<%BANNER%>

The Mechanism of Collagen Self-Assembly

Permanent Link: http://ufdc.ufl.edu/UFE0024315/00001

Material Information

Title: The Mechanism of Collagen Self-Assembly Hydrophobic and Electrostatic Interactions
Physical Description: 1 online resource (134 p.)
Language: english
Creator: Li, Yuping
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2009

Subjects

Subjects / Keywords: collagen, periodicity, salts, self, surface, surfactant
Materials Science and Engineering -- Dissertations, Academic -- UF
Genre: Materials Science and Engineering thesis, Ph.D.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract: Collagen molecules assemble into fibrils with ordered structure is a spontaneous and thermally driving process which is favored by a large positive entropy contribution due to the displacement of structured water around collagen molecules. The underlying principle on forming fibrils with periodicity is still unclear. Our goal is to study the mechanism of collagen fibril formation by analyzing the intermolecular interactions during fibril formation. By introducing the anionic surfactant, we found those surfactants can accelerate the fibrillogenesis even though there are no remarkable interactions before fibril formation. The denaturation of collagen occurred with increasing concentration of surfactants. It is possible that the adsorption of surfactants on collagen through hydrophobic interactions destabilize the collagen triple helix when collagen molecules are dehydrated during incubation. Collagen fibril formation is influenced by environmental factors especially the pH and electrolytes. We found collagen fibrils can be formed at pH from 6.6 to 9.2. Zeta potential measurements of soluble collagen indicate that the surface net charge of collagen is not only affected by the pH of medium but also by the presence of added salts. The acceleration of fibrillogenesis rate with increasing pH from 6.6 to 9.2 is consistent with a reduction of surface net charge since the isoelectric point of soluble collagen is approaching. The native D-periodicity of 62 nm was found except at pH 7.1 where collagen molecules form short banding of 50-60 nm in the early stage of fibrillogenesis which might be caused by unusual alignment of collagen molecules in fibrils. We also introduced different monovalent and divalent electrolytes into fibril formation instead of PBS buffer. It was found the fibril formation was facilitated by divalent ions. Divalent ions can shift the isoelectric point of collagen by adsorption. We proposed the adsorbed divalent ions form the salt bridges between collagen which facilitate fibril formation. The unusual periodicities along the collagen fibril were also found which might be come from altered alignment of molecules due to the change of surface charge.
General Note: In the series University of Florida Digital Collections.
General Note: Includes vita.
Bibliography: Includes bibliographical references.
Source of Description: Description based on online resource; title from PDF title page.
Source of Description: This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Statement of Responsibility: by Yuping Li.
Thesis: Thesis (Ph.D.)--University of Florida, 2009.
Local: Adviser: Douglas, Elliot P.

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2009
System ID: UFE0024315:00001

Permanent Link: http://ufdc.ufl.edu/UFE0024315/00001

Material Information

Title: The Mechanism of Collagen Self-Assembly Hydrophobic and Electrostatic Interactions
Physical Description: 1 online resource (134 p.)
Language: english
Creator: Li, Yuping
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2009

Subjects

Subjects / Keywords: collagen, periodicity, salts, self, surface, surfactant
Materials Science and Engineering -- Dissertations, Academic -- UF
Genre: Materials Science and Engineering thesis, Ph.D.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract: Collagen molecules assemble into fibrils with ordered structure is a spontaneous and thermally driving process which is favored by a large positive entropy contribution due to the displacement of structured water around collagen molecules. The underlying principle on forming fibrils with periodicity is still unclear. Our goal is to study the mechanism of collagen fibril formation by analyzing the intermolecular interactions during fibril formation. By introducing the anionic surfactant, we found those surfactants can accelerate the fibrillogenesis even though there are no remarkable interactions before fibril formation. The denaturation of collagen occurred with increasing concentration of surfactants. It is possible that the adsorption of surfactants on collagen through hydrophobic interactions destabilize the collagen triple helix when collagen molecules are dehydrated during incubation. Collagen fibril formation is influenced by environmental factors especially the pH and electrolytes. We found collagen fibrils can be formed at pH from 6.6 to 9.2. Zeta potential measurements of soluble collagen indicate that the surface net charge of collagen is not only affected by the pH of medium but also by the presence of added salts. The acceleration of fibrillogenesis rate with increasing pH from 6.6 to 9.2 is consistent with a reduction of surface net charge since the isoelectric point of soluble collagen is approaching. The native D-periodicity of 62 nm was found except at pH 7.1 where collagen molecules form short banding of 50-60 nm in the early stage of fibrillogenesis which might be caused by unusual alignment of collagen molecules in fibrils. We also introduced different monovalent and divalent electrolytes into fibril formation instead of PBS buffer. It was found the fibril formation was facilitated by divalent ions. Divalent ions can shift the isoelectric point of collagen by adsorption. We proposed the adsorbed divalent ions form the salt bridges between collagen which facilitate fibril formation. The unusual periodicities along the collagen fibril were also found which might be come from altered alignment of molecules due to the change of surface charge.
General Note: In the series University of Florida Digital Collections.
General Note: Includes vita.
Bibliography: Includes bibliographical references.
Source of Description: Description based on online resource; title from PDF title page.
Source of Description: This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Statement of Responsibility: by Yuping Li.
Thesis: Thesis (Ph.D.)--University of Florida, 2009.
Local: Adviser: Douglas, Elliot P.

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2009
System ID: UFE0024315:00001


This item has the following downloads:


Full Text

PAGE 1

1 THE MECHANISM OF COL LAGEN SELF ASSEMBLY: HYDROPHOBIC AND ELECTROSTATIC INTERACTIONS By YUPING LI A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR T HE DEGREE OF DOCTOR OF PHILOSOPHY UNIVERSITY OF FLORIDA 2009

PAGE 2

2 2009 Yuping Li

PAGE 3

3 To my grandma, the most influential person in my life

PAGE 4

4 ACKNOWLEDGMENTS I would like to thank Dr. Elliot P. Douglas, my a dvisor and committee chair sincerely for his guidance, understanding, patience, encouragement and support, most importantly, his conscientiousness during my graduate studies at University of Florida. I also thank my doctoral committee members: Dr. John Mec holsky, Dr. Christopher D. Batich and Dr. Richard Dickinson for their input, valuable discussion and accessibility. I want to thank all the members in Dr. Douglass group: ChyhChyang Luo and Lewei Pu for helping me get through the hard time s for the first two years at Gainesvil le; Andrew Stewart for his music ; Sangwon Choi and Sangjun Lee for their humor and entertainment; Changhua Liu, Jie Li for listening and discussion; Amran Asadi, Margo R. Monroe, Chelsea E. Catania and Sasha L. Perkins for their coll aboration. Especially, I want to thank Jennifer L.Wrighton for her assistance and help. I would want to thank Dr. Laure Gower for her assistance and guidance in my graduate career. Besides, I am very grateful for the friendship with all of the members in her research group: Sangsoo Jee and Taili Thula, with whom I worked closely and figured out many of the research problems; Mark Bewernitz for training me on polarized optical microscopy and took my samples for testing; Chih Wei Liao for providing suggestion and information on many technical problems. Moreover, I thank the Department of Materials Science and Engineering, the University of Florida for providing me a great place for studying and living. Additionally, I thank the Major Analysis Instrumentation Center (MAIC), the Particle Engineering Research Center (PERC) and the staffs for their hard work, expertise and assistance. Next, I want to acknowledge many friends I met at UF: Jianlin Li, Xiaobo Li and Hongxia Yan, Aijie Chen, Jessie Choi and Ying Zheng. Without you, I couldnt have such a joyful life in UF.

PAGE 5

5 Finally, and most importantly, I would like to thank my parents and my brother for the ir support, encouragement, unselfish love and faith.

PAGE 6

6 TABLE OF CONTENTS page ACKNOWLEDGMENTS ...............................................................................................................4 LIST OF TABLES ...........................................................................................................................9 LIST OF FIGURES .......................................................................................................................10 CHAPTER 1 INTRODUCTION ..................................................................................................................14 2 AN OVERVIEW OF COLLAGEN ........................................................................................18 Collagen and Its Applications .................................................................................................18 Structure of Collagen Triple Helix .........................................................................................20 Stabilization of Collagen Fibrils by Covalent Cross Linking ................................................22 Fibril Structure Model of Collagen .........................................................................................24 Structure and Mechanical Properties of t he CollagenMineral Composite in Bone ..............26 Mechanism of Collagen Fib rillogenesis .................................................................................28 Collagen Initial Aggregates in Lag Phase ..............................................................................29 Lateral Arrangement of Collagen Molecules in Fibrils ..........................................................30 Purpose of Our Study ..............................................................................................................30 3 KINETICS STUDIES OF COLLAGEN FIBRIL FORMATION ..........................................32 Introduction .............................................................................................................................32 Materials and Methods ...........................................................................................................33 Materials ..........................................................................................................................33 Polyacrylamide Gel Electrophoresis In SDS (SDS PAGE) ............................................33 TurbidityTime Measurement .........................................................................................33 Morphology Studies of Fibrils .........................................................................................34 Mathematical Analysis of Kinetics of Fibril Formation ..................................................35 Activation Energies for Fibril Formation ........................................................................36 Results .....................................................................................................................................37 Collagen Characterization ...............................................................................................37 Concentration Effects ......................................................................................................38 Temperature Effects ........................................................................................................42 Effects o f Phosphate a nd Ionic Strength .........................................................................44 Discussion ...............................................................................................................................47 Conclusion ..............................................................................................................................48 4 INTERACTIONS BETWEEN IONIC SURFACTANT AND COLLAGEN ON FIBRIL FORMATION .........................................................................................................................49 Introduction .............................................................................................................................49

PAGE 7

7 Surface Tension Measurement ...............................................................................................50 Materials and Methods ...........................................................................................................51 Materials ..........................................................................................................................51 Surface Tension Measurement of Surfactants .................................................................51 Turbidity Measurement ...................................................................................................51 SEM Measurements of Collagen Fibrils .........................................................................52 Results .....................................................................................................................................52 Surface Tension Measurements .......................................................................................52 Turbidity Measurements ..................................................................................................55 The Effect of Temperature and Activation Energy .........................................................59 Morphology of Collagen Fibrils ......................................................................................62 Discussio n ...............................................................................................................................65 Interactions between Collagen and Surfactants ...............................................................65 Conclusion ..............................................................................................................................67 5 PH EFFECTS ON COLLAGEN FIBRILLOGENESIS IN VITRO: ELECTROSTATIC INTERACTIONS AND PHOSPHATE BINDING ................................................................68 Introduction .............................................................................................................................68 Zeta Potential and E lectrical Double Layer ............................................................................70 Electrical Double Layer ...................................................................................................70 Zeta Potential ...................................................................................................................72 Sm oluchowskis Equation ...............................................................................................72 Hckels Equation ...........................................................................................................73 Materials and Methods ...........................................................................................................74 Materials ..........................................................................................................................74 Turbidity Measurements of Fibrillogenesis .....................................................................74 Electron Microscopy of Collagen Fibrils ........................................................................75 Image Analysis of Fibrils from TEM ..............................................................................75 Zeta Potential Measurement of Collagen Solution ..........................................................76 Results .....................................................................................................................................76 Turbidity Measurement at Different pH ..........................................................................76 Fibril Morphologies .........................................................................................................77 D Periodicity Mea surements ...........................................................................................81 Surface Charge Measurement ..........................................................................................82 Discussion ...............................................................................................................................83 Collagen Fibrill ogenesis ..................................................................................................83 Electrostatic Effects .........................................................................................................84 D periodicity ....................................................................................................................86 Conclusion ..............................................................................................................................87 6 THE FORMATION OF NATIVE FIBRILS INDUCED BY DIVALENT IONS .................88 Introduction .............................................................................................................................88 Protein Salt Interactions .........................................................................................................90 Salting out/Salting in Theory ..........................................................................................90 SurfaceTension Increment Effect ..................................................................................91

PAGE 8

8 Preferential Ion Binding Interactions ..............................................................................91 Materials and Methods ...........................................................................................................93 Materials ..........................................................................................................................93 Zeta Potential Measurement ............................................................................................93 Turbidity Measurements ..................................................................................................93 Electron Microscopy Measurement .................................................................................93 Small Angle X ray Scattering (SAXS) ............................................................................94 Circular Dichroism Measurement ...................................................................................94 Results .....................................................................................................................................95 Surface Charges Measurement ........................................................................................95 Turbidity Measurements ..................................................................................................96 Circular Dichrois m Spectroscopy ....................................................................................98 TEM Morphologies Of Fibrils Formed In Salts ..............................................................99 Kinetics Of Fibrillogenesis In Salts ...............................................................................103 D periodicity ..................................................................................................................104 Discussion .............................................................................................................................107 CollagenIon Complex ..................................................................................................107 Stability of Collagen and Fibril Formation ...................................................................107 Multivalent Ions Effect on Regulate Banding Pattern ...................................................108 Conclusion ............................................................................................................................110 7 SUMMARY AND FUTURE WORK ..................................................................................111 Summary ...............................................................................................................................111 Surfactant Effects ..........................................................................................................112 pH Effects ......................................................................................................................112 Salts Effects ...................................................................................................................113 Future work ...........................................................................................................................114 LIST OF REFERENCES .............................................................................................................117 BIOGRAPHICAL SKETCH .......................................................................................................134

PAGE 9

9 LIST OF TABLES Table page 21 List of the Collagen Types and Information on Chain Composition, Structure, Tissue Loca tion and Related Information ..........................................................................19 22 Commercially available co llagen based medical devices ..................................................20 23 Amino acid composition of type I collagen .......................................................................21 41 CMC values of SDS determined in salt solution. ..............................................................54 51 D periodicity of collagen fibrils at various pHs and fibrillogenesis times ........................82 61 D periodicity of collagen fibrils in presence of different salts .......................................105

PAGE 10

10 LIST OF FIGURES Figure page 21 Model of the collagen triple helix. .....................................................................................22 22 Chemical stru ctures of collagen crosslinks in vivo. ...........................................................24 23 One dimensional packing arrangement model of molecules in a fibril.. ..........................25 24 Sketch of the arrangement of collagen and hydroxyapatite composite .............................27 31 Coomassie blue stained SDS PAGE of collagen.. .............................................................38 32 A typical turbi dity curve presenting an increase in light absorbance as collagen fibrils are formed ................................................................................................................39 33 Effects of collagen concentration on collagen fibril formation. ........................................40 34 Linear dependence of A on c oncentration of collagen type I ..........................................41 35 Effect of collagen concentrat ion on rate of fibril formation .............................................42 36 Turbidity measure ments of temperature effects on collagen fibril formation.. .................43 37 TEM images of collagen fibrils formed at different temperatures ...................................44 38 TEM images and corresponding fibril diameter distributions of collagen fibrils formed at various phosphate buffer concentrations and ionic strengths ............................46 41 Illustration of surface tension measurement by Wilhelmy plate method. .........................50 42 Surface tension of SDS in water, SDS in PBS and SDS in PBS mixed with collagen as a function of the concentration. .....................................................................................55 43 Turbidity measurements of collagen gelation (2.5mg/mL) in sodium dodecylsulfonate at the different concentrations ...............................................................56 44 Turbidity measurements of collagen gel ation (2.5 mg/mL) in sodium dodecylbenzenesulfonate at the different concentrations ..................................................57 45 Turbidity measurements of collagen gelation in (o.5mg/ml) the presence of sodium dodecylsulfonate (SDS) .....................................................................................................58 46 Turbidity measurements of collagen gelation (0.5mg/mL) in sodium dodecylbenzenesulfonate (SDBS) at the different concentrations. ....................................59 47 Temperature effects on collagen fibril formation in SDS ..................................................60

PAGE 11

11 48 Temperature effects on collagen fibril formation in SDBS ..............................................61 49 Photograph of shaked collagen gels formed in surfactants ...............................................62 410 SEM images of collagen fibrils formed in the presence of surfactants .............................64 51 Illustration of an electrical double layer at a solid liquid interface ...................................71 52 Turbidity measurements of collagen fibrillogenesis at different pH. ................................77 53 TEM images of self assemble collagen fibrils after three days of fibrillogenesis .............78 54 SEM images of collagen fibrils obtained in different pH conditions. ...............................79 55 Histograms of diameter distribution from pH 6.6 to 8.0 ....................................................81 56 Zeta potential measurement of surface charge of soluble collagen as a function of pH. ......................................................................................................................................83 61 Zeta potential measurement of surface charges of collagen affected by salts. ..................95 62 Effects of salt typ e and ionic strength on the collagen gelationat pH 7.4 ..........................97 63 Effects of salt type and ionic strength on the collagen gelation degree at pH 9.0.. ...........98 64 Circular dichroism spectra of collagen. .............................................................................99 65 TEM images of type I collagen fibrils in monovalent ions of NaCl and KCl ................101 66 TEM images of type I collagen fibrils formed in di valent cations of Na2HPO4 and Na2SO4. ............................................................................................................................102 67 TEM images of type I collagen fibrils formed in di valent cations of CaCl2 and M gCl2. ..............................................................................................................................103 68 Kinetics of collagen gelation in salts determined by turbidity measurements. ................104 69 Small angle X ray diffraction mea surement of collagen fibrils ......................................106

PAGE 12

12 Abstract of Dissertation Presented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy MECHANISM OF COLLAGE N SELF ASSEMBLY INTO FIBRILS: HYDROPHOBIC AND ELECTROSTATIC INTERA CTIONS By Yuping Li May 2009 Chair: Elliot P. Douglas Major: Materials Science and Engineering Collagen molecules assemble into fibrils with ordered structure is a spontaneous and therma lly driving process which is favored by a large positive entropy contribution due to the displacement of structured water around collagen molecules. The underlying principle on forming fibrils with periodicity is still unclear. Our goal is to study the mec hanism of collagen fibril formation by analyzing the intermolecular interactions during fibril formation. By introducing the anionic surfactant, we found those surfactants can accelerate the fibrillogenesis even though there are no remarkable interactions before fibril formation. The denaturation of collagen occurred with increasing concentration of surfactants. It is possible that the adsorption of surfactants on collagen through hydrophobic interactions destabilize the collagen triple helix when collagen molecules are dehydrated during incubation. Collagen fibril formation is influenced by environmental factors especially the pH and electrolytes. We found collagen fibrils can be formed at pH from 6.6 to 9.2. Zeta potential measurements of soluble collagen indicate that the surface net charge of collagen is not only affected by the pH of medium but also by the presence of added salts. The acceleration of fibrillogenesis rate with increasing pH from 6.6 to 9.2 is consistent with a reduction of surface

PAGE 13

13 net cha rge since the isoelectric point of soluble collagen is approaching. The native D periodicity of 62 nm was found except at pH 7.1 where collagen molecules form short banding of 5060 nm in the early stage of fibrillogenesis which might be caused by unusual alignment of collagen molecules in fibrils. We also introduced different monovalent and divalent electrolytes into fibril formation instead of PBS buffer. It was found the fibril formation was facilitated by divalent ions. Divalent ions can shift the isoel ectric point of collagen by adsorption. We proposed the adsorbed divalent ions form the salt bridges between collagen which facilitate fibril formation. The unusual periodicities along the collagen fibril were also found which might be come from altered al ignment of molecules due to the change of surface charge. These studies provide new insight on understanding the basis for collagen fibril formation and help for designing new biomaterials.

PAGE 14

14 CHAPTER 1 INTRODUCTION Collagen is the most abundant protein pr esent in connective tissue. It is biodegradable, biocompatible, and can enhance cell attachment and cellular penetration [1] The prevalence of collagen also makes it a natural choice as a polymer for biomedical materials and tissue engineering matrices. The potential value of collagen as a biomaterial has led to research on use in scaffolds for wound repair, collagen hydrogel as drug delivery, and scaffolds of tissue engineering[2, 3] Collagen is also a target for study in disease s involving extensive collagen remodeling, including aortic heart valve repair and bone repair [4, 5] A better understanding of interactions between collagen allow for the more rational design and use of collagen scaffolds for biomedical application as well as understanding the biological system. There are over 20 known types of collagen[6] Type I collagen is the major fibril component in bone, skin, tendons, ligaments, cornea, and internal organs accounting for 90% of body collagen; type II collagen comprising fibril lar cartilage, and vitreous humor of the eye; type III collagen comprising fibril skin, blood vessels, and internal organs. For all collagen type s each collagen molecule is composed of three left form the right handed triple helix. Each helix chain has about 1024 amino acids and each of collagen molecule is about 300 kDa, composed of about 10% each of proline and hydroxyproline, and has a glycine present at every third amino acid positions. Collagen molecules can self assembl e into fibrils with an ordered structure along the fibril in vitro The mechanism of collagen self assembly has been studied a lot. However, it is still unclear. From a thermodynamic point of view, collagen self a ssembly is an entropy driving process. When temperature is increased, entropy of water is increased due to the dissociation of structured water around the collagen and entropy of collagen is decreased a little compare d to the

PAGE 15

15 water. Even though it is an endothermic process, the free energy of the whole process is negative which suggests collagen fibril formation is a spontaneous reaction. When structured water molecules are dissociated during heating, the exposed side groups of amino acids are attract ed with each other. During the fibril formation, there exist many interactions, such as van der Waal s attraction, electrostatic repulsion and hydrophobic attr action. Which interaction force play a major role during fibril formation? Why collagen molecules can align along the fibril to form ordered structure? Does collagen recognize the specific regions of nearby one to form the native fibrils? In this dissertation, I tried to comprehensively study the molecular interac tions during fibril formation in order to provide insight in to the mechanisms of collagen self assembly into fibrils. A literature review of the collagen molecules self assemble into fibrils is descr ibed in Chapter 2. It starts at the hierarchical structur e of the collagen from the amino acids composition, triple helix structure to the collagen fibrils. The models of collagen fibrils from electron microscopy and XRD are also reported. T he mechanisms of collagen fibrils formation studied from kinetics, and m orphologies are also summarized. In Chapter 3, the kinetics of collagen fibril formation was studied by turbidity measurement. The effects of collagen concentration, temperature, and ionic strength of phosphate buffer on fibril formation were investigated. The activation energy of collagen fibril formation was calculated. TEM was used to examine the morphologies of collagen fibrils in order to understand the fundamental mechanism about fibril formation. By introducing anionic surfactants, sodium dodecylsulf ate (SDS) or sodium dodecylbenzenesulfonate (SDBS), we hope to know if there are specific interactions between surfactants and collagen molecules as well as how those interactions influence the fibril

PAGE 16

16 formation (Chapter 4). The anionic surfactants have bot h a hydrophobic hydrocarbon chain and negative charged head gr oup. When collagen molecules are dehydrated during fibril formation, the exposed side groups of collagen interact with surfactants. The rate of collagen fibril formation was found to be accelera ted remarkably in the presence of 0.10.5 mM SDS or SDBS while the unfolding of the collagen triple helix also occurred when up to 1 mM SDS was added or more than 0.35 mM SDBS was added. The morphology studies from SEM indicated partial unfolding occurred and non fibrillar collagen gel formed within fibrils. The binding and denatur ation effects of surfactants were also discussed. In Chapter 5, Collagen self assembly in vitro was conducted in the pH range from 6.0 to 10.5 in order to investigate the electrostatic interactions that occurred during fibril formation. Collagen fibril morphologies imaged by TEM and SEM present bundling of fibrils with a small amount of nonfibrillar collagen. Even though the rat e of fibrillogenesis accelerated with increasing pH in this range, the s ize distribution of fibrils did nt change significantly. We also found that the net surface charge of collagen molecules is not only affected by the pH of medium but also by the presence of added salts. It was found that electrosta tic interactions are important for fibril formation and the electrostatic interactions can be controlled through changing the pH of the medium. In Chapter 6 variant electrolytes instead of phosphate buffer is introduced in order to facilitate fibrillogenesis. In our studies, the facilitation of collagen aggregation through reducing surface net charge by divalent ions binding was observed. We also found the competition between unfolding and aggregation in collagen at high pH which inhibited the collagen fibril f ormation. The divalent ions binding to the collagen molecules not only change the surface charge but also create fibrils with native D periodic banding pattern. It is possible that the binding divalent ions induce the

PAGE 17

17 like charge attraction during collagen self assembly into fibrils The findings can help understanding the mechanism of fibrillgenesis. In Chapter 7, a summary for the dissertation and the potential research work for the future were described.

PAGE 18

18 CHAPTER 2 AN OVERVIEW OF COLLA GEN Collagen and It s Applications As one of the major components of connective tissue, collagens account for approximately 30% of all proteins in the human body[7, 8] They are often found in every major tissue that requires strength and flexibility, such as tendon, skin, bone, and fascia. There are over 20 types of collagens (Table 2 1), with the most abundant type being type I collagen (more than 90% of all fibrous protein). The word collagen describes a family of structurally related proteins that are located in the extracellular matrix of connective tissue. Collagen has played a critical role in the evolution of large complex organisms, where it acts as an insoluble scaffold for providing shape and form, for attaching biopolymers inorganic ions, and cells. They occur as supra molecular assemblies that range in morphology from rope like fibrils that provide the fibrous scaffold maintaining the integrity of tendons, ligaments and bone to net like sheets in the base discs that underlie epithelial and endothelial cells[9] Type I c ollagen has become the biomaterial of choice for a number of important medical applications due to[10] : well developed technique for obtaining large quantities of medical grade c ollagen, a number of established collagen product s, some of which are well known, a good safety profile as a biomaterial, easily used in minimally invasive procedures, the improved understanding of collagen s role in wound healing, metaboli sm, catabolism, and the interaction between cells and collagen

PAGE 19

19 Table 2 1. List of the Collagen Types and Information on Chain Composition, Structure, Tissue Location and Related Information[42] Types Chain Composition Structural Details Localization Notes I [a1(I)]2[a(I)] 300 nm, 67 nm banded fibrils Skin, tendon, bone, etc. 90% of all collagen of the human body. Scar tissue the end product when tissue heals by repair. II [a1(II)]3 300 nm, small 67 nm fibrils Cartilage, vitreous humor Articular cartilage III [a1(III)]3 300 nm, small 67 nm fibrils Skin, muscle, frequently with type I Collagen of granulation tissue, and produced quickly by young fibroblasts. IV [a1(IV)2[a2(IV)] 390 nm C term globular domain, nonfibrillar All basal lamina Basal lamina V [a1(V)][a2(V )][a3(V)] 390 nm N term globular domain, small fibers Most interstitial tissue,assoc. with type I Most interstitial tissue, assoc. with type I VI [a1(VI)][a2(VI)][a3(VI)] 150 nm, N1C term. globular domains, microfibrils, 100nm banded fibrils Most interstitial tissue, assoc. with type I Most interstitial tissue, assoc.with type I VII [a1(VII)]3 450 nm, dimer Epithelia VIII [a1(VIII)]3 130 nm, N1C term. Globular domains Some endothelial cells IX [a1(IX)][a2(IX)][a3(IX)] 200 nm, N term Globular domain, bound proteoglycan Cartilage, assoc. with type II X [a1(X)]3 150 nm, C term. Globular domain Hypertrophic and mineralizing cartilage XI [a1(XI)][a2(XI)][a3(XI)] 300 nm, small fibers Cartilage XII a1(XII) 75 nm triple he lical tail, central globule, three 60 nm globule arms Interacts with types I and III Mainly type I IV have been utilized to varying degrees in tissue engineering related biomaterials studies, with some efforts on the other types shown.

PAGE 20

20 The potentia l value of collagen as a biomaterial has led to research on its use in scaffolds for ligament repair, collagen grafts for scar and burn repair, and the engineering of osterochondral tissue [2, 1113] The commercial application s of collagen are listed in Table 22. Table 2 2. Commercially available collagen based medical devices[10] Medical Specialty Application General surgery hemostastis Dermatology Soft tissue augmentation Dentistry Oral woun ds; periodontal ligament attachment Ophthalmology Corneal shields Cardiovascular Anti infectious catheter cuffs; Arterial puncture repair Plastic and reconstructive surgery Wound dressings; Artificial skin Orthopedics Bone repair Urology Bulking agent for incontinence Drug delivery Cancer therapeutics; growth factors Structure o f Collagen Triple Helix Type I collagen, the principal component of the organic matrix of bone, as well as other connective tissues, is a large fibrous protein with a highly repetitive amino acid sequence [Gly X Y]n where X and Y can be any amino acids but are frequently the imino acids proline and hydroxyproline, respectively(Figure 21). The three peptide subunits are chain (Ta ble 23), composed of about 1024 amino acid residues. They can fold into a unique triple helical structure which consists of three domains: the NH2 terminated helical, the triple helical, and the COOH terminated helical domains. It is also called telopeptide collagen or tropocollagen. When the helical domains are cleaved by pepsin digestion, the collagen is called atelopeptide collagen The length of collagen molecule is about 300 nm, and the diameter is about 1.5nm [14] .The single uninterrupted triple helical

PAGE 21

21 domain represents more than 95% of the molecule. The triple helical conformation based on highangle X ray diffraction patterns, consists of three polypeptide chains entwined into a lefthanded helix with approximately 3.33 residues per turn and a unit twist of about 108, in which adjacent chains are staggered by one resi due. Successive units are translated by approximately 0.29 nm parallel to the helix axis (h) and chains are held together by two hydrogen bonds per triplet [15] One hydrogen bond is formed between the NH group of the residue of the glycyl residue and the CO group of the residue in the second position of the triplet in the neighboring chains. Another one is formed via a water molecule participating in the formation of additional hydrogen bonds with the help of the hydroxyl gr oup of 4hydroxyproline in the third position of the triplet (GlyX Y) [16] Table 2 3. Amino acid composition of type I collagen [6] Amino acid 3 Hydroxyproline 1 1 4 Hydroxyproline 108 93 Aspartic acid 42 44 Threonine 16 19 Serine 34 30 Glutamic acid 73 68 Proline 124 113 Glycine 333 338 Alanine 115 102 Half cystine 0 0 Valine 21 35 Methi onine 7 5 Isoleucine 6 14 Leucine 19 30 Tyrosine 1 4 Phenylalanine 12 12 Hydroxylysine 9 12 Lysine 26 18 Histine 3 12 Arginine 50 50 All 1000 1000

PAGE 22

22 Figure 21. Model of the collagen triple helix. The structure is shown for (GlyPro Pro)n in which glycine is designated by 1, proline in X position by 2 and proline in Y position by 3:A) and B) are side views. C) is top view in the direction of the helix axis. The three chains are connected by hydrogen bonds between the backbone NH of glycine and the backbone CO of proline in Y position. Arrows indicate the directions in which side chains other than proline rings emerge from the helix. The residueto residue distances are indicated in nm. [43] Stabilization of Collagen Fibrils b y Covalent Cross Linking The assembly of type I, II and III collagen into fibrils is usually accompanied by formation of inter and intra molecular covalent cross chains [17] The crosslinks in the fibrils provide the high tensile and mechanical strength needed for tissue integrity. These cross links are based on aldehyde formation and condensation

PAGE 23

23 involving specific peptidyl lysine and hydroxylysine residues [18] (F igure 22). The process is catalyzed by a single enzyme, lysyloxidase, which oxidatively deaminates th e amino group of certain lysine and hydroxylysine residues in the telopeptide regions of collagen molecules to form reactive allysyl and hydroxyallysyl aldehydes, respectively. These aldehyde groups then react amino group of lysine and hydroxylysine residues and other aldehydes to form a variety of di triand tetrafunctional cross links. Increasing the intermolecular cross links can increase the biodegradation time by making collagen less susceptible to enzymatic degradation; decreasing the capacity of the collag en to absorb water; and increasing the tensile strength of the collagen fibers. For longterm biomimetic applications, the thermal a nd mechanical stability of collagen must be achieved by crosslinking. Over the years, a number of well described physical or chemical techniques have been used such as de hydrothermal treatment, UV radiation and chemical cross linking agents [1, 19, 20] However, those methods might induce partial denatu ration and substantially alter its in vivo degradation. Especially, some cross linking reagents, such as aldehydes, polyepoxides, and isocyanates, are usually toxic and bioincompatible.

PAGE 24

24 H N H C C O C H2 C H2 H C R1 C H2 N H2 H N H C C O C H2 C H2 H C R1 C l ys y l oxi da s e R1= H l y s i ne R1= O H hyd r ox yl y s i ne O H N H C C O C H2 C H2 H C R1 C H C C H2 C O C H2 C H C H N O H H H N H C C O C H2 C H2 H C R1 C H2 C H2C H2 C H C H N O N C H H C R2 R2= H o r O H H N H C C O C H2 C H2 H C R1 C H H C C H2 C O C H2 C H C H N O H N N H2C H C N H C O hi s t i d i ne a l do c o nde ns a t i on H N H C C O C H2 C H2 C H2C H2 C H C H N O H2C H C R1 N+ O H H2C C H N H C O H y dr o xyp yr i di n i um c r os s l i nkA B C D E Figure 22. Chemical structures of collagen crosslinks in vivo. (A) oxidation of lysine and hydroxylysine via lysyloxidase, (B)intramolecular aldol condensation type of crosslink, (C) inter molecular aldimine type crosslink, (D) condensation of aldo condensation with hydroxyproline, (E) hydroxypyridinium type crosslink. Fibril Structure Model o f Collagen X ray scattering and electron microscopy are the principal techniques used to probe the structure of fibrils. The D periodicity in collagen fibrils w as observed by X ray scattering and electron microscopy. An explanation of the D periodicity in fibrils requires accurate determination of the molecular length of collagen and knowledge of the re lationship between the collagen molecule and the fibril banding pattern. Measurements

PAGE 25

25 performed in collagen show that the molecular length, l, of type I collagen is approximately 4.4D. In the fibril, molecules are staggered with respect to one another. Whe n they are regularly staggered by D (Figure 2 3), because of the nonintegral ratio of l/D, regions of gap and overlap are produced with the fibril; the overlap will be about 0.4D in axial extent and the gap will be approximately 0.6D [21, 22] The D periodicity of collagen fibrils is now known to an accuracy of one residue. Using a value of 234 residues for the 67nm D period length, the residue spacing, h, computes to 0.286 nm and is in good agreement with diffraction measurements of the axial translation of GlyX Y repeat units. Figure 23. One dimensional packing arrangement model of molecules in a fibril. The lateral packing arrangement of collagen molecules shown here (collagen molecules mutually staggered by 1D 67nm length) is only one of many possible arrangements. The gap overlap zones represent 0.6D a nd 0.4D, respectively. TEM image of a fibril negatively stained with phosphotungstic acid (1%, pH 7.4) is shown on the bottom

PAGE 26

26 Several models for the molecular packing of type I collagen fibrils have been proposed[23] either the tetragonal or hexagonal models. A better model is the quasi hexagonal p acking arrangement which fit s with the X ray diffraction data. This model is related with both the positions and intensities of the near equatorial reflections and molecules tilted by 4 to the fibr il axis to account for the reflections. In each unit, there are five molecules. Struc ture and Mechanical Properties o f t he Collagen Mineral Composite in Bone Bone has a very complex hierarchical structure and is optimized to achieve remarkable mechanical p erformance[24] The basic material of bone is the collagen mineral composite, containing nano sized mineral platelets (essentially carbonated hydroxyapatite), protein (predominantly collagen type I), and water. These components have different mechanical properties: the mineral is stiff and brittle while the protein is m uch softer but also more ductile than the mineral. Remarkably, the composite combines the optimal properties of the components, both the st iffness and the toughness. This rather unusual combination of material properties provides both rigidity and resistance against fracture. A sketch of the most probable arrangement of the mineral particles with respect to the collagen molecules[25, 26] is shown in Figure 24. Bone mineral is a poorly crystal line hydroxyapatite phase. The reflections obtained by selected area diffraction (SAD) confirmed the hydroxyapatite nature of the mineral crystals. The strongest reflection co mes from the 002 lattice planes indicating of the c axis of the crystal which was preferentially oriented to the longitudinal axis of the fibrils [27] Recent investigation on individual bone trabeculae revealed a predominant parallel orientation of the c axis of

PAGE 27

27 the mineral crystals with respect to the longitudinal axis of the trabeculae, whereby the crystals follow closely to the plane of the lamellae[27] Figure 24. Sketch of the arrangement of collagen and hydroxyapatite composite. Hydroxyapatite is typically v ery thin crystals (2 4nm) and aligned with the colla gen matrix. Hydroxyapatite is believed to be appeared within gap zone of fibrils (Reproduced from Rho et al.[44] ) Native type I collagen is hydrated moiety, greatly contributing to the viscoelastic properties of bone [28] The mineralized collagen fibril about 100 nm in diameter is the

PAGE 28

28 basic building block of the bone material. When the osteoblasts deposite the triple helical collagen molecules into the extracellular space, they build fibrils by a self assembly process. Adjacent molecules are staggered in their long axis by 67 nm in hydrated condition, generating a characteristic banding pattern of gap zones with 40 nm length and overlap zones with 27 nm length within the fibril [29] There is evidence t hat the mineral particles start in the gap zones, growing initially in the gap zone and then further into the overlap zone, forming aggregates of staggered plate like motifs [25, 26, 30, 31] Thereby, the long axis (c axis) of the mineral crystals is oriented parallel to the longitudinal axis of the fibrils. Mechanism o f Collagen Fibrillogenesis Reconstituted collagen fibrils can also be self assembled from purified collagen in vitro. The simplest system usually uses purified type I collagen extracted from skin or tendon. Typically, a solution of type I collagen in acidic solution is warmed and neutralized with buffer (pH 7.4, I 0.1) generating a gel of fibrils. The mechanism by which collagen molecules self assemble into fibrils has been a topic of intense research since the 1950s. Early studies evaluate d the kinetics of the transition of collagen from a soluble to a solid phase suggest that it was controlled by the addition of molecules at the fibril surface at temperatures above 16C and limited by the rate of diffusion of collagen molecules at temperatures less than 16C [32]. Thermodynamic studies indicated that native collagen fibril formation was an endothermic process that is thermodynamically favorable by the large increase in mobility of the water molecules as they are released Early work using electron microscopy suggests that fibril growth proceeds via nucleation of monomers and growth to a fibril at a constant shape [33] Further electron microscopic studies and examination from light scattering techniques indicate, however, that

PAGE 29

29 fibrillogenesis in vitro is a multistep process in which the formation of linear aggregates precedes lat eral growth [34] Comparison of theoretically derived translational diffusion coefficients of the collagen monomer with those obtained experimentally from quasi elastic light scattering studies of collagen solutions suggest that 4.4D staggered interactions occur in the early stages of assembly [35] Electron microscopic observations on the early stages of assembly support the existence of 4.4D staggered dimers, but also show that the conditions under which fibril formation occurs influence s the morphology of intermediate aggregates of the assembly p rocess[36] Telopeptide sequences at the N and C termina l of the triple helical domains are important in stabilizing initial aggregates by formation of aldehyde adducts in vitro [37] whereby molecules in reconstituted fibrils have a memorial influence on the dissociation and reformation of fibrils [38] Collagen Initial Aggregates in Lag Phase Early turbidimetric studies of collagen self assembly indicate that the lag period before fibril growth was important in determining the final fibril size. Factors such as pH, ionic strength, ion types, and temperature were found to influence the length of the lag period in manner that has only been explained phenomenologically. The mechanism of initiation, as well as a suitable explanation for lag phase and the influence of soluti on parameters should be important in understanding collagen self assembly. Inelastic light scattering data indicate that the first formed aggregates result from linear rather that lateral association of collagen molecules [39] i.e. molecules tend to associate with minimal overlap and hence maximal stagger. Gelman & Piez claim that this predominantly linear mode of assembly filaments is an intermediate step [34] However, recent evidence does suggest that no unique assembly pathway exists [40]

PAGE 30

30 Other methods used to initiate assembly can lead to the earlier onset of simultaneous linear and lateral association and th e more rapid formation of native fibrils. Lateral Arrangement o f Col lagen Molecules i n Fibrils Within a fibril, there is considerable long range order in the axial direction, with neighboring collagen molecules staggered axially by integral multiples of D (i.e., nD, n=1, 2, 3, or 4, where D~67 nm, depending on tissue sourc e). The lateral arrangement of molecules (i.e., in a plane perpendicular to the fibril axis) is less well established. In the fibrils from some tissues, x ray diffraction indicates only liquidlike short range order in the lateral packing, as shown by the diffuse equatorial scattering ( i.e., perpendicular to the fibril axis) with a maximum in the region of the intermolecular interference function[41] How size and form of collagen fibrils are regulated is not well understood. The regulation of fibril growth occurs in the growth step in vivo Although collagen fibrils have broad diameter and length distribution, individual fibrils tend to be f airly uniform in diameter except at their ends. All the features of the fibrils observed in vivo have been observed in reconstituted fibrils, except the uniformity of diameter. Therefore, the components that regulate fibril size and length may be lost when collagen is extracted or when fibrils are generated from pure solutions [7] However, studies on protease damaged collagen that lacks the telopeptide sequences suggest that those short extrahelical sequences are critical for the formation of the uniform diameter. P urpose o f Our Study It has been demonstrated that bone mineralization can be duplicated in Dr. Gowers group which involves the liquid liquid phase separation of a fluidic amorphous precur sor induced by the presence of small amount of an acidic polymer in the crystallizing

PAGE 31

31 solution. This process is termed the Polymer Induced Liquid Precursor (PILP) process. It has been proposed that mineralization of collagen occurs by PILP process in whic h the liquid precursor phase is drawn into the collagen fibrils through capillary action. Therefore, the structure of collagen fibrils may influence the ability of the liquid precursor to be drawn into the fibrils. The alignment of collagen molecules in fi brils results in hole and overlap zones where it is thought the liquid precursor may enter the hole zones. During collaboration on the project of mimicking bone, we found that forming collagen fibrils with native D periodicity is critical for the mineraliz ation process. In this dissertation, we hope to le arn how collagen molecules self assemble into the fibrils and what are the major factors lead ing to native fibril formation. T o answer the questions, we tried to examine the molecular interactions between c ollagen molecules during formation by introducing ionic surfactant, salts or controlling the surface charges of collagen. We hope our studies can help in understanding the biological system, and furthermore, bring new clues on the design and development o f new biomimetic materials. The interactions between collagen during fibril formation have been studied a lot. However, due to the technique deficiency, there is no systematical observation on the fibril structure at nano scale which casts a limitation on understanding the mechanism of fibril formation. In our studies, we focused on examining fibril structure by electron miscroscopy (TEM and SEM), especially measuring the banding length in the fibrils. We hope these examinations can bring new insight to understand the intermolecular interactions during fibril formation.

PAGE 32

32 CHAPTER 3 KINETICS STUDIES OF COLLAGEN FIBRIL FORMATION Introduction It is well known that collagen molecules form different types of aggregates in vitro, such as segment long spacing type (SLS), fibrous long spacing type (FLS), native type fibrils, fibrils without periodic striation, under select conditions [45 47] as observed by electron microscopy, which might come from collagen specific intermolecular interaction s The factors affecting the interaction between collagen molecules are divided into three classes. The first class is concerned with the environment in the solution, such as the pH, ionic strength, ionic species, other substances, temperature and pressure. The second class is the molecular structure of the collagen itself, such as collagen sources, existence of non helical regions, the formation of cross links. The third class is the factors affecting collagen fibrillogenesis due to biological considerations, such as the concentration of collagen. A number of studies have been carried on the effect of pH, ionic strength, and temperature on collagen fibril formation [40, 48, 49] Turbidity time examination s for type I collagen are commonly used to study the collagen fibril formation behavior. Turbidi metric curves are composed of lag and growth phases. During the lag phase, aggregation occurs primarily by linear addition of collagen molecules for forming end overlapped or 4.4D stagge red dimers and trimers [39, 50] The time of the lag phase is dependent on both the presence of helical and nonhelical ends [51, 52] and several studies suggest that charged and hydr ophobic residues are important for aggregation. However, there are some disagreements on whether the aggregates that form during the lag phase are unstable nuclei that grow by monomer addition or subfibrillar components that grow in a multiple steps [50, 52] It is clear from laser light scattering studies that physical changes can be measured during the lag phase[52] Near the end

PAGE 33

33 of the lag phase and during the growth phase lateral aggregation of elements is formed. Studies also suggest that lateral growth occurs via the formation of a subfibrillar unit composed of five trimers which can linearly and laterally associate with other subfibrillar components [39] Linear and lateral aggregation of these units may be controlled by interactions between attractive charged pairs [35] Since the experimental condition s and collagen source always influence fibrillogenesis kinetics and structure of fibrils, it is important to study the basic factors on fibril formation as wel l as the mechanism of self assembly in order to design and fabricate collagen fibrillar matrices. In this chapter, the kinetics of fibril formation, the effects of collagen concentration, temperature, and ionic strength of phosphate buffer were investigate d. Materials a nd Methods Materials Purified type I collagen (99%) was purchased as a solution of pepsinsolubilized adult bovine dermal col lagen dissolved in hydrochloric acid (0.012N) with the concentration of 2.9 mg/mL (Vitrogen; Cohesion Technologies, I nc., Palo Alto, CA).10fold phosphate buffered saline, NaOH (0.1N) and HCl (0.1N) were purchased from Sigma. Polyacrylamide Gel Electrophoresis In SDS ( SDS PAGE) Collagen in 0.12N HCl w ith different concentrations was used for gel electrophoresis by dial yzing them overnight against a sample buffer containing 3 to 8% NuPA GE Tris acetate Gels with Tris a cetate SDS buffer and stained with Coomassie blue. All samples were reduced and denatured before gel electrophoresis. Turbidity Time Measurement Collagen f ibrillogenesis in the range of 28C to 34C was studied by turbidity measurements. Chilled pepsin solubilized collagen solution was mixed with 10fold phosphate

PAGE 34

34 buffered saline and 0.1 M NaOH in an 8:1:1 volume ratio in an ice bath to give a final composit ion of 10 mM phosphate and 168 mM of NaCl. The mixture was poured into spectrophotometer cells, which were sealed and transferred to the cell compartment of a UV/Vis spectrometer (Perkin Elmer Lambda 800 UV/Vis spectrometer, Fremont, CA), equipped with a t hermostated cell holder, and maintained at the desired temperature by water circulation. The process of fibrillogenesis was monitored by recording the light transmittance at 400 nm wavelength as a function of time. The turbidity curves were plotted as foll ows: the absorbance at 400 nm was calculated by A=2 log (Transmittance), and the turbidity as an increase in absorbance was plotted versus time in minutes. Usually, a bsorbance was converted into turbidity by multiplying with 2.303[53] In our study, the absorbance was taken as an indirect indication of turbidity. The curve consisted of an initial lag phase with no turbidity change, a growth phase during which there was an increase in turbidity and a plateau where no further change in turbidity was observed. The t1/2 was defined reciprocal of t1/2 was taken as an estimation of the apparent rate of collagen fibril formation in order to obtain the activation energy of the overall process. Morphology Studies of Fibrils Typical studies of the morphologies of collagen fibrils were conducted by TEM, in which image contrast is caused by the relative ability of regions of the fibril to scatter or transmitted electrons. Sample staining with heavy, electron scattering elements is often carried out in order to enhance the contrast within org anic samples such as collagen. Within a fibril, the heavy metals which were deposited in the gap zone appear dark and the overlap zone, which has less metal deposition appear bright under the TEM examination. When preparing a sample, a small portion of col lag en fibrils washed with MiliQ water were first placed on a cooper grid with mesh size of

PAGE 35

35 300. After draining for 1 minute, it was negatively stained at room temperature for 15 second with 1% phosphotungstic acid at pH 7.4. Then, the grid was washed with water for 1 minute and air dried. The prepared specimens were examined in a Jeol TEM 200CX with acceleration voltage of 80 kV, and digital micrographs were taken at magnification of 37, 000X Mathematical Analysis of Kinetics of Fibril Formation The turbi dity of a macromolecule solution depends both on the size (molecular weight) as well as shape (particle dissipation factor). Therefore, turbidity increases result from size and shape changes which can be studied using a mathematical approach. Under ideal scattering conditions the turbidity (T) is given by equation 31, where T is directly proportional to the weight concentration of macromolecules in solution, C; the molecular weight, Mr; and the particle dissipation factor, Q [54, 55] The particle dissipation factor is inversely related to the largest macromolecular dimension and can be obtained for a given s hape from tabulated values [54, 55] As the particle gets bigger Q decreases and approaches zero. The cons tant of proportionality, H is a function of the index of refraction in solution n0, the refractive index equation 2. T=HMrCQ (3 1) 3n0 2(dn/dc)24) (3 2) As for a mixture of macromolecular species having different values of M an d Q, the average turbidity (T) is proportional to the average molecular weight (Mr) and average Q (Q) were used for analysis [56] By differentiating equation (31) with respect to time, the result of dif ferentiation is given by equation 33.

PAGE 36

36 = + + (3 3) When dc/dt is isolated on the left hand side equation 33 becomes: = ln + ln (3 4) When both sides of equation (34) are divided by the aver age molecular weight, equation (3 4) becomes a rate expression. Since the molecular weight increases as fibrils are formed, the left side of the equation (34) represents the decrease in the moles of collagen as fibrils are formed. A first order rate expre ssion for the disappearance of moles of collagen would have the following form: [ ] = [ ] (3 5) A plot of d[c]/dt versus [c] based on equation (35) is a straight line for a first order reaction and has a slop e equal to k, the rate constant. By integrating equation (3 5), the equation of (3 6) can be given below: ln[c]/[c]0= kt (3 6) As for the fi rst order reaction, the reaction rate constant can be expressed by t1/2, the time to reach the middle point of the final opacity k=ln2/t1/2 (3 7) However, a more simplified equation has been proposed where the turbidity is proportional to the concentration. It turns out that t1/2 in turbidity time measurement can be used in studying the kinetics of fibril formation. Activation Energies f or Fibril Formation Since the apparent rate constant can be treated as t1/2 as discussed above, the activation energy, Ea, for the fibril formation can be determined. The activation energy is obtained from the temperature dependence of the apparent rate constant (k) using the Arrhenius relationship where

PAGE 37

37 T is the absolute temperature in K, R is the gas constant and A is the rate constant as 1/T approaches 0. = / (3 8) A plot of ln( k) versus 1/T has a slope equa l to the activation energy divided by the negative gas constant. Therefore, by plotting lnt1/2 versus 1/T, the activation energies for fibril formation can b e obtained. Results Collagen Characterization The SDSPAGE analysis of collagen is shown in Figure 31. Line 1 is the protein markers. Lines 2 to 6 are the collagen content from 5 g to 40 g. A triple helix of type I collagen molecule is composed of tw PAGE monomer. [57] kDa and the type I collagen triple helix has a molecular weight about 360 kDa.

PAGE 38

38 Figure 31. Coomassie blue stained SDS PAGE of collagen. It was separated in 3 to 8% NuPAGE Tris Acetate Gels wit h Tris Acetate SDS buffer. La ne 1, mass marker; Lane 2, 5 g collagen; Lane 3, 10 g; Lane 4, 20 g; Lane 5, 30 g; Lane 6, 40 g. Concentration Effects Light scattering from collagen fibril formation is complex as described in the experimental section. I t is proportional to the particle dissipation factor and concentration of fibrils. However,

PAGE 39

39 it has been found that turbidity is proportional to the amount of precipitation and can be used for kinetics studies[58, 59] A typical turbidity time curve for type I collagen is shown in Figure 32. This curve is composed of three phases as descr ibed in the introduction part. I n the lag phase, th ere is no d etectable change in turbidity; in the growth phase, the turbidity changes rapidly, and at the plateau phase the turbidity remains constant again where the curve can be characterized by the time to reach the m iddle point of the final absorbance. 0 10 20 30 40 50 60 0.0 0.5 1.0 1.5 2.0 2.5 A at 400 nmTime (min) t 1/2A Figure 32. A typical turbidity curve presenting an increase in light absorbance as collagen fibrils are formed. The collagen concentration was 2.5 mg/mL and the temperature was 34C. The f ibrillogenesis of collagen in different concentrations were conducted from 0.5 mg/mL to 2.5 mg/mL at 34C (Figure 33). The rate of fibril forma tion and the final turbidity were

PAGE 40

40 highly affected by the concentration. By increasing the concentration, the rat e and the turbidity increased. versus concentration (Figure 3 4) indicates that the final turbidity is proportional to the collagen concentration, which is consistent with previous reports [34, 49, 56] The dependence of t1/2 on concentration was plotted with the format of log (1/t ) versus log concentration in Figure 35. The rate of fibril formation by measuring turbidity is propor tional to collagen concentration on the first power which suggests the mechanism of fibril formation is the same at different concentration s of collagen, as reported before[58] 0 10 20 30 40 50 60 70 80 90 0.0 0.5 1.0 1.5 2.0 2.5 A at 400 nmT (min)2.5mg/mL 2mg/mL 1.5mg/ml 1mg/mL 0.5mg/mL Figure 33. Effect s of collagen concentration on collagen fibril formation. Condition: 34C, pH 7.4; ionic strength 0.2; Na2HPO40.02M, NaCl 0.14N.

PAGE 41

41 0.5 1.0 1.5 2.0 2.5 0.5 1.0 1.5 2.0 2.5 C (mg/mL) Figure 34. Linear dependence of A on concentration of collagen type I. The relationship is consis tent with the theoretical relationship in equation (3 1).

PAGE 42

42 -0.4 -0.3 -0.2 -0.1 0.0 0.1 0.2 0.3 0.4 0.5 -1.8 -1.7 -1.6 -1.5 -1.4 Log (1/t1/2)LogC Figure 35. Effect of collagen concentration on rate of fibril formation. Condition: 34C, pH 7.4, 10 mM phospate.168 mM NaCl. Temperature Effects It is well known that temperature is of primary importance in collagen fibril formation. The results in Figure 36 indicate that the fibrillogenesis rate is influenced by the change of temperature. The higher the temperature, the faster the collagen fibrillogenesis proceeds. An Arrhenius plot for collagen fibr il formation gi ve an activation energy of 167.6 kJ/mol, which is similar to the result of fibrillogenesis from telopeptides collagen [53, 58] It has been found that the removal of nonhelical ends inhibited the initiation of the self assembly of collagen molecules. [59] As for the atelopeptides which was used in studies here, it should have a high activation energy compared to telopeptides collagen in the same condition. However, the activat ion energy of fibrillogenesis can be also a result of the experimental conditions and

PAGE 43

43 collagen source. Therefore, i t is difficult to compare the activation energy to literature data directly. 32.4 32.6 32.8 33.0 33.2 -4.5 -4.0 -3.5 -3.0 -2.5 Ea=167.6kJ/molln(1/t1/2)(1/T) x10 4(K-1) B Figure 36. Turbidity measurements of temperature effects on collagen fibril formation. A) Turbiditytime measurements of fibril formation in different temperature. B) An Arrhenius plot for collagen fibril formation from A. Collagen 2.5 mg/mL, pH 7.4. 0 20 40 60 80 100 120 140 0.0 0.5 1.0 1.5 2.0 35C 34C 32C 28CA at 400 nmTime (min) 30C A

PAGE 44

44 Temperature not only affect s the fibri llogenesis rate, it also affect s the final fibrils morphology. As the temperature is decreased to less than 16C collagen molecular movements are inhibited, resulting in a reduced rate of fibril formation and most of the collagen appear s as a mixture of filamentous aggregates with occasional thin fibrils. Moreover, at lower temperatures (e.g. room temperature), it is hard to form fibrils even after 10 days of storage while most of the collagen appears as a mixture of amorphous gel (data not show) Collagen thermal stability can also affect the fibril formation. When temperatures are higher than 37C the collagen molecules undergo helixcoil transition, and no fibrils are formed. In TEM measurements (Figure 37), with increasing temperature from 28C to 37C, the fibril size is decreased and the small fibrils with unclear D periodicity appeared at 37C. Figure 37. TEM imag es of collagen fibrils formed at different temperature s A) Collagen fibrils are formed at 28C. B) Fibrils are formed at 30C. C) Fibrils are formed at 37C. Collagen concentration is 2.5 mg/mL, scale bar 1 m. Effects of Phosphate a nd Ionic Strength The effects of phosphate and ionic strength on collagen fibrillogenesis were also conducted at a constant pH, including 10mM phosphate w ith 168mM sodium chloride, 18mM phosphate

PAGE 45

45 with 302mM sodium chloride, and 25mM phosphate with 421mM sodium chloride. At low phosphate concentrations, native fibrils were formed (Figure 38A). The fibrils show ed the D periodicity consisting of a fine polari zed band pattern superimposed on alternate light and dark regions. The fibrils had diameters of about 130 to 200 nm and had a constant diameter along the fibrils direction but tend ed to bend and entangle d with each other. At phosphate concentrations near 18 mM (Figure 3 8B), the native D periodicity appeared clearly and fibrils had diameters of about 120 to 250 nm. At high phosphate concentrations, the fibrils had sizes of about 150 nm, which was smaller than in low phosphate concent rations and the D perio dicity was unclear (Figure 3 8C). Fibril formation in the absence of the PBS was also conducted (data no t show), as a result, amorphous collagen gels with occasional fibrils were presented.

PAGE 46

46 Figure 38. TEM images and corresponding fibril diameter distributions of collagen fibrils formed at various phosphate buffer concentrations and ionic strengths. A) 10mM phosphate and 168mM NaCl. B) 18mM phosphate and 296mM NaCl. C) 25mM phosphate and 402mM NaCl. Average fibril widths are in nm and standard deviations are shown. The number of fibrils measured was at least 50 for each sample. Scale bar is 0.5 m.

PAGE 47

47 Discussion Turbiditytime studies of collagen fibrillogenesis have shown that collagen aggregation is a multiple step process which is composed of a lag phase, growth phase and plateau phase. It has been proposed that collagen fibril formation is a nucleation and growth process. But the power dependence of t1/2 on collagen concentration of one indicates it is not the typical nucleation growth process. Since the nucleation growth process of keratin assembly has a power value of about 3[60] and hemoglobin S assembly is 30 to 40[61] A power dependence of one suggests simple grow th by accretion[58] Time lapse AFM studies on fibril formation by Cisneros, DA et al. suggest that collagen self assembly is a multistep process: at the first step, the collagen molecules assemble with each other to form aggregates; at the second step, those aggreg ates turn into microfibrils which have diameters of about 4 nm then these microfibrils self assemble into fibrils through longitudinal and lateral packing.[62] These multiple processes can t be directly examined by turbidity measurement since turbidity measurement cant measure the formation of mi crofibrils due to its small size. Based on the microfibril model of the collagen fibril, the lag phase may be dominated by forming intermediate and linear growth of microfibrils, which are too small to be detected by turbidity while the turbidity increase measures primarily lateral assembly of microfibrils to fibrils. Self assembly of collagen type I molecules into fibrils is a temperature favored process which is thermodynamically favorable by a large positive entropy contribution, presumably arising from structural reorganization of the water during fibril formation [63, 64] The importance of structured water in molecular interactions has been recognized and the interactions involving both polar and nonpolar molecules in water can be stabilized by the large positive entropy contributions coming from releasing the structured water. Our studies on changing temperatu re for fibrillogenesis indicate low and high temperature inhibiting fibril formation. At

PAGE 48

48 the low temperature, it is diff icult to break the structured water and the repulsion between collagen molecules could not be overcome to form the fibrils. At the high temperature, partial denaturation of collagen triple helix also inhibits fibril formation. Temperature from 28 to 37C i s the suitable range for fibril formation without interrupting collagens triple helix structure. Phosphate appear s to play a critical role in fibril formation beyond its capacity as a buffer [58]. Although phosphate inhibit ed the rate of fibril formation [65] its role cannot be only related to a rate effect, since 18mM is sufficient to produce well ordered native banded fibrils. The nature of its interaction with collagen is not known, but it is necessary to add phosphate to trigger fibrils with native banding formation. Based on the se results, phosphate at 18mM and 302mM sodium chloride was included in all further experiments. The specific role of phosphate on fibril formation will be studied in more detail in Chap ter 5. Conclusion The collagen fibril formation in pepsin digested collagen extracted from bovine dermis has been followed by turbidimetric examination under a range of experimental conditions. All fibrillogenesis curves showed a lag period, followed by a quick growth phase. Both phases were accelerated by increasing collagen concentration and temperature. Collagen fibril formation is a temperature favored first order reaction and the activation energy is 16 7 kJ/mol. It is a multistep process which involves of nucleation, linear and lateral growth. Phosphate ions play a key role on forming native banded fibrils which can not simply be explained by buffer solution. The specific effects of phosphates will be investigated in Chapter 5.

PAGE 49

49 CHAPTER 4 INTERACTIONS BETWEEN IONIC SURFACTANT AND COLLAGEN ON FIBRIL FORMATION Introduction The surfactants, ionic or nonionic in nature, interact with protein which leads to significant changes in the bulk properties of protein solution. Sodium dodecyl sulfate (SDS), an ionic surfactant, has been generally used to denature protein for SDS PAGE. SDS breaks up the twoand three dimensional structure of the proteins by adding negative charge to the amino acids during heating. Through the like charge repulsion, the proteins are m ore or less straightened out, immediately rendering them functionless. Alkyl sulphates, at concentrations below their critical micelle concentrations, (CMC) form complexes with human serum albumin and bovine serum albumin[66] It is generally known that the formation of these complexes is accompanied by unfolding of the protein structure [66 68] However, in some cases, the occurrence of the assoc iation between surfactants and bovine serum albumin has been observed without drastic conformational change [69] It has been reported that ionic surfactants can accelerate the rate of collagen fibrillogenesis [70] However, the acceleration effects and how surfactants can influence the morp hologies of collagen fibrils are not well understood. The addition of anionic surfactants (e.g. SDS) stimulated the initial stage of aggregation and the growth, while non ionic surfactants with oxyethylene groups stimulated all stages of fibril formation and formed the thinner collagen fibrils at low concentrations of surfa ctants[70, 71] Investigations on the adsorption of anionic and nonionic surfactants to collagen suggest multiple adsorption occurred and the adsorption of nonionic surfactant on collagen through the hydrophobic interactions [72] The systematic study of the role of SDS in conformational transitions of collagen suggest that under the non isoelectric

PAGE 50

50 conditions, both the electrostatic and hydrophobic interactions happen while at isoelectric conditions, hydrophobic interactions are dominant.[73] Surface Tension Measurement There are many methods which can be used to measure the surface tension. A s imple one is the Wilhelmy plate method[74] The b asic mechanism of Wilhelmy plate method is tha t a thin plate, su ch as a microscope cover glass or platinum foil will sup port a meniscus whose weight is measured by detachment given very accurately by the ideal equation (assuming zero contact angle): cos = ( ) 2 (4 1) Wtotal is the total weight to be measured, Wpl ate is the weight of platinum plate, b is the buoyance force, l is the length of plate (Figure 4 1). Normally, a the plate just touches liquid so the buoyancy is small which can be ignored. Figure 41 Illustration of s urface tension measurement by Wilhelmy plate method. In this chapter, sodium dodecylsulfate (SDS) or sodium dodecylbenzenesulfonate (SDBS) were used to study the effects of ionic surfactants on collagen fibril formation. Th e interactions between surfactant and collagen were observed by turbidity and surface tension measurements.

PAGE 51

51 The denaturation effect of surfactants on collagen and the specific interactions bet ween surfactant and collagen were discussed Materials and Metho ds Materials Purified type I collagen (97%) with the remainder being comprised of type III collagen was purchased as a solution of pepsinsolubilized adult bovine dermal collagen dissolved in hydrochloride acid with a concentration of 3.0 mg/mL (PureCol, INAMED, Inc., Fremont, CA). 10x concentrate phosphate buffered saline (PBS) pH 7.4, sodium dodecylsulfate (SDS) and sodium dodecylbenzenesulfonate (SDBS) were purchased from Sigma. Surface Tension Measurement of Surfactants Surface tension was measured usi ng the Wilhelmy platinum plate method (~5 cm perimeter) at 171C. The Wilhelmy plate was washed by de ionized water, methanol, acetone and water, then burned for 15 second to remove impurities. All aqueous surfactant solutions were kept at room temperatur e for at least 30 minutes before measurement in order to obtain stable data. Critical micelle concentrations (CMCs) were determined from the inflection points of the respective surface tension curves. The standard deviation for the experimental surface ten sion was less than 0.02mN/m. Surface tensions of surfactant in collagen solution were also measured in the same process. Turbidity Measurement To produce collagen fibrils, the chilled type I collagen solution was mixed with PBS at the ratio of 8:1 v/v and self assembled with incubation at 30C for 2 hours. The pH of the solutions was mediated by adding 0.1M NaOH. The degree of the fibrillar assembly with the incubation time was recorded by measuring the transmittance at 400nm using a UV Vis spectrometer

PAGE 52

52 (P erkin Elmer Lambda 800 UV/Vis spectrometer, Fremont, CA), equipped with a thermostat cell holder, and maintained at the desired temperature by water circulation. SEM Measurements o f Collagen Fibrils The morphology of collagen gel was examined by scanning electron microscopy (SEM, Jeol, JSM6335F). The freeze dried collagen samples were mounted on a stub and carbon coated. The morphologi es were observed at an accelerating voltage of 15kV. Results Surface Tension Measurements In order to investigate the inte ractions bewteen surfactant and collagen molecules, the surface tension s of surfactants solution with/without collagen at room temperature were measured. It is known that surfactants can reduce the surface tension of water by absorbing at the air/water int erface. They can also assemble into aggregates in the bulk solution that are known as micelles. The concentration at which surfactants begin to form micelles is known as the critical micelle concentration (CMC). As reported in the literature, the CMC value of surfactants decreases as t he concentration of electrolyte increased since the electrolytes neutralize the charge around the micelle surface and reduces the thickness of the ionic atmosphere around the surfactant ionic heads and the electrostatic repuls ions between them [75]. The decrease in the CMC of an ionic surfactant due to the effects of the electrolyte counter ion can be quantified through the following empirical expre ssion, developed by Corrin and Harkins [76] : ic b a CMC log log (4 2) where a and b are constants that mainly depend on the surfactant species and the temperature, and ci is the total counterion concentration. For SDS, Corrin et al. gave the following equation at 25C based on the results determined by a s pectrophotometric method[76]:

PAGE 53

53 ic CMC log 458 0 249 3 log (4 3) Fuguet et al. obtained a similar equation using a different method (conductometric measurements) [77] They found that this equation can accurately predict the CMC of SDS obtained from experimental measurements: ic CMC log 486 0 230 3 log (4 4) The surface tensions of SDS in MilliQ water and PBS were measured at room temperature (Figure 4 2). It display s the typical surfactant behavior in aqueous solution that lowers the surface tension of water by absorbing at the air water interface. The reduction of the surface tension of SDS after the addition of PBS was observed. In order to check the accuracy of the measurements, the CMC values of SDS which was calculated from inflection points in the figure are listed in Table 1. Compared to the literature data, the CMC value is matched well with that of the literature: the CMC of SDS in MilliQ water is 10.5mM and in PBS is 1.1mM. The surface tensions of collagen solutions in PBS were 72.4mN/m (0.5mg/mL), 72.12mN/m (1.0mg/mL) and 69.43mN/m (2.5mg/mL) respectively, which is close to MilliQ water (72.88mN/m at 20C). After adding collagen wi th concentration of 0.5, 1.0 and 2.5mg/mL respectively, a small reduction in the surface tension of SDS in collagen was observed without changing the CMC value. The surface tension of SDBS in water was also conducted and the critical micelle concentration of SDBS in PBS was found at 0.1mM.

PAGE 54

54 Table 4 1. CMC values of SDS determined in salt solution. Electrolyte concentration (mM) CMC (mM) SDS a SDS b SDS c SDS d SDS e 0 8.08 8.13 10.5 5 6.09 6.38 7.73 10 4.61 4.65 5.52 15 3.84 3.86 4.53 20 3.27 3.38 3.94 30 2.73 3.12 2.81 3.24 40 2.30 2.46 2.81 50 1.99 2.22 2.53 100 1.33 1.62 1.8 168 1.1 1.28 1.4 200 0.90 1.18 1.29 400 0.55 0.86 0.92 500 0.52 0.77 8.25 a All values of CMC are from data in ref. [77] SDS in phosphate buffer (pH 7.0). b Value of CMC is from data in ref. [86] SDS in the presence of NaCl The micelles in these solutions exist in the presence of monom eric detergent ions of concentration equal to the added electrolyte concentration plus the CMC. This only has an appreciable effect on the results for the solution containing no added salt. c Value of CMC is from our data measured by surface tension. d Cal culated by equation (43) e Calculated by equation (44)

PAGE 55

55 -4.5 -4.0 -3.5 -3.0 -2.5 -2.0 -1.5 -1.0 30 35 40 45 50 55 60 65 70 75 Surface tension (mN/m)LogC (mol/L) water PBS 0.5mg/mL collagen 1.0mg/mL collagen 2.0mg/mL collagen Figure 42. S urface tension of SDS in water SDS in PBS and SDS in PBS mixed with collagen as a function of the concentration. Turbidity Measurements The effects of surfactants on the collagen fibril formation measured by turbidimetric method are shown from Figure 43 to Figure 46 In high concentration of collagen (2.5 mg/mL), the fibrillogenesis rate was accelerated by S DS from 0.1 to 0.5 mM(Figure 43). By increasing th e concentratio n of SDS, the fibrillogenesis was significantly promoted and the final turbidity did not change with surfactant concentration. Moreover, the acceleration occurred in both lag and growth phases. Compare to the SDS on accelerating collagen fibr il formation, SDBS also promoted the fibrillogenesis rate with increasing concentratio n from 0.1 to 0.5mM. (Figure 44 ) The early precipitation was also observed before incubation w hen the concentration of SDBS was higher than 0.1 mM. Since SDBS is more hy drophobic than SDS and has the CMC at 0.1

PAGE 56

56 mM in PBS buffer it is reasonable that precipitation results from the formation of SDBS micelles. During incubation, collagen fibrillogenesis occurred and the turbidity increased again in the presence of SDBS. Low er final turbidity was observed in high concentration of SDBS which might be resulted from collagen denatuation. 0 5 10 15 20 25 30 0 20 40 60 80 100 T% at 400 nmTime (min) 0mM 0.1mM 0.2mM 0.5 mM 0.35mM Figure 43. Turbidity measurements of collagen gelation (2.5mg/mL) in sodium dodecylsulfonate at the different conc entrations Thermometer temperature was 34C.

PAGE 57

57 0 5 10 15 20 25 30 0 20 40 60 80 100 T% at 400 nmTime (min) 0mM 0.1mM 0.2mM 0.35mM 0.5mM Figure 44. Turbidity measurements of collagen gelation (2.5 mg/mL) in sodium dodecylbenzenesulfonate at the different concentrations Thermometer temperature was 34C. The effects of surfactant on collagen fibril formation at collagen concentration of 0.5 mg/mL were also investigated. It was found that SDS accelerated the collagen fibril formation with increasing surfactant concentration from 0.1 mM to 1 mM but the final turbidity was appreciably decreased at high surfactant concentrations. (Figure 4 5) In 1 mM SDS, where the molal ratio of SDS to collagen is 600 considering the molecular weight of collagen of about 300,000 g/mol, the collagen fibril formation occurred at the beginning of 15 minutes and then dissolved. Thes e results indicate that the high concentration of SDS inhibits fibril formation by unfolding the collagen triple helix, even though the incubation temperature, 30C, was below the body temperature. In the presence of SDBS, the stronger denaturation effect on collagen was observed (Figure 46) and collagen molecules could not form fibrils when the molal ratio of

PAGE 58

58 SDBS to collagen was up to 200. At ratios below 200, a decrease on final turbidity was also observed even though acceleration on fibrillogenesis rate occurred. When in the high concentration of collagen (2.5 mg/mL) where the ra tio of surfactant to collagen was in the range from 13 to 63, the final turbidity was not noticeably changed. As the molal ratio of surfac tant to collagen increased, the inhibition on final turbidity significantly appeared. This i nhibition effect on turbidity was also stronger in the presence of more hydrophobic surfactant. Hayashi and Nagai have observed similar results that 1.7 mM SDS inhi bited the collagen fibril formation in the concentration of 1.2 mg/mL by denaturing collagen molecule s while 0.14 mM SDS promoted fibril formation [71] 0 10 20 30 40 50 60 70 80 90 10 20 30 40 50 60 70 80 90 100 0.1mM 0mM 0.35mM 0.5mM 1mMT% at 400nmTime (min) Figure 45. Turbidity measurements of collagen gelation in the presence of sodium dodecylsulfonate (SDS). Collagen concentration was 0.5mg/ml, thermometer temperature was 34 C.

PAGE 59

59 0 10 20 30 40 50 60 70 80 30 40 50 60 70 80 90 100 T% at 400nm Time (min) 0 mM 0.02mM 0.1mM 0.35mM 0.5mM Figure 46. Turbidity measurements of collagen gelation (0.5mg/mL) in sodium dodecylbenzenesulfonate (SDBS) at the different concentrations Thermometer temperature was 34C. The Effect of Temperature a nd Activation Energy The effects of surfactant on collagen fibrillogenesis under different tem perature s were conducted and plotted in Figure 47 and 4 8. The increase of temperature accelerated the c ollagen fibrillogenesis which is consistent with data in Chapter 3, since collagen fibril formation is an endothermic process which is made thermodynam ically favorable by large positive entropy of precipitation associated with structural changes. The final turbidity of fibrils did not change with increasing temperature indicating no inhibition of fibril content was detected. The Arrehenius equation for the effect of temperature on the rate of fibril formation under th e surfactants of SDS and SDBS was plotted and activation energies were calculated from the slopes. Without surfactant, the activation energy of fibril formation of type I collagen is 167.6

PAGE 60

60 kJ/mol. By adding 0.1 mM SDS or SD BS, the activation energies are reduced to 153.6 and 114.7 kJ/mol respectively. Figure 47. Temperature effects on collagen fibril formation in SDS. The concentration of SDS is 0.1 mM, Collagen is 2.5 mg/mL, pH7.4

PAGE 61

61 Fig ure 48. Temperature effects on collagen fibril formation in SDBS. The concentration of SDBS is 0.1 mM, Collagen is 2.5 mg/mL. pH 7.4

PAGE 62

62 Figure 49. Photograph of shaked collagen gels formed in surfactants. A) Collagen fibril bundles formed without surfact ant. B) Collagen fibril bundles formed in 0.1 mM SDS. C) Collagen gel clots formed in 0.35 mM SDS. D) Collagen gel clots formed in 0.5 mM of SDS. E) Collagen gel clots formed in 0.1 mM SDBS. F) Collagen gel clots formed in 0.35 mM SDBS. G) Collagen gel clots formed in 0.5 mM SDBS. Morphology of Collagen Fibrils R econstituted collagen fibrils are generally composed of a random mesh of collagen fibrils and more than 95% of excess fluid. It has been observed that the fluid was a result of the casting, and can be expelled out by plastic compression[78] Except for the fibrils and fluid, insoluble collagen, insoluble random coiled network and the soluble random coiled collagen can also appear depending on the incubation temperature and stability of collagen [73] In the conditions where collagen was destabilized, the increase of temperature led to fibrils redissolving, collagen unfolding and partial de naturation. The appearance of nonfibrillar collagen influences the mechanical properties of collagen gel. In our experiments, we got the similar collagen gels when surfactants were added before fibril formation. However, when the gels were shook, the broke collagen gels presented the different macro scale morphologies as showed in Figure 4 9. The bundles of fibrils can be directly visualized when no surfactant was added, as well as in the

PAGE 63

63 presence of 0.1mM SDS. However, with increa sing concentrations of surfactant SDS and SDBS, the collagen gel clots appeared instead of the fibrils bundles. The difference of shake n gels in the presence of surfactant might be the result of nonfibrillar collagen. The ultrastructure of collagen fibrils with different surfactants were also analyzed by SEM (Figure 4 10 ). Native fibrils were present in all the samples and nonfibrillar collagen was also observed. The addition of the surfactants made the collagen fibrils more compact due to the appearance of nonfibrillar collagen. In the magnifie d collagen fibrils in Figure 4 10H, the characteristic D periodicity was well observed which suggests that surfactant s do not change the mechanism of collagen self assembly into native fibrils.

PAGE 64

64 Figure 410. SEM images of collagen fibrils formed in the presence of surfactants. A) Fibrils formed in PBS buffer without surfactant. B), D) and F) Collagen fibrils formed in 0.1mM, 0.35mM and 0.5 mM SDS respectively. C), E) and G) Fibrils formed in 0.1mM, 0.35 mM and 0.5 mM SDBS respectively. H)The magnified collagen fibrils formed in 0.5 mM SDS. Scale bars in H 100 nm, others are 1 m.

PAGE 65

65 Discussion Interactions between Collagen a nd Surfactants The effect of the presence of various electrostatic and hydrophobic surfactants on collagen fibrillogenesis has already been extensively studied in literature [65, 70, 71, 73, 79] However, how could the interactions influence the collagen fibrils formation is not well understood. It has been reported that the rate of collagen fibril formation was accelerated remarkably in the presence of 0.1 mM SDS or 110 mM nonionic surfactants with polyoxyethylene gr oups [70] The acceleration on the lag phase of fibril formation and subsequent acceleration of the growth of collagen fibrils, and fibrillogenesis rate with more hydrophilic polyoxyethene chains suggests the occurrence of interactions between the hydrophilic chains of the nonionic materials and collagen molecules[70, 80] The electrostatic and hydrophobic interactions between ionic surfactants and collagen have been repor ted [70, 72] .The interactions between ionic surfactant and collagen depend on the charges of collagen[81] In high or low pH solution when collagen molecules are highly ionized, the interactions between collagen and ionic surfactant molecules are attributed primarily to the electrostatic attraction between head groups of surfactant and lysine and arginine residues of the collagen molecules. In the condition where collagen molecules are neutralized, e.g. at its isoelectric point, the m ajor interactions between collagen and ionic surfactants are the hydrophobic interactions between hydrocarbon chains of surfactant and hydrophobic amino acid residues of collagen. In our studies, collagen molecules were dissolved at pH7.4 in PBS, the hydrophobic interactions between surfactants and collagen molecules might be the major one during fibril formation. The existence of variant amino acid side groups on collagen molecules suggests the possibility of interactions between them and small molecules containing hydrocarbon chains or hydrophilic groups. That binding effect regulates the collagen fibril formation and fibrils

PAGE 66

66 structure. Through the study of surface tension of SDS, only weak interactions between the surfactant and the collagen were seen bef ore fibrillogenesis, despite the fact that the SDS/collagen molal ratio from the solution concentrations is relatively high (12 or greater). These results are consistent with results from the literature which observed no interaction between collagen and no nylphenol poly(oxyethylene) in aqueous solution [72] or between collagen and pluronic PE 6800 in PBS solution[82] Before collagen fibrillogenesis, collagen molecules are hydrated, the helical structure of collagen around H2O can not be interrupted by the surfactant. However, during collagen fibril formation where the struct ured water molecules around collagen are broke n, the exposed amino acid side groups interact with nearby surfactants or collagen. The turbidity measurements in 2.5 mg/mL collagen indicate that the interaction was more related to the hydrophobic property si nce SDBS has a stronger acceleration effect than SDS. Even though surfactants accelerated the fibrillogenesis rate both in the lag and growth phase in the high concentration of collagen (2.5mg/mL), the denaturation effects of surfactants on low concentrati on of collagen have also been observed. The adsorption of s urfactants on collagen destabilizes the collagen triple helix. Collagen molecules within the fibril are substantially more thermally stable than the same molecules in diluted solution [83, 84] Collagen molecules in solution also undergo unfolding close to the body temperature of the species from which the molecules are extracted [85] It is possible that denaturation of collagen molecules in the presence of surfactant occurred which influences the kinetics of fibril formation. In fact, the remarkable acceleration effect of surfactants on collagen fibril formatio n suggests there are some specific interactions between surfactants and collagen molecules. Since a very small amount of surfactant can greatly affect collagen fibril formation, surfactant molecules may bind strongly to

PAGE 67

67 collagen molecules during fibril for mation as in the case of most globular proteins [69] The morphology studies indicate the non fibrillar collagen appeared and increased with increasing surfactant concentration. Moreover, collagen molecules are completely denatured at a high ratio of surfactants, where no fibrils were formed. It is possible collagen molecul es which are bound by ionic surfactants are more easily denatured and the denatured collagen molecules do not take part in forming native fibrils. Conclusion Self assembly of Collagen molecules into fibrils is an endothermic process and it has been thought that hydrophobic interactions play the dominant role. We intended to introduce ionic surfactant to collagen fibril formatio n. At room temperature, there was no significant appearance of interactions between surfactants and collagen since collagen molecules were fully hydrated and neutralized. As the temperature increased, we propose that surfactant molecules bind to collagen which accelerated the fibril formation. The strong binding occurred through hydrophobic interactions between surfactants and collagen molecules by forming collagensurfactant complexes. Those collagen molecules bound by surfactants are possibly denatured and do not form the native fibrils.

PAGE 68

68 CHAPTER 5 PH EFFECTS ON COLLAG EN FIBRILLOGENESIS I N VITRO: ELECTROSTATIC INTERACTIONS AND PHOSPH ATE BINDING Introduction As the primary structural protein in connective tissues, type I collagen is a major component of tendon, cartilage, ligament, skin, cornea and bone. It forms insoluble networks of fibrous bundles which act as scaffolding, providing shape and support in the body. Moreover, it is capable of being mineralized in vertebrate tissues, offering mechanical support and strain energy storage in bone [44, 87, 88] As a potential biomimetic material, design and mimicry of mineralized collagen has attracted much attention [89 92] Comprehensive information about the mechanism of collagen fibrillogenesis is required with respect to improving the design of collagen based materials, as well as exploring their functional properties in biological systems. Type I collagen is a triple helix consisting of three left handed polyproline peptide chains intertwined in a righthanded manner. Self assembly of soluble collagen into fibrils with characteristic D periodicity in vitro was observed more than 50 years ago [93] The D periodicity is described as consecutive domains consisting of 234 amino acid residues, and can be observed via electron microscopy [94] and X ray diffraction [95, 96] It has been found that the conditions for the formation of native type fibrils are in the pH range of 5.08.5, ionic strength between 0.1 and 0.8, a nd temperatures between 15 and 37C [40, 49, 97, 98] Subsequent studies also indicate pH, temperature, ionic strength, ion species, surfactants, saccharides, and the removal of the nonhelical ends of collagen have a strong influence on modifying collagen fibril formation[49, 65, 99, 100] Mechanisms of forming D periodicity during fibril formation are still not well under stood. Collagen fibrillogenesis is a thermally driven process which is favored by a large positive entropy contribution due to the displacement of structured water around collagen molecules[64] Hydrophobic interactions between nonpolar regions of adjacent molecules are

PAGE 69

69 the predominant effect governing collagen fibril formation due to the negative temperature coefficient of collag en solubility and the endothermic nature of in vitro fibrillogenesis [64, 101] It has also been repor ted that the collagen fibrillogenesis is driven by hydrogen bonding between polar residues through direct measurement of forces between collagen molecules[102] Early electron microscopic investigations and theoretical models indicate the highest polar and hydrophobic contact is achieved when molecules were shifted against each other by the distance of 234 amino acid residues [ 103] In fact, the ionized residues along collagen regulate the stability of the collagen triple helix; the pH of the medium affects the stability of collagen fibrils, as well as the diameter and the D periodicity of fibrils [65, 104108] However, the effect of electrostatic interactions on collagen self assembly is still unclear. It has been reported that dival ent phosphate ions binding on collagen molecules appear to form salt bridges within regions of high excess positive charge in collagen fibrils[109] Charges on ionizable groups whose charging is pH dependent take part in electrostatic interactions. The salts which bind to the collagen might also affect the electrostatic interactions. The occurrence of salt bridges indicates that the effect of pH on fibrillogenesis is not due to electrostatic interactions alone. Therefore, in order to choose proper conditions to create the desired morphology of collagen fibrils and study the underlying mechanism of collagen fibrillogenesis, the kinetics of fibril formation and the subsequent morphologies of fibrils were examined under the different pH conditions of the medium. Further, the surface charge of soluble collagen as a function of pH was measured and correlated with the fib rillogenesis process. The results were analyzed in relation to the possible intermolecular interactions.

PAGE 70

70 Zeta Potential and Electrical Double Layer Electrical Double Layer It is well known that most solid surfaces in aqueous are electrostatic charged, i.e. an electrical surface potential. When the liquid contains a certain amount of ions, the electrostatic charges on the solid surface will attract the counterions in the liquid. Thus, an electrical double layer is formed in the region of the solid liquid in terface. As illustrated in Figure 5 1, this double layer consists of two layers: a stern layer that includes counter ions bound relatively tightly to the surface, normally a bout several angstroms thick. Because of the electrostatic attraction, the counter ion concentration near the solid surface is higher than that in the bulk liquid far away from the solid surface. The concentration of coions is lower than that in the bulk liquid due to the electrical repulsion. So there is a net charge in the region clos e to the surface. From the stern layer to the uniform bulk liquid, the net charge density gradually reduces to zero. Ions in this region are less affected by electrostatic interactions and are mobile. This layer is called the diffuse layer. The thickness o f the diffuse layer depends on the concentration of bulk ions and the electrical properties of the liquid and usually ranging from several nano meters to several micrometers. The boundary between the compact layer and diffuse layer is called as shear plane. The electrical potential at the shear plane is called the zeta potential experimentally since the electrical potential at the solid liquid interface is difficult to measure directly. = 20 2 = 11 / 2 (5 1)

PAGE 71

71 r is the relati 0 is the permittivity of free space, kB is the Boltzmann constant, T is the absolute temperature, e is the elementary charge, NA is Avogadros number, Ci is the molar concentration of the i th ion (mol/L), and Zi is the charge number of the i th ion. Figure 51. Illustration of an electrical double la yer at a solid liquid interface. A) ion distribution; B) electrical potential distribution

PAGE 72

72 In an electric field, such as in microelectropho resis, each particle and its most closely associated ions move through the solution as a unit, and the potential at the surface of shear plane between this unit and the surrounding medium is known as the zeta potential. Zeta potential is therefore a functi on of the surface charge of the particle, any adsorbed layer at the interface, and the nature and composition of the surrounding suspension medium. It can be experimentally determined and, because it reflects the effective charge on the particles and is th erefore related to the electrostatic repulsion between them, the zeta potential has proven to be extremely relevant to the practical study and control of colloidal stability and flocculation processes. Zeta Potential Zeta Potential is the electrical potent ial that exists at the "shear plane" of a particle, which is some small distance from its surface. Zeta Potential is derived from measuring the mobility distribution of a dispersion of charged particles as they are subjected to an electric field. Mobility is defined as the velocity of a particle per electric field unit and is measured by applying an electric field to the dispersion of particles and measuring their average velocity. Depending on the concentration of ions in the solution, either the Smoluchow ski (for high ionic strengths) or Huckel (for low ionic strengths) equations are used to obtain the Zeta potent ial from the measured mobility. Smoluchowskis Equation The electrophoretic mobility of a particle moving with a velocity U in an electrolyte solution in an applied electric field E is given by the ratio U/E. The most widely employed Smoluchowskis formula, =0 (5 2)

PAGE 73

73 r 0 is the permittivity of a vacuum. The zeta po potential at the plane where the liquid velocity relative to the particle is zero. This plane is called the slipping plane or shear plane. The slipping plane does not necessarily coincide with the particle surface. Only if the slipping plane is located at the particle surface, does the zeta 0 which happens at low concentration of 0. Smoluchowskis equation was derived on the basis of approximations This equation is valid for large particles irrespective of their shape and the dimension of the particle is much planar. For a sphere with ra Hckels Equation ckels equation: =2 03 (5 3) The difference between Smoluchowskis equation and Hckels equation is by a factor of 2/3. Most electrolyte ions in the double layer experience an undistorted original field for the in the double layers experience a distorted field. This is the reason why Smoluchowskis equation differs from the Hckel equation by 2/3. For a cylindrical particle, the electrophoretic mobility is related to the orientation of the particle with respect to the applied electric field. When the cylinder is oriented parallel to the applied electric field, its electrophoretic mobility is given by Smoluchowskis equation (51). When the cylinder is oriented perpendicularly to the applied field, then the mobility depends not

PAGE 74

74 =0 ( ) (5 4) With ( ) =1 2 1 +2 1 +2 .55 ( 1 + exp ( ) ) 2 (5 5) and 1.5nm in diameter as reported, the collagen in aqueous solution are relatively flexible [110] and we are going to treat as particle and use Smoluchowskis equation for the measurements. Materials and Methods Materials Purified type I pepsin treated adult bovine dermal collagen (97%, with the remainder comprised of type III collagen) dissolved in 0.012N hydrochloric acid was purchased as a solution with concentration of 2.9 mg/ml (PureCol, INAMED, Inc., Fremont, CA). 10fold concentrated phosphate buffered saline (PBS) pH 7.4 was purchased from Sigma. Turbidity Measurem ents o f Fib rillogenesis To produce collagen fibrils, the chilled type I collagen solution was mixed with 10fold PBS at the ratio of 4:1 v/v on ice. The pH of each solution was adjusted by adding enough 0.1 M NaOH to reach the desired pH, resulting in a final concentration of 2.1 mg/ml collagen in buffer solution with 18 mM sodium phosphate and 283 mM sodium chloride. The samples were then sealed in 1.5 ml vials and thermostated at 30 C. The degree of fibril formation with incubation time was recorded by measuring t he absorbance at 400 nm using a UV Vis spectrometer (Perkin -

PAGE 75

75 Elmer Lambda 800 UV/Vis spectrometer, Fremont, CA), equipped with a thermostat cell holder, and maintained at the desired temperature by water circulation. Electron Microscopy of Collagen Fibrils The collagen fibrils with D periodicity were viewed by transmission electron microscopy (TEM; TEM 200CX, JEOL) using an accelerating voltage of 80 kV. TEM samples were prepared by placing the collagen suspensions on copper grids with 200 mesh size and removing excess water by placing a piece of filter paper at the edge of the grid. Then, the fibrils were negatively stained with 1% phosphotungstic acid solution at pH 7.4 for 15 seconds. The stained grids were rinsed with MilliQ water (deionized water, R=18 M and air dried. The collagen fibrils were also examined by scanning electron microscopy (SEM, JEOL, JSM 6400). The freeze dried collagen samples were mounted on stub and carbon coated. The morphologies ware observed at an acce lerating voltage of 15 kV. Image Analysis of Fibrils from TEM Image analysis of collagen fibrils was performed using NIH ImageJ software version 1.38 developed at the National Institutes of Health (Bethesda, MD). For each sample, at least five TEM images w ere randomly captured with magnification of 20,000 and at least 30 different fibrils were selected from each image. The fibril size was measured to generate histograms of fibril frequency vs. fibril diameter with x axis representing the fibril diameter and yaxis representing the number of fibrils in each interval. The D periodicity of fibrils was also estimated from the TEM images. To obtain average D periodicities, 150 measurements were taken for each fibrillogenesis condition and averaged. Each measureme nt consisted of averaging 5 periodicities on a fibril.

PAGE 76

76 Zeta Potential Measurement of Collagen Solution solution of type I collagen was diluted to 0.5 mg/ml in 12 mM HCl with MilliQ water and sonicated for 5 min. The pH of the solution was measured with pH paper and adjusted to the desired value within the range of 3 to 11 by adding 0.1 M NaOH. The zeta potential measurement was carried out at 17 C immediately after pouring 2 mL of freshly prepared solution into a plastic cuvette in order to avoid gel formation. Zeta potenti al was calculated using the Smoluchowskis equation. Results Turbidity Measurement at Different pH The kinetics of type I collagen fibrillogenesis was determined by turbidity measurement. As shown in Figure 52 the turbidity vs. time curves of collagen fi brillogenesis were sigmoidal curves with a lag phase before the onset of the turbidity increase and a plateau phase after the growth phase. In our study, it takes more than one day to achieve completion of fibrillogenesis due to the relatively high concent ration of ions while a short time is needed in one fold PBS [49, 56] An apparent retardation in growth phase, as evidenced by an intermediate plateau, was found from pH 6.9 to pH 9.2 and the highest fibrillogenesis rate and optical density was achieved at pH 9.2. At low pH (pH 6.5) and high pH (pH10.5) where fibril formation was strongly inhibited, the fibrillar gel was of relatively weak strength.

PAGE 77

77 Figure 52. Turbidity measurements of collagen fibrillogenesis at different pH. Collagen concentration is 2.1 mg/ml in 18mM phosphate, 283 mM sodium chloride. Fibril Morphologies The morphologies of collagen fibrils varied widely depending on the pH condition. Fibrils with characteristic D periodicity were present in TEM images (Figure 5 3). Fibrils with variable size were formed at low pHs, where in the presence of the large fibrils, many small fibrils with no D periodicity were detected. With increasing pH, the fibrils were more uniform and interwo ven into a network. SEM data were consistent with TEM images (Figu re 5 4). When fibrillogenesis was highly inhibited, in the range from pH 6.0 to pH 6.9, the gels were composed of more nonfibrillar collagen. The more highly fibrillogenes is was inhibited, the greater the amount of nonfibrillar collagen. From pH 7.1 to pH 10.0, no apparent difference was found in SEM results.

PAGE 78

78 Figure 53. TEM images of self assemble collagen fibrils after three days of fibrillogenesis. The inserts in the lower right hand corner of each image show a magnified view of a single fibril to highlig ht the banding pattern. Scale bar 500 nm.

PAGE 79

79 Figure 54. SEM images of collagen fibrils obtained in different pH conditions. The fibrils form branched networks with nonfibrillar collagen. Scale bar is 10m for all the images.

PAGE 80

80 Fibril diameter distributions measured from TEM images are summarized in Figure 5 5 and only fibrils with size larger than 45 nm are counted. No data were collected at pH 6.0 for one day since the fibrils can not be visualized. Small fibrils with diameter approximately 85 nm were obta ined at pH 6.6 after three days of fibrillogenesis. The final fibrils have a constant diameter of approximately 200 nm at pHs between 6.9 and pH 8.0 in spite of the remarkable difference in the fibrillogenesis rate. An increase of diameter as a function of time was found at pHs between 6.9 and 7.3 while no obvious change in fibril size was found at pH 8.0 since the fibrillogenesis was finished within one day.

PAGE 81

81 Figure 55. Histograms of diameter distribution from pH 6.6 to 8.0. X axis is diameter in nm, Y axis is the frequency. Average diameters and standard deviations are in unit of nm. D Periodicity Measurements It is worth measuring D periodicity of reconstituted fibrils since the presence of native D periodicity is believed to be important for the mec hanical and biological functions of collagenous matrix [111, 112] Fibrils with native D periodicity also play an important role in mineralization [112] The reported D periodicity is near 67.0 nm for wet tissues and 64.0 nm for air dried samples [113] Skin tissues have a banding length of 65 nm in the wet state [114] As summarized in Table 5 1, the average D periodicity after three days of fibrillogenesis was

PAGE 82

82 approximately 62 nm which matches well with native D periodic banding length since our samples are dehydrated collagen fibrils reconstituted from calf skin. Two unusual D periodicities appeared at pH 7.1 after one and two days, which had an average D periodicity of 50 nm and 54 nm, respectively. Table 5 1 D periodicity of collagen fibrils at various pHs and fibrilloge nesis times. pH D periodicity (nm) One day Two days Three days 6.9 60.47.8 64.93.8 62.15.7 7.1 50.54.5 54.39.1 62.36.0 7.3 65.87.6 64.06.7 62.23.6 8.0 62.45.3 62.55.1 61.64.4 Electron micrographs were taken at a constant magnification o f 37,000. Each value in the table is the mean standard deviation for 150 D periodicity measurements. Surface Charge Measurement To study the electrostatic interactions of collagen during fibrillogenesis, we measured the zeta potential of collagen molecu les in aqueous solution. Collagen molecules are less soluble at the isoelectric po int (pI) and therefore favor fibril formation. Collagen is reported to have an isoelectric point (pI) of pH 9.3, which was determined in the absence of other electrolytes [115] Howe ver, it has been found that the pI was affected by adsorption of ions. Jackson and Neuberger observed the shift of the isoelectric point to lower pH with increasing ionic strength due to the preferential binding of the anions of aldolase in gelatin[116] Freudenberg et al. [117] found a shift from pH 7.5 to 5.3 with increasing ionic strength of KCl while, in the presence of CaCl2, the isoelectric point shifts to more basic pH. The shift of pI in different electrolyte conditions indicates it is difficult to quantify the surface net charge directly. As shown in Figure 56, the pH corresponding to the pI of soluble collagen was 9.2 in 12 mM of NaCl which is consisten t with the pI in the absence of salts[116] However, when 10 mM Na2HPO4 was added, the pI shifted to pH 7.5. Since in zeta potential measurement the surface charge is measured at

PAGE 83

83 the slip plane, the shift of pI with added salts might be due to the preferential adsorption of phosphate ions to the surface of collagen. Figure 56. Zeta potential measurement of surface charge of soluble collagen as a function of pH. Discussion Collagen Fibrillogenesis The growth of collagen fibrils in vitro i s a timedependent process. When monitored by changes in turbidity, it consists of three major phases: a lag phase, in which there is no change in turbidity; a growth phase, when turbidity rapidly increases; and a plateau phase where turbidity stops increa sing. Ultrastructural studies of the lag and earlygrowth phase using laser light scattering and atomic force microscopy have revealed the fibril formation is a multistep process[52, 62] where a number of different intermediates are formed during the early growth

PAGE 84

84 phase. These ultrastructural studies have further sugges ted that addition of these intermediate to the ends of microfibrils results in linear growth, while lateral growth occurs through fusion of microfibrils. These two processes appear to occur simultaneously. In our study, the process of how collagen molecule s self assemble into fibrils was well observed by the combination of turbidity measurement and morphology examination. In the turbidity measurement, lag, growth and plateau phases are observed from pH 6.6 to pH 9.2. Further examination of fibril size indic ated that obvious lateral growth occurred during the growth phase, as evidenced by either the increase in average diameter with time or a shift in the distribution of diameters. Surprisingly, an inhibition in the growth phase was observed as an intermediat e plateau during growth for some fibrillogenesis conditions. A similar phenomenon has also been reported where no difference before and after the intermediate plateau was found in AFM examination[118] Collage n fibril formation is a multistep process which involves molecular packing into an ordered structure. Type I collagen is unstable and partial denatured within a couple of days at body temperature [119] The occurrence of denaturation during fibrillogenesis seems to inhibit fibril growth in all our case since a longer fibrillogenesis time was used. In addition, our TEM and SEM images show the presence of amorphous collagen which further indicates some inhibition of fibrillogenesis. Electrostatic Effects A number of studies have been carried out on the mechanism of collagen fibril formation in vivo and in vitro[49, 104, 108, 120122] The aggregation of collagen monomers is controlled to a significant degree by electrostatic interactions. To discuss the effect of electrostatic interactions, the length scale over which electrostatic interactions occur (Debye screening length 1) is taken into account. The Debye length is the characteristic distance over which the electrostatic potential near a charged entity decays. Using Equation 5 1 of

PAGE 85

85 soluble collagen is reduced from a theoret ical maximum of ~970 nm in pure water to less than 0.6 nm under our electrolyte condition. The distance between the surfaces of two collagen monomers in fullyhydrated reconstituted collagen fibrils is approximately 0.5 nm [123] and so the electrostatic interactions only become important when the monomers are aggregated to form fibrils. Fibrillogenesis is a thermally driven process which is favored by hydrophobic interactions. When the pH approaches the pI, the surface charge of the collagen monomers is reduced, which would minimize the electrostatic repulsion and favor collagen molecule aggregation. A strong influence of the ion species on the surface charge on collagen was found in our zeta potential measurement. This shift mi ght be the result of preferential adsorption of phosphate ions. Weinstock et al observed binding of ~6 phosphates per collagen molecule at 100 mM NaCl and ~20 phosphates at low NaCl concentration[124] In our system, the pI of collagen is pH 9.2 in 12 mM NaCl while it is at pH 7.4 in 12 mM NaCl and 10 mM Na2HPO4. Correspondingly, in t he presence of 18 mM phosphate and 283 mM NaCl, which we used for fibril formation, it is reasonable that the pI of collagen is around pH 9.2 due to less phosphate binding in the presence of high concentration of NaCl, as shown by Weinstock et al [124] Unfortunately, zeta potential measurements could not be conducted under the conditi ons used for fibrillogenesis because the high ion concentration interferes with the measurement. The fibrillogenesis rate accelerated by increasing pH from 6.0 to 9.2 and then decreased, which is consistent with a reduction in the net charge on the collage n molecules as the pI is approached. When fibril formation was accelerated by increasing pH from pH 6.9 to 8.0, there is no difference in the final fibril diameter. This phenomenon suggests that fibril size has no direct relationship with the rate of fibr illogenesis. Even though there is a difference in pH conditions

PAGE 86

86 from pH 6.9 to pH 9.2, it does not appear that the resulting change in surface charge affects the size of the fibrils. D periodicity The presence of a periodic banding pattern along a fibril i s the distinguishing feature of collagen fibrils from other structures in connective tissues. The observed D periodicity after fibrillogenesis in our results is in agreement with that of native collagen fibrils, which typically have a D periodicity ranging from 62 to 67 nm depending on the source of collagen and sample preparation[125, 126] Smaller D periodicity was observed at pH 7.1 at one and two days. Less than 56 nm banding was also reported from tendon, cornea, skin, liver, and reconstituted fibrils [127131] The presence of a small amount (15%) of type III collagen in skin was proposed to account for the unusual periodicity, since type III is associated with type I collagen in skin while bone consists of almost 100% type I collagen [113, 114] However, the presence of short D periodicity reported in the early growth of tendon and reconstituted fibrils [127, 129] cast doubt on this proposal since tendon consists of 100% of type I collagen while reconstituted collagen consists of 97% type I collagen. The binding of chondroitin sulfate associated with rat tail tendon[132] the binding of dermatan sulfate associated with skin, [114] and the cleavage of nonhelical telopeptid e by pepsin treatment of reconstituted fibrils [127] have also been propose d to account for the unusual D periodicity. Since pepsin treated collagen molecules have a shorter length than native collagen due to partial cleavage of nonhelical telopeptides, the specific binding of phosphate in a similar manner as chondroitin sulfate and dermatan sulfate may alter the charge distributions and water binding, permitting a somewhat shorter length for the molecule leading to a shorter banding. Our data also shows that normal banding was found at the final stage of fibril formation, sugges ting that native D periodicity is the most stable state and that the shorter D -

PAGE 87

87 periodicity is a transient structure that occurs during the growth process. However, we can not yet provide a specific structural model to explain the short banding. Conclusion The rate of collagen fibrillogenesis is a ffected to a significant degree by electrostatic interactions and its optimal fibrillogenesis was achieved when pI of soluble collagen was approached. However, those electrostatic interactions have a weak effect on fibril size and morphology. The surface charge measurement indicates that phosphate binding has a great effect on electrostatic interactions. The average D periodicity was 62 nm with two notable exceptions at pH 7.1. We propose that the unusual D periodic ity (50 and 54 nm) might come from altered alignment of molecules due to the change of surface charge.

PAGE 88

88 CHAPTER 6 THE FORMATION OF NAT IVE FIBRILS INDUCED BY DIVALENT IONS Introduction For more than 50 years, the fibrillogenesis of type I collagen has been intensively studie d in vitro [49, 97, 133] This has lead to the reproducible procedures for the production of collagen fibrils with the essential native banding pa ttern found in biological tissues. Collagen fibril formation is a thermally driven process which is favored by a large positive entropy contribution due to the release of structured water molecules[64] Non covalent forces including Coulombic, van der Waals and hydrophobic forces play important role s in fibril formation. Although a neutral pH and a physiological condition are general ly used for collagen fibril formation, the underlying effects of salts on fibril formation are not completely understood. Our surfactan t studies on collagen self assembly into fibrils indicate that the introduction of hydrophobic interaction could not facilitate fibril formation. The pH and phosphate studies have indicated that pH and phosphate play a significant role on forming native fi brils. The question is whether the phosphate can be re p laced by another buffer or salt ? What is the critical factor on forming the native fibrils? Salts can modulate the conformation of collagen molecules and the electrostatic interactions [117, 134] The importance of pH and ionic strength for the thermal stability of collagen has been demonstrated by several authors. The stability of collagen in solution corre lates with both the type of salt and the pH [135137] For CaCl2 a decrease in thermal stability of soluble collagen with increasing salt concentration was observed in the physiological pH range while the thermal stability was reported to increase with increasing ionic strength at pH 2.3. In the case of KCl at pH 6, a slight decr ease of the thermal stability was found with increasing salt at low concentration. Arktas reported an increase of the thermal stability with

PAGE 89

89 increasing the ionic strength of NaCl at pH 3.7 and pH 5.7, but a decrease for an addition of the CaCl2 at pH 5.7[135] By changing the solution pH in invariant electrolyte, a maximum thermal stability of the collagen type I molecules was determined in the physiological pH range [137] The surface charge which is related with salts and the pH of the solution is an important parameter for collagen self assembly into fibrils. To facilitate the collagen self assembly, the triple helix of collagen should be stable long enough to begin assembly. Moreover, the surface charge of collagen should be screened in order to reduce the electrostatic repulsion. Collagen has th e IEP at 9.3 as we determined in Chapter 5. It is expe cted to accelerate fibrillogenesis with increasing the pH to isoelectric point. However, Wood et al. found the fibrillogenesis rate to increase with decreasing pH from 7.4 to 6 while to decrease with increasing pH from 7.4 to 8[49] This might come from the binding of salt which can change the surface charge of collagen and intermolecula r interaction. It has been reported that preferential adsorption of ions can shift the IEP and influence the conformational stability of the collagen[117] These binding ions should also affect the surface charges of collagen molecules and change the fibrillogenesis rate. However, due to the affinity of ions and the difficulty in surface charge measur ement, there are no systematic studies of salts on co llagen fibrillogenesis related to surface charge. S alts influence the stability of collagen, conformation and intermolecular interactions through saltingin, salting out and binding[134, 138] It has been proposed that d ivalent phosphate and sul f ate ions binding to collagen molecules appear to form salt bridges within regions of high excess positive charges in fibril s [109] During fibril formation, the attraction between the hydrophobic side groups as well as opposite charges groups is thought to be the major interactions. The electrostatic interactions between charged amino residues strongly contribute to the molecular organization of D staggered array model of collagen fibrils.

PAGE 90

90 Electrostatic interactions in aqueous media are commonly understood in terms of screening Coulomb interact ions, wher e like charged objects, such as polyelectrolytes, always repel in the PoissonBoltzmann theory (PB theory) However, a series of experiments on charged biopolymers, including DNA, F actin fibers, microtubules and aggregating viruses indicate that in the presence of small amounts of polyvalent salts, attractive forces of different origin s (e.g., hydrophobic or hydration interactions) overwhelm electrostatic repulsion [139141] Protein Salt Interactions Salting out/S alting in Theory The protein solubility can be expressed in terms of the solvation free energy of the protein molecule in the equilibrated fluid phase [142] When the solubility is small, fluid phase proteinprotein interactions can be neglected. The phase equilibrium criterion (the chemical potential of the protein in the crystal is equal to that in the fluid) at a given salt conc entration reduces to: = (6 1) Where and are the chemical potential of the protein in the solid phase and the infini te dilution standard state chemical potential of the protein, respectively, and is protein solubility relative to a standard state solubility. The major assumption of the salting out theory is that the protein is a pure phase and the s alting out behavior is determined from the dependence of the standardstate chemical potential on salt concentration. Protein salt interactions can be divided into three main groups: 1) effect of salt on charged groups, 2) effect of salt on exposed peptide groups, and 3) effect of salt on nonpolar groups [143, 144] At low salt concentrations there is a salting in effect due to the favorable interaction between the protein and the surrounding ion atmosphere. At higher salt concentrations, protein solubility is determined by the balance of unfavorable interactions between the salt and the

PAGE 91

91 nonpolar surface of the protein and favorable weak ionbinding interactions between the salt and the protein. In most cases, because the u nfavorable hydrophobic interactions are greater than the attractive weak ion binding interactions, saltingout is observed. Surface Tension Increment Effect The surface free energy of a nonpolar group of the protein in contact with water is related to the surface tension of the water. Because all salts increase the surface tension of water, similarly, they increase the surface free energy of nonpolar groups. As a result, the magnitude of the salt protein interaction is related to the molal surface tension i ncrement of the salt. The molal surface tension increment of the salt is correlated with the ions position in the lyotropic series that was originally developed to describe the saltingout effectiveness of various ions for globular proteins. For anions, t he series in decreasing order of the molal surfacetension increment is > 4 2 > > > > > ; the corresponding series for cations is +> +> +> +> +> +. High lyotropic series salts (kosmotropes) are good saltingout agents because th ey interact strongly with water; water molecules surrounding the salt ions are more structured relative to bulk water. Low lyotropic series salts (chaotropes) break the structure of water of the surrounding water molecules. Chaotropes are weak salting out agents due to weak interaction with water [145] Preferential Ion Binding Interactions Studies on the solubility of proteins in salt solutions have shown that there is salting in effect due to the electrostatic interaction between the salt ions and the peptide group [146, 147] which is attributed to the large dipole moment of the peptide group. Since amino groups of protein carry a partial positive charge and the carbonyl oxygen groups carry a partial negative charg e, it is possible that anions bind at or near the nitrogen atoms and cations bind at or near the oxygen atoms. In addition, divalent cations have stronger binding affinities to the peptide group,

PAGE 92

92 as indicated by measurements of the retention times of salt on columns containing a stationary phase of polyacrylamide [143] In these studies, chaotropic anions or divalent cations were retarded due to interaction with the peptide group, whereas kosmotropic anions were not retarded due to unfavorable interaction with the nonpolar backbone of polyacrylamide. Anion binding to the positively charged groups of protein molecules has been observed in studies concerning stabilization of folded structures of protein molecules at low pH. The strength of this interaction is related to the ions position in the electroselectivity series [148] as measured by the affinity of the ion in an anion exchange resin. The series adsorption dependen dence on the resin was exploited, and the general trend is in the order of > > > > For monovalent anions, the electroselectivity series is the inverse of the lyotropic series. It is likely that higher binding affi nities of the chaotropic anions reflect weaker unfavorable interactions with the nonpolar backbones of the resins. However, a divalent charge interacts more strongly with the charged resin than the monovalent charge, as is observed with In fact, the intensity of anion binding to positively charged surfaces is the reverse of lyotropic series dependence of the solubility of basic proteins for monovalent anions [149] This reverse solubility dependence is attributed to the for mation of insoluble proteinanion complexes. It was commonly thought favorable protein salt interactions should favor solubilization of the protein molecule. However, in addition to protein salt interactions, changes in protein protein interactions must al so be considered; a decline in solubility occurs if there is an increase in the net attraction between the proteins. In the present study, we found that polyvalent electrolytes adsorb on collagen molecules whi ch shift collagens isoelectric point. There is a competition between fibrillogenesis and denaturation at pH condition. We also found that collagen fibril formation is not solely a salting-

PAGE 93

93 out process which is regulated by screening surface charge. The likely charge attraction originated by divalent ions bridging play a critical role on forming native fibrils. Materials and Methods Materials Purified type I pepsin solubi li zed adult bovine dermal collagen (97%, with the remainder comprised of type III collagen) dissolved in hydrochloride acid was purcha sed as a solution with a concentration of 2.9 mg/ml (PureCol, INAMED, Inc., Fremont, CA). CaCl2, NaCl, KCl, Na2HPO4, Na2SO4, K2HPO4, MgCl2, ethylenediaminetetraacetic acid ( EDTA) and 0.1M NaOH were purchased from Fischer Scientific. Zeta Potential Measur ement solution of type I collagen was diluted to 0.5 mg/ml in 12 mM HCl with desired salts and sonicated for 5 min utes The pH of the initial solution was measured with pH paper and adjusted within the range of 3 to 11 by 0.1 M NaOH. Fresh solutions were used for zeta potential measurements at each pH. Turbidity Measurements The turbidity was recorded by measuring the transmittance at 400 nm using a UV Vis spectrometer (Be ckman DU 640 Spectrophotometer). In a typical experiment, 1.2 ml collagen was mixed with 0.1M NaOH to desired pH. After adding the desired salt, the mixture was incubated at 30C for 20 hours before turbidity measurement. Electron Microscopy Measurement In order to verify collagen fibril formation, electron microscopy of the final collagen gel was taken. The collagen fibrils with native banding were viewed by transmission electron microscopy (TEM; TEM 200CX, JEOL) using an accelerating voltage of 80 kV. TEM samples

PAGE 94

94 were prepared by placing the collagen clot on a 200 mesh copper grid and staining with phosphotungstic acid (1%, pH7.4). Small Angle X ray Scattering (SAXS) pinhole instrument (NanoStar from Bruker AXS, Karlsruhe, Germany) in Dr. Peter Fra t zls lab in Germany. Radial averaging of the twodimensional scattering data gave the int ensity I as a function of the modulus of the scattering vector, which is defined as = 4 / (6 2) Ci rcular Dichroism Measurement After turbidity measurement, samples of collagen with low turbidity were used for CD measurement in order to determine the thermal stability of the triple helix in specific salts condition. Collagen samples in 350 mM CaCl2, MgC l2 and NaCl at pH 7.4 were diluted with DI water to give a concentration of about 0.27 mg/ml. The samples then were placed in a jacket e d cell at 25C with a 1 mm light path and the CD spectra were obtained by using a Cir cular Dichroism Spectrometer (Mo del 400, Biomedical, Inc. Lakewood, NJ, USA) in Dr. Joanna Longs Lab. For comparison, the CD spectra of helical and random coil form of collagen were also obtained. The random coil form was obtained after raising the temperature of solution to 50 C where mel ting is completed. Samples were scanned at 25C from 250 nm to 180 nm in 1nm steps using 1sec time constant. 32 dmol1 L is the path length in millimeters, C is the concentration in milligrams per milliliter, and M is the average residue molecular weight of collagen and its value is 91.2.

PAGE 95

95 Results Surface Charges Measurement To inves tigate the influence of salt type on the thermally driven self assembly of collagen, the surface net charge of collagen molecules were measured by Zeta potential in different salt type (Figure 6 1) It can be seen that not only the surface charge of proteins but also the isoelectric point can be affected by salt specificity. In the control solution, t he pH corresponding to the isoele ctric point of collagen was found to be approximately 8.9, which is the approximate isoionic point [116] After adding 10 mM Na2HPO4, the isoelectric point of collagen shifted to 7.5. Sim ilarly, adding Na2SO4 causes the IEP shift to acid side. On the other side by adding CaCl2, the IEP of collagen shifted to 9.4. In other words at the fixed pH, the surface of coll agen is more positive charged with multivalent cations than with monovalent ions; collage n is more negatively charged with multivalen t anions than with monovalent ions. 6 7 8 9 10 -8 -6 -4 -2 0 2 4 6 8 pHZeta Potential (mV) NaCl CaCl2 Na2HPO4 Na2SO4 Figure 61. Zeta potential measurement of surface charges of collagen affected by salts.

PAGE 96

96 Turbidity Measurements C ollagen f ibril formation was conducted at physiological pH and salt condition in order to generate native collagen fibrils. To manipulate surface net charge of collagen, the specific salt, ionic strength and pH were used in our studies. Based on the surface charge results desc ribed above it is expected collagen monomers are less surface charged at physiological pH in multivalent anions than monovalent ions, while more positively charged in multivalent cations than monovalent ions. Figure 62 and 63 compare turbidity of colla gen fibrils data for various salts at pH 7.4 and 9.0, respectively. The trends are different at both pH conditions At pH 7.4, with increasing ionic strength the difference in gelation degree increases in divalent anions of Na2SO4, Na2HPO4 and K2HPO4. In divalent cations of MgCl2 and CaCl2, the gelation degree is highly reduced when ionic strength increase s to 250 mM. For the monovalent ions of NaCl and KCl, the gelation degree is reduced at ionic strength of 250 mM. The specific nature of the ion is less important at an ionic strength less than 250 mM since there is no remarkable difference in gelation degree. The difference in gelation degree becomes significant for different salt types above ionic strength 250 mM. At pH 7.4, the surface charges of collag en in divalent anions are very low from the zeta potential measuremen t while highly positively charged in monovalent ions and divalent cations condition.

PAGE 97

97 KCl NaCl MgCl2 CaCl2 K2HPO4 Na2HPO4 Na2SO40 20 40 60 80 100 25mM 50mM 75mM 100mM 150mM 250mM 350mMTransmittance % Figure 62. Effects of salt type and ionic strength on the collagen gelati onat pH 7.4. The gelation degree is increased with increasing ionic strength while gelation is inhibited in CaCl2, MgCl2, KCl and NaCl when ionic strength is about 250 mM. Figure 63 show s gelation data for salts and ionic strength at pH 9.0. Compare d to results at pH 7.4, gelation is highly inhibited at pH 9.0 in the range of ionic strength investigated since low turbidities were measured However, in the divalent cations of MgCl2 and CaCl2, a high degree of gelation was obtained at ionic strength from 50 mM to 150mM. In the divalent anions of K2HPO4 and Na2HPO4, a medium degree of gelation was obtained by increasing the ionic strength which might result from the buffer effect of HPO4 2 -.

PAGE 98

98 MgCl2 CaCl2 KCl NaCl Na2SO4 K2HPO4 Na2HPO40 20 40 60 80 100 25mM 50mM 75mM 100mM 150mM 250mM 350mMTransmittance % Figure 63. Effects of salt type and ionic strength on the collagen gelation degree at pH 9.0. Low gelation is obtained in all salt type except in divalent cations of CaCl2 and MgCl2. Circular Dichroism Spectroscopy Turbidity measurement s indicated gelation was inhibited by the salts concentration and pH. It has been found that high ionic strength destabilize s collagen s triple helix structure [143] which might inhibit the gelation. It is necessary to determine the helical structure of collagen after incubation The CD of collagen was performed in 350 mM of CaCl2, MgCl2 and NaCl at pH 7.4 where there is no gel formed at all (Figure 6 4) The helical and random coil forms of collagen were also conducted for comparison. It can be observed that t he helical form of collagen has a characteristic positive band at 221 nm and a large negative band at 197 nm while the

PAGE 99

99 random coil form, the positive and negative band s have almost disappeared T he spectra of collagen in the presence of CaCl2, MgCl2 or NaCl have a similar profile with respect to the spectrum of helical form of collagen but the intensity of the bands decreased which m ight be attributed to partial loss of helical form. 190 200 210 220 230 240 250 -8 -6 -4 -2 0 2 Wavelength (nm)[ ]x10-4(degree cm2 decimole-1)Denatured collagen In MgCl2In NaCl In CaCl2Collagen Figure 64 Circular dichroism spectra of collagen. The concentration of collagen was 0.27mg/ml. Collagen (native) has characteristic bands at 197 nm and 212 nm. The bands disappeared after denaturation. In 350 mM salts of CaCl2, MgCl2 and NaCl, two bands are decreased due to the partial denaturation. TEM Morphologies Of Fibrils Formed In Salts Electr on microscopy images were obtained for the collagen gelation after incubation at 30C for 20 hours in 100 mM and 200 mM ionic strength of salt, pH 7.4. In 100 mM NaCl and KCl, collagen monomers tend to associate into microfibrils (Figure 6 5). No banding pattern can

PAGE 100

100 be observed within these loosely formed microfibrils while nonfibrillar collagen monomers coexist with fibrils. However, when ionic strength is increased to 200 mM, the collagen monomers form more highlyordered fibrils. Alongside the microfibrils, larger fibrils with D periodic banding pattern are also found. When in the presence of divalent anions of Na2SO4 and Na2HPO4, the lateral alignment of the fibrils is clear and the characteristic D banding pattern of collagen fibrils across the whole collagen fibrils bundle (Figure 66) is well demonstrated. Even in 100 mM of Na2HP O4, small fibrils also display ed the characteristic D periodic banding pattern. C ompared to the fibrils formed with monovalent ions, the collagen fibrils with divalent anions have more highly ordered structure. In the presence of CaCl2 and MgCl2 at ionic s trength of 200 mM and pH 9.0, the TEM images of collagen fibrils in divalent cations at 100 mM are showed in Figure 67. Large and bundled fibrils with D periodicity appeared in the whole image. Since the lateral growth of fibrils is a diffusion control pr ocess, we though the slow rate of gelation in divalent cations might be the reason for large fibrils and decided to study the kinetics of fibrillogenesis in salts

PAGE 101

101 Figure 65. T EM images of type I collagen fibrils in monovalent ions of NaCl and KCl Sc ale bar is 1 m.

PAGE 102

102 Figure 66. TEM images of type I collagen fibrils formed in di valent cations of Na2HPO4 and Na2SO4. Scale b ar is 500 nm.

PAGE 103

103 Figure 67. TEM images of type I collagen fibrils formed in di valent cations of CaCl2 and MgCl2Scale b ar is 500 nm. Kinetics o f Fibrillogenesis i n Salts The rate of collagen fibrillogenesis is affected by the intermolecular interactions. When collagen molecules are highly positive or negative charged, it is difficult to form the fibrils due to the electrostatic re pulsion. When collagen molecules are neutralized, f ibrils are easier to be formed. Based on the zeta potential measurement, at pH 7.4, collagen molecules are less charged in Na2HPO4 and Na2SO4 than in NaCl, CaCl2 and MgCl2. The effects of charge density on rate of fibril formation are observed in Figure 6 8. The order of fibrillogenesis rate is Na2HPO4>NaCl>Na2SO4>MgCl2>CaCl2 which is consistent with reducing the surface charges of collagen except in the presence of NaCl.

PAGE 104

104 0 100 200 300 400 500 600 0 20 40 60 80 100 Transmittance at 400 nmTime (min) MgCl2 CaCl2 Na2SO4 Na2HPO4 NaCl Figure 68. Kinetics of collagen g elation in salts determined by turbidity measurements.I onic Strength 100 mM, pH 7.4, 30C. D periodicity Collagen exhibits a number of longrange interactions resulting from axial order and possible substructures within the fibrils. The most characteristic long rang order is D periodicity. The D periodicity was calculated from TEM images and summarized in Table 6 1. Normal D periodicities which are around 60 nm was measured in MgCl2, CaCl2 and Na2HPO4, but unusual bandings were found in Na2SO4 and EDTA.

PAGE 105

105 Table 6 1. D periodicity of collagen fibrils in presence of different salts Salts Ionic Strength (mM) Banding Length (nm) Count Number CaCl 2 100 59.43.9 89 MgCl 2 100 60.37.7 85 Na 2 HPO 4 200 58.74.4 81 Na 2 SO 4 200 49.23.0 178 EDTA 200 55.75.5 316 The D periodicity of collagen fibrils were also measured by small angle X ray scattering. The electron density profile of the d spacing produces a diffraction series of sharp reflections on the small angle scattering image as show n in Figure 69. These sharp reflections can be used to determine a change in the d spacing of the collagen molecules as the alteration in electron density along the fibril axis are reflected in changes to the intensity of these peak. The orders of Bragg r eflections due to the electron density distribution of the gap and overlap interactions of collagen molecules in the axial direction and the D periodicity of the fibrils were calculated by th e slope of the insert plot using periodicity of fibrils formed in KCl and Na2HPO4 is 61 nm which is same as the D periodicity of natural bone.

PAGE 106

106 Figure 69. Small angle X ray diffraction measurement of collagen fibrils. A)Small angle X ray diffraction patte rn of collagen fibrils formed in 100 mM KCl. B) A linear intensity profile of the small angle diffraction image of reconstituted collagen fibrils. The orders of Bragg reflections due to the electron density distribution of the gap and overlap interactions of the collagen molecules in the axial direction have been included.

PAGE 107

107 Discussion Collagen Ion Complex Collagensalt interactions can be divided into two main groups: binding counterions on charged groups which can reduce the effective charge on the collagen ; free ions of both charges which act to screen the charge of collagen ion complex. The binding effects are attributed to the large dipole moment of the peptide group. The amino groups on collagen carry partial negative charges, suggesting anion binding at or near the nitrogen atom. Divalent cations have larger binding affinities to the peptide group than monovalent ions, as indicated by measurements of the retention times of salts on columns containing a stationary phase of polyacrylamide [143] The binding effects of ions change the surface charge of collagen molecules. Moreo ver, the shifting of isoelectric point in different salts measured by zeta potential indicates collagen molecules can be overcharged by binding ions. The overcharging phenomenon was found in many other areas of macro ion electrostatic interaction because o f the strong overneutralizing effect of the counterion condensation [150] For the monovalent salts, they have equal binding affinity on charged groups of collagen molecules and the isoelectric point ca n not be shifted. However, for the divalent ions, overcharging occurred through the divalent ions binding which reverse the collagen charges. Stability o f Collagen and Fibril Formation An important phenomenon in our studies is the inhibition of fibril formation at the isoionic point of collagen (pH 9). The for mation of collagen fibrils requires the temperature increasing to break the structured water. The hydrophobic attraction and electrostatic attraction overcome hydration to aggregate the collagen molecules. It has been proposed the specific recognition and highest intermolecular interactions occurred in native fibrils driving fibril formation. The instability of collagen in isoionic condition at high temperature is one factor of inhibiting fibril

PAGE 108

108 formation due to the denaturation. Our CD results also confirm ed the occurrence of partial denaturation of collagen at high ionic strength. While in pH 9.0, the pH is close to isoionic point of collagen, it is possible that the unfolding of the collagen helix is too fast and prevents fibrillogenesis. However, because of divalent cations, which shift the IEP to base side and make collagen slightly positive charged, the destabilization of collagen is overcome by Fibril formation. Multivalent Ions Effect on Regulate Banding Pattern Type I collagen monomers self assemble into native D periodic fibrils by a high degree of specific intermolecular interactions. Firstly, the collagen monomers exhibit a high degree of parallel alignment to each other to form microfibrils. Then the microfibrils associate laterally to forming fibrils. When self assembly occurs in the absence of divalent ions, the fibrils are smooth and do not show clear longitudinal periodicity In the presence of divalent ions, collagen fibrils were formed with periodicity of 62 nm which lies well within the characteristic values of around 64 nm It is said that the characteristic native periodicity of collagen occurs in the buffer solutions containing potassium [121, 151154] Our results indicate that even without potassium, collagen also forms fibrils with native banding pattern. Compare d to the salt type, fibril with nat ive banding pattern formed with the divalent ions. It is possible s alts with divalent ions induced the like charge attraction during fibrillogenesis through bridging. Short D periodicities were found in the presence of Na2SO4 and EDTA. We found the short D periodicities on early stage of fibrillogenesis in PBS. We proposed the misalignment due to the electrostatic interactions between collagen molecules which might also happen in those two salts. The D periodicity depends on the state of hydration of the fibril and decreases from 67 nm from the hydrated fibril to around 64 nm in air dried samples, and down to 60 nm after dehydrothermal treatments at 12 0C [155, 156] Short D periodicities around 57 nm and 62 nm

PAGE 109

109 from air dried collagen fibrils has also been reported from AFM measurement[157] Since dehydration induces structural disorde r and mechanical stresses, it is possible that the collagen fibrils are destabilized and the D periodicities are decreased after water evaporation. Usually, the two identical collagenion complexes always repel each other, or the magnitude of repulsion is reduced in the electrolytes. It has been know for over a decade, from the simulation and integral equation approaches, the short range correlations between the counterions can lead to attractive corrections to the electrostatic interaction [140] The occurrence B /b, where lB=e2 0BT is the Bjerrum length (lB=7.1 A at room tempera ture in water). According to the Manning criterion[158, 159] because of condensation of the small mobile counterions onto the rion valency). The fraction 1 on them. An experimental rule for the disappearance of electrostatic repulsion between DNA strands is that the effective charge per unit len gth should be less that 10% of its bare value. Take the collagen as a rod shape molecule, the electrostatic interaction between collagen molecules is the sum of two terms: The first term is repulsive, and originates from the net charge of each rod, which i s nonzero because not all counterions have condensed. The second term is attractive, and originates from charge fluctuations along the rods, due to the free exchange between condensed and free counterions. When the two collagenion complexes approach each other during increasing temperature, attractions are induced by hydrophobic attraction. At small distances, the binding ions can undergo dramatic rearrangement that renormalized the effective charge distribution. It is possible that the renormalization of charge distribution of collagensalt

PAGE 110

110 complexes and subsequently searching a particular target sequence to form the characteristic D periodic fibrils. Conclusion The degree of collagen fibrillogenesis is highly affected by salt type and concentration. Due to the stronger tendency of multivalent ions over monovalents to bind on collagen surface, salts with high binding affinities can screen off the Coulombic repulsion and promote intermolecular interactions. Fibrils with a native banding pattern formed in s olutions containing salts with divalent ions. It is thought that these divalent salts could facilitate fibrillogenesis by providing like charge attraction and forming salt bridges between collagen molecules.

PAGE 111

111 CHAPTER 7 SUMMARY AND FUTURE W ORK Summary Self a ssembly of solubilized collagen into fi brils was first observed about 60 years ago[106] Under the physiological condition, collagen molecules aggregate spontaneously to form fibrils with the characteristic D periodic pattern. Subsequent studies indicate that ions, alcohol, and other substances which influence electrostatic, hydrophobic and covalent bonding were able to modify self assembly behavior. However, the mechanisms by which collagen molecules assembly into fibrils are still not well understood, especially th e mechanisms that drive the D periodic packing of monomers to form native fibril. It has been proposed that specific distribution of charged and hydrophobic amino acids along collagen molecules plays an important role in the molecular packing of collagen[160] The telopeptides on the collagen also play a c ritical role in the formation of collagen fibrils [59] but other factors driving collagen assembly into D periodic fibrils are not known. The aim of this dissertation is to understand the mechanisms of the collagen assembly into D periodic fibrils, the interactions between the collagen molecules during fibril formation as well as the underlying forces driving the formation of D perio dic pattern. To accomplish these goals, we investigated the kinetics of collagen molecules assemble into fibrils, the temperature, concentration, pH, salts and surfactants effects on controlling the alignment of collagen monomers into fibrils. We employed three kinds of molecular interactions: 1) hydrophobic interactions which were introduced by surfactants of SDS and SDBS; 2) Electrostatic interactions which were controlled by changing the pH of the solution; 3) specific likely charge attractions induced b y weak bridges of divalent ions. Morphology studies by electron microscopy suggest that the fundamental mechanism of the

PAGE 112

112 characteristic fibrils is mediated by the electrostatic interactions and salts bridges play an important role on molecular recognition to form D periodic pattern. Surfactant Effects The effect s of the presence of ionic surfactants on collagen fibrillogenesis indicated that t he rate of collagen fibril formation was accelerated remarkably in the presence of 0.1 0.5 mM sodium dodecylsulfate (SDS) or sodium dodecylbenzenesulfonate (SDBS) while the unfolding of collagen triple helix also occurred when up to 1 mM SDS was added or more than 0.35 mM SDBS was added. The morphology studies from SEM confirmed occurrence of partial unfolding and nonf ibrillar collagen gel within fibrils. There is a weak interaction between collagen and SDS at room temperature and stronger binding appeared between collagen and ionic surfactant during collagen dehydration above room temperature It is possible surfactant s bind to collagen which promote s collagen fibril formation. pH Effects Collagen self assembly in vitro was conducted in the pH range from 6.0 to 10.5 at 30 C in order to investigate the electrostatic interactions that occur during fibril formation. A sigmoidal curve was observed in the growth rate of fibrils. Collagen fibril morphologies imaged by transmission electron microscopy (TEM) and scanning electron microscopy (SEM) present bundling of fibrils with a small amount of nonfibrillar collagen. At a low pH of 6.6, collagen molecules form small fibrils with a diameter of 85 nm. In the pH range from 6.9 to 8.0, they form fibrils with diameter of approximately 200 nm, even though the rate of fibrillogenesis accelerates with increasing pH in this range. Ze ta potential measurements of soluble collagen indicate that the net surface charge of collagen molecules is not only affected by the pH of medium but also by the presence of added salts. The acceleration of fibrillogenesis rate with increasing pH from 6.6 to 9.2 is consistent with a reduction of surface net charge since the

PAGE 113

113 isoelectric point of soluble collagen is approach ed The native D periodicity of 62 nm was found except at pH 7.1 where collagen molecules form short banding of 5060 nm in the early sta ge of fibrillogenesis which might be caused by unusual alignment of collagen molecules in fibrils. Salts Effect s The behavior of collagen molecules self assembly into fibrils is commonly understood in terms of hydrophobic and electrostatic interactions in the short range by releasing the structured water. As an expectation, the electrostatic interactions are repulsive based on mean field theories, such as the Poisson Boltzmann theory (PB theory). In our studies, the facilitation of collagen aggregation through reducing surface net charge by adding salts was observed. We also found the competition between unfolding and aggregation of collagen at high pH. The divalent ions binding to the collagen molecules not only change the surface net charge but also facili tate the formation of fibrils with native D periodic banding pattern. It is possible that the bound divalent ions induce the like charge attraction during collagen self assembly through coupling and renormalization charge distribution between counter ion c orrelation and collagen. We believe these findings can fundamentally help understanding the mechanisms of fibrillgenesis and may initiate new developments of biological self assembly. Even though hydrophobic attraction is the major force for collagen aggre gation, the formation of native fibrils depends on the appearance of the divalent ions which induce the like charge attraction and bridging the collagen for fibril formations. As it is know, the self assembly of molecules has attracted a lot of attention f or design and development of new materials. Major mechanism for self assembly is general the hydrophobic attractions. Our finding of like charge attraction brings a new sight for developing self assembly system.

PAGE 114

114 Future Work The adsorption of collagen at s olid surfaces is of importance in a wide variety of applications, including in vitro cell growth, membrane fouling, protein purification, and biosensor design. Microfabrication technology offers the capability to control cell surface, cell cell, and cell medium interactions on a micro or nanometer scale. Various studies proved that cells are affected by the topography of the surface on which they were seeded. Cells were reported to elongate in the direction of the micrometersize grooves and migrate as gu ided by the grooves [161] Therefore, micro patterned substrates are expected to better maintain cell morphology, differentiation and functionality over long periods of time. Our studies have indicated that surfactant s bind to collagen by hydrophobic interactions. It has also repor ted adsorption of collagen on the substrate through hydrophobic interactions. One of our future works is to fabricate collagen fibrils on the substrates and control their position by microcontact patterning. To achieve this goal, experiments are designed as: 1) D esign applicable patterns and fabricate the substrates which are suitable for collagen fibril adsorption. 2) Prepare collagen fibril dispersion and print on the substrate. 3) Examine its function on cell attachment and proliferation. Usually, c ollagen fibrils are arranged in complex three dimensional arrays in vivo often in an aligned manner to fulfill certain biomechanical functions such as resisting high tensile stress. Collagen can be found as parallel fiber bundles in tendon and ligaments [162] as concentric waves in bone [163] or as orthogonal lattices in cornea [164] The spatial organization of collagen fibers in vivo is believed to play an important role in directing cell behavior and providing mechanical support. Several approache s have been introduced to reconstitute spatial ordered collagen matrices in vitro Such as exposing collagen solution to a strong magnetic field which

PAGE 115

115 aligns collagen fibrils due to the diamagnetic properties of collagen molecules [165] Aligned collagen nanofiber matrices have also been produced by electrospinning or use of a mica surface in combination with hydrodynamic flow [166, 167] We have tried to make the oriented native collagen fibril with concentrate d collagen solution. However, the occurrence of nonfibrillar collagen during fibril formation makes it infeasible. Electrospinning method is usually involved in collagen denaturation. We have found that collagen can form native fibrils in divalent ion sol ution and fibrils can be stabilized after long term storage. Those stabilized fibrils can be dispersed by stirring. Normally, the collagen fibrils formed from gelation are entangled with each other. Strong mechanical agitation causes fibril dissociation. S ince we can make the stable and well dispersed fibrils, it is possible that the fibrils can be aligned by hydrodynamic flow to form the 3D matrices. We found collagen can form the unusual periodicity less than 67 nm. Short periodicities have also been repo rted but the mechanisms are unclear. Correct longitudinal alignment of collagen monomers in a fibril is important for the mechanical and biological functions of collageneous matrices. Specifically, a precise alignment of monomers facilitates the formation of fibril stabilizing inter molecular chemical crosslinks between specific lysine residues present within telopeptide regions of one monomer and specific lysine residues present with a triple helical region of interaction monomers. It is worthy of studying the specific factors which cause the short periodicities in order to control the periodicities and help understanding the biological system and collagen synthesis in vivo For this goal, the research is designed as: 1) Produce the collagen fibrillar matrices with shorter periodicity. 2) Examine its mechanical properties. It has been thought native fibrils with D periodicity are stabilized by intermolecular chemical crosslinks, is it possible that unusual fibrils with short periodicity would be less stable. To study the

PAGE 116

116 mechanical properties, the mechanical strength of single fibril will be measured by AFM. 3) It has been reported that hydrophobic interactions and electrostatic interactions bring a substantial contribution the formation of D periodicity. Calculating the hydrophobic and electrostatic interactions between collagen molecules in unusual periodicities would be performed.

PAGE 117

117 LIST OF REFERENCES [1] Friess W. Collagen biomaterial for drug delivery. Eur J Pharm Biopharm 1998;45(2):113136. [2] Burke JF, Yannas IV, Quinby WC, Bondoc CC, Jung WK. Successful Use Of A Physiologically Acceptable Artificial Skin In The Treatment Of Extensive Burn Injury. Annals Of Surgery 1981;194(4):413428. [3] Eaglstein WH, Falanga V. Tissue engine ering for skin: An update. Journal Of The American Academy Of Dermatology 1998;39(6):10071010. [4] Blair HC, Zaidi M, Schlesinger PH. Mechanisms balancing skeletal matrix synthesis and degradation. Biochem J 2002;364:329341. [5] Brekken RA, Sage EH. SPAR C, a matricellular protein: at the crossroads of cell matrix. Matrix Biology 2000;19(7):569 580. [6] Piez KA, Reddi AH. Extracellular Matrix Biochemistry: New York, 1984. [7] Kadler K. Extracellular Matrix.1. Fibril Forming Collagens. Protein Profile 1995; 2(5):491619. [8] Cunningam W, Frederiksen DW. Methods in Enzymology. New York: Academic Press, 1982. [9] Hulmes DJS. The collagen superfamily diverse structures and assemblies. Essays In Biochemistry, Vol 27 1992;27:4967. [10] Pachence JM. Collagen bas ed devices for soft tissue repair. J Biomed Mater Res 1996;33(1):3540. [11] Meaney Murray M, Rice K, Wright RJ, Spector M. The effect of selected growth factors on human anterior cruciate ligament cell interactions with a three dimensional collagen GAG scaffold. J Orthop Res 2003;21(2):23844.

PAGE 118

118 [12] Angele P, Kujat R, Nerlich M, Yoo J, Goldberg V, Johnstone B. Engineering of osteochondral tissue with bone marrow mesenchymal progenitor: Cells in a derivatized hyaluronan gelatin composite sponge. Tissue Engineering 1999;5(6):545553. [13] Eaglstein WH, Falanga V. Tissue engineering and the development of Apligraf a human skin equivalent. Adv Wound Care 1998;11(4 Suppl):1 8. [14] Ananthan.S, Veis A. Molecular Parameters Of Monomeric And Acid Soluble Collagens Low Shear Gradient Viscosity And Electric Birefringenc e. Biopolymers 1972;11(7):13651373. [15] Ramachandran GN. Treatise on Collagen. New York: Academic Press, 1967. [16] Fraser RDB, Macrae TP, Suzuki E. Chain Conformation In The Collagen Molecule. J Mol Biol 1979;129(3):463481. [17] Akeson WH, Amiel D, Mechanic GL, Woo SLY, Harwood FL, Hamer ML. Collagen Cross Linking Alterations In Joint Contractures Changes In Reducible Cross Links In Periarticular Connective Tissue Collagen After 9 Weeks Of Immobil ization. Connect Tissue Res 1977;5(1):1519. [18] Yamauchi M, Mechanic G. Crosslinking of collagen. Boca Raton, FL: CRC Press, 1988. [19] Kuijpers AJ, Engbers GHM, Krijgsveld J, Zaat SAJ, Dankert J, Feijen J. Crosslinking and characterisation of gelatin m atrices for biomedical applications. J Biomater SciPolym Ed 2000;11(3):225243. [20] Khor E. Methods for the treatment of collagenous tissues for bioprostheses. Biomaterials 1997;18(2):95105. [21] Hodge AJ. Molecular Models Illustrating The Possible Distributions Of Holes In Simple Systematically Staggered Arrays Of Type I Collagen Molecules In Native Type Fibrils. Connect Tissue Res 1989;21(14):467477.

PAGE 119

119 [22] Veis A, Yuan L. Structure Of Collagen Microfibril 4Strand Overlap Model. Biopolymers 1975;14( 4):895900. [23] Miller A. The molecular packing in collagen fibrils. New York: Plenum Press, 1976. [24] Currey JD. How well are bones designed to resist fracture? J Bone Miner Res 2003;18(4):591598. [25] Landis WJ. Mineral characterization in calcifying tissues: Atomic, molecular and macromolecular perspectives. Connect Tissue Res 1996;35(14):1 8. [26] Landis WJ, Hodgens KJ, Arena J, Song MJ, McEwen BF. Structural relations between collagen and mineral in bone as determined by high voltage electron microscopic tomography. Microscopy Research And Technique 1996;33(2):192202. [27] Jaschouz D, Paris O, Roschger P, Hwang HS, Fratzl P. Pole figure analysis of mineral nanoparticle orientation in individual trabecula of human vertebral bone. Journal Of Applied Crystallography 2003;36:494498. [28] Yamauchi M. Calcium and phosphorus in Health and Disease. New York: CRC Press, 1996. [29] Hodge A. J. PJA. Aspects of protein structure. New York: Academic, 1963. [30] Fratzl P, Fratzlzelman N, Klaushofer K, Vogl G, K oller K. Nucleation And Growth Of Mineral Crystals In Bone Studied By SmallAngle X Ray Scattering. Calcif Tissue Int 1991;48(6):407413. [31] Fratzl P, Groschner M, Vogl G, Plenk H, Eschberger J, Fratzlzelman N, et al. Mineral Crystals In Calcified Tissue s A Comparative Study By Saxs. J Bone Miner Res 1992;7(3):329334.

PAGE 120

120 [32] Cassel JM, Christen.R.G. Volume Change On Formation Of Native Collagen Aggregate. Biopolymers 1967;5(5):431438. [33] Wood GC. Formation Of Fibrils From Collagen Solutions.2. Mechani sm Of Collagen Fibril Formation. Biochem J 1960;75:598605. [34] Gelman RA, Piez KA. Collagen Fibril Formation Invitro A Quasi Elastic Light Scattering Study Of Early Stages. J Biol Chem 1980;255(17):80988102. [35] Silver FH. A Molecular Model For Linea r And Lateral Growth Of TypeI Collagen Fibrils. Collagen And Related Research 1982;2(3):219229. [36] Holmes DF, Capaldi MJ, Chapman JA. Reconstitution Of Collagen Fibrils Invitro The Assembly Process Depends On The Initiating Procedure. Int J Biol Macr omol 1986;8(3):161 166. [37] Brennan M, Davison PF. Role Of Aldehydes In Collagen Fibrillogenesis Invitro. Biopolymers 1980;19(10):18611873. [38] Haworth RA, Chapman JA. Study Of Growth Of Normal And Iodinated Collagen Fibrils Invitro Using Electron Micro scope Autoradiography. Biopolymers 1977;16(9):18951906. [39] Silver FH. Type I Collagen Fibrillogenesis Invitro Additional Evidence For The Assembly Mechanism. J Biol Chem 1981;256(10):4973 4977. [40] Gross J, Kirk D. Heat Precipitation Of Collagen From Neutral Salt Solutions Some Rate Regulating Factors. J Biol Chem 1958;233(2):355 360. [41] Fratzl P, Fratzlzelman N, Klaushofer K. Collagen Packing And Mineralization An X Ray Scattering Investigation Of Turkey Leg Tendon. Biophys J 1993;64(1):260266.

PAGE 121

121 [42] Abraham LC, Zuena E, Perez Ramirez B, Kaplan DL. Guide to collagen characterization for biomaterial studies. Journal Of Biomedical Materials Research Part B Applied Biomaterials 2008;87B(1):264285. [43] Engel J, Bachinger HP. Structure, stability a nd folding of the collagen triple helix. Collagen. Berlin: Springer Verlag Berlin, 2005. p. 733. [44] Rho JY, KuhnSpearing L, Zioupos P. Mechanical properties and the hierarchical structure of bone. Med Eng Phys 1998;20(2):92102. [45] Gross J, Highberger JH, Schmitt FO. Collagen Structures Considered As States Of Aggregation Of A Kinetic Unit The Tropocollagen Particle. Proc Natl Acad Sci U S A 1954;40(8):679688. [46] Highberger JH, Gross J, Schmitt FO. The Interaction Of Mucoprotein With Soluble Col lagen An Electron Microscope Study. Proc Natl Acad Sci U S A 1951;37(5):286291. [47] Schmitt FO, Gross J, Highberger JH. A New Particle Type In Certain Connective Tissue Extracts. Proc Natl Acad Sci U S A 1953;39(6):459470. [48] Bianchi E, Conio G, Cif erri A. Helix Coil Transformation For Tropocollagen Solutions And Its Relationship To Transformations Involving Crystalline Form Of Prot ein. Biopolymers 1966;4(9):957963. [49] Wood GC, Keech MK. Formation Of Fibrils From Collagen Solutions.1. Effect Of Ex perimental Conditions Kinetic And ElectronMicroscope Studies. Biochem J 1960;75:588598. [50] Silver FH, Langley KH, Trelstad RL. Type I Collagen Fibrillogenesis Initiation Via Reversible Linear And Lateral Growth Steps. Biopolymers 1979;18(10):25232535.

PAGE 122

122 [51] Brennan M, Davison PF. Influence Of The Telopeptides On Type I Collagen Fibrillogenesis. Biopolymers 1981;20(10):2195 2202. [52] Gelman RA, Williams BR, Piez KA. Collagen Fibril Formation Evidence For A Multistep Process. J Biol Chem 1979;254(1):180186. [53] Delorenzi NJ, Sculsky G, Gatti CA. Effect of monovalent anions on type I collagen fibrillogenesis in vitro. Int J Biol Macromol 1996;19(1):1520. [54] Doty P, Edsall JT. Light Scattering In Protein Solutions. AdvProtein Chem 1951;6:35121. [55] Doty P, Steiner RF. Light Scattering And Spectrophotometry Of Colloidal Solutions. Journal Of Chemical Physics 1950;18(9):1211 1220. [56] Silver FH, Birk DE. Kinetic Analysis Of Collagen Fibrillogenesis.1. Use Of Turbidity Time Data. Collage Relat R es 1983;3(5):393 405. [57] Miller EJ, Rhodes RK. Preparation And Characterization Of The Different Types Of Collagen. Methods Enzymol 1982;82:3364. [58] Williams BR, Gelman RA, Poppke DC, Piez KA. Collagen Fibril Formation Optimal Invitro Conditions And Preliminary Kinetic Results. J Biol Chem 1978;253(18):65786585. [59] Veis A. Collagen Fibrillogenesis. Connect Tissue Res 1982;10(1):11 24. [60] Steinert PM, Idler WW, Zimmerman SB. Self Assembly Of Bovine Epidermal Keratin Filaments Invitro. J Mol Biol 1976;108(3):547 567. [61] Hofrichter J, Ross PD, Eaton WA. Supersaturation In Sickle Cell Hemoglobin Solutions. Proc Natl Acad Sci U S A 1976;73(9):30353039.

PAGE 123

123 [62] Cisneros DA, Hung C, Franz CA, Muller DJ. Observing growth steps of collagen self assembly b y time lapse highresolution atomic force microscopy. J Struct Biol 2006;154(3):232 245. [63] Cassel JM. Collagen Aggregation Phenom ena. Biopolymers 1966;4(9):989997. [64] Cooper A. Thermodynamic Studies Of Assembly InVitro Of Native Collagen Fibrils. Bi ochem J 1970;118(3):355365. [65] Hayashi T, Nagai Y. Factors Affecting Interactions Of Collagen Molecules As Observed By In Vitro Fibril Formation.2. Effects Of Species And Concentration Of Anions. J Biochem (Tokyo) 1973;74(2):253262. [66] Steinhar.J, St ocker N, Carroll D, Birdi KS. Nonspecific Large Binding Of Amphiphiles By Proteins. Biochemistry 1974;13(21):44614468. [67] Nozaki Y, Reynolds JA, Tanford C. Interaction Of A Cationic Detergent With Bovine Serum Albumin And Other Proteins. J Biol Chem 1974;249(14):4452 4459. [68] Steinhar.J, Krijn J, Leidy JG. Differences Between Bovine And Human Serum Albumins Binding Isotherms, Optical Rotatory Dispersion, Viscosity, Hydrogen Ion Titration, And Fluorescence Effects. Biochemistry 1971;10(22):40054015. [69] Reynolds JA, Herbert S, Polet H, Steinhar.J. Binding Of Divers Detergent Anions To Bovine Serum Albumin. Biochemistry 1967;6(3):937947. [70] Honya M, Mizunuma H. Collagen Fibrillogenesis In Vitro Accelerative Effect Of Surfactants On Collagen Fibril Formation. J Biochem (Tokyo) 1974;75(1):113121. [71] Hayashi T. Factors Affecting Interactions Of Collagen Molecules As Observed By In Vitro Fibril Formation.1. Effects Of Small Molecules, Especially Saccharides. J Biochem (Tokyo) 1972;72(3):749758.

PAGE 124

124 [7 2] Maldonado F, Almela M, Otero A, Costalopez J. The Binding Of Anionic And Nonionic Surfactants To Collagen Through The Hydrophobic Effect. Journal Of Protein Chemistry 1991;10(2):189192. [73] Henriquez M, Lissi E, Abuin E, Ciferri A. Assembly Of Amphiphilic Compounds And Rigid Polymers.1. Interaction Of Sodium Dodecyl Sulfate With Collagen. Macromolecules 1994;27(23):68346840. [74] Holmberg K. Handbook of applied surface and colloid chemistry. New York: Wiley and Sons, 2002. [75] Benito I, Garcia MA, Mo nge C, Saz JM, Marina ML. Spectrophotometric and conductimetric determination of the critical micellar concentration of sodium dodecyl sulfate and cetyltrimethylammonium bromide micellar systems modified by alcohols and salts. Colloids And Surfaces A Physicochemical And Engineering Aspects 1997;125(23):221 224. [76] Corrin ML, Harkins WD. The Effect Of Salts On The Critical Concentration For The Formation Of Micelles In Colloidal Electrolytes. J Am Chem Soc 1947;69(3):683 688. [77] Fuguet E, Rafols C, Roses M, Bosch E. Critical mic elle concentration of surfactants in aqueous buffered and unbuffered systems. Analytica Chimica Acta 2005;548(12):95100. [78] Brown RA, Wiseman M, Chuo CB, Cheema U, Nazhat SN. Ultrarapid engineering of biomimetic materials and tissues: Fabrication of na noand microstructures by plastic compression. Advanced Functional Materials 2005;15(11):17621770. [79] Nezu T, Nishiyama N, Nemoto K, Terada Y. The effect of hydrophilic adhesive monomers on the stability of type I collagen. Biomaterials 2005;26(18):38013808. [80] Honya M. Collagen Fibrillogenesis Invitro Functional Role Of Nonionic Materials In Collagen Fibril Formation. J Biochem (Tokyo) 1974;75(5):10371045.

PAGE 125

125 [81] Miller DD, Lenhart W, Antalek BJ, Williams AJ, Hewitt JM. The Use Of Nmr To Study Sodi um Dodecyl Sulfate Gelatin Interactions. Langmuir 1994;10(1):68 71. [82] De Cupere VM, Van Wetter J, Rouxhet PG. Nanoscale organization of collagen and mixed collagenpluronic adsorbed layers. Langmuir 2003;19(17):69576967. [83] Na GC, Phillips LJ, Freire EI. Invitro Collagen Fibril Assembly Thermodynamic Studies. Biochemistry 1989;28(18):71537161. [84] Tiktopulo EI, Kajava AV. Denaturation of type I collagen fibrils is an endothermic process accompanied by a noticeable change in the partial heat capaci ty. Biochemistry 1998;37(22):81478152. [85] Miles CA, Ghelashvili M. Polymer in a box mechanism for the thermal stabilization of collagen molecules in fibers. Biophys J 1999;76(6):32433252. [86] Mysels KJ, Princen LH. Light Scattering By Some Lauryl Sulf ate Solutions. Journal Of Physical Chemistry 1959;63(10):16961700. [87] Katz EP, Li S. Structure And Function Of Bone Collagen Fibrils. J Mol Biol 1973;80(1):115. [88] Reilly DT, Burstein AH. Review Article Mechanical Properties Of Cortical Bone. J Bon e Joint SurgAm Vol 1974;A 56(5):10011022. [89] Landis WJ, Silver FH, Freeman JW. Collagen as a scaffold for biomimetic mineralization of vertebrate tissues. J Mater Chem 2006;16(16):14951503. [90] Olszta MJ, Cheng XG, Jee SS, Kumar R, Kim YY, Kaufman MJ, et al. Bone structure and formation: A new perspective. Mater Sci Eng R Rep 2007;58(35):77 116.

PAGE 126

126 [91] Olszta MJ, Douglas EP, Gower LB. Scanning electron microscopic analysis of the mineralization of type I collagen via a polymer induced liquidprecursor (PILP) process. Calcif Tissue Int 2003;72(5):583591. [92] Song J, Malathong V, Bertozzi CR. Mineralization of synthetic polymer scaffolds: A bottom up approach for the development of artificial bone. J Am Chem Soc 2005;127(10):3366 3372. [93] Schmitt F, H all C, Jakus M. Electron microscope investigations of the structure of collagen. J Cell Comp Physiol 1942;20(1):1133. [94] Schmitt FO, Hall CE, Jakus MA. Electron microscope investigations of the structure of collagen. J Cell Comp Physiol 1942;20(1):1133 [95] Bear RS. Long xray diffraction spacings of collagen. J Am Chem Soc 1942;64:727727. [96] Bear RS. Complex formation between starch and organic molecules. J Am Chem Soc 1944;66:21222123. [97] Bensusan HB, Hoyt BL. The Effect Of Various Parameters O n The Rate Of Formation Of Fibers From Collagen Solutions. J Am Chem Soc 1958;80(3):719 724. [98] Cassel JM, Christen.R.G. Volume Change On Formation Of Native Collagen Aggregate. Biopolymers 1967;5(5):431437. [99] Hayashi T, Nagai Y. Factors Affecting In teractions Of Collagen Molecules As Observed By Invitro Fibril Formation.3. NonHelical Regions Of Collagen Molecules. J Biochem (Tokyo) 1974;76(1):177186. [100] Helseth DL, Veis A. Collagen Self Assembly Invitro Differentiating Specific Telopeptide Dep endent Interactions Using Selective Enzyme Modification And The Addition Of Free Amino Telopeptide. J Biol Chem 1981;256(14):7118 7128.

PAGE 127

127 [101] Cassel JM. Collagen Aggregation Phenomena. Biopolymers 1966;4(9):989997. [102] Leikin S, Rau DC, Parsegian VA. Te mperature Favored Assembly Of Collagen Is Driven By Hydrophilic Not Hydrophobic Interactions. Nat Struct Biol 1995;2(3):205210. [103] Hofmann H, Fietzek PP, Kuhn K. Role Of Polar And Hydrophobic Interactions For Molecular Packing Of Type I Collagen 3Di mensional Evaluation Of AminoAcid Sequence. J Mol Biol 1978;125(2):137165. [104] Bard JBL, Chapman JA. Diameters Of Collagen Fibrils Grown InVitro. Nature New Biology 1973;246(151):8384. [105] Hayes RL, Allen ER. Electron Microscopic Studies On A Doubl e Stranded Beaded Filament Of Embryonic Collagen. J Cell Sci 1967;2(3):419434. [106] Noda H, Wyckoff RWG. The Electron Microscopy Of Reprecipitated Collagen. Biochim Biophys Acta 1951;7(4):494506. [107] Raspanti M, Alessandrini A, Ottani V, Ruggeri A. Di rect visualization of collagen bound proteoglycans by tappingmode atomic force microscopy. J Struct Biol 1997;119(2):118122. [108] Vanamee P, Porter KR. Observations With The Electron Microscope On The Solvation And Reconstitution Of Collagen. J Exp Med 1951;94(3):255273. [109] Mertz EL, Leikin S. Interactions of inorganic phosphate and sulfate anions with collagen. Biochemistry 2004;43(47):1490114912. [110] Sun YL, Luo ZP, Fertala A, An KN. Direct quantification of the flexibility of type I collagen mo nomer. Biochem Biophys Res Commun 2002;295(2):382 386. [111] Fratzl P, Gupta HS, Paschalis EP, Roschger P. Structure and mechanical quality of the collagen mineral nano composite in bone. J Mater Chem 2004;14(14):21152123.

PAGE 128

128 [112] Glimcher MJ. Molecular Bio logy Of Mineralized Tissues With Particular Reference To Bone. Rev Mod Phys 1959;31(2):359393. [113] Brodsky B, Eikenberry EF, Cassidy K. Unusual Collagen Periodicity In Skin. Biochim Biophys Acta 1980;621(1):162166. [114] Stinson RH, Sweeny PR. Skin Col lagen Has An Unusual D Spacing. Biochim Biophys Acta 1980;621(1):158 161. [115] Hattori S, Adachi E, Ebihara T, Shirai T, Someki I, Irie S. Alkali treated collagen retained the triple helical conformation and the ligand activity for the cell adhesion via a lpha 2 beta 1 integrin. J Biochem (Tokyo) 1999;125(4):676684. [116] Jackson DS, Neuberger A. Observations On The Isoionic And Isoelectric Point Of AcidProcessed Gelatin From Insoluble And Citrate Extracted Collagen. Biochim Biophys Acta 1957;26(3):638639. [117] Freudenberg U, Behrens SH, Welzel PB, Muller M, Grimmer M, Salchert K, et al. Electrostatic interactions modulate the conformation of collagen I. Biophys J 2007;92(6):21082119. [118] Goh MC, Paige MF, Gale MA, Yadegari I, Edirisinghe M, Strzelczy k J. Fibril formation in collagen. Physica A 1997;239(13):95 102. [119] Leikina E, Mertts MV, Kuznetsova N, Leikin S. Type I collagen is thermally unstable at body temperature. Proc Natl Acad Sci U S A 2002;99(3):13141318. [120] Feng D, Knight DP. The Ef fect Of Ph On Fibrillogenesis Of Collagen In The Egg Capsule Of The Dogfish, Scyliorhinus Canicula. Tissue Cell 1994;26(5):649659. [121] Jiang FZ, Horber H, Howard J, Muller DJ. Assembly of collagen into microribbons: effects of pH and electrolytes. J Str uct Biol 2004;148(3):268278.

PAGE 129

129 [122] Ripamonti A, Roveri N, Braga D, Hulmes DJS, Miller A, Timmins PA. Effects Of Ph And Ionic Strength On The Structure Of Collagen Fibrils. Biopolymers 1980;19(5):965975. [123] Katz EP, Li ST. Intermolecular Space Of Recon stituted Collagen Fibrils. J Mol Biol 1973;73(3):351369. [124] Weinstock A, King PC, Wuthier RE. IonBinding Characteristics Of Reconstituted Collagen. Biochem J 1967;102(3):983987. [125] Brodsky B, Eikenberry EF. Characterization Of Fibrous Forms Of Col lagen. Methods Enzymol 1982;82:127174. [126] Brodsky B, Eikenberry EF, Belbruno KC, Sterling K. Variations In Collagen Fibril Structure In Tendons. Biopolymers 1982;21(5):935951. [127] Wu TJ, Huang HH, Lan CW, Lin CH, Hsu FY, Wang YJ. Studies on the micr ospheres comprised of reconstituted collagen and hydroxyapatite. Biomaterials 2004;25(4):651658. [128] Scott JE, Orford CR. Dermatan Sulfate Rich Proteoglycan Associates With Rat Tail Tendon Collagen At The D Band In The Gap Region. Biochem J 1981;197(1): 213 216. [129] Gillard GC, Merrilees MJ, Bellbooth PG, Reilly HC, Flint MH. Proteoglycan Content And Axial Periodicity Of Collagen In Tendon. Biochem J 1977;163(1):145151. [130] Smith JW, Frame J. Observations On Collagen And Proteinpolysaccharide Complex Of Rabbit Corneal Stroma. J Cell Sci 1969;4(2):421436. [131] Tzaphlidou M. Measurement of the axial periodicity of collagen fibrils using an image processing method. Micron 2001;32(3):337 339. [132] Scott JE, Orford CR, Hughes EW. ProteoglycanCollagen A rrangements In Developing Rat Tail Tendon An Electron Microscopical And Biochemical Investigation. Biochem J 1981;195(3):573581.

PAGE 130

130 [133] Vanamee P, Porter KR. Observations With The Electron Microscope On The Solvation And Reconstitution Of Collagen. J Exp Med 1951;94(3):255268. [134] Vonhippe.Ph, Schleich T. Ion Effects On Solution Structure Of Biological Macromolecules. Accounts Of Chemical Research 1969;2(9):257263. [135] Aktas N. The effects of pH, NaCl and CaCl2 on thermal denaturation characteristic s of intramuscular connective tissue. Thermochim Acta 2003;407(12):105 112. [136] Usha R, Ramasami T. Effect of pH on dimensional stability of rat tail tendon collagen fiber. J Appl Polym Sci 2000;75(13):15771584. [137] Bianchi E, Conio G, Ciferri A, Pue tt D, Rajagh L. Role Of Ph Temperature Salt Type And Salt Concentration On Stability Of Crystalline Helical And Randomly Coiled Forms Of Collagen. J Biol Chem 1967;242(7):13611369. [138] Brown EM, Farrell HM, Wildermuth RJ. Influence of neutral salts on t he hydrothermal stability of acid soluble collagen. J Protein Chem 2000;19(2):8592. [139] Wong GCL, Lin A, Tang JX, Li YL, Janmey PA, Safinya CR. Regulation of F actin architecture using simple multivalent ions. Biophys J 2001;80(1):436. [140] GronbechJensen N, Mashl RJ, Bruinsma RF, Gelbart WM. Counterioninduced attraction between rigid polyelectrolytes. Phys Rev Lett 1997;78(12):24772480. [141] Podgornik R, Rau DC, Parsegian VA. Parametrization Of Direct And Soft Steric Undulatory Forces Between Dna Double Helical Polyelectrolytes In Solutions Of Several Different Anions And Cations. Biophys J 1994;66(4):962 971. [142] Melander W, Horvath C. Salt Effects On Hydrophobic Interactions In Precipitation And Chromatography Of Proteins Interpretation Of Lyot ropic Series. Archives Of Biochemistry And Biophysics 1977;183(1):200 215.

PAGE 131

131 [143] Vonhippe.Ph, Schleich T. Ion Effects On Solution Structure Of Biological Macromolecules. Accounts Of Chemical Research 1969;2(9):257265. [144] Robinson DR, Jencks WP. Effect Of Concentrated Salt Solutions On Activity Coefficient Of Acetyltetraglycine Ethyl Ester. J Am Chem Soc 1965;87(11):24702479. [145] Collins KD, Washabaugh MW. The Hofmeister Effect And The Behavior Of Water At Interfaces. Quarterly Reviews Of Biophysics 1 985;18(4):323422. [146] Nandi PK, Robinson DR. Effects Of Salts On Free Energy Of Peptide Group. J Am Chem Soc 1972;94(4):12991308. [147] Nandi PK, Robinson DR. Effects Of Salts On Free Energies Of Nonpolar Groups In Model Peptides. J Am Chem Soc 1972;94(4):13081315. [148] Gjerde DT, Schmuckler G, Fritz JS. Anion Chromatography With Low Conductivity Eluents.2. Journal Of Chromatography 1980;187(1):35 45. [149] Rieskautt MM, Ducruix AF. Relative Effectiveness Of Various Ions On The Solubility And Crystal Growth Of Lysozyme. J Biol Chem 1989;264(2):745748. [150] Angelini TE, Liang H, Wriggers W, Wong GCL. Like charge attraction between polyelectrolytes induced by counterion charge density waves. Proc Natl Acad Sci U S A 2003;100(15):86348637. [151] Agarwa l G, Kovac L, Radziejewski C, Samuelsson SJ. Binding of discoidin domain receptor 2 to collagen I: An atomic force microscopy investigation. Biochemistry 2002;41(37):1109111098. [152] Paige MF, Rainey JK, Goh MC. Fibrous long spacing collagen ultrastructure elucidated by atomic force microscopy. Biophys J 1998;74(6):32113216.

PAGE 132

132 [153] Lin H, Clegg DO, Lal R. Imaging real time proteolysis of single collagen I molecules with an atomic force microscope. Biochemistry 1999;38(31):99569963. [154] Suzuki Y, Someki I, Adachi E, Irie S, Hattori S. Interaction of collagen molecules from the aspect of fibril formation: Acid soluble, alkali treated, and MMP1 digested fragments of type I collagen. J Biochem (Tokyo) 1999;126(1):54 67. [155] Bella J, Brodsky B, Berman HM. Hydration Structure Of A Collagen Peptide. Structure 1995;3(9):893906. [156] Wess TJ, Orgel JP. Changes in collagen structure: drying, dehydrothermal treatment and relation to long term deterioration. Thermochim Acta 2000;365(12):119128. [157] Habelitz S, Balooch M, Marshall SJ, Balooch G, Marshall GW. In situ atomic force microscopy of partially demineralized human dentin collagen fibrils. J Struct Biol 2002;138(3):227236. [158] Stigter D. Evaluation Of The Counterion Condensation Theory Of Polyelectro lytes. Biophys J 1995;69(2):380388. [159] Ray J, Manning GS. An Attractive Force Between 2 Rodlike Polyions Mediated By The Sharing Of Condensed Counterions. Langmuir 1994;10(7):24502461. [160] Trus BL, Piez KA. Molecular Packing Of Collagen 3Dimensio nal Analysis Of Electrostatic Interactions. J Mol Biol 1976;108(4):705732. [161] Weiss P. Experiments On Cell And Axon Orientation Invitro The Role Of Colloidal Exudates In Tissue Organization. Journal Of Experimental Zoology 1945;100(3):353386. [162] Elliott DH. Structure And Function Of Mammalian Tendon. Biological Reviews Of The Cambridge Philos ophical Society 1965;40(3):392401.

PAGE 133

133 [163] Weiner S, Traub W. Bone Structure From Angstroms To Microns. Faseb J 1992;6(3):879885. [164] Holmes DF, Gilpin CJ Baldock C, Ziese U, Koster AJ, Kadler KE. Corneal collagen fibril structure in three dimensions: Structural insights into fibril assembly, mechanical properties, and tissue organization. Proc Natl Acad Sci U S A 2001;98(13):73077312. [165] Torbet J, Mal bouyres M, Builles N, Justin V, Roulet M, Damour O, et al. Orthogonal scaffold of magnetically aligned collagen lamellae for corneal stroma reconstruction. Biomaterials 2007;28(29):42684276. [166] Baker BM, Mauck RL. The effect of nanofiber alignment on t he maturation of engineered meniscus constructs. Biomaterials 2007;28(11):19671977. [167] Cisneros DA, Friedrichs J, Taubenberger A, Franz CM, Muller DJ. Creating ultrathin nanoscopic collagen matrices for biological and biotechnological applications. Sma ll 2007;3(6):956963.

PAGE 134

BIOGRAPHICAL SKETCH Yuping Li was born at suburban district of Chengdu, Sichuan province, China in 1978. She had a happy childhood and her parents provided her the nice living environment. She enjoyed playing with friends and the younger brother; learning knowledge. She enjoyed inventing new colors with different paints. During her middle school studies, she found she likes analysis and logical thinking. In high school, she chose to study in science and engineering. In 1997, she enrolled in Zhejiang University to study for a bachelors degree in the Department of Polymer Science and Engineering. In August 2001, she entered the graduate school in the same university under the advisors of Liqun Wang and Kehua Tu. During her graduate study, she worked on the synthesis and characterization of temperature responsive graft polymer for target drug delivery. In March 2004, she was awarded the masters degree from Zhejiang University. In August 2004, she came to the Department of Materials Science and Engineering at the University of Florida to pursue a Ph. D. degree. She worked under Dr. Elliot P. Douglas to study the mechanism of collagen fibril formation and develop coll agen scaffold for biomineralization. She received the Doctor of Phil osophy in May, 2009.