k bT [ p  1 4(1 −  x 2 ) Lc  −  1 x + ] (Equation 6.1) 4 Lc  in which, F and x stand for force and distance, respectively. kb is Boltzmann constant, T is absolute temperature, p is persistence length (the shortest rigid segment length) and Lc is contour length (the fully extended length). Since the CCK peptide sequence has 39 amino acids and each amino acid is 0.36nm, the Lc for CCK sequence equals 0.36nm/aa×39aa=14.4nm. Since the p is the shortest rigid segment length, the p of CCK sequence is set as 0.36nm [179, 366]. Thus, the WLC model for CCK becomes: 164  Fwlc (d ) =  4 .1 [ p  1 4(1 −  d 2 ) 14.4  −  d 1 + ] (Equation 6.2), 4 14.4  in which p and d are in nm and Fwlc in pN.  6.3 Results and Discussions 6.3.1 Design Considerations A tandem modular protein with a CFP/YFP FRET pair (FP) separated by a linker sequence named CCK (Figure 6.1), (GR)2-CFP-CCK-YFP-(GR)2, was designed and constructed. In the tandem modular protein, G and R represent GB1 domain and resilin structured sequences, respectively. And they are employed to provide mechanical integrity and sites for cross-linking, respectively. The CFP/YFP FP is a widely used FP in FRET measurments [290]. CCK is a peptide sequence with 39 amino acid and it is capable of coiling with CCE (another 39 amino acid peptide sequence) to form α-helix coiled coil [303]. The CCK is used here to serve as a flexible linker, capable of reflecting the distance between the FP fluorophores. As the length of CCK is changed under certain conditions, the distance between the CFP/YFP FP will change resulting in a change in the E. For example, CCK can form coiled coil with CCE, and end to end distance of the resultant coiled coil is longer than that of the expected end-to-end distance of CCK [303]. This will 165  give rise to an increased distance between FP chromophores, and thus a decreased E. This is also the reason that CCK is used as the linker between CFP/YFP FP in this study. There are also alternative methods to induce changes of CCK length, such as hydrogel swelling force.  Figure 6.1. Schematic illustration of how the FRET efficiency depends on the length of CCK linker sequences. Cyan and yellow cylinders stand for cyan fluorescent protein (CFP) and yellow fluorescent protein (YFP), respectively. Blue and green curves stand for CCK and CCE peptides, respectively.  (GR)2-CFP-xtz3-YFP-(GR)2 was constructed for calibration of the E, where xtz3 is a sequence with a different length from that of CCK and also serves 166  as a flexible linker between CFP/YFP FP[290]. Protein CCE was constructed to form coiled coil with (GR)2-CFP-CCK-YFP-(GR)2 for calibration of the E. (GR)2-CFP-(GR)2 was constructed as a control and also for calibration of the E at zero. An SDS-PAGE gel picture of these proteins was shown in Figure 6.2. All proteins are expressed with correct molecular weights. Polyprotein GRG5RG4R which has been reported previously [237] was also constructed. GRG5RG4R was used as the bulk matrix of the hydrogels to dilute the concentrations of CFP/YFP FP in the ultimate hydrogels so that FP are well separated from each other to avoid inter-molecule energy transfer. It is of note that with the current concentration of CFP/YFP FP in the hydrogels, there might still be inter-molecule energy transfer occurring, which might be a limitation of this study.  167  Figure 6.2 Coomassie blue stained SDS-PAGE picture for the constructed proteins (from left to right) (GR)2-CFP-xtz3-YFP-(GR)2, the broad range molecular weight marker (New England Biolabs), (GR)2-CFP-CCK-YFP-(GR)2, (GR)2-CFP-(GR)2, (GR)2-YFP-(GR)2, CCE and GRG5RG4R.  6.3.2 The designed tandem modular proteins are fluorescent and capable of functioning as FPs. The first thing to do is to check whether the fusion process with G and R will affect the fluorescence of CFP and YFP. Fluorescence spectroscopy experiments were performed on all of the tandem modular proteins containing fluorescence proteins.  168  Fluorescence Intensity  (GR)2-CFP-xtz3-YFP-(GR)2 (GR)2-CFP-CCK-YFP-(GR)2 (GR)2-CFP-CCK-YFP-(GR)2 coiled with CCE (GR)2-CFP-(GR)2  460  480  500  520  540  560  580  600  Wavelength (nm)  Figure 6.3. Fluorescence spectroscopy of (GR)2-CFP-xtz3-YFP-(GR)2, (GR)2-CFP-CCK-YFP-(GR)2,(GR)2-CFP-CCK-YFP-(GR)2 after coiling with CCE, (GR)2-CFP-(GR)2.  As shown in Figure 6.3, the CFP and YFP remain fluorescent. The FPs with different linker sequences exhibit different fluorescence profiles, indicating different E. In the present study, the intensity ratio of YFP emission at 530nm over CFP emission at 480nm (Iratio) is calculated and used as a parameter for comparison of the E (summarized in Table 6.1).  169  Table 6.1 The intensity ratio values. Protein  Intensity ratio  (GR)2-CFP- xtz3-YFP-(GR)2  1.20±0.00  (GR)2-CFP-CCK-YFP-(GR)2  0.59±0.05  (GR)2-CFP-CCK-YFP-(GR)2 coiled with CCE  0.47±0.01  (GR)2-CFP-(GR)2  0.37±0.01  The Iratio of (GR)2-CFP-CCK-YFP-(GR)2 is calculated to be ~0.59. Since the Iratio of CFP is around ~0.37, the ratio bigger than that indicates the existence of FRET. This suggests that in PBS buffer solution the length of the linker CCK is in the range (<10nm) in which FRET can happen. It is of note that for (GR)2-CFP-(GR)2, we got a Iratio of 0.37, lower than previously reported 0.5 [290], this might be due to differences in measurement conditions. Nevertheless, the fact that the ratio of (GR)2-CFP-CCK-YFP-(GR)2 is higher than that of (GR)2-CFP-(GR)2, proves the occurrence of FRET. FRET also occurs in (GR)2-CFP-xtz3-YFP-(GR)2, and because the linker sequence xtz3 (20aa) is shorter than CCK (39 aa), (GR)2-CFP- xtz3-YFP-(GR)2 has a higher E than (GR)2-CFP-CCK-YFP-(GR)2 as it would be expected. There is an advantage of using CCK peptide as the linker between CFP and YFP in our study. There have been studies showing that the CCK can form 170  coiled-coil with CCE and the end-to-end distance of the CCK will be extended after forming coiled-coil. Fluorescence measurement was also performed on (GR)2-CFP-CCK-YFP-(GR)2 in the presence of CCE. It was observed that the Iratio of (GR)2-CFP-CCK-YFP-(GR)2 droped to ~0.49 after adding CCE. Meanwhile, CCE did not have any effect on (GR)2-CFP-(GR)2 (Figure 6.4) or (GR)2-YFP-(GR)2. This suggests the decreasing E is due to increase in the distance between the FP. The decreasing ratio indicates lower E, and thus longer linker length. This agrees with previous reports that the end-to-end distance of CCK will be extended after forming coiled-coil with CCE [303]. Both the difference in Iratio of FPs with different linker and the change in Iratio after coiled coil formation demonstrates that the designed  Fluorescence Intensity  tandem modular proteins containing FP are capable of proper functioning.  (GR)2-CFP-(GR)2 (GR)2-CFP-(GR)2 with CCE  460  480  500  520  540  560  580  600  Wavelength (nm)  Figure 6.4. The Iratio of (GR)2-CFP-(GR)2 remains the same after addition of CCE, indicating CCE did not have any effect on (GR)2-CFP-(GR)2. 171  6.3.3 Hydrogels constructed from the tandem modular protein containing FP are fluorescent. Next step is to make hydrogels from these designed tandem modular protein containing FPs. A prerequisite is that the fluorescence profile of the (GR)2-CFP-CCK-YFP-(GR)2 will not be affected by the components for cross-linking reactions or by the fraction of the cross-linked matrix. Fluorescence spectroscopy of (GR)2-CFP-CCK-YFP-(GR)2 was measured before and after inducing cross-linking with APS and the Ru complex (Figure 6.5). The Iratio was calculated to be ~0.62 before and ~0.63 after adding cross-linking matrix, indicating that cross-linking did not affect the fluorescence intensity ratio very much. It is of note that a broad emssion peak appears due to the Ru(II) complex [367]. However, due to different emission wavelengths, the effect of this emission on the Iratios measurement is negligible.  172  Fluorescence Intensity  (GR)2-CFP-CCK-YFP-(GR)2 (GR)2-CFP-CCK-YFP-(GR)2 with APS, Ru(II) added  450  500  550  600  650  700  Wavelength (nm)  Figure 6.5. Fluorescence spectroscopy of (GR)2-CFP-CCK-YFP-(GR)2 before and after adding cross-linking matrix. Fluorescence intensity was normalized to give similar emission at 480nm. The Iratio was calculated to be ~0.63 after adding cross-linking matrix, close to that of Iratio (~0.62) before adding cross-linking matrix. It is of note that a broad emission peak appears due to the Ru(II) complex [367].  After confirming that cross-linking matrix did not affect fluorescence, we used the Ru(II)-mediated photochemical cross-linking method to cast the materials into solid hydrogels. And the hydrogels are fluorescent under CFP and YFP excitation (Figure 6.6). Fluorescence measurements were performed on 15 samples at 3 different positions for each sample. The average Iratio is calculated to be 0.55±0.03. This Iratio of hydrogels containing (GR)2-CFP-CCK-YFP-(GR)2 in PBS is lower than that of 173  (GR)2-CFP-CCK-YFP-(GR)2 protein solution. As mentioned previously in Chapter 1, the E is dependent on both distance and orientation of fluorophores. Therefore, the decrease in Iratio might be caused by an increase in the end-to-end distance of the CCK sequence. The decrease in Iratio may be also be caused by the change in orientation of the CFP/YFP FP since the orientation might be constrained by the network structure of the hydrogel as well as swelling force generated by the hydrogel. Researchers using other fluorescent protein pairs under tension reported that protein conformational change did not affect the E [269]. There might still be other orientation constraints. Nevertheless, decrease in the E was related to elongation of linker sequences [269]. Since FRET is highly sensitive to the distance between fluorophores, the decrease in Iratio observed here indicates that the CCK sequence is extended, to some extent, to a longer end-to-end distance. In previous studies, it was noticed that the hydrogels made from GRG5RG4R showed some degree of swelling [237]. It is widely accepted that polymer chains will be extended during swelling. Therefore, the end-to-end distance of the CCK will be elongated due to hydration/swelling of the hydrogels, which is consistent with the decreased FRET E in the hydrogels. This suggests (GR)2-CFP-CCK-YFP-(GR)2 is somehow extended in hydrogels due to hydration. And this gives rise to the so called pre-extension force. It is 174  of note that the Iratio of hydrogels containing (GR)2-CFP-CCK-YFP-(GR)2 in PBS is between that of (GR)2-CFP-CCK-YFP-(GR)2 protein solution before and after coiling with CCE. Considering our previous assumption that FRET is relatively highly sensitive to the distance between fluorophores, the end-to-end distance of the CCK in hydrogels in PBS would be between that of the CCK in (GR)2-CFP-CCK-YFP-(GR)2 protein solution before and after coiling with CCE.  Figure 6.6. Photographs of hydrogels under fluorescence.  6.3.4 Force on single polymer chain of hydrogel can be reflected by the E. Force generated by a polymer chain is distance dependent. In order to measure the pre-extension force, it is important to know the pre-extension length of polymer chain. It is known that the E change as the distance 175  between FP changes. Thus, it is possible to derive length from the E. Therefore, through length, a relation between the E and force can be obtained.  6.3.4.1 Estimating length from the E Distance r between FPs can be derived precisely from E based on the relation r=R0[(1/E)-1]1/6. However, for present, we were only able to calculate the FRET Iratio as an indication of the E, we could not calculate exact value of the E and thus could not calculate exact value of the r. This is another limitation of this study. A possible solution to this problem is to estimate the lengths of the linker sequence in certain extreme situations under some assumptions to get approximate values, and then correlated the approximate length with the calculated FRET Iratio. In order to estimate the end-to-end distance of the CCK sequence, a Gaussian Chain Model is applied. It is of note that the Gaussian Chain Model describes the behaviour of ideal chain (freely joint chain) of polymers, which treats a polymer as a true “random coil” (freely jointed) without any interactions between monomers [368]. It might not be appropriate to apply the Gaussian Chain Model to the CCK sequence since joining of adjacent amino acid residues of the CCK sequence is fixed by peptide bonds. Even 176  though the Gaussian Chain Model might not be a good model to estimate the end-to-end distance of the CCK sequence, considering the relative flexibility of the CCK sequence, the Gaussian Chain Model might be used under the assumption that the CCK sequence presents as a true “random coil” (freely joint chain) in certain extreme situations. According to the Gaussian Chain Model, for an ideal chain composed of N segment with length l, the root mean square end-to-end distance equals to l× N . Thus, the end-to-end distance of CCK peptide equals 0.36nm× 39 =2.25nm. It is certain that in the CFP/YFP FP fused protein, the CCK sequence might present in a different confirmation other than a true “random coil”. Therefore, the actual end-to-end distance of the CCK sequence might deviate from the estimated 2.25nm. For the present study, we used 2nm as an approximate end-to-end distance of CCK before coiling with CCE. In order to estimate the length of CCK sequence after coiling with CCE, it is assumed that the coiled coil of CCE and CCK presents as α-helix [303]. Each turn of α-helix contains 3.6 amino acid residues and the lengh of each turn is ~0.54nm. Thus, the length of the CCE and CCK coiled coil equals  39residues 0.54nm × = 5.85nm . There might be other secondary 3.6residues turn turn structural components present in the coiled coil of CCE and CCK. Therefore, 177  the actual length of the coiled coil of CCE and CCK might deviate from the estimated 5.85nm. For the present study, we used 6nm as an approximate end-to-end distance of CCK after coiling coil with CCE. As mentioned above, the Iratio of hydrogels containing (GR)2-CFP-CCK-YFP-(GR)2 in PBS is between that of (GR)2-CFP-CCK-YFP-(GR)2 protein solution before and after coiling with CCE, suggesting that the end-to-end distance of CCK in hydrogels in PBS would be of some value between 2nm and 6nm. The result is shown in Figure 6.7. Red spots stand for proteins with CFP/YFP separated by xtz3, CCK, CCK/CCE coiled coil and protein with only CFP. Black triangles stand for the Iratio of hydrogels containing (GR)2-CFP-CCK-YFP-(GR)2.  Intensity ratio (530nm/430nm)  0  2  4  6  8  10  1.2  1.2 xtz3 1.6nm  1.0  1.0  0.8  0.8 (GR)2-CFP-CCK-YFP-(GR)2 hydrogels in PBS  0.6  0.6  CCK 2nm  0.4  0.4  CCK/CCE 6nm  0.2  0.2  CFP only set as10nm  0.0  0.0 0  2  4  6  8  10  Distance between CFP and YFP (nm)  Figure 6.7. Relation of Iratio of (GR)2-CFP-xtz3-YFP-(GR)2, (GR)2-CFP-CCK-YFP-(GR)2, (GR)2-CFP-CCK-YFP-(GR)2 after coiling with CCE, (GR)2-CFP-(GR)2 with linker lengths. 178  6.3.4.2 Estimating Force from length WLC Model is widely used to describe how the behavior of polymers upon stretching (Figure 6.8). From the WLC model, the force of CCK at extensions of 2nm and 6nm was calculated to be 3pN and 10pN, respectively. Since the root mean square end to end distance of CCK in hydrogels in PBS would be between 2nm and 6nm, we can refer that the force on (GR)2-CFP-CCK-YFP-(GR)2 in hydrogels would be in the range of [3,10] pN. It is of note that since the end to end distance of CCK was estimated under certain assumptions, the force calculated based on the approximate end to end distance of CCK might deviate from actual force. In addition, the force was calculated based on the WLC model which might not be exactly the real force to arise from the linker sequences. Further studies are needed to obtain more accurate force values. Nevertheless, our results suggest an approximate range of the force on single polymer chain in the designed swollen hydrogels.  179  400  Force (pN)  300  200  100  0 0  2  4  6  8  10  12  14  Extension (nm)  Figure 6.8. Force-extension relation of CCK peptide sequence derived from WLC model.  Through the study of a FRET pair embedded hydrogel system, we investigated the swelling effect on single polymer chain length and demonstrated the pre-extension force on single polymer chain due to hydration. The change of the E upon swelling of hydrogels not only demonstrates that the tandem modular protein based hydrogels are fluorescently functionalized, it also provides direct evidence of swelling effect on single polymer chain length upon swelling of bulk materials, for the first time, to our knowledge. In order to optimally design hydrogels, extensive research has been done toward understanding the mechanism of swelling and drying kinetics. Previous hydrogel studies have explored the volume change of gels during 180  swelling and drying. For example, fluorescence techniques based on fluorophores attached to a monomer [369] or absorbed/ desorbed by porous gels [370] have been developed for visualization of deformation during swelling and drying. There were also studies on the effect of a solvent on a polymer and orientations of polymer chains using Nuclear Magnetic Resonance (NMR) [371, 372] and Magnetic Resonance Imaging (MRI) techniques [373]. It is widely accepted that during swelling polymer chains will be extended. However, there is limited experimental information available regarding the effects of swelling on single polymer chain length. In this study, FRET Iratio was measured, as an indication of the E, on swollen materials. Distances between FPs embedded in the hydrogels were reflected by the E. It is known that the E changes as the distance between FP changes following the equation r=R0[(1/E)-1]1/6. Thus, as the hydrogels swelled, the distance between FP increased, which corresponding to the extension of the end-to-end distance of the CCK linker sequence. This result, for the first time, provides direct evidence of extended polymer chain upon swelling. The FRET pair cross-linked into the hydrogel is essential for studying the swelling effect on single polymer chain. There have been studies on polymer swelling/de-swelling using dyes [369, 370, 374]. But most often the dyes are uncross-linked, which may introduce the problem of diffusion and such dyes 181  would not be stretched through the gel during gel swelling experiment. Therefore, the present study with a FP cross-linked into the hydrogel is a better approach than previous studies on polymer swelling/de-swelling. Our results also suggest that for materials with swelling properties (such as hydrogels), when the materials are kept in buffer solutions, the  materials  might have already been extended due to the swelling force before the actual mechanical tests (pre-extension). The demonstration of pre-extension force is very important in studies on swelling effects on mechanical properties. First, one assumption in previous studies, that materials are under minimal residual strain in fully hydrated samples [375], might no longer stand true for swollen samples considering the pre-extension force. Second, our study provides a solution to one difficulty in previous studies on swelling effect on mechanical properties, which is measurement of rest length (Linitial) of samples in different hydration status. Researchers have to use a regression procedure to get Linitial values [315].The pre-extension length derived from the E might provide another method to measure Linitial. Third, including the pre-extension force, when comparing mechanical properties of materials in different hydration status, might provide better understanding of swelling effect. Most importantly, the demonstration of pre-extension force here is the first report of a swelling effect on a single polymer chain, to our knowledge. 182  6.4 Conclusions Through a well developed photochemical cross-linking method, a hydrogel constructed from tandem modular protein containing CFP/YFP FRET pair was designed. The designed hydrogels showed fluorescence and FRET upon swelling. In this study of the FRET pair embedded hydrogels, we investigated the swelling effect on single polymer chain length and demonstrated the pre-extension force on single polymer chain due to hydration. This study not only provides further evidence of designing functional and multi-functional hydrogels based on tandem modular functional proteins, but also provides a novel method for hydrogel immobilization of functional proteins that remain active. In addition, this study, providing direct evidence of swelling effect on single polymer chain length of the first time, is a "proof of concept" study of the widely accepted concept that during swelling polymer chains will be extended. Moreover, our results suggest that the E can reflect the force applied on single polymer chain. This result implies possible applications of the FRET pair embedded hydrogels as force sensors. Since FRET strategy can be used to study biological phenomenon involving changes in the distance between the fluorophores [281-283], we also anticipate to seeing applications of the fluorescent hyrogels as biosensors. 183  CHAPTER 7 Conclusions and Future Directions 7.1 Conclusions Protein-based hydrogels have attracted great interest in developing biomaterials for biomedical applications including drug delivery and tissue engineering [1-9]. Tandem modular construction is one of the common features of natural elastomeric proteins, making tandem modular proteins that consist of many individually folded functional domains promising building blocks for constructing novel biomaterials. Incorporating tandem modular folded proteins into hydrogels will be very important in order to design optimal biomaterials for a variety of biomedical applications.  We have developed a novel two-component approach to engineer tandem modular protein-based extracellular matrix-mimetic hydrogels. This method makes use of two complementary leucine zipper sequences (CCE andCCK). The two sequences do not self-associate, but upon mixing they will assemble into heterodimeric coiled-coils. We constructed two tandem modular proteins, one carrying CCE and the other one carrying CCK. These two proteins can be expressed and purified under native conditions with relatively high yield. The two proteins can form hydrogels upon mixing. The resultant hydrogel can undergo reversible sol-gel transitions dependent on 184  temperature. And the erosion properties are improved comparing to our first generation of tandem modular protein-based hydrogels. This method provides a new approach to tune properties of hydrogels and allows the use of large native extracellular proteins to engineer extracellular matrix-mimetic hydrogels [214].  Going a step further, we employed a photochemical cross-linking strategy is to covalently cross-link engineered artificial elastomeric protein to produce tandem modular protein-based biomaterials. We designed artificial elastomeric proteins that combine folded tandem modular proteins GB1 with random coil-like sequences from resilin mimicking the molecular architecture of the giant muscle protein titin. We showed that these artificial elastomeric proteins can be photochemically cross-linked and cast into solid biomaterials. The mechanical properties of these biomaterials are comparable to the passive elasticities of muscles in the physiological range of sarcomere length. In addition, the mechanical properties of these biomaterials can be tuned by adjusting the formation conditions and the composition of the elastomeric proteins. These biomaterials represent a new class of muscle-mimetic biomaterial. We anticipate that these biomaterials will find applications in tissue engineering [237]. 185  The photochemical cross-linking strategy to design tandem modular protein-based hydrogels also offers the possibility of using natural tandem modular ECM proteins in order to construct novel hydrogels for biomedical applications. We used the third fibronectin type III (FnIII) domain of tenascin-C (TN-C, TNCfn3)) as building blocks to construct ECM mimetic hydrogels. Our results demonstrate that tandem modular proteins containing TNfn3 can be cross-linked into hydrogels. And the hydrogel can support the spread of human lung fibroblast cells. Our hydrogel provides a platform to provide an artificial environment for cells by mimicking the ECM [354].  Our results on TNfn3-based hydrogels also indicate that the RGD motifs of the protein remain active in hydrogels, which makes the photochemical cross-linking strategy promising method to incorporate functional protein domain(s) into hydrogels using a modular approach to design functional hydrogels. To prove this, we designed an enzymatic hydrogel using BCX which catalyzes the hydrolysis of xylans. We designed a polyprotein G-R-G-BCX-G4-R. The designed tandem modular protein fused with BCX can retain partial enzymatic activity of non-fusion BCX. The protein can easily be cross-linked to hydrogel using the Ru(II)-mediated covalent 186  cross-linking method. The resultant hydrogels exhibit enzymatic activity and can retain activity after repeatedly tests, indicating their potential in industrial applications.  All of the above results demonstrated that the tandem modular protein-based technique making use of tandem modular protein domains as building blocks to construct novel hydrogels offers a possible method to design hydrogels with multiple functionalities. We further proved this feasibility using hydrogels constructed from fluorescent protein pairs as force sensors that are capable of measuring swelling forces on single peptide chains at the picoNewton level. We construct a tandem modular protein based on Cyan Fluorescent Protein (CFP) and Yellow Fluorescent Protein (YFP) with a flexible linker in between. Hydrogels are constructed through the Ru(II)-mediated covalent cross-linking method. The designed hydrogels showed fluorescence and FRET upon swelling. The force on single polymer chain due to hydration was estimated. Our study is a "proof of concept" study of the widely accepted concept that during swelling polymer chains will be extended and provides direct evidence of swelling effect on single polymer chain length for the first time to our knowledge.  187  In conclusion, we explored the feasibility and present five examples of using tandem modular proteins for constructing protein-based biomaterials using artificial tandem modular proteins as model systems. This new tandem modular protein-based approach provides a new tool set in the design of biomaterials that better mimic the natural environment of ECM for various biomedical applications and basic biological studies. However, there are still some questions unresolved which will need more studies in the future.  7.2 Future Directions 7.2.1 Direct proof of GB1 unfolding As we discussed in Chapter 3, we observed hysteresis in the stress-strain curves. The hysteresis indicates that there is non-covalent bond breaking involved during stretching at higher strain. Meanwhile, the stress-strain curves during consecutive cycles are identical, suggesting reversible bond-breaking. Through analogy to Ig domains in muscle proteins, we proposed that the observed hysteresis is probably a result of the unfolding of a small number of GB1 domains. However, direct proof is still needed. One way to provide direct evidence is to investigate the hysteresis and the time course of hysteresis and try to match the hysteresis recovery rate to the refolding rate of GB1 domains. The problem now is the refolding of GB1 188  domains is so fast that it is beyond the dead time of the Instron tensile test instrument we are using. As a result, the hysteresis recovery rate of the resultant materials could not be measured. So what we will do is to slow down the refolding process. Studies show that a GB1 mutant G(D46A) shows a folding rate ~20-fold slower (from 412 s-1 to 21.8 s-1) but an almost unaffected unfolding rate [376]. Toward this goal, we are going to construct similar artificial titin-mimeticpolyproteins replacing GB1 with G(D46A). Toward this goal, we are going to construct similar artificial titin-mimetic polyproteins replacing GB1 with the mutant. If the hysteresis recovery rate could be measured and matched to the refolding rate of the mutant, it will be direct evidence towards proving that the hysteresis is indeed caused by to unfolding and refolding of a small number of GB1 domains.  7.2.2 Improvement of the activity of the enzymatic hydrogels As we discussed in Chapter 5, the fusion and cross-linking process affect the enzymatic activity of the enzyme BCX. However, the reason for the decreasing in activity is not clear for now. First, the kcat/Km of our designed protein G-R-G-BCX-G4-R is smaller than previously reported kcat/Km of wild type BCX. The decrease in the kcat/Km might be caused by the interaction of GB1 and resilin with the BCX. The decrease might also be 189  caused by steps in protein preparation, such as lyophilization and redissolving with vortex. In order to test for the possibility of lyophilization, control assays using non-lyphilized protein samples will be needed. If the kcat/Km value of non-lyophilized G-R-G-BCX-G4-R is comparable to that of previously reported kcat/Km of BCX, the result will suggest that the lyophilization step in protein preparation will indeed cause a decrease in the kcat/Km. Otherwise, the decrease in the kcat/Km is caused by other reasons. Second, the enzyme reaction rate of BCX was apparently decreased with a smaller rate constant value after being cross-linked to hydrogels. As discussed previously, the decrease could be caused by limitations arised from diffusion in the hydrogel network, changes in enzyme conformations as well as changes in optimum conditions for enzymatic reactions. In order to test the possibility of limited diffusion, experiments can be performed to monitor the diffusion rates of the substrate and product using UV/Vis spectroscopy since both the substrate and product have characteristic absorbance. An alternative method may be to fluorescently lable substrate and product, then fluorescent microscopy can be performed to determine the distribution of the substrate and product, and thus the reaction site of the enzymes in the hydrogels. In order to test the possibility of changes in optimum conditions for enzymatic reactions, activity assays can be performed at different pH and 190  temperature. And the activity at different conditions can be compared to find the optimum reaction conditions for the enzymatic hydrogels.  7.2.3 Increase in the precision of force estimate using the FRET based force sensor As we discussed in Chapter 6, there are several limitations of our designed force sensors based on FRET hydrogels. First, even though we have tried to avoid inter-molecule energy transfer through using GRG5RG4R as bulk matrix of the hydrogels to dilute the concentrations of CFP/YFP FP in the ultimate hydrogels so that FP is well separated from each other, there might still be inter-molecule energy transfer occurring with the current concentration of CFP/YFP FP in the hydrogels. More studies will be needed in order to find the concentration at which there will be no inter-molecule energy transfer. For example, based on the concentration of the proteins in the hydrogels, assuming the proteins are evenly distributed in the hydrogels, the distance between protein molecules can be estimated. The protein concentrations that can eliminate inter-molecule FRET would be the concentrations at which the distance between protein molecules is larger than 10 nm since FRET is only sensitive within [2, 10] nm. Second, for the study describe in Chapter 6, we used intensity ratio as a parameter for comparison 191  of the FRET efficiency. Exact values of FRET efficiency will be important in order to derive accurate distances between fluorophores. In this way, the distance will not have to be estimated using Gaussian Chain Model or other models which might not be appropriate for the linker sequences. Third, the force generated by the linker sequences was calculated based on Worm Like Chain Model which might not be exactly the real force to arise from the linker sequences. It will be important to measure the force generated by the linker sequence using single molecule force spectroscopy, such as AFM, to provide more accurate force values.  192  REFERENCES  [1] Banta S, Wheeldon IR, Blenner M. Protein Engineering in the Development of Functional Hydrogels. Annual Review of Biomedical Engineering 2010;12:167-86. [2] Langer R, Tirrell DA. Designing materials for biology and medicine. Nature 2004;428:487-92. [3] Maskarinec SA, Tirrell DA. Protein engineering approaches to biomaterials design. Current Opinion in Biotechnology 2005;16:422-6. [4] van Hest JCM, Tirrell DA. Protein-based materials, toward a new level of structural control. Chemical Communications 2001:1897-904. [5] C. Xu JK. Self-Assembling Hydrogels. Polymer Bulletin 2007;58:53-63. [6] Petka WA, Harden JL, McGrath KP, Wirtz D, Tirrell DA. Reversible Hydrogels from Self-Assembling Artificial Proteins. Science 1998;281:389-92. [7] Wang C, Stewart RJ, KopeCek J. Hybrid hydrogels assembled from synthetic polymers and coiled-coil protein domains. Nature 1999;397:417-20. [8] Kopecek J. Hydrogel biomaterials: A smart future? Biomaterials 2007;28:5185-92. [9] Kopecek J, Yang J. Peptide-directed self-assembly of hydrogels. Acta Biomaterialia 2009;5:805-16.  193  [10] Balakrishnan B, Banerjee R. Biopolymer-Based Hydrogels for Cartilage Tissue Engineering. Chemical Reviews 2011;111:4453-74. [11] Drury JL, Mooney DJ. Hydrogels for tissue engineering: scaffold design variables and applications. Biomaterials 2003;24:4337-51. [12] Ratner B HA, Schoen F, Lemons J. . Biomaterials Science: An Introduction to Materials in Medicine.: San Diego, CA: Elsevier Academic Press; 1996. [13] Liu H, Lin J, Roy K. Effect of 3D scaffold and dynamic culture condition on the global gene expression profile of mouse embryonic stem cells. Biomaterials 2006;27:5978-89. [14] Battista S, Guarnieri D, Borselli C, Zeppetelli S, Borzacchiello A, Mayol L, et al. The effect of matrix composition of 3D constructs on embryonic stem cell differentiation. Biomaterials 2005;26:6194-207. [15] Baharvand H, Azarnia M, Parivar K, Ashtiani SK. The effect of extracellular matrix on embryonic stem cell-derived cardiomyocytes. Journal of Molecular and Cellular Cardiology 2005;38:495-503. [16] Silvia Sihui Chen RPR, Joshua Zimmerberg and Leonid Margolis. Differentiation of Rhesus Monkey Embryonic Stem Cells in Three-Dimensional Collagen Matrix Methods in Molecular Biology 2006;330:431-43. [17] Engler AJ, Sen S, Sweeney HL, Discher DE. Matrix Elasticity Directs Stem Cell Lineage Specification. Cell 2006;126:677-89.  194  [18] Humphries JD, Byron A, Humphries MJ. Integrin ligands at a glance. Journal of Cell Science 2006;119:3901-3. [19] Vogel V, Sheetz MP. Cell fate regulation by coupling mechanical cycles to biochemical signaling pathways. Current Opinion in Cell Biology 2009;21:38-46. [20] Baneyx G, Baugh L, Vogel V. Fibronectin extension and unfolding within cell matrix fibrils controlled by cytoskeletal tension. Proceedings of the National Academy of Sciences 2002;99:5139-43. [21] Vogel V. Mechanotransduction Involving Multimodular Proteins: Converting Force into Biochemical Signals. Annual Review of Biophysics and Biomolecular Structure 2006;35:459-88. [22] Indik Z, Yeh H, Ornstein-Goldstein N, Sheppard P, Anderson N, Rosenbloom JC, et al. Alternative splicing of human elastin mRNA indicated by sequence analysis of cloned genomic and complementary DNA. Proceedings of the National Academy of Sciences 1987;84:5680-4. [23] Urry DW. Entropic elastic processes in protein mechanisms. II. Simple (passive) and coupled (active) development of elastic forces Journal of Protein Chemistry 1988;7:81-114. [24] Gosline J, Lillie M, Carrington E, Guerette P, Ortlepp C, Savage K. Elastic proteins: biological roles and mechanical properties. Philosophical Transactions of the Royal Society of London Series B: Biological Sciences 2002;357:121-32.  195  [25] Vrhovski B, Weiss AS. Biochemistry of tropoelastin. European Journal of Biochemistry 1998;258:1-18. [26] Urry DW, Hugel T, Seitz M, Gaub HE, Sheiba L, Dea J, et al. Elastin: a representative ideal protein elastomer. Philosophical Transactions of the Royal Society of London Series B: Biological Sciences 2002;357:169-84. [27] M. Kajitani YW, H. Xie, M. T. Hinds, S. W. Shalaby,K. R. Swartz and K. W. Gregory. ASAIO J 2000;46:409-14. [28] Schoen FJ, Levy RJ. Tissue heart valves: Current challenges and future research perspectives. Journal of Biomedical Materials Research 1999;47:439-65. [29] Kakisis JD, Liapis CD, Breuer C, Sumpio BE. Artificial blood vessel: The Holy Grail of peripheral vascular surgery. Journal of vascular surgery : official publication, the Society for Vascular Surgery 2005;41:349-54. [30] Gilad E. Amiel MK, Oz Shapira, James J. Yoo, Saami Yazdani, Joel Berry, Sunjay Kaushal, Joyce Bischoff, Anthony Atala, and Shay Soker. Engineering of Blood Vessels from Acellular Collagen Matrices Coated with Human Endothelial Cells Tissue Engineering 2006;12:2355-65. [31] Daamen WF, Veerkamp JH, van Hest JCM, van Kuppevelt TH. Elastin as a biomaterial for tissue engineering. Biomaterials 2007;28:4378-98. [32] Hinds MT, Courtman DW, Goodell T, Kwong M, Brant-Zawadzki H, Burke A, et al. Biocompatibility of a xenogenic elastin-based biomaterial in a murine  196  implantation model: The role of aluminum chloride pretreatment. Journal of Biomedical Materials Research Part A 2004;69A:55-64. [33] Paule WJ, Bernick S, Strates B, Nimmi ME. Calcification of implanted vascular tissues associated with elastin in an experimental animal model. Journal of Biomedical Materials Research 1992;26:1169-77. [34] Nimni ME, Myers D, Ertl D, Han B. Factors which affect the calcification of tissue-derived bioprostheses. Journal of Biomedical Materials Research 1997;35:531-7. [35] Urry DW. Characterization of soluble peptides of elastin by physical techniques. Methods in Enzymology 1982;82: 673–716. [36] Urry DW, Luan CH, Parker TM, Gowda DC, Prasad KU, Reid MC, et al. Temperature of polypeptide inverse temperature transition depends on mean residue hydrophobicity. Journal of the American Chemical Society 1991;113:4346-8. [37] McPherson DT, Morrow C, Minehan DS, Wu J, Hunter E, Urry DW. Production and Purification of a Recombinant Elastomeric Polypeptide, G-(VPGVG)19-VPGV, from Escherichia coli. Biotechnology Progress 1992;8:347-52. [38] Urry DW, Gowda DC, Parker TM, Luan C-H, Reid MC, Harris CM, et al. Hydrophobicity scale for proteins based on inverse temperature transitions. Biopolymers 1992;32:1243-50.  197  [39] Yamaoka T, Tamura T, Seto Y, Tada T, Kunugi S, Tirrell DA. Mechanism for the Phase Transition of a Genetically Engineered Elastin Model Peptide (VPGIG)40 in Aqueous Solution. Biomacromolecules 2003;4:1680-5. [40] Fong E, Tzlil S, Tirrell DA. Boundary crossing in epithelial wound healing. Proceedings of the National Academy of Sciences;107:19302-7. [41] Di Zio K, Tirrell DA. Mechanical Properties of Artificial Protein Matrices Engineered for Control of Cell and Tissue Behavior. Macromolecules 2003;36:1553-8. [42] Vincent JFV, Wegst UGK. Design and mechanical properties of insect cuticle. Arthropod Structure &amp; Development 2004;33:187-99. [43] Welsh ER, Tirrell DA. Engineering the extracellular matrix: a novel approach to polymeric biomaterials. I. Control of the physical properties of artificial protein matrices designed to support adhesion of vascular endothelial cells. Biomacromolecules 2000;1:23-30. [44] Zhang K, Diehl MR, Tirrell DA. Artificial Polypeptide Scaffold for Protein Immobilization. Journal of the American Chemical Society 2005;127:10136-7. [45] Lim DW, Nettles DL, Setton LA, Chilkoti A. In Situ Cross-Linking of Elastin-like Polypeptide Block Copolymers for Tissue Repair. Biomacromolecules 2007;9:222-30.  198  [46] Andrew Mackay J, Chilkoti A. Temperature sensitive peptides: Engineering hyperthermia-directed therapeutics. International Journal of Hyperthermia 2008;24:483-95. [47] Shamji MF, Betre H, Kraus VB, Chen J, Chilkoti A, Pichika R, et al. Development and characterization of a fusion protein between thermally responsive elastin-like polypeptide and interleukin-1 receptor antagonist: Sustained release of a local antiinflammatory therapeutic. Arthritis & Rheumatism 2007;56:3650-61. [48] Chilkoti A, Dreher MR, Meyer DE, Raucher D. Targeted drug delivery by thermally responsive polymers. Advanced Drug Delivery Reviews 2002;54:613-30. [49] Holmes TC. Novel peptide-based biomaterial scaffolds for tissue engineering. Trends in biotechnology 2002;20:16-21. [50] Dreher MR, Simnick AJ, Fischer K, Smith RJ, Patel A, Schmidt M, et al. Temperature Triggered Self-Assembly of Polypeptides into Multivalent Spherical Micelles. Journal of the American Chemical Society 2007;130:687-94. [51] Straley KS, Heilshorn SC. Independent tuning of multiple biomaterial properties using protein engineering. Soft Matter 2009;5:114-24. [52] Li L, Charati MB, Kiick KL. Elastomeric polypeptide-based biomaterials. Polymer Chemistry 2010;1:1160-70. [53] Dana L. Nettles KK, Neil A. Hanson, Charlene M. Flahiff, Brian A. Mata, Edward W. Hsu, Ashutosh Chilkoti, and Lori A. Setton. In Situ Crosslinking  199  Elastin-Like Polypeptide Gels for Application to Articular Cartilage Repair in a Goat Osteochondral Defect Model Tissue Engineering Part A 2008;14:1133-40. [54] Martin L, Alonso M, Girotti A, Arias FJ, Rodriguez-Cabello JC. Synthesis and Characterization of Macroporous Thermosensitive Hydrogels from Recombinant Elastin-Like Polymers. Biomacromolecules 2009;10:3015-22. [55] Aigner T, Stove J. Collagens-major component of the physiological cartilage matrix, major target of cartilage degeneration, major tool in cartilage repair. Advanced Drug Delivery Reviews 2003;55:1569-93. [56] Dietrich F, Lelkes P. Fine-tuning of a three-dimensional microcarrier-based angiogenesis assay for the analysis of endothelial-mesenchymal cell co-cultures in fibrin and collagen gels. Angiogenesis 2006;9:111-25. [57] Jürgen Brinckmann HNaPKM. Collagen: Primer in Structure, Processing and Assembly: Springer-Verlag:Berlin; 2005. [58] Jonker AM, Lowik DWPM, van Hest JCM. Peptide- and Protein-Based Hydrogels. Chemistry of Materials 2012;24:759-73. [59] Kawamura S, Wakitani S, Kimura T, Maeda A, Caplan AI, Shino K, et al. Articular cartilage repair. Rabbit experiments with a collagen gel-biomatrix and chondrocytes cultured in it. Acta Orthop Scand 1998;69:56-62.  200  [60] Wakitani S, Kimura T, Hirooka A, Ochi T, Yoneda M, Yasui N, et al. Repair of rabbit articular surfaces with allograft chondrocytes embedded in collagen gel. Journal of Bone & Joint Surgery, British Volume 1989;71-B:74-80. [61] Shigeyuki Wakitani TG, Randell G. Young, Joseph M. Mansour, Victor M. Goldberg, and Arnold I. Caplan. Repair of Large Full-Thickness Articular Cartilage Defects with Allograft Articular Chondrocytes Embedded in a Collagen Gel. Tissue Engineering Part A 1998;4:429-44. [62] Fujisato T, Sajiki T, Liu Q, Ikada Y. Effect of basic fibroblast growth factor on cartilage regeneration in chondrocyte-seeded collagen sponge scaffold. Biomaterials 1996;17:155-62. [63] Kimura T, Yasui N, Ohsawa S, Ono K. Clin Orthop 1984;186. [64] Nehrer S, Breinan HA, Ramappa A, Shortkroff S, Young G, Minas T, et al. Canine chondrocytes seeded in type I and type II collagen implants investigated In Vitro. Journal of Biomedical Materials Research 1997;38:95-104. [65] Nehrer S, Breinan HA, Ramappa A, Young G, Shortkroff S, Louie LK, et al. Matrix collagen type and pore size influence behaviour of seeded canine chondrocytes. Biomaterials 1997;18:769-76. [66] Schuman L, Buma P, Versleyen D, de Man B, van der Kraan PM, van den Berg WB, et al. Chondrocyte behaviour within different types of collagen gel in vitro. Biomaterials 1995;16:809-14.  201  [67] van Susante JL, Buma P, van Osch GJ, Versleyen D, vander Kraan PM, van der Berg WB, et al. Culture of chondrocytes in alginate and collagen carrier gels. Acta OrthopScand 1995;66:549-56. [68] Yasui N, Osawa S, Ochi T, Nakashima H, Ono K. Primary culture of chondrocytes embedded in collagen gels. ExpCell Biol 1982;50:92-100. [69] Galois L, Hutasse S, Cortial D, Rousseau CF, Grossin L, Ronziere M-C, et al. Bovine chondrocyte behaviour in three-dimensional type I collagen gel in terms of gel contraction, proliferation and gene expression. Biomaterials 2006;27:79-90. [70] Nehrer S, Breinan HA, Ramappa A, Hsu HP, Minas T, Shortkroff S, et al. Chondrocyte-seeded collagen matrices implanted in a chondral defect in a canine model. Biomaterials 1998;19:2313-28. [71] DA Wahl JC. Collagen-Hydroxyapatite Composites for Hard Tissue Repair. European Cells and Materials Journal 2006;11:43-56. [72] Bradt J-H, Mertig M, Teresiak A, Pompe W. Biomimetic Mineralization of Collagen by Combined Fibril Assembly and Calcium Phosphate Formation. Chemistry of Materials 1999;11:2694-701. [73] Gelinsky M, Welzel PB, Simon P, Bernhardt A, K  枚 nig U. Porous  three-dimensional scaffolds made of mineralised collagen: Preparation and properties of a biomimetic nanocomposite material for tissue engineering of bone. Chemical Engineering Journal 2008;137:84-96.  202  [74] Heinemann S, Ehrlich H, Douglas T, Heinemann C, Worch H, Schatton W, et al. Ultrastructural Studies on the Collagen of the Marine Sponge Chondrosia reniformis Nardo. Biomacromolecules 2007;8:3452-7. [75] Lin Z SK, Zhang X, Pavlos NJ, Abel T, Willers C, Dai K, Xu J, Zheng Q, Zheng M. In vitro Evaluation of Natural Marine Sponge Collagen as a Scaffold for Bone Tissue Engineering. Int J Biol Sci 2011;7:968-77. [76] Yunoki S, Nagai N, Suzuki T, Munekata M. Novel biomaterial from reinforced salmon collagen gel prepared by fibril formation and cross-linking. Journal of Bioscience and Bioengineering 2004;98:40-7. [77] Hoyer B, Bernhardt A, Heinemann S, Stachel I, Meyer M, Gelinsky M. Biomimetically Mineralized Salmon Collagen Scaffolds for Application in Bone Tissue Engineering. Biomacromolecules;13:1059-66. [78] Bosnakovski D, Mizuno M, Kim G, Takagi S, Okumura M, Fujinaga T. Chondrogenic differentiation of bovine bone marrow mesenchymal stem cells (MSCs) in different hydrogels: Influence of collagen type II extracellular matrix on MSC chondrogenesis. Biotechnology and Bioengineering 2006;93:1152-63. [79] Nimni ME, Cheung D, Strates B, Kodama M, Sheikh K. Chemically modified collagen: A natural biomaterial for tissue replacement. Journal of Biomedical Materials Research 1987;21:741-71.  203  [80] Yang F, Cho S-W, Son SM, Hudson SP, Bogatyrev S, Keung L, et al. Combinatorial Extracellular Matrices for Human Embryonic Stem Cell Differentiation in 3D. Biomacromolecules;11:1909-14. [81] Skrzeszewska PJ, de Wolf FA, Werten MWT, Moers APHA, Cohen Stuart MA, van der Gucht J. Physical gels of telechelic triblock copolymers with precisely defined junction multiplicity. Soft Matter 2009;5:2057-62. [82] Werten MWT, Wisselink WH, Jansen-van den Bosch TJ, de Bruin EC, de Wolf FA. Secreted production of a custom-designed, highly hydrophilic gelatin in Pichia pastoris. Protein Engineering 2001;14:447-54. [83] Hu Y, Wang H, Wang J, Wang S, Liao W, Yang Y, et al. Supramolecular hydrogels inspired by collagen for tissue engineering. Organic & Biomolecular Chemistry 2010;8:3267-71. [84] Lee KY, Mooney DJ. Hydrogels for Tissue Engineering. Chemical Reviews 2001;101:1869-80. [85] Feijen AJKGHMEJKSAJZJDJ. Cross-linking and characterisation of gelatin matrices for biomedical applications. Journal of Biomaterials Science, Polymer Edition  2000;11:225 -- 43  [86] Lien S-M, Li W-T, Huang T-J. Genipin-crosslinked gelatin scaffolds for articular cartilage tissue engineering with a novel crosslinking method. Materials Science and Engineering: C 2008;28:36-43.  204  [87] Kim U-J, Park J, Li C, Jin H-J, Valluzzi R, Kaplan DL. Structure and Properties of Silk Hydrogels. Biomacromolecules 2004;5:786-92. [88] Altman GH, Diaz F, Jakuba C, Calabro T, Horan RL, Chen J, et al. Silk-based biomaterials. Biomaterials 2003;24:401-16. [89] Meinel L, Hofmann S, Karageorgiou V, Zichner L, Langer R, Kaplan D, et al. Engineering cartilage-like tissue using human mesenchymal stem cells and silk protein scaffolds. Biotechnology and Bioengineering 2004;88:379-91. [90] Fini M, Motta A, Torricelli P, Giavaresi G, Nicoli Aldini N, Tschon M, et al. The healing of confined critical size cancellous defects in the presence of silk fibroin hydrogel. Biomaterials 2005;26:3527-36. [91] Aoki HT, N.; Morita, Y.; Hattori, K.; Harada, Y.;Sonobe, M.; Wakitani, S.; Tamada, Y. . Biomed Mater Eng 2003;13. [92] Morita Y, Tomita N, Aoki H, Sonobe M, Wakitani S, Tamada Y, et al. Frictional properties of regenerated cartilage in vitro. Journal of biomechanics 2006;39:103-9. [93] Morita YT, N.; Aoki, H.; Wakitani, S.; Tamada, Y.;Suguro, T.; Ikeuchi, K. Biomed Mater Eng 2002;12. [94] J. M. Gosline MEDaMWD. Endeavour 1986;10:37-43.  205  [95] Hayashi CY, Lewis RV. Evidence from flagelliform silk cDNA for the structural basis of elasticity and modular nature of spider silks. Journal of Molecular Biology 1998;275:773-84. [96] Zhou Y, Wu S, Conticello VP. Genetically Directed Synthesis and Spectroscopic Analysis of a Protein Polymer Derived from a Flagelliform Silk Sequence. Biomacromolecules 2001;2:111-25. [97] Bereczky ZK, E.; Muszbek, L. . Pathophysiol HaemostThromb 2003;33. [98] Sidelmann JJG, J.; Jespersen, J.; Kluft, C. Semin Thromb Hemost 2000;26. [99] Helgerson SLS, T.; DiOrio, J.; Tawil, B.; Bittner, K. M.; Spaethe, R. . Encyclopedia of Biomaterials and Biomedicine Engineering: Marcel Dekker: New York; 2004. [100] Kjaergard HK, Weis-Fogh US. Important Factors Influencing the Strength of Autologous Fibrin Glue; the Fibrin Concentration and Reaction Time --Comparison of Strength with Commercial Fibrin Glue. European Surgical Research 1994;26:273-6. [101] Kheirabadi BS, Acheson EM, Deguzman R, Crissey JM, Delgado AV, Estep SJ, et al. The Potential Utility of Fibrin Sealant Dressing in Repair of Vascular Injury in Swine. The Journal of Trauma and Acute Care Surgery 2007;62:94-103 10.1097/01.ta.0000251595.45451.d0.  206  [102] Frisman I, Shachaf Y, Seliktar D, Bianco-Peled H. Stimulus-Responsive Hydrogels Made From Biosynthetic Fibrinogen Conjugates for Tissue Engineering: Structural Characterization. Langmuir;27:6977-86. [103] Giuseppe M. Peretti J-WX, Lawrence J. Bonassar, Carl Hendrick Kirchhoff, Michael J. Yaremchuk, and Mark A. Randolph. Review of Injectable Cartilage Engineering Using Fibrin Gel in Mice and Swine Models Tissue Engineering 2006;12:1151-68. [104] van Susante JLC, Buma P, Homminga GN, van den Berg WB, Veth RPH. Chondrocyte-seeded hydroxyapatite for repair of large articular cartilage defects. A pilot study in the goat. Biomaterials 1998;19:2367-74. [105] Anderson SO, and Weis-Fogh, T. . Resilin. A rubber-like protein in arthropod cuticle. Adv Insect Physiol 1964;2:1-65. [106] Weisfogh T. A Rubber-like Protein in Insect Cuticle Journal of Experimental Biology 1960;37. [107] Gorb SN. Serial Elastic Elements in the Damselfly Wing: Mobile Vein Joints Contain Resilin. Naturwissenschaften 1999;86:552-5. [108] Burrows M. Biomechanics: Froghopper insects leap to new heights. Nature 2003;424:509-.  207  [109] Rothschild M, Schlein J, Parker K, Neville C, Sternberg S. Execution of the Jump and Activity. Philosophical Transactions of the Royal Society of London B, Biological Sciences 1975;271:499-515. [110] Neff D, Frazier SF, Quimby L, Wang R-T, Zill S. Identification of resilin in the leg of cockroach, Periplaneta americana: confirmation by a simple method using pH dependence of UV fluorescence. Arthropod Structure &amp; Development 2000;29:75-83. [111] Andersen. O. in Elastomeric Proteins Cambridge University Press, Cambridge, Editonedn.; 2003. [112] Weis-Fogh T. Molecular interpretation of the elasticity of resilin, a rubber-like protein. Journal of Molecular Biology 1961;3:648-67. [113] Haas F, Gorb S, Wootton RJ. Elastic joints in dermapteran hind wings: materials and wing folding. Arthropod Structure &amp; Development 2000;29:137-46. [114] Haas F, Gorb S, Blickhan R. The function of resilin in beetle wings. Proceedings of the Royal Society of London Series B: Biological Sciences 2000;267:1375-81. [115] Young D, Bennet-Clark H. The role of the tymbal in cicada sound production. Journal of Experimental Biology 1995;198:1001-20.  208  [116] Skals N, Surlykke A. Sound production by abdominal tymbal organs in two moth species: the green silver-line and the scarce silver-line (Noctuoidea: Nolidae: Chloephorinae). JOURNAL OF EXPERIMENTAL BIOLOGY 1999;202:2937-49. [117] Ardell DH, Andersen SO. Tentative identification of a resilin gene in Drosophila melanogaster. Insect Biochemistry and Molecular Biology 2001;31:965-70. [118] Weisfogh KBaT. Biochim Biophys Acta 1961;48. [119] Bochicchio B, Pepe A, Tamburro AM. Investigating by CD the molecular mechanism of elasticity of elastomeric proteins. Chirality 2008;20:985-94. [120] Nairn KM, Lyons RE, Mulder RJ, Mudie ST, Cookson DJ, Lesieur E, et al. A Synthetic Resilin Is Largely Unstructured. Biophysical journal 2008;95:3358-65. [121] Tamburro AM, Panariello S, Santopietro V, Bracalello A, Bochicchio B, Pepe A. Molecular and Supramolecular Structural Studies on Significant Repetitive Sequences of Resilin. ChemBioChem 2010;11:83-93. [122] Elvin CM, Carr AG, Huson MG, Maxwell JM, Pearson RD, Vuocolo T, et al. Synthesis and properties of crosslinked recombinant pro-resilin. Nature 2005;437:999-1002. [123] Kim M, Elvin C, Brownlee A, Lyons R. High yield expression of recombinant pro-resilin: Lactose-induced fermentation in E. coli and facile purification. Protein Expression and Purification 2007;52:230-6.  209  [124] Lyons RE, Lesieur E, Kim M, Wong DCC, Huson MG, Nairn KM, et al. Design and facile production of recombinant resilin-like polypeptides: gene construction and a rapid protein purification method. Protein Engineering Design and Selection 2007;20:25-32. [125] Lyons RE, Nairn KM, Huson MG, Kim M, Dumsday G, Elvin CM. Comparisons of Recombinant Resilin-like Proteins: Repetitive Domains Are Sufficient to Confer Resilin-like Properties. Biomacromolecules 2009;10:3009-14. [126] Charati MB, Ifkovits JL, Burdick JA, Linhardt JG, Kiick KL. Hydrophilic elastomeric biomaterials based on resilin-like polypeptides. Soft Matter 2009;5:3412-6. [127] Nie T, Akins Jr RE, Kiick KL. Production of heparin-containing hydrogels for modulating cell responses. Acta Biomaterialia 2009;5:865-75. [128] Yamaguchi N, Zhang L, Chae B-S, Palla CS, Furst EM, Kiick KL. Growth Factor Mediated Assembly of Cell Receptor-Responsive Hydrogels. Journal of the American Chemical Society 2007;129:3040-1. [129] Nie T, Baldwin A, Yamaguchi N, Kiick KL. Production of heparin-functionalized hydrogels for the development of responsive and controlled growth factor delivery systems. Journal of Controlled Release 2007;122:287-96. [130] Hersel U, Dahmen C, Kessler H. RGD modified polymers: biomaterials for stimulated cell adhesion and beyond. Biomaterials 2003;24:4385-415.  210  [131] Liu S, Julian J, Carson DD. A Peptide Sequence of Heparin/Heparan Sulfate (HP/HS)-interacting Protein Supports Selective, High Affinity Binding of HP/HS and Cell Attachment. Journal of Biological Chemistry 1998;273:9718-26. [132] Rizzi SC, Ehrbar M, Halstenberg S, Raeber GP, Schmoekel HG, Hagenmuller H, et al. Recombinant Protein-co-PEG Networks as Cell-Adhesive and Proteolytically Degradable Hydrogel Matrixes. Part II: Biofunctional Characteristics. Biomacromolecules 2006;7:3019-29. [133] Liu S, Zhou F, Hook M, Carson DD. A heparin-binding synthetic peptide of heparin/heparan sulfateinteracting protein modulates blood coagulation activities. Proceedings of the National Academy of Sciences 1997;94:1739-44. [134] Cao Q, Wang Y, Bayley H. Sequence of abductin, the molluscan 'rubber' protein. Current Biology 1997;7:R677-R8. [135] Tatham AS, Shewry PR. Comparative structures and properties of elastic proteins. Philosophical Transactions of the Royal Society of London Series B: Biological Sciences 2002;357:229-34. [136] KIKUCHI Y, TAMIYA N. Methionine Sulfoxide in the Resilium Protein of Surf Clams. Journal of Biochemistry 1981;89:1975-6. [137] Bochicchio B, Jimenez-Oronoz F, Pepe A, Blanco M, Sandberg LB, Tamburro AM. Synthesis of and Structural Studies on Repeating Sequences of Abductin. Macromolecular Bioscience 2005;5:502-11.  211  [138] Shewry PR, Tatham AS. The prolamin storage proteins of cereal seeds: structure and evolution. Biochem J 1990;267:1-12. [139] Tatham AS, Hayes L, Shewry PR, Urry DW. Wheat seed proteins exhibit a complex mechanism of protein elasticity. Biochimica et Biophysica Acta (BBA) Protein Structure and Molecular Enzymology 2001;1548:187-93. [140] Tatham AS, Drake AF, Shewry PR. Conformational studies of synthetic peptides corresponding to the repetitive regions of the high molecular weight (HMW) glutenin subunits of wheat. Journal of Cereal Science 1990;11:189-200. [141] Van Dijk AA, Van Wijk LL, Van Swieten E, Robillard GT, Vliet AV, Tesser GI, et al. Structure characterization of the central repetitive domain of high molecular weight gluten proteins. I. Model studies using cyclic and linear peptides. Protein Science 1997;6:637-48. [142] Gilbert SM, Wellner N, Belton PS, Greenfield JA, Siligardi G, Shewry PR, et al. Expression and characterisation of a highly repetitive peptide derived from a wheat seed storage protein. Biochimica et Biophysica Acta (BBA) - Protein Structure and Molecular Enzymology 2000;1479:135-46. [143] Slaughter BV, Khurshid SS, Fisher OZ, Khademhosseini A, Peppas NA. Hydrogels in Regenerative Medicine. Advanced Materials 2009;21:3307-29.  212  [144] Sargeant TD, Guler MO, Oppenheimer SM, Mata A, Satcher RL, Dunand DC, et al. Hybrid bone implants: Self-assembly of peptide amphiphile nanofibers within porous titanium. Biomaterials 2008;29:161-71. [145] Adams DJ, Holtzmann K, Schneider C, Butler MF. Self-Assembly of Surfactant-like Peptides. Langmuir 2007;23:12729-36. [146] Guler MO, Stupp SI. A Self-Assembled Nanofiber Catalyst for Ester Hydrolysis. Journal of the American Chemical Society 2007;129:12082-3. [147] Williams BAR, Lund K, Liu Y, Yan H, Chaput JC. Self-Assembled Peptide Nanoarrays: An Approach to Studying Protein–Protein Interactions. Angewandte Chemie International Edition 2007;46:3051-4. [148] O'Shea EK, Lumb KJ, Kim PS. Peptide  Velcro : Design of a heterodimeric  coiled coil. Current biology : CB 1993;3:658-67. [149] O'Shea E, Rutkowski R, Stafford W, Kim P. Preferential heterodimer formation by isolated leucine zippers from fos and jun. Science 1989;245:646-8. [150] O'Shea E, Klemm J, Kim P, Alber T. X-ray structure of the GCN4 leucine zipper, a two-stranded, parallel coiled coil. Science 1991;254:539-44. [151] Harbury P, Zhang T, Kim P, Alber T. A switch between two-, three-, and four-stranded coiled coils in GCN4 leucine zipper mutants. Science 1993;262:1401-7.  213  [152] Lumb K, Kim P. Measurement of interhelical electrostatic interactions in the GCN4 leucine zipper. Science 1995;268:436-9. [153] Malashkevich VN, Kammerer RA, Efimov VP, Schulthess T, Engel J. The Crystal Structure of a Five-Stranded Coiled Coil in COMP: A Prototype Ion Channel? Science 1996;274:761-5. [154] Liu J, Zheng Q, Deng Y, Cheng C-S, Kallenbach NR, Lu M. A seven-helix coiled coil. Proceedings of the National Academy of Sciences 2006;103:15457-62. [155] Xu C, Breedveld V, Kopecek J. Reversible Hydrogels from Self-Assembling Genetically Engineered Protein Block Copolymers. Biomacromolecules 2005;6:1739-49. [156] Banwell EF, Abelardo ES, Adams DJ, Birchall MA, Corrigan A, Donald AM, et al. Rational design and application of responsive [alpha]-helical peptide hydrogels. Nat Mater 2009;8:596-600. [157] Pandya MJ, Spooner GM, Sunde M, Thorpe JR, Rodger A, Woolfson DN. Sticky-End Assembly of a Designed Peptide Fiber Provides Insight into Protein Fibrillogenesis. Biochemistry 2000;39:8728-34. [158] Guler MO, Soukasene S, Hulvat JF, Stupp SI. Presentation and Recognition of Biotin on Nanofibers Formed by Branched Peptide Amphiphiles. Nano Letters 2004;5:249-52.  214  [159] Bull SR, Guler MO, Bras RE, Meade TJ, Stupp SI. Self-Assembled Peptide Amphiphile Nanofibers Conjugated to MRI Contrast Agents. Nano Letters 2004;5:1-4. [160] Niece KL, Hartgerink JD, Donners JJJM, Stupp SI. Self-Assembly Combining Two Bioactive Peptide-Amphiphile Molecules into Nanofibers by Electrostatic Attraction. Journal of the American Chemical Society 2003;125:7146-7. [161] Hartgerink JD, Beniash E, Stupp SI. Peptide-amphiphile nanofibers: A versatile scaffold for the preparation of self-assembling materials. Proceedings of the National Academy of Sciences 2002;99:5133-8. [162] Hung AM, Stupp SI. Simultaneous Self-Assembly, Orientation, and Patterning of Peptide-Amphiphile Nanofibers by Soft Lithography. Nano Letters 2007;7:1165-71. [163] Hwang JJ, Iyer SN, Li L-S, Claussen R, Harrington DA, Stupp SI. Self-assembling biomaterials: Liquid crystal phases of cholesteryl oligo(l-lactic acid) and their interactions with cells. Proceedings of the National Academy of Sciences 2002;99:9662-7. [164] Davis ME, Motion JPM, Narmoneva DA, Takahashi T, Hakuno D, Kamm RD, et al. Injectable Self-Assembling Peptide Nanofibers Create Intramyocardial Microenvironments for Endothelial Cells. Circulation 2005;111:442-50.  215  [165] Kisiday J, Jin M, Kurz B, Hung H, Semino C, Zhang S, et al. Self-assembling peptide hydrogel fosters chondrocyte extracellular matrix production and cell division: Implications for cartilage tissue repair. Proceedings of the National Academy of Sciences 2002;99:9996-10001. [166] Zhang S. Fabrication of novel biomaterials through molecular self-assembly. Nat Biotech 2003;21:1171-8. [167] Santoso S, Hwang W, Hartman H, Zhang S. Self-assembly of Surfactant-like Peptides with Variable Glycine Tails to Form Nanotubes and Nanovesicles. Nano Letters 2002;2:687-91. [168] Carlos E. Semino JK, Yasunori Hayashi, and Shuguang Zhang. Entrapment of Migrating Hippocampal Neural Cells in Three-Dimensional Peptide Nanofiber Scaffold. Tissue Engineering 2004;10:643-55. [169] Yan C, Pochan DJ. Rheological properties of peptide-based hydrogels for biomedical and other applications. Chemical Society Reviews 2010;39:3528-40. [170] Gosal WS, Clark AH, Ross-Murphy SB. Fibrillar β-Lactoglobulin Gels: Part 1. Fibril Formation and Structure. Biomacromolecules 2004;5:2408-19. [171] Akkermans C, Venema P, van der Goot AJ, Gruppen H, Bakx EJ, Boom RM, et al. Peptides are Building Blocks of Heat-Induced Fibrillar Protein Aggregates of  β-Lactoglobulin Formed at pH 2. Biomacromolecules 2008;9:1474-9.  216  [172] Durand D, Christophe Gimel J, Nicolai T. Aggregation, gelation and phase separation of heat denatured globular proteins. Physica A: Statistical Mechanics and its Applications 2002;304:253-65. [173] Gosal WS, Clark AH, Pudney PDA, Ross-Murphy SB. Novel Amyloid Fibrillar Networks Derived from a Globular Protein: β-Lactoglobulin. Langmuir 2002;18:7174-81. [174] Yan H, Saiani A, Gough JE, Miller AF. Thermoreversible Protein Hydrogel as Cell Scaffold. Biomacromolecules 2006;7:2776-82. [175] Yan H, Nykanen A, Ruokolainen J, Farrar D, Gough JE, Saiani A, et al. Thermo-reversible protein fibrillar hydrogels as cell scaffolds. Faraday Discussions 2008;139:71-84. [176] Wheeldon IR, Calabrese Barton S, Banta S. Bioactive Proteinaceous Hydrogels from Designed Bifunctional Building Blocks. Biomacromolecules 2007;8:2990-4. [177] Vakonakis I, Campbell ID. Extracellular matrix: from atomic resolution to ultrastructure. Current Opinion in Cell Biology 2007;19:578-83. [178] Tatham AS, Shewry PR. Elastomeric proteins: biological roles, structures and mechanisms. Trends in Biochemical Sciences 2000;25:567-71.  217  [179] Rief M, Gautel M, Oesterhelt F, Fernandez JM, Gaub HE. Reversible Unfolding of Individual Titin Immunoglobulin Domains by AFM. Science 1997;276:1109-12. [180] Carrion-Vazquez M, Oberhauser AF, Fisher TE, Marszalek PE, Li H, Fernandez JM. Mechanical design of proteins studied by single-molecule force spectroscopy and protein engineering. Progress in Biophysics and Molecular Biology 2000;74:63-91. [181] Linke WA, Rudy DE, Centner T, Gautel M, Witt C, Labeit S, et al. I-Band Titin in Cardiac Muscle Is a Three-Element Molecular Spring and Is Critical for Maintaining Thin Filament Structure. The Journal of Cell Biology 1999;146:631-44. [182] Linke WA, Popov VI, Pollack GH. Passive and active tension in single cardiac myofibrils. Biophysical journal 1994;67:782-92. [183] Minajeva A, Kulke M, Fernandez JM, Linke WA. Unfolding of Titin Domains Explains the Viscoelastic Behavior of Skeletal Myofibrils. Biophysical journal 2001;80:1442-51. [184] Mutungi G, Ranatunga KW. The viscous, viscoelastic and elastic characteristics of resting fast and slow mammalian (rat) muscle fibres. The Journal of Physiology 1996;496:827-36. [185] Wang K, McCarter R, Wright J, Beverly J, Ramirez-Mitchell R. Viscoelasticity of the sarcomere matrix of skeletal muscles. The titin-myosin  218  composite filament is a dual-stage molecular spring. Biophysical journal 1993;64:1161-77. [186] Kellermayer MSZ, Smith SB, Bustamante C, Granzier HL. Mechanical Fatigue in Repetitively Stretched Single Molecules of Titin. Biophysical journal 2001;80:852-63. [187] Labeit S, Kolmerer B. Titins: Giant Proteins in Charge of Muscle Ultrastructure and Elasticity. Science 1995;270:293-6. [188] Bullard B, Garcia T, Benes V, Leake MC, Linke WA, Oberhauser AF. The molecular elasticity of the insect flight muscle proteins projectin and kettin. Proceedings of the National Academy of Sciences of the United States of America 2006;103:4451-6. [189] Oberhauser AF, Marszalek PE, Erickson HP, Fernandez JM. The molecular elasticity of the extracellular matrix protein tenascin. Nature 1998;393:181-5. [190] Dietz H, Rief M. Exploring the energy landscape of GFP by single-molecule mechanical experiments. Proceedings of the National Academy of Sciences of the United States of America 2004;101:16192-7. [191] Cao Y, Lam C, Wang M, Li H. Nonmechanical Protein Can Have Significant Mechanical Stability. Angewandte Chemie International Edition 2006;45:642-5.  219  [192] Yang G, Cecconi C, Baase WA, Vetter IR, Breyer WA, Haack JA, et al. Solid-state synthesis and mechanical unfolding of polymers of T4 lysozyme. Proceedings of the National Academy of Sciences 2000;97:139-44. [193] Brockwell DJ, Beddard GS, Paci E, West DK, Olmsted PD, Smith DA, et al. Mechanically Unfolding the Small, Topologically Simple Protein L. Biophysical journal 2005;89:506-19. [194] Cecconi C, Shank EA, Bustamante C, Marqusee S. Direct Observation of the Three-State Folding of a Single Protein Molecule. Science 2005;309:2057-60. [195] Ainavarapu SRK, Li L, Badilla CL, Fernandez JM. Ligand Binding Modulates the Mechanical Stability of Dihydrofolate Reductase. Biophysical journal 2005;89:3337-44. [196] Junker JP, Hell K, Schlierf M, Neupert W, Rief M. Influence of Substrate Binding on the Mechanical Stability of Mouse Dihydrofolate Reductase. Biophysical journal 2005;89:L46-L8. [197] Best RB, Li B, Steward A, Daggett V, Clarke J. Can Non-Mechanical Proteins Withstand Force? Stretching Barnase by Atomic Force Microscopy and Molecular Dynamics Simulation. Biophysical journal 2001;81:2344-56. [198] Wang C, Kopecek J, Stewart RJ. Hybrid Hydrogels Cross-Linked by Genetically Engineered Coiled-Coil Block Proteins. Biomacromolecules 2001;2:912-20.  220  [199] Cao Y, Li H. Engineering tandem modular protein based reversible hydrogels. Chemical Communications 2008:4144-6. [200] Li P-C, Makarov DE. Ubiquitin-like Protein Domains Show High Resistance to Mechanical Unfolding Similar to That of the I27 Domain in Titin: Evidence from Simulations. The Journal of Physical Chemistry B 2003;108:745-9. [201] Carrion-Vazquez M, Li H, Lu H, Marszalek PE, Oberhauser AF, Fernandez JM. The mechanical stability of ubiquitin is linkage dependent. Nat Struct Mol Biol 2003;10:738-43. [202] Chyan C-L, Lin F-C, Peng H, Yuan J-M, Chang C-H, Lin S-H, et al. Reversible Mechanical Unfolding of Single Ubiquitin Molecules. Biophysical journal 2004;87:3995-4006. [203] Schwaiger I, Schleicher, M., Noegel, A. A. & Rief, M. . The folding pathway of a fast-folding immunoglobulin domain revealed by single-molecule mechanical experiments. . EMBO 2005;Rep. 6:46-51 [204] Li H, Linke WA, Oberhauser AF, Carrion-Vazquez M, Kerkvliet JG, Lu H, et al. Reverse engineering of the giant muscle protein titin. Nature 2002;418:998-1002. [205] Park S-H, Shastry MCR, Roder H. Folding dynamics of the B1 domain of protein G explored by ultrarapid mixing. Nat Struct Mol Biol 1999;6:943-7. [206] Cao Y, Li H. Polyprotein of GB1 is an ideal artificial elastomeric protein. Nat Mater 2007;6:109-14.  221  [207] Becker N, Oroudjev E, Mutz S, Cleveland JP, Hansma PK, Hayashi CY, et al. Molecular nanosprings in spider capture-silk threads. Nat Mater 2003;2:278-83. [208] Shen W, Lammertink RGH, Sakata JK, Kornfield JA, Tirrell DA. Assembly of an Artificial Protein Hydrogel through Leucine Zipper Aggregation and Disulfide Bond Formation. Macromolecules 2005;38:3909-16. [209] Shen W, Zhang K, Kornfield JA, Tirrell DA. Tuning the erosion rate of artificial protein hydrogels through control of network topology. Nat Mater 2006;5:153-8. [210] Shen W, Kornfield JA, Tirrell DA. Dynamic Properties of Artificial Protein Hydrogels Assembled through Aggregation of Leucine Zipper Peptide Domains. Macromolecules 2007;40:689-92. [211] Fischer SE, Liu X, Mao H-Q, Harden JL. Controlling cell adhesion to surfaces via associating bioactive triblock proteins. Biomaterials 2007;28:3325-37. [212] Mi L, Fischer S, Chung B, Sundelacruz S, Harden JL. Self-Assembling Protein Hydrogels with Modular Integrin Binding Domains. Biomacromolecules 2005;7:38-47. [213] Helmes M, Trombitas K, Centner T, Kellermayer M, Labeit S, Linke WA, et al. Mechanically Driven Contour-Length Adjustment in Rat Cardiac Titin's Unique N2B Sequence. Circulation Research 1999;84:1339-52.  222  [214] Lv S, Cao Y, Li H. Tandem modular protein-based hydrogels constructed using a novel two-component approach. Langmuir 2012;28:2269-74. [215] Bode F, da Silva MA, Drake AF, Ross-Murphy SB, Dreiss CA. Enzymatically Cross-Linked Tilapia Gelatin Hydrogels: Physical, Chemical, and Hybrid Networks. Biomacromolecules 2011;12:3741-52. [216] 1. The diversity of covalent cross-links in protein. Acta Physiologica Scandinavica 1965;66:9-13. [217] Carbinatto FM, de Castro AD, Cury BSF, Magalhaes A, Evangelista RC. Physical properties of pectin-high amylose starch mixtures cross-linked with sodium trimetaphosphate. International Journal of Pharmaceutics 2012;423:281-8. [218] Koob TJ, Hernandez DJ. Mechanical and thermal properties of novel polymerized NDGA-gelatin hydrogels. Biomaterials 2003;24:1285-92. [219] Hoare TR, Kohane DS. Hydrogels in drug delivery: Progress and challenges. Polymer 2008;49:1993-2007. [220] Zhao X, Huebsch N, Mooney DJ, Suo Z. Stress-relaxation behavior in gels with ionic and covalent crosslinks. Journal of Applied Physics 2010;107:063509-5. [221] Fancy DA, Kodadek T. Chemistry for the analysis of protein-protein interactions: Rapid and efficient cross-linking triggered by long wavelength light. Proceedings of the National Academy of Sciences 1999;96:6020-4.  223  [222] Elvin CM, Brownlee AG, Huson MG, Tebb TA, Kim M, Lyons RE, et al. The development of photochemically crosslinked native fibrinogen as a rapidly formed and mechanically strong surgical tissue sealant. Biomaterials 2009;30:2059-65. [223] Tskhovrebova L, Trinick J. Titin: properties and family relationships. Nat Rev Mol Cell Biol 2003;4:679-89. [224] Fürst DO, Osborn, M., Nave, R. and Weber, K. . The organization of titin filaments in the half-sarcomere revealed by monoclonal antibodies in immunoelectron microscopy: a map of ten nonrepetitive epitopes starting at the Z line extends close to the M-line. J Cell Biol 1988;106:1563-72. [225] Itoh Y, Suzuki, T., Kimura, S., Ohashi, K., Higushi, H., Sawada, H.,, Shimizu T, Shibata, M. and Maruyama, K. . Extensible and less extensible domains of connectin filaments in stretched vertebrate skeletal muscle sarcomeres as detected by immunofluorescence and immunoelectronmicroscopy using monoclonal antibodies.  . J Biochem 1988;104:504-8.  [226] Trombitás KaP, G. H. . Elastic properties of the titin filament in the Z-line region of vertebrate striated muscle. . J Muscle ResCell Motil 1993;14:416-22. [227] Horowits R, Kempner, E. S., Bisher, M. E. and Podolsky, R. J. A physiological role for titin and nebulin in skeletal muscle. . Nature 1986;323:160-4. [228] Wang S-MaJ, C.-J. . Interaction between titin and  α-actinin.  . Biomed Res 1992;13:197-202.  224  [229] Linke WA, Ivemeyer M, Mundel P, Stockmeier MR, Kolmerer B. Nature of PEVK-titin elasticity in skeletal muscle. Proceedings of the National Academy of Sciences 1998;95:8052-7. [230] Granzier H, Helmes, M. and Trombitas, K. . Nonuniform elasticity of titin in cardiac myocytes: a study using immunoelectron microscopy and cellular mechanics. . Biophys J 1996;70:430-42. [231] Linke WA, Ivemeyer, M., Olivieri, N., Kolmerer, B., Rüegg, J. C. and Labeit, S. Towards a molecular understanding of the elasticity of titin. . J Mol Biol 1996;261:62-71. [232] Maruyama K. Connectin/titin, giant elastic protein of muscle. The FASEB Journal 1997;11:341-5. [233] Wang K. Titin/connectin and nebulin: giant protein rulers of muscle structure and function. . Adv Biophys 1996;33. [234] Granzier HL, Irving TC. Passive tension in cardiac muscle: contribution of collagen, titin, microtubules, and intermediate filaments. Biophysical journal 1995;68:1027-44. [235] Kushner AM, Gabuchian V, Johnson EG, Guan Z. Biomimetic Design of Reversibly Unfolding Cross-Linker to Enhance Mechanical Properties of 3D Network Polymers. Journal of the American Chemical Society 2007;129:14110-1.  225  [236] Kushner AM, Vossler JD, Williams GA, Guan Z. A Biomimetic Modular Polymer with Tough and Adaptive Properties. Journal of the American Chemical Society 2009;131:8766-8. [237] Lv S, Dudek DM, Cao Y, Balamurali MM, Gosline J, Li H. Designed biomaterials to mimic the mechanical properties of muscles. Nature 2010;465:69-73. [238] Jones FS, Jones PL. The tenascin family of ECM glycoproteins: Structure, function, and regulation during embryonic development and tissue remodeling. Developmental Dynamics 2000;218:235-59. [239] Denda S, Muller U, Crossin KL, Erickson HP, Reichardt LF. Utilization of a Soluble Integrin-Alkaline Phosphatase Chimera To Characterize Integrin α8β1 Receptor Interactions with Tenascin: Murine α8β1 Binds to the RGD Site in Tenascin-C Fragments, but Not to Native Tenascin-C. Biochemistry 1998;37:5464-74. [240] Peng Q, Zhuang S, Wang M, Cao Y, Khor Y, Li H. Mechanical Design of the Third FnIII Domain of Tenascin-C. Journal of Molecular Biology 2009;386:1327-42. [241] Leahy D, Hendrickson W, Aukhil I, Erickson H. Structure of a fibronectin type III domain from tenascin phased by MAD analysis of the selenomethionyl protein. Science 1992;258:987-91.  226  [242] Bernstein H, Langer R. Ex vivo model of an immobilized-enzyme reactor. Proceedings of the National Academy of Sciences 1988;85:8751-5. [243] Lu HD, Wheeldon IR, Banta S. Catalytic biomaterials: engineering organophosphate hydrolase to form self-assembling enzymatic hydrogels. Protein Engineering Design and Selection 2010;23:559-66. [244] Miyata T, Uragami T, Nakamae K. Biomolecule-sensitive hydrogels. Advanced Drug Delivery Reviews 2002;54:79-98. [245] Wheeldon IR, Gallaway JW, Barton SC, Banta S. Bioelectrocatalytic hydrogels from electron-conducting metallopolypeptides coassembled with bifunctional enzymatic building blocks. Proceedings of the National Academy of Sciences 2008;105:15275-80. [246] Feng L, Wang L, Hu Z, Tian Y, Xian Y, Jin L. Encapsulation of horseradish peroxidase into hydrogel, and its bioelectrochemistry. Microchimica Acta 2009;164:49-54. [247] Li Z, Konno T, Takai M, Ishihara K. Fabrication of polymeric electron-transfer mediator/enzyme hydrogel multilayer on an Au electrode in a layer-by-layer process. Biosensors and Bioelectronics 2012;34:191-6. [248] Alpeeva IS, Vilkanauskyte A, Ngounou B, Csoregi E, Sakharov IY, Gonchar M, et al. Bi-Enzyme Alcohol Biosensors Based on Genetically Engineered Alcohol Oxidase and Different Peroxidases. Microchimica Acta 2005;152:21-7.  227  [249] Yan J, Pedrosa VA, Simonian AL, Revzin A. Immobilizing Enzymes onto Electrode Arrays by Hydrogel Photolithography to Fabricate Multi-Analyte Electrochemical Biosensors. ACS Applied Materials & Interfaces 2010;2:748-55. [250] Kotanen C, Tlili C, Guiseppi-Elie A. Bioactive Electroconductive Hydrogels: The Effects of Electropolymerization Charge Density on the Storage Stability of an Enzyme-Based Biosensor. Applied Biochemistry and Biotechnology 2012;166:878-88. [251] Heo J, Crooks RM. Microfluidic Biosensor Based on an Array of Hydrogel-Entrapped Enzymes. Analytical Chemistry 2005;77:6843-51. [252] Jang E, Kim S, Koh W-G. Microfluidic bioassay system based on microarrays of hydrogel sensing elements entrapping quantum dot-enzyme conjugates. Biosensors and Bioelectronics 2012;31:529-36. [253] Ivekovic D, Milardovic S, Grabaric BS. Palladium hexacyanoferrate hydrogel as a novel and simple enzyme immobilization matrix for amperometric biosensors. Biosensors and Bioelectronics 2004;20:872-8. [254] Dumitriu S, Magny P, Montane D, Vidal PF, Chornet E. Polyionic Hydrogels Obtained by Complexation between Xanthan and Chitosan: Their Properties as Supports for Enzyme Immobilization. Journal of Bioactive and Compatible Polymers 1994;9:184-209.  228  [255] Tan H, Rubin JP, Marra KG. Injectable in situ forming biodegradable chitosan-hyaluronic acid based hydrogels for adipose tissue regeneration. Organogenesis 2010;6:173-80. [256] Henrissat B, Bairoch A. New families in the classification of glycosyl hydrolases based on amino acid sequence similarities. Biochem J 1993;293:781-8. [257] Campbell R, Rose, D., Wakarchuk, W., To, R., Sung, W., & Yaguchi, M. . In: Suominen P, & Reinikainen, T., editor. Proceedings of the 2nd TRICEL symposium on Trichoderma reesei cellulases and other hydrolases 1993. p. 63-72. [258] Wakarchuk WW, Campbell RL, Sung WL, Davoodi J, Yaguchi M. Mutational and crystallographic analyses of the active site residues of the bacillus circulans xylanase. Protein Science 1994;3:467-75. [259] Gebler J, Gilkes NR, Claeyssens M, Wilson DB, Beguin P, Wakarchuk WW, et al. Stereoselective hydrolysis catalyzed by related beta-1,4-glucanases and beta-1,4-xylanases. Journal of Biological Chemistry 1992;267:12559-61. [260] Joshi MD, Sidhu G, Nielsen JE, Brayer GD, Withers SG, McIntosh LP. Dissecting the Electrostatic Interactions and pH-Dependent Activity of a Family 11 Glycosidase. Biochemistry 2001;40:10115-39. [261] Gebler J, Gilkes NR, Claeyssens M, Wilson DB, B  茅 guin P, Wakarchuk WW,  et al. Stereoselective hydrolysis catalyzed by related beta-1,4-glucanases and beta-1,4-xylanases. Journal of Biological Chemistry 1992;267:12559-61.  229  [262] Sinnott ML. Catalytic mechanism of enzymic glycosyl transfer. Chemical Reviews 1990;90:1171-202. [263] Miao S, Ziser L, Aebersold R, Withers SG. Identification of Glutamic Acid 78 as the Active Site Nucleophile in Bacillus subtilis Xylanase Using Electrospray Tandem Mass Spectrometry. Biochemistry 1994;33:7027-32. [264] Lawson SL, Wakarchuk WW, Withers SG. Positioning the Acid/Base Catalyst in a Glycosidase: Studies with Bacillus circulans Xylanase. Biochemistry 1997;36:2257-65. [265] Jollymore A, Li H. Measuring "unmeasurable" Folding Kinetics of Proteins by Single-Molecule Force Spectroscopy. Journal of Molecular Biology 2010;402:610-7. [266] Sadoshima J. The cellular and molecular response of cardiac myocytes to mechanical stress. Annual review of physiology 1997;59. [267] Ingber DE. Tensegrity: the architectural basis of cellular mechanotransduction. Annual review of physiology 1997;59. [268] Liu M, Tanswell AK, Post M. Mechanical force-induced signal transduction in lung cells. American Journal of Physiology - Lung Cellular and Molecular Physiology 1999;277:L667-L83.  230  [269] Grashoff C, Hoffman BD, Brenner MD, Zhou R, Parsons M, Yang MT, et al. Measuring mechanical tension across vinculin reveals regulation of focal adhesion dynamics. Nature 2010;466:263-6. [270] Shimomura O JF, Saiga Y. J Cell Comp Physiol 1962;59:223-39  [271] Yang F, Moss LG, Phillips GN. The molecular structure of green fluorescent protein. Nat Biotech 1996;14:1246-51. [272] Ormo M, Cubitt AB, Kallio K, Gross LA, Tsien RY, Remington SJ. Crystal Structure of the Aequorea victoria Green Fluorescent Protein. Science 1996;273:1392-5. [273] Gottfried J. Palm AZ, George A. Gaitanaris, Roland Stauber, George N. Pavlakis, Wlodawer A.  . The structural basis for spectral variations in green  fluorescent protein Nature Structural Biology 1997;4:361-5 [274] Phillips Jr GN. Structure and dynamics of green fluorescent protein. Current Opinion in Structural Biology 1997;7:821-7. [275] Brejc K, Sixma TK, Kitts PA, Kain SR, Tsien RY, Ormo M, et al. Structural basis for dual excitation and photoisomerization of the Aequorea victoria green fluorescent protein. Proceedings of the National Academy of Sciences 1997;94:2306-11.  231  [276] Wachter RM, King BA, Heim R, Kallio K, Tsien RY, Boxer SG, et al. Crystal Structure and Photodynamic Behavior of the Blue Emission Variant Y66H/Y145F of Green Fluorescent Protein. Biochemistry 1997;36:9759-65. [277] Tsien RY. THE GREEN FLUORESCENT PROTEIN. Annual Review of Biochemistry 1998;67:509-44. [278] Chalfie M, Tu Y, Euskirchen G, Ward W, Prasher D. Green fluorescent protein as a marker for gene expression. Science 1994;263:802-5. [279] Zacharias DA, Violin JD, Newton AC, Tsien RY. Partitioning of Lipid-Modified Monomeric GFPs into Membrane Microdomains of Live Cells. Science 2002;296:913-6. [280] Forster VT. Zwischenmolekulare energiewanderung und fluoreszenz. .  Ann Phys 1948;6:54-75. [281] Stryer L. Fluorescence Energy Transfer as a Spectroscopic Ruler. Annual Review of Biochemistry 1978;47:819-46. [282] Tsien RY, Bacskai BJ, Adams SR. FRET for studying intracellular signalling. Trends in Cell Biology 1993;3:242-5. [283] JR L. Principles of Fluorescence Spectroscopy: New York: Plenum; 1983. [284] Mitra RD, Silva CM, Youvan DC. Fluorescence resonance energy transfer between blue-emitting and red-shifted excitation derivatives of the green fluorescent protein. Gene 1996;173:13-7.  232  [285] Krafft GA, Wang GT, Lawrence C. Kuo JAS. [6] Synthetic approaches to continuous assays of retroviral proteases.  Methods in Enzymology: Academic  Press; 1994. p. 70-86. [286] Graham Knight C, Alan JB. [2] Fluorimetric assays of proteolytic enzymes. Methods in Enzymology: Academic Press; 1995. p. 18-34. [287] PeriasamyA KS, Day RN. . In Functional Imaging and OpticalManip-ulation of LivingCells,Proc. SPIE 2983: Belling-ham, WA: SPIE; 1997. [288] Romoser VA, Hinkle PM, Persechini A. Detection in Living Cells of Ca2+-dependent Changes in the Fluorescence Emission of an Indicator Composed of Two Green Fluorescent Protein Variants Linked by a Calmodulin-binding Sequence. Journal of Biological Chemistry 1997;272:13270-4. [289] Campbell RE. Fluorescent-Protein-Based Biosensors: Modulation of Energy Transfer as a Design Principle. Analytical Chemistry 2009;81:5972-9. [290] Cheng Z, Campbell RE. Assessing the Structural Stability of Designed β-Hairpin Peptides in the Cytoplasm of Live Cells. ChemBioChem 2006;7:1147-50. [291] Cheng Z, Miskolzie M, Campbell RE. In Vivo Screening Identifies a Highly Folded β-Hairpin Peptide with a Structured Extension. ChemBioChem 2007;8:880-3. [292] Rizzo MA, Springer GH, Granada B, Piston DW. An improved cyan fluorescent protein variant useful for FRET. Nat Biotech 2004;22:445-9.  233  [293] Griesbeck O, Baird GS, Campbell RE, Zacharias DA, Tsien RY. Reducing the Environmental Sensitivity of Yellow Fluorescent Protein. Journal of Biological Chemistry 2001;276:29188-94. [294] Lee J, Macosko CW, Urry DW. Elastomeric Polypentapeptides Cross-Linked into Matrixes and Fibers. Biomacromolecules 2001;2:170-9. [295] Nagarsekar A, Crissman J, Crissman M, Ferrari F, Cappello J, Ghandehari H. Genetic Engineering of Stimuli-Sensitive Silkelastin-like Protein Block Copolymers. Biomacromolecules 2003;4:602-7. [296] Urry DW, Pattanaik A. Elastic Protein-based Materials in Tissue Reconstructiona. Annals of the New York Academy of Sciences 1997;831:32-46. [297] Wright ER, Conticello VP. Self-assembly of block copolymers derived from elastin-mimetic polypeptide sequences. Advanced Drug Delivery Reviews 2002;54:1057-73. [298] Wheeldon IR, Campbell E, Banta S. A Chimeric Fusion Protein Engineered with Disparate Functionalities-enzymatic Activity and Self-assembly. Journal of Molecular Biology 2009;392:129-42. [299] Burkhard P, Stetefeld J, Strelkov SV. Coiled coils: a highly versatile protein folding motif. Trends in Cell Biology 2001;11:82-8. [300] Yu YB. Coiled-coils: stability, specificity, and drug delivery potential. Advanced Drug Delivery Reviews 2002;54:1113-29.  234  [301] Lupas A. Coiled coils: new structures and new functions. Trends in Biochemical Sciences 1996;21:375-82. [302] Gronenborn A, Filpula D, Essig N, Achari A, Whitlow M, Wingfield P, et al. A novel, highly stable fold of the immunoglobulin binding domain of streptococcal protein G. Science 1991;253:657-61. [303] Yang J, Xu C, Wang C, Kopecek J. Refolding Hydrogels Self-Assembled from N-(2-Hydroxypropyl)methacrylamide Graft Copolymers by Antiparallel Coiled-Coil Formation. Biomacromolecules 2006;7:1187-95. [304] Wu K, Yang J, Liu J, Kopecek J. Coiled-coil based drug-free macromolecular therapeutics: In vivo efficacy. Journal of Controlled Release 2012;157:126-31. [305] Yang J, Wu K, Konak C, Kopecek J. Dynamic Light Scattering Study of Self-Assembly of HPMA Hybrid Graft Copolymers. Biomacromolecules 2008;9:510-7. [306] Dusek K, Duskova-Smrckova M, Yang J, Kopecek J. Coiled-Coil Hydrogels: Effect of Grafted Copolymer Composition and Cyclization on Gelation. Macromolecules 2009;42:2265-74. [307] Alexander P, Fahnestock S, Lee T, Orban J, Bryan P. Thermodynamic analysis of the folding of the streptococcal protein G IgG-binding domains B1 and B2: why small proteins tend to have high denaturation temperatures. Biochemistry 1992;31:3597-603.  235  [308] Shin H, Quinten Ruhe P, Mikos AG, Jansen JA. In vivo bone and soft tissue response to injectable, biodegradable oligo(poly(ethylene glycol) fumarate) hydrogels. Biomaterials 2003;24:3201-11. [309] Burdick JA, Anseth KS. Photoencapsulation of osteoblasts in injectable RGD-modified PEG hydrogels for bone tissue engineering. Biomaterials 2002;23:4315-23. [310] Kurisawa M, Chung JE, Yang YY, Gao SJ, Uyama H. Injectable biodegradable hydrogels composed of hyaluronic acid-tyramine conjugates for drug delivery and tissue engineering. Chemical Communications 2005:4312-4. [311] Lee F, Chung JE, Kurisawa M. An injectable enzymatically crosslinked hyaluronic acid-tyramine hydrogel system with independent tuning of mechanical strength and gelation rate. Soft Matter 2008;4:880-7. [312] Silva EA, Kim E-S, Kong HJ, Mooney DJ. Material-based deployment enhances efficacy of endothelial progenitor cells. Proceedings of the National Academy of Sciences 2008;105:14347-52. [313] Higuchi H. Viscoelasticity and function of connectin/titin filaments in skinned muscle fibers. . Adv Biophys 1996;33:159-71. [314] Carrion-Vazquez M, Oberhauser AF, Fowler SB, Marszalek PE, Broedel SE, Clarke J, et al. Mechanical and chemical unfolding of a single protein: A comparison. Proceedings of the National Academy of Sciences 1999;96:3694-9.  236  [315] Lillie MA, Chalmers GWG, Gosline JM. The Effects of Heating on the Mechanical Properties of Arterial Elastin. Connective Tissue Research 1994;31:23-35. [316] Bell E, Gosline J. Mechanical design of mussel byssus: material yield enhances attachment strength. JOURNAL OF EXPERIMENTAL BIOLOGY 1996;199:1005-17. [317] Rief M, Fernandez JM, Gaub HE. Elastically Coupled Two-Level Systems as a Model for Biopolymer Extensibility. Physical Review Letters 1998;81:4764-7. [318] Bell G. Models for the specific adhesion of cells to cells. Science 1978;200:618-27. [319] Evans E. Probing the Relation Between Force-lifetime-and Chemistry in Single Molecular Bonds. Annual Review of Biophysics and Biomolecular Structure 2001;30:105-28. [320] Watanabe K, Nair P, Labeit D, Kellermayer MSZ, Greaser M, Labeit S, et al. Molecular Mechanics of Cardiac Titin's PEVK and N2B Spring Elements. Journal of Biological Chemistry 2002;277:11549-58. [321] Li H, Oberhauser AF, Redick SD, Carrion-Vazquez M, Erickson HP, Fernandez JM. Multiple conformations of PEVK proteins detected by single-molecule techniques. Proceedings of the National Academy of Sciences 2001;98:10682-6.  237  [322] Gosline BBAJM. Optical properties of single elastin fibres indicate random protein conformation. Nature 1980;287:865-7. [323] Savage KN, Gosline JM. The role of proline in the elastic mechanism of hydrated spider silks. JOURNAL OF EXPERIMENTAL BIOLOGY 2008;211:1948-57. [324] Dudek DM, Elvin, C. M. & Gosline, J. .  Annual Meeting of the  Society-for-Integrative-and-Comparative-Biology. Boston, MA2009. [325] Amin AFMS, Lion A, Sekita S, Okui Y. Nonlinear dependence of viscosity in modeling the rate-dependent response of natural and high damping rubbers in compression and shear: Experimental identification and numerical verification. International Journal of Plasticity 2006;22:1610-57. [326] Sheldon RA. Enzyme Immobilization: The Quest for Optimum Performance. Advanced Synthesis & Catalysis 2007;349:1289-307. [327] Tischer W, Wedekind F, Fessner W-D, Archelas A, Demirjian D, Furstoss R, et al. Immobilized Enzymes: Methods and Applications Biocatalysis - From Discovery to Application. Springer Berlin / Heidelberg; 1999. p. 95-126. [328] Chockalingam K, Blenner M, Banta S. Design and application of stimulus-responsive peptide systems. Protein Engineering Design and Selection 2007;20:155-61.  238  [329] Bhak G, Lee S, Park JW, Cho S, Paik SR. Amyloid hydrogel derived from curly protein fibrils of alpha-synuclein. Biomaterials 2010;31:5986-95. [330] Abbate V, Kong X, Bansal SS. Photocrosslinked bovine serum albumin hydrogels with partial retention of esterase activity. Enzyme and Microbial Technology 2012;50:130-6. [331] McCall JD, Anseth KS. Thiol-ene Photopolymerizations Provide a Facile Method To Encapsulate Proteins and Maintain Their Bioactivity. Biomacromolecules 2012;13:2410-7. [332] J.P.  Joseleau JC, and K.  Ruel. Xylans and Xylanases. Amsterdam:  Elsevier; 1992. [333] A.M.  Nissen LA, N. Munk,  and N.K.  Lange. Xylans  andxylanases. Amsterdam: Elsevier; 1992. [334] Ziser L, Setyawati I, Withers SG. Syntheses and testing of substrates and mechanism-based inactivators for xylanases. Carbohydrate Research 1995;274:137-53. [335] McIntosh LP, Hand G, Johnson PE, Joshi MD, Korner M, Plesniak LA, et al. The pKa of the General Acid/Base Carboxyl Group of a Glycosidase Cycles during Catalysis: A 13C-NMR Study of Bacillus circulans Xylanase. Biochemistry 1996;35:9958-66.  239  [336] Sung WL, Luk CK, Zahab DM, Wakarchuk W. Overexpression of the Bacillus subtilis and circulans Xylanases in Escherichia coli. Protein Expression and Purification 1993;4:200-6. [337] Lawson SL, Wakarchuk WW, Withers SG. Effects of both Shortening and Lengthening the Active Site Nucleophile of Bacillus circulans Xylanase on Catalytic Activity. Biochemistry 1996;35:10110-8.  [338] Armstrong  Z, Reitinger  S, Kantner  T, Withers SG. Enzymatic Thioxyloside  Synthesis: Characterization of Thioglycoligase Variants Identified from A Site-Saturation Mutagenesis Library of Bacillus Circulans Xylanase. ChemBioChem 2010;11:533-8. [339] Shchukin DG, Sukhorukov GB, Price RR, Lvov YM. Halloysite Nanotubes as Biomimetic Nanoreactors. Small 2005;1:510-3. [340] Inoue N, Otsuka H, Wada S-I, Takahara A. (Inorganic Nanofiber/Enzyme) Hybrid Hydrogel: Preparation, Characterization, and Enzymatic Activity of Imogolite/Pepsin Conjugate. Chemistry Letters 2006;35:194-5. [341] Kennedy JFaC, J. M. S. Enzyme immobilization. Weinheim: VCH  Verlagsgesellschaft; 1987.  240  [342] Leon JV, Jose M. . Effect of Immobilization on the Kinetic and Stability Properties of O-Acetyl-L-Serine Sulfhydrylase from: Chlamydomonas reinhardtii. Biocatalysis and Biotransformation 1992;7:29-35. [343] Laidler KJ, Bunting PS, Daniel LP. [9] The kinetics of immobilized enzyme systems.  Methods in Enzymology: Academic Press; 1980. p. 227-48.  [344] Withers SG, Rupitz K, Street IP. 2-Deoxy-2-fluoro-D-glycosyl fluorides. A new class of specific mechanism-based glycosidase inhibitors. Journal of Biological Chemistry 1988;263:7929-32. [345] Eneyskaya EV, Brumer Iii H, Backinowsky LV, Ivanen DR, Kulminskaya AA, Shabalin KA, et al. Enzymatic synthesis of β-xylanase substrates: transglycosylation reactions of the β-xylosidase from Aspergillus sp. Carbohydrate Research 2003;338:313-25. [346] Martinek K MV. Immobilization of enzymes: an approach to fundamental studies in biochemistry. Adv Enzymol Relat Areas Mol Biol 1985;57:179-249. [347] Cheetham PSJ. Principles  of industrial enzymology: Basis  utilization of soluble and immobilized enzymes in processes.  of  industrial  : Chichester Ellis Horwood Ltd.; 1985.  [348] Mozhaev VV, Melik-nubarov NS, Sergeeva MV, siksnis V, Martinek K. Strategy for Stabilizing Enzymes Part One: Increasing Stability of Enzymes via their  241  Multi-Point Interaction with a Support. Biocatalysis and Biotransformation 1990;3:179-87. [349] Wang Q, Yang Z, Gao Y, Ge W, Wang L, Xu B. Enzymatic hydrogelation to immobilize an enzyme for high activity and stability. Soft Matter 2008;4:550-3. [350] Wang Q, Yang Z, Wang L, Ma M, Xu B. Molecular hydrogel-immobilized enzymes exhibit superactivity and high stability in organic solvents. Chemical Communications 2007:1032-4. [351] Allcock HR, Pucher SR, Visscher KB. Activity of urea amidohydrolase immobilized within poly[di(methoxyethoxyethoxy)phosphazene] hydrogels. Biomaterials 1994;15:502-6. [352] Grosová Z, Rosenberg M, Rebroš M, Šipocz M, Sedláčková B. Entrapment of β-galactosidase in polyvinylalcohol hydrogel. Biotechnology Letters 2008;30:763-7. [353] Tanaka A KT. Cell and enzyme immobili-zation. Washington DC:  ASM Press; 1999. [354] Lv S, Bu T, Kayser J, Bausch A, Li H. Towards constructing extracellular matrix-mimetic hydrogels: An elastic hydrogel constructed from tandem modular proteins containing tenascin FnIII domains. Acta Biomaterialia. [355] Brazel CS, Peppas NA. Mechanisms of solute and drug transport in relaxing, swellable, hydrophilic glassy polymers. Polymer 1999;40:3383-98.  242  [356] Laaksonen H, Hirvonen J, Laaksonen T. Cellular automata model for swelling-controlled drug release. International Journal of Pharmaceutics 2009;380:25-32. [357] Gerald Gerlach MG, Joerg Sorber, Gunnar Suchaneck, Karl-Friedrich Arndt, Andreas Richter. Chemical and pH sensors based on the swelling behavior of hydrogels. Sensors and Actuators B: Chemical 2005;111-112:555-61. [358] Zhang L, Seitz R. A pH sensor based on force generated by pH-dependent polymer swelling. Analytical and Bioanalytical Chemistry 2002;373:555-9. [359] Lin G, Chang S, Hao H, Tathireddy P, Orthner M, Magda J, et al. Osmotic swelling pressure response of smart hydrogels suitable for chronically implantable glucose sensors. Sensors and Actuators B: Chemical 2010;144:332-6. [360] Johnson BD, Beebe DJ, Crone WC. Effects of swelling on the mechanical properties of a pH-sensitive hydrogel for use in microfluidic devices. Materials Science and Engineering: C 2004;24:575-81. [361] Obukhov SP, Rubinstein M, Colby RH. Network Modulus and Superelasticity. Macromolecules 1994;27:3191-8. [362] Horkay F, Han M-H, Han IS, Bang I-S, Magda JJ. Separation of the effects of pH and polymer concentration on the swelling pressure and elastic modulus of a pH-responsive hydrogel. Polymer 2006;47:7335-8.  243  [363] Arnold JC, Li J, Isaac DH. The effects of pre-immersion in hostile environments on the ESC behaviour of urethane-acrylic polymers. Journal of Materials Processing Technology 1996;56:126-35. [364] Am Ende MT, Bell CL, Peppas NA, Massimo G, Colombo P. Measurement of the swelling force in ionic polymeric networks: II. Swelling force and disintegration of controlled release dosage formulations using pH-sensitive components. International Journal of Pharmaceutics 1995;120:33-40. [365] Elvin CM, Carr AG, Huson MG, Maxwell JM, Pearson RD, Vuocolo T, et al. Synthesis and properties of crosslinked recombinant pro-resilin. Nature 2005;437:999-1002. [366] Bustamante C, Marko J, Siggia E, Smith S. Entropic elasticity of lambda-phage DNA. Science 1994;265:1599-600. [367] Kalyanasundaram K. Photophysics, photochemistry and solar energy conversion with tris(bipyridyl)ruthenium(II) and its analogues. Coordination Chemistry Reviews 1982;46:159-244. [368] Kuhn W, Grun F. Beziehungen zwischen elastischen Konstanten und Dehnungsdoppelbrechung hochelastischer Stoffe. Kolloid-Zeitschrift 1942;101:248-71. [369] Kamrunnahar M, Huang B, Fisher DG. Estimation of Markov parameters and time-delay/interactor matrix. Chemical Engineering Science 2000;55:3353-63.  244  [370] Erdoan M, Pekcan O. Fast transient fluorescence method for measuring swelling and drying activation energies of a polystyrene gel. Polymer 2004;45:2551-8. [371] Sommer J-U, Chasse W, Valentin JL, Saalwachter K. Effect of excluded volume on segmental orientation correlations in polymer chains. Physical Review E 2008;78:051803. [372] Mareci TH, Donstrup S, Rigamonti A. NMR imaging and relaxation study of polymer swelling and chain dynamics. Journal of Molecular Liquids 1988;38:185-206. [373] Paris R, Barrales-Rienda JM, Quijada-Garrido I. Dynamic swelling of hydrogels based on random terpolymers of N-isopropylacrylamide, methacrylic acid and poly(ethylene glycol) macromonomer. Polymer 2009;50:2065-74. [374] Achilleos EC, Prud'homme RK, Christodoulou KN, Gee KR, Kevrekidis IG. Dynamic deformation visualization in swelling of polymer gels. Chemical Engineering Science 2000;55:3335-40. [375] Lillie MA, Chalmers GWG, Gosline JM. Elastin dehydration through the liquid and the vapor phase: A comparison of osmotic stress models. Biopolymers 1996;39:627-39. [376] Erika L. McCallister EA, David Baker. Critical role of βετα-hairpin formation in protein G folding. Nature Structural Biology 2000; 7:669 - 73.  245  

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
http://iiif.library.ubc.ca/presentation/dsp.24.1-0073719/manifest

Comment

Related Items