<%BANNER%>

A Model of Arthrofibrosis Using Intra-Articular Gene Delivery of Transforming Growth Factor Beta 1

Permanent Link: http://ufdc.ufl.edu/UFE0041397/00001

Material Information

Title: A Model of Arthrofibrosis Using Intra-Articular Gene Delivery of Transforming Growth Factor Beta 1
Physical Description: 1 online resource (140 p.)
Language: english
Creator: Watson, Rachael
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2010

Subjects

Subjects / Keywords: adhesive, arthrofibrosis, capsulitis, chondrometaplasia, ctgf, differentiation, factor, fibrosis, growth, mmp, tgfbeta1, transforming
Genetics (IDP) -- Dissertations, Academic -- UF
Genre: Medical Sciences thesis, Ph.D.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract: Idiopathic adhesive capsulitis (IAC) of the shoulder is a disease of unknown etiology characterized by painful, chronic fibrotic expansion of the synovium and joint capsule, gradually leading to loss of joint motion. Although IAC affects approximately 3-5% of the population, and 20% of diabetics, little is known about its pathogenesis. While the underlying causes are diverse, it is likely that many of the harmful aspects of fibrosis are mediated by transforming growth factor-beta1 (TGF-beta1), a pleiotropic cytokine. In an effort to establish an animal model of IAC and develop an understanding of the cellular and molecular events contributing to arthrofibrosis, we first used an adenovirus to deliver and over-express TGF-beta1 cDNA (Ad.TGF-?1) in the knee joints of athymic nude rats. By 5 days, TGF-beta1 induced a rapid increase in knee diameter and the complete encasement of the joints in dense scar-like tissue, locking the joints at 90o of flexion. Histologically, massive proliferation of synovial fibroblasts was seen, followed by their differentiation into myofibroblasts. By day 30 the phenotype of the expanded fibrotic tissue had undergone chondrometaplasia, indicated by tissue and cellular morphology, and matrix composition. Pre-labeling of the articular cells by injection of recombinant lentivirus containing the cDNA for eGFP showed the cells comprising the fibrotic/cartilaginous tissues appeared to arise almost entirely from the local proliferation and differentiation of resident fibroblasts. These results indicate that TGF-beta1 is a potent inducer of arthrofibrosis, and resident articular fibroblasts have immense proliferative potential and are highly plastic. Reduced viral dose and delivery of Ad.TGF-beta1 into the joints of immunocompetent animals resulted in a fibrotic pathology that more closely resembled IAC. This model was less severe and gradually resolved over 120 days. To examine the effects of diabetes on this arthrofibrotic model, streptozotocin was used to induce diabetes in animals. Ten days after onset of diabetes, diabetic animals were intra-articularly injected with Ad.TGF-beta1 which induced a fibrotic event similar to non-diabetic animals; however, there was markedly less chondrogenic differentiation, and tissue appeared to be less resolved at the end of the experiment. Alterations in ECM and adhesion genes were seen in the joint synovium, suggesting a reason why diabetics are prone to development of IAC and why diabetic animals may take longer to resolve fibrosis.
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 Rachael Watson.
Thesis: Thesis (Ph.D.)--University of Florida, 2010.
Local: Adviser: Ghivizzani, Steven.
Electronic Access: RESTRICTED TO UF STUDENTS, STAFF, FACULTY, AND ON-CAMPUS USE UNTIL 2011-04-30

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2010
System ID: UFE0041397:00001

Permanent Link: http://ufdc.ufl.edu/UFE0041397/00001

Material Information

Title: A Model of Arthrofibrosis Using Intra-Articular Gene Delivery of Transforming Growth Factor Beta 1
Physical Description: 1 online resource (140 p.)
Language: english
Creator: Watson, Rachael
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2010

Subjects

Subjects / Keywords: adhesive, arthrofibrosis, capsulitis, chondrometaplasia, ctgf, differentiation, factor, fibrosis, growth, mmp, tgfbeta1, transforming
Genetics (IDP) -- Dissertations, Academic -- UF
Genre: Medical Sciences thesis, Ph.D.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract: Idiopathic adhesive capsulitis (IAC) of the shoulder is a disease of unknown etiology characterized by painful, chronic fibrotic expansion of the synovium and joint capsule, gradually leading to loss of joint motion. Although IAC affects approximately 3-5% of the population, and 20% of diabetics, little is known about its pathogenesis. While the underlying causes are diverse, it is likely that many of the harmful aspects of fibrosis are mediated by transforming growth factor-beta1 (TGF-beta1), a pleiotropic cytokine. In an effort to establish an animal model of IAC and develop an understanding of the cellular and molecular events contributing to arthrofibrosis, we first used an adenovirus to deliver and over-express TGF-beta1 cDNA (Ad.TGF-?1) in the knee joints of athymic nude rats. By 5 days, TGF-beta1 induced a rapid increase in knee diameter and the complete encasement of the joints in dense scar-like tissue, locking the joints at 90o of flexion. Histologically, massive proliferation of synovial fibroblasts was seen, followed by their differentiation into myofibroblasts. By day 30 the phenotype of the expanded fibrotic tissue had undergone chondrometaplasia, indicated by tissue and cellular morphology, and matrix composition. Pre-labeling of the articular cells by injection of recombinant lentivirus containing the cDNA for eGFP showed the cells comprising the fibrotic/cartilaginous tissues appeared to arise almost entirely from the local proliferation and differentiation of resident fibroblasts. These results indicate that TGF-beta1 is a potent inducer of arthrofibrosis, and resident articular fibroblasts have immense proliferative potential and are highly plastic. Reduced viral dose and delivery of Ad.TGF-beta1 into the joints of immunocompetent animals resulted in a fibrotic pathology that more closely resembled IAC. This model was less severe and gradually resolved over 120 days. To examine the effects of diabetes on this arthrofibrotic model, streptozotocin was used to induce diabetes in animals. Ten days after onset of diabetes, diabetic animals were intra-articularly injected with Ad.TGF-beta1 which induced a fibrotic event similar to non-diabetic animals; however, there was markedly less chondrogenic differentiation, and tissue appeared to be less resolved at the end of the experiment. Alterations in ECM and adhesion genes were seen in the joint synovium, suggesting a reason why diabetics are prone to development of IAC and why diabetic animals may take longer to resolve fibrosis.
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 Rachael Watson.
Thesis: Thesis (Ph.D.)--University of Florida, 2010.
Local: Adviser: Ghivizzani, Steven.
Electronic Access: RESTRICTED TO UF STUDENTS, STAFF, FACULTY, AND ON-CAMPUS USE UNTIL 2011-04-30

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2010
System ID: UFE0041397:00001


This item has the following downloads:


Full Text

PAGE 1

A MODEL OF ARTHROFIBROSIS USING INTRA-ARTICULAR GENE DELIVERY OF TRANSFORMING GROWTH FACTOR BETA 1 By RACHAEL SUSAN WATSON A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORID A IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY UNIVERSITY OF FLORIDA 2010 1

PAGE 2

2010 Rachael Susan Watson 2

PAGE 3

To my family and friends for their uncond itional love, support and friendship! 3

PAGE 4

ACKNOWLEDGMENTS I would first like to thank my mentor, Dr. Steven Ghivizzani, for accepting me into his lab and presenting me with the necessary skills to become a research scientist. His encouragement, patience, guidance and mo tivation will not be forgotten. Dr. Ghivizzanis office door was always open for discussions, scientific or otherwise, and his willingness to meet regularly and provide support and insight fo r my project exemplifies his dedication. I would also like to thank the members of the Gh ivizzani lab. The assistance and mentorship from the postdoctoral associate, Padraic Levings, was critical. I would additionally like to thank my fellow graduat e students, Marsha Bush and Jesse Kay, for their friendshi p and camaraderie over the years. Furthermore, I would like to thank the other member s of our lab, Anthony Dac anay for his constant humor, Elvire Gouze, Jean-Noel Gouze, and Celine Theodore. I am especially grateful to my committ ee members, Dr. Alfr ed Lewin, Dr. Gregory Schultz, and Dr. Bryon Petersen, for their helpful suggestions, feedback and expertise, and for generously providing materials, res ources, and time necessary for me to complete these studies. I would particularly li ke to thank our collabo rators, above all, Dr. Thomas Wright for providing the inspiration for this project as well as for providing invaluable materials and guidance and Dr. Pa rker Gibbs, Dr. John Reith, and Marda Jorgensen. Lastly, none of this would be possible without the support of my loving family and friends: my parents, Robert and Susan, my brothers, Alexander and Robert Jr., my sister, Melissa, and my dearest friends Padraic, Lisa, LeeAnn, and Christy; all of whom provided an endless flow of love, humor and guidance. 4

PAGE 5

TABLE OF CONTENTS page ACKNOWLEDGMENTS ..................................................................................................4 LIST OF TABLES ............................................................................................................8 LIST OF FIGURES ..........................................................................................................9 ABSTRACT ...................................................................................................................14 CHAPTER 1 INTRODUC TION....................................................................................................16 Idiopathic Adhesive Capsulitis (IAC) .......................................................................16 Stages of IAC ..........................................................................................................17 Treatments of IAC are Controversial ......................................................................18 Tissue Repair and Fibrosis .....................................................................................19 TGF-beta 1 and Fibrogenesis .................................................................................21 Analysis of TGF1 and CTGF Expression in IAC Patients ....................................23 Diabetes, TGF-beta1 and Fibrosis ..........................................................................24 Gene Transfer to Capsular Fibroblasts ...................................................................27 Significance ............................................................................................................29 2 MATERIALS A ND METHOD S................................................................................34 Vector Production ...................................................................................................34 Tissue Culture .........................................................................................................34 ASO Transfection ...................................................................................................35 TGF1 ELISA ........................................................................................................35 Cell Migration Assay ...............................................................................................35 Animal Experimentation ..........................................................................................36 For Anti-sense oligonucleot ides (ASO) Experiments ........................................36 Induction of Diabetes in Animals ......................................................................36 Preparation of Total RNA ........................................................................................37 PCR Array ...............................................................................................................37 Immunohistochemistry ............................................................................................38 Detection of GFP ..............................................................................................39 Detection of Reporter ASO ...............................................................................39 Microarray ...............................................................................................................40 Sample Preparation ..........................................................................................40 RNA Amplification, Labeli ng and Array Hybridization .......................................40 3 GENERATION AND CHARACTERIZA TION OF ADENOVIRAL VECTOR ............42 In vitro Expression of Adenoviral Vector .................................................................42 5

PAGE 6

Overexpression of TGF1 In vivo Induces Fibrosis ...............................................44 4 GENE DELIVERY OF TGF-BETA1 INDUCES ARTHROFIBROSIS AND CHONDROMETAPLASIA OF SYNOVIUM IN VIVO ...............................................51 Introduction .............................................................................................................51 Results ....................................................................................................................54 Intra-articular Delivery of Ad.TGF1 Induces a Dose-Dependent Fibrotic Response: .....................................................................................................54 Sustained Overexpression of TGF1 in the Joints of Nude Rats Induces Severe Arthrofibrosis and Chondrometaplasia: .............................................55 TGF1 Stimulates Expression of Genes for Matricellular Proteins, MMPs, Collagens and Adhesion Molecules ..............................................................57 Resident Synovial/Capsular Fibroblasts Proliferate to form the Fibrotic Mass .60 Discussion ..............................................................................................................61 Expression Profiling is Consistent with an Aggressive Fibrotic and Chondrometaplastic Phenotype ....................................................................62 Synovial and Capsular Fibroblasts hav e a High Proliferative Capacity and Innate Plasticity .............................................................................................65 5 EXAMINING THE EFFECTS OF LONG TERM ARTHROFIBROSIS IN IMMUNOCOMPETEN T ANIMAL S..........................................................................76 Introduction .............................................................................................................76 Results ....................................................................................................................78 Delivery of Ad.TGF1 to the Joints of Immunocompetent Rats Induces Arthrofibrosis and Chondrometaplas ia that Resolves with Time ...................79 Expression Profiles of Genes Stimulated by TGFDisplay Altering Patterns of Expression Consistent with the Phenotype of the Tissue ...........80 Discussion ..............................................................................................................82 6 EXAMINING THE RELATIONSHIP BETWEEN DIABETES AND ARTHROFIBROSIS................................................................................................89 Introduction .............................................................................................................89 Results ....................................................................................................................91 D-glucose Stimulates Cell Migration .................................................................91 Streptozotocin Diabetic Induction .....................................................................92 Diabetes Induced Alterations in Gene Expression in Synovial Tissue ..............93 Diabetic Joints Receiving Ad.TGF1 Resulted in an Overall Less Severe Fibrosis ..........................................................................................................94 Decreases in Gene Expression at Onset of Fibrosis were seen between Diabetic and Non-Diabetic Animals Receiving Ad.TGF1 ............................96 Discussion ..............................................................................................................96 7 ANTISENSE-OLIGONUCLEOTIDES LESSEN SEVERITY OF JOINT FIBROSIS WHEN DELIEVERED AT ONSET OF DISEASE ................................111 6

PAGE 7

Introduction ...........................................................................................................111 Results ..................................................................................................................113 TGFKnockdown by Specific ELISA ..........................................................113 Intra-articular Injection of Reporter ASO is Retained in the Joint Space ........113 ASOs were Effective at Inhibiting Arthrofibrosis in Animal Model ...................114 Discussion ............................................................................................................115 8 SUMMARY AND FUTURE DIRECTIONS............................................................122 LIST OF REFERENCES .............................................................................................127 BIOGRAPHICAL SKETCH ..........................................................................................140 7

PAGE 8

LIST OF TABLES Table page 1-1 Stages of Idiopathi c Adhesive Capsulitis ............................................................31 4-1 Relative Expression Values of ECM and Associated Genes in the Joints of Nude Rats Receiving Ad.TGF1 .......................................................................68 5-1 Relative Expression Values of ECM and Associated Genes in the Joints of Wistar rats receiving 5.0*107 Ad.TGF1 ............................................................86 6-1 Mean of blood glucose in diabetic and normal Wistar rats. ..............................100 6-2 Relative signal values of ECM and adhesion genes in the joints of Diabetic Wistar rats compared to non-di abetic normal Wistar joints ...............................101 6-3 Relative signal values of ECM and adhesion genes in the joints of Diabetic Wistar rats receiving Ad.TGF1 compared to untreated diabetic Wistar rats ..103 6-4 Relative signal values of ECM and adhesion genes in the joints of Diabetic Wistar rats receiving Ad.TGF1 compared to non-diabetic Wistar rats receiving Ad.TGF1 ........................................................................................105 6-5 Arbitrary scale measuring severity of fibrosis by a blinded observer. Scale represents 0 (Normal)-4 (severe fibrosis). ........................................................107 7-1 Arbitrary scale in which a blinded observer measured severity of fibrosis. Scale represents 0 (normal)-4 (severe fibrosis). ...............................................118 8

PAGE 9

LIST OF FIGURES Figure page 1-1 TGFSmad signaling pathway. ........................................................................32 1-2 Immunohistochemistry analysis for CTGF expression in tissue from patients with adhesive capsulitis ......................................................................................33 3-1 Western blot analysis of CTGF protein ...............................................................46 3-2 hCTGF specific ELISA. .......................................................................................47 3-3 Cell proliferation assay. ......................................................................................48 3-4 Percentage change in knee diameter re lative to day zero of adenovirus injected Wistar rats. ............................................................................................49 3-5 Histolgical H&E stain of Wistar rats knees, four days post intra-articular injection.. ............................................................................................................50 4-1 Transgene expression following infecti on of rat synovial fibroblasts with Ad.TGF1.. ........................................................................................................70 4-2 Intra-articular delivery of Ad.TGF1 induces joint swelling ................................71 4-3 Local overexpression of Ad.TGF1 in the knee joint induce severe arthrofibrosis. ......................................................................................................72 4-4 Capsular fibrosis and chondrogenesis induced by Ad.TGF1. ..........................73 4-5 Immunohistochemical staining for MMPs 9 and 13, and smooth muscle actin ( SMA). .....................................................................................................74 4-6 Arthrofibrosis and chondrometaplas ia arise from resident synovial and capsular fibroblasts. ............................................................................................75 5-1 Capsular fibrosis and chondrogenesis induced by Ad.TGF1 resolves with time. ....................................................................................................................88 6-1 Glucose stimulates rat synovial fibroblast migration. ........................................108 6-2 Glucose stimulates rat synovial fibroblast migration. ........................................109 6-3 Capsular fibrosis induced by Ad.TGF1 in diabetic animals. ...........................110 9

PAGE 10

7-1 Transgene expression following transducti on of rat synovial fibroblasts with Ad.TGF1 and/or transfection with an antisens e oligonucleotide specific for TGF. ............................................................................................................119 7-2 Immunohistochemical staining of reporter anti-sense oligonucleotide ..............120 7-3 Capsular fibrosis induced by Ad.TGF1 in Wistar rats is inhibited by co-intraarticular injection of ASO. .................................................................................121 10

PAGE 11

LIST OF ABBREVIATIONS alpha SMA alpha smooth muscle actin Ad Adeno virus ASO antisense o ligonucleotide beta C degrees centigrade cDNA complementary DNA CMV cytomegalovirus CO 2 carbon dioxide CCN2/CTGF connective tissue growth factor CsCl cesium chloride DMEM Delbeccos modified Eagles medium DNA deoxyribonucleic acid ECM extracellular matrix EDTA ethylenediaminet etraacetic acid ELISA enzyme-linked immunosorbant assay FBS fetal bovine serum FGFR-1 fibroblast growth factor receptor 1 GFP green fluorescent protein H&E hematoxylin and eosin stain hr hour IAC idiopathic adhes ive capsulitis IHC immunohistochemistry IL-1 interleukin-1 beta 11

PAGE 12

IL-1Ra Interleukin-1 receptor antagonist mg milligram g microgram mL milliliter l microliter mM millimolar MMPs matrix metalloproteases ng nanogram nM nanomolar NSAID non steroidal anti-inflammatory drug OA osteoarthritis OPN osteopontin PBS phosphate buffered solution PCR polymerase chain reaction PDGFR platelet derived growth factor receptor alpha RA rheumatoid arthritis RNA ribonucleic acid ROM range of motion RSF rat synovial fibroblast SEM standard error of the mean siRNA small interfering RNA STZ streptozotocin TGF transforming growth factor TIMP tissue inhibitor of MMP TNFtumor necrosis factor alpha 12

PAGE 13

UV ultraviolet Vp viral particles VSV-G vesicular stomatitis virus G protein 13

PAGE 14

Abstract of Dissertation Pr esented to the Graduate School of the University of Florida in Partial Fulf illment of the Requirements for t he Degree of Doctor of Philosophy A MODEL OF ARTRHOFIBROSIS USIN G INTRA-ARTCULAR DELVERY OF TRANSFORMNG GROWTH FACTOR BETA-1 By Rachael Susan Watson May 2010 Chair: Steven C. Ghivizzani Major: Medical Sciences Genetics Idiopathic adhesive capsulitis (IAC) of the shoulder is a disease of unknown etiology characterized by painful, chronic fi brotic expansion of the synovium and joint capsule, gradually leading to loss of joint motion. Although IAC affects approximately 35% of the population, and 20% of diabetics, little is known about its pathogenesis. While the underlying causes are diverse, it is likely that many of the harmful aspects of fibrosis are mediated by transforming growth factor1 (TGF1), a pleiotropic cytokine. In an effort to establish an animal model of IAC and develop an understanding of the cellular and molecular events contributing to arthrofibrosis, we first used an adenovirus to deliver and over-express TGF1 cDNA (Ad.TGF-1) in the knee joints of athymic nude rats. By 5 days, TGF1 induced a rapid increase in knee diameter and the complete encasement of the joints in dense scar-like tissue, locking the joints at 90 o of flexion. Histologically, massive prolif eration of synovial fibroblasts was seen, followed by their differentiation into myofibroblasts. By day 30 the phenotype of the expanded fibrotic tissue had undergone chondrometaplasia, indicat ed by tissue and cellular morphology, and matrix composition. Pre-l abeling of the articular cells by injection of recombinant lentivirus containing the cDNA for eGFP showed the cells comprising the 14

PAGE 15

fibrotic/cartilaginous tissues appeared to arise almost entirely from the local proliferation and differentiation of resident fibroblasts These results indicate that TGF1 is a potent inducer of arthrofibrosis, and resident articu lar fibroblasts have immense proliferative potential and are highly pl astic. Reduced viral dose and delivery of Ad.TGF1 into the joints of immunocompetent animals resulted in a fibrotic pathology that more closely resembled IAC. This model was less severe and gradually resolved over 120 days. To examine the effects of diabetes on this arthro fibrotic model, streptozotocin was used to induce diabetes in animals. Ten days after onset of diabetes, diabetic animals were intra-articularly injected with Ad.TGF1 which induced a fibrotic event similar to nondiabetic animals; however, there was markedly less chondr ogenic differentiation, and tissue appeared to be less resolved at the en d of the experiment. Alterations in ECM and adhesion genes were seen in the joint synovium, suggesting a reason why diabetics are prone to developm ent of IAC and why diabetic animals may take longer to resolve fibrosis. 15

PAGE 16

CHAPTER 1 INTRODUCTION Idiopathic Adhesive Capsulitis (IAC) Also known as Frozen Shoulder Syndrome, IAC is a disease of unknown etiology that is characterized by both primary and se condary capsulitis. Primary capsulitis is manifested as intense pain, shoulder stiffnes s and a loss of passive and active shoulder rotation due to a thickening contraction and adhesion of the capsule, synovium, and surrounding structures. Secondary capusilits is histopathologically similar, but associated with an extrinsic or intrinsi c condition such as diabetes, autoimmune disease, stroke or myocardial infarction 1 Affecting 2-5% of the population and 20% of diabetics, this disease is commonly recognized, but poorly understood. It usually begins in the sixth decade of life and affects more women than men 2 This idiopathic disease is typically self-limiting, lasting 1-3 years, and the non-dominant shoulder is most commonly afflicted. Less than 20% of patients incur similar disease in the opposite shoulder within five year s of the resolution of the originally affected shoulder 2 In this disease, synovitis, dense adhesions and capsular constriction are found intra-articularly, causing severe restricti on of motion. The diagnosis for IAC is mainly clinical, and shoulders are examined thr ough radiographs and scapular rotation. Normal scapular rotation is 90 degrees; however, the scapular rotation of an affected individual is at 60 degrees with active abduction of the shoulder 3 IAC is associated with significant morbidit y; the shoulder becomes so stiff that everyday movements, such as raising the arm or rotating the humerus, ar e difficult to perform. Patients also suffer considerable loss of productivity due to pai n and the inability to sleep, and are often unable to adequately perform at work. After 20-30 months, the fibr osis will thaw and 16

PAGE 17

many patients will have minimal pain and will have regained most, but not all, of the use of the affected shoulder 4 However, some individuals, us ually diabetic patients, never work their way out of the painful and sti ff stages and remain permanently disabled. There is some disagreement whether t he underlying pathologic process is an inflammatory condition or one of fibrosis. Pa tient histories are consistent with the view that IAC arises initially from a persistent synovial inflammation which, in turn, chronically stimulates tissue repair pathways whose d ysregulation leads to the development of capsular fibrosis. Stages of IAC Based on clinical and arthroscopic exam ination, and histol ogic appearance of capsule biopsy specimens, adhesive capsulitis can be divided into four stages. These do not represent discrete, well-defined steps, but rather a continuum of disease progression ( Table 1-1 ). The following description is based on that provided by Hannafin et al 1 In Stage 1 symptoms have typically been present for less than 3 months. Patients present with an aching pain in the shoulder which becomes sharp at the extreme ranges of motion. Pain is usually associated with a loss of internal rotation, forward flexion and adduction, as well as a more subtle loss of external rotation in adduction. Arthroscopic and histologic examinations of the joint tissues demonstrate the presence of a hypertrophic vascular synovitis coveri ng the entire capsular lining which itself appears normal. Under anesthesia the range of motion (ROM) is nearly normal, indicating that at this early stage, the loss in ROM is prim arily attributable to painful synovitis rather than actual capsular contraction. In Stage 2 symptoms have been present for 3 to 9 months, with progressive loss of ROM and persistence of the pain 17

PAGE 18

pattern described above. Manipulation of the joint under anesthesia reveals that the loss of motion is increasingly attributable to loss of capsular volume. Examination of the joint tissues reveals a hypervascular synovitis with capsular fibroplasia and deposition of disorganized collagen fibrils, and a hypercellu ar appearance. Typically an inflammatory infiltrate is not present wit h either Stage 1 or 2. In Stage 3 symptoms have been present for 10-20 months or greater, and have changed noticeably with time. Patients present with a history of painful stiffening of the shoulder and a significant loss of ROM. Often reported is an extremely painful phase that has resolved, resulting in a relatively pain-free but very stiff shoul der. Increased ROM is not achieved under anesthesia reflecting a persistent loss of capsular volume, and fibrosis of the glenohumeral joint capsule. Examination of t he joint tissues reveals a syn ovial thickening and a dense hypercellular collagenous tissue. Stage 4 is marked by a thawing of IAC, and is characterized by the slow, steady recovery of ROM resulting from capsular remodeling. Little histologic information is available r egarding these patients as they rarely have surgery or capsular biopsy. Treatments of IAC are Controversial Despite its prevalence and debilitating effects, frozen shoulder has received little research attention because it is generally se lf-limiting, often disappearing within 2-3 years without specific medical or physical action. Since the underlying causes of IAC remain unknown, no specific treatment has b een shown to cure IAC or have a long-term advantage. Treatments are aimed at reduci ng pain and maintaining shoulder mobility; therefore, all stages of IAC require physical therapy 5-7 to enhance/preserve ROM and relax the muscles. NSAIDS 8, 9 have been shown to reduce inflammation and relieve pain, and oral corticosteroids 10 as well as intra-articular injections 6, 11-13 aid in physical 18

PAGE 19

therapy and also pain relief. For patients in more advanced stages of IAC, surgical intervention, particularly manipulations under anesthesia to break up the adhesions, and post-surgical rehabilitations, are the most co mmon treatments. In addition, arthroscopic release is recommended for patients who have not improved after several months of physical therapy. To date, there have been no studies in which patients were assigned randomly to treatment groups and followed fr om diagnosis, through treatment, and for a significant time downstream. Tissue Repair and Fibrosis Little is known of the etiology or the biol ogy that drives the pathologic fibrosis of IAC; however, tissue fibrosis is a prominent feature of progressive disease in several other organs such as the skin, liver, lung and kidney. Important clues to the pathogenesis of IAC are likely to come from par allels with these conditions. In many fibrotic diseases, pathology is thought to ar ise from tissue repair responses to chronic injury such as from alcohol abuse and viral hepatitis in the liver. Fibrogenesis is a normal part of wound hea ling and occurs in response to tissue injury. Many factors that mediate normal ti ssue repair also participate in pathologic fibrosis. Following injury, there is an ear ly inflammatory step characterized by hemorrhage of the vasculature and coagulati on of the blood. During this phase, platelets degranulate and rel ease numerous growth factors into the local environment that help to attract inflamma tory cells, particularly mono cytes and granulocytes, to the site of injury. These infiltrating cells rel ease protein factors that initiate the repair process, including platelet derived growth fact or, insulin-like growth factor, endothelins, angiotensin I, transforming grow th factor beta 1, among others 14 Elevated levels of these cytokines stimulate the recruitment, ac tivation and proliferation of fibroblasts and 19

PAGE 20

fibroblast-like cells as well as the synthesis by these cells of extracellular matrix (ECM) components, which serve to replace the prov isional fibrin matrix of the coagulated blood 15 Fibroblasts present early in the repa ir process appear immature; however, as the repair proceeds, a number of the fibroblas tic cells acquire features of smooth muscle cells and are termed myofibroblasts. T hese cells contain cytoplasmic bundles of microfilaments consisting of alpha smooth muscle-actin ( SMA) which, with mechanisms similar to smooth muscl e, enable cellular contraction. During the granulation phase of healing, the contractile properties of myofibroblasts are responsible for facilitating wound closure 16 Adhesions from the myofibroblasts in the ECM allow them to pull on the matrix and transmit force to neighboring cells. This retraction is stabilized by new deposition of matrix components such as collagens I and III resulting in overall tissue shortening. A direct correlation exists between the level of SMA in fibrotic tissue and its contractility. Myofibroblasts also express other SM cyto skeletal proteins including ca ldesmon, desmin, and SMmyosin heavy chains, but their roles in these cells are not fully understood. In the resolution phase of healing, there is a reduction in the overall cellularity of the repair tissue, particularly of myofibrobl asts and inflammatory cells. Much of this reduction is attributable to apoptosis; however, a portion of the myofibroblastic cells may undergo a reversion to the normal fibroblasti c phenotype. The fibroblasts that remain after the wound is closed possess a quie scent, noncontractile phenotype lacking the SMA microfilament bundles. The overall phenotype of the repair tissue changes from a synthetic/contractile mode with reduced matrix metall oproteinases (MMPs) and increa sed tissue inhibitors of 20

PAGE 21

metalloproteinases (TIMPS) to one of re modeling, with increased MMPs and reduced TIMP production. The signal that triggers th e loss of myofibroblastic cells is unknown but may be related to reduced levels of gr owth factors in the resolving wound or changes in the balance of MMPs and TIMPs 16 Pathologic fibrosis may similarly begin with a tissue insult and initiation of repair mechanisms. However, as the process lose s its inflammatory component, there is no accompanying reduction in myofibroblastic cells and thus the contractile fibrotic state persists. It is thought that mechanisms t hat continually stim ulate myofibroblast differentiation or conversely those that s pecifically inhibit apoptosis or phenotypic reversion in these cells are responsible for their persistence. TGF-beta 1 and Fibrogenesis Several cytokines have been found to be centra l to the process of fibrogenesis, in other tissues including transfo rming growth factor-beta 1(TGF), connective tissue growth factor (CCN2/CTGF), platelet deriv ed growth factor (PDGF) and fibroblast growth factor-2 (FGF-2, or basic FGF), among others 14, 17-20 While the underlying causes are diverse, it is likely that many of the harmful aspects of fibrosis are mediated by t he effects of Transforming Growth Factorbeta 1 (TGF1) 17 TGF1 is a secreted homodimeric protein that participates in a broad array of biological activities, such as norma l development, wound repair and regulation of inflammation. It directs numerous cellula r responses, including proliferation, differentiation, migration and apoptosis 21 It also is a potent inducer of ECM protein synthesis. TGF1 is the prototypic member of the TGFsuperfamily which consists of over 35 structurally related pleiot ropic cytokines that includes TGF2, TGF3, the 21

PAGE 22

Bone Morphogenetic Proteins, the Activins and Nodals. Members of the TGFsuperfamily have crucial roles in developm ent and tissue homeostasis. Perturbation of their signaling has been implicated in numerous developmental disorders and in a variety of human diseases including cancer, autoimmune disease, and fibrosis as discussed here 22, 23 Most of TGFis secreted as homodimer comp lexed with latency associated peptide (LAP) and is stored in the ECM. Activation of TGFinvolves a complex process 24 involving the cleavage of LAP by various proteases, or the physical interaction of LAP with proteins, such as thrombospondin-1. Cell signaling by TGFfamily members, diagrammed in Figure 1-1 occurs through type I and type II serine/threonine kinase receptors (TGFRI and TGFRII, respectively). Upon binding of its TGFligands, TGFRII recruits TGFRI into an activated heterotetrameric receptor complex. TGFRII phosphorylates TGFRI leading to subsequent phosphorylation of its intracellular effector molecules, receptor-regulated Smad proteins 2 and 3. Following phosphorylation, Smads -2 and -3 bind a co-Smad (Smad-4) and translocate to the nucleus, where they interact with transcription factors and coactivators to stimulate tran scription of responsive target genes 24-26 Inhibitory Smads (I-Smads: Smad-6 and Smad-7) act in an opposing manner to receptor-regulated Smads to regulate TGFactivity. They recruit ubiqui tin ligases (Smurfs) to the activated type I receptor, which sequentially leads to receptor ubiquitination, degradation and the termination of signaling. On a molecular level, TGFis known to regulate several genes involved in fibrogenesis including connective ti ssue growth factor (CCN2/CTGF) 27 plasminogen 22

PAGE 23

activator inhibitor-1 (PAI-1) 28 JunB 29 c-Jun 29 Smad-7 30 platelet-derived growth factor -chain 31 and integrin 5 32 Moreover, several ECM encoding genes are direct Smad targets, including collagen types I, III, V, VI, VII and X, fibronectin and proteoglycans 3335 ECM production by TGFis thought to be enhanced by it s inhibitory effect on MMP synthesis and stimulation of increased production of protease i nhibitors such as TIMP-1 and PAI-1 35 TGFmRNA and protein are increased in tiss ue biopsies of fibrotic lesions in most fibrotic conditions. Because TGFcontrols both the expression of components of the ECM network and the expression of protease inhibitors, it is a crucial regulator of ECM deposition, and a key growth factor in the development of tissue fibrosis. Indeed, type I collagen and ECM deposition are the unifying histopathologic hallm arks of fibrotic disorders, such as renal sclerosis, liver cirrhosis, keloid scars, and systemic sclerosis. TGFis also a chemoattractant fo r fibroblasts and myofibroblasts 36 It serves to maintain the fibrotic state by stimulating myofibroblast differentiation and suppressing apoptosis in these cells 37, 38 Analysis of TGF1 and CTGF Expression in IAC Patients Few studies have addressed the involvement of specific growth factors in IAC; however, Rodeo et. al in 1997, demons trated positive staining for TGFin synovial cells from tissues of patients with IAC, whereas none was detected in synovial cells derived from normal tissues 39 As TGF1 has been suggested to be a primary mediator of fibrotic pathology in tissues throughout the body, its positive detection in frozen shoulder tissues implicates its involvement in the fibrotic development of IAC. 23

PAGE 24

Additionally, our research colleagues, Drs. Wright and Schultz, initiated preliminary experiments to determine the contribution of CTGF, believed to be the downstream pro-fibrotic mediator of TGF1, in the pathophysiology of IAC ( Figure 1-2 ). For this, tissue biopsies from patients under going surgery for IAC were processed for immunohistochemical staining, using a m onoclonal antibody directed against human CTGF. As seen in Figure 1-2 A and B, CTGF was found at high levels in the fibroblastic cells of the synovium of IAC patients. Diffu se staining was also seen in the ECM of these tissues. Similar tissue samples obtained from patients without IAC did not exhibit detectable CTGF staining (Figure 1-2, C and D). To quantify CTGF protein levels, biopsy tissues were homogenized and analyzed for CT GF content by specific ELISA. In the tissue samples analyzed to date, CTGF was found to be ex pressed between 3 and 10 ng/mg protein in capsular tissues of IA C patients. Those fr om control patients without IAC showed less than 0.02ng/mg The potential for CTGF and TGF1 to induce pathologic fibrosis in the joints of rats will be examined in the present study. These studies implicate TGF1 and CTGF in the pathogenesis of IAC and their roles in arthrofibrosis will be further ex amined in the present study. Diabetes, TGF-beta1 and Fibrosis As previously mentioned, approximately 20% of diabetic patients not only suffer with IAC, but have a more persistent and severe case. It is possible that their inability to properly regulate blood glucose may be a cont ributing factor. Diabetes mellitus is a medical condition associated with abnormally high levels of glucose in the blood. Normally, blood glucose levels are controlled by insulin; however, destruction (type I) or dysfunction (type II) of beta cells, leads to diabetes. Type I diabetes, insulin-dependent 24

PAGE 25

diabetes, accounts for 10% of all cases of diabetes and results from the autoimmune destruction of insulin-producing beta cells in the pancreas by CD4+ and CD8+ T cells and macrophages infiltrating the islets 42 Afflicting over 150 mill ion people world wide, type II diabetes, non-insulindependent diabetes, NIDDM, is an incurable metabolic disorder characterized by insulin re sistance, decreased bet a-cell function, and hyperglycemia 43 Wound healing deficiencies in diabetic patients are common. Normal wound healing typically takes 3-14 days and occurs in three phases: inflammation, proliferation and remodeling. During the inflamma tory phase, macrophages and neutrophils phagocytize bacteria and other debris. During the proliferative phase, fibroblasts lay down a collagen matrix, and blood vessels move into the new granulation tissue. During the remodeling phase fibroblasts reorganize the collagen matrix and many transform into myofibroblasts involved in wound contraction. Diabetics have impaired wound healing, due to decreased growth factor production 44 macrophage function 45 collagen accumulation 46 fibroblast migration and pr oliferation, bone healing 47 and an imbalance between accumulation and remodeling of the ECM 48 among others. Significant research efforts have gone in to studying diabetes, its associated morbidities, and the effects of hyperglycemia. Much of this research has focused on examining the relationship between high glucose and the cytokines involved in fibrogenesis both in vivo and in vitro including TGF1. TGFis upregulated in diabetes and has been proven to mediate m any of the pathological changes in the diabetic kidney 63 TGFpromotes production of excess ECM in kidney cells and causes renal cell hypertrophy. Once activated TGFinduces mRNA and protein 25

PAGE 26

production of ECM molecules and impairs ECM degradation. A significant activation of the intracellular Smad pathway, which transduces the TGFsignal, was noted in the diabetic kidney, as well as upregulation of renal TGFtype II receptor 54, 64 High levels of ambient glucose have been shown to stimulate the production of extracellular matrix components, via TGF1 signaling 46 Both TGF1 and CCN2/CTGF have been shown to be elevated in plasma 49 secreted in the urine 50 and found in kidney tissues 49, 51, 52 of patients who suffer from diabetic nephropathy: a fibrotic condition which causes kidney func tion impairment, leading to end-stage renal disease. Research in both type I and type II diabetic animal models further demonstrates the involvement of TGF1 and CTGF in diabetes. Increased levels of both CTGF 49, 51, 52 and TGF1 53 mRNA and protein have been found in kidney tissues of experimental diabetic animal models, along with an activated Smad signaling pathway, transducing the TGFsignal 53, 54 Additionally, both cell culture and anima l studies have shown that a high glucose environment induces many of the signaling ca scades involved in cell proliferation and fibrogenesis. Decreased expression of MMP s, increased expression of TIMPs 55 an accumulation of fibronectin, laminin, and types I and IV collagen have all been shown to occur in high glucose environments when co mpared with cells grown in the presence of low glucose 56-58 In most renal cell types, treatment with a TGF1 antagonist, such as a neutralizing monoclonal antibody 57 or anitisense oligonucleotides 59 reverses the rise in ECM expression due to high gluc ose. This indicates that TGF1 is a mediator of the profibrotic effects of high glucose on the kidney 57 Similar studies have been performed in vivo that show treatment with neutralizin g monoclonal antibodies against TGF-1 26

PAGE 27

prevents mRNA increases of TGF-1, type IV collagen, and fibronectin in diabetic mice, 60 and anti-TGF1 antibody therapy prevented mes angial matrix expansion and renal insufficiency 61 Many features of the di abetic state stimulate TGFproduction. A selfmaintaining cycle has been proposed whereby an overexpression of Glut-1 leads to an increase in glucose uptake and activation of metabolic pathways that result in an excess of TGF1 production, which in turn maintains overexpression of Glut-1 62 Glut-1 is a ubiquitously expressed mole cule that resides on the cell plasma membrane and mediates the rate of glucose transport into t he cell. It is a high-affinity and low capacity transporter that is near saturati on at physiologic glucose leve ls. Therefore, an increase in basal glucose uptake would be the result of an increase in t he number of Glut-1 molecules. High glucose concentrations in mesangial cells increase Glut-1 expression via a TGF1-dependent mechanism 61 Gene Transfer to Capsular Fibroblasts Significant portions of the laboratory component of this dissertation involve the delivery of exogenous cDNAs and expression cassettes to fibroblastic cells of the synovium and sub-synovium in vitro and in vivo procedures with which our research group has considerable experience. Our efforts in this area were originally directed toward the development of an improved dr ug delivery system for the treatment of rheumatoid arthritis 73, 74 It was reasoned that if exogenous cDNAs encoding antiarthritic proteins could be delivered to cells t hat line the joint capsule, these cells could be adapted to become factories for the su stained, localized production of the therapeutic gene products. Indeed, we and others have exhaustively demonstrated the 27

PAGE 28

proof-of-concept, and have established its effe ctiveness in several experimental models of arthritis and two clinical trials. The first embodiment of intraarticular gene transfer involved an ex vivo approach whereby fibroblastic cells from the synovium and capsular tissues were genetically modified in vitro then injected into the joint space where they engrafted and secreted the transgene products into the synovial fluid and surrounding tissues 75, 76 In subsequent studies our lab has found that seve ral viral vector systems provide efficient gene delivery following direct intra-articular injection, enabling genetic modification of a high percentage of fibroblastic cells within the synovium and joint capsule. Following injection of a recombinant adenovirus contai ning a marker gene (cell associated alkaline phosphatase) into the knee joint of a rabbit, the virus is capable of transducing cells several layers deep within the lining 74 We have found this technology to be extremely useful for evaluating the effe cts of putative anti-arthritic protein products in diseased joints. We have also used it to constitutively overexpress gene products associated with the pathogenesis of disease as a means to study the effects of their sustained presentation on the biology of the articular environment 77 Although we found that we can generate exc eptionally high, biologically relevant levels of transgenic expression following intra-articular gene transfer, the window of expression was brief, persisting only for several days. We have since dedicated a considerable research effort into understanding the biology of the join t tissues as targets for gene delivery, and the factors that limit transgenic persis tence. We have found that the joint tissues are highly immune sensitive to the expression of foreign antigens, both non-homologous transgene products as well as vi ral proteins. Thus, in normal animals, 28

PAGE 29

transduced cells that expre ss non-native gene produc ts are rapidly e liminated by T-cell mediated cytotoxicity. In recent work we have found that in the joints of immunocompromised animals, such as in at hymic nude rats, which lack the T-cell arm of the immune system, expression of fo reign transgene products will persist for the lifetime of the animals 78 These animals thus provide an opportunity to examine the effects of long-term expression of foreign gene products on the joint tissues. Within this system, both integrating lentiv iral vectors which encode no viral proteins, and first generation adenoviral vectors whose genome remains episomal, are similarly capable of sustaining transgenic expression intraarticularly in the nude rat. In the work described herein, these vector systems will be used for gene transfer of TGF1, and CCN2/CTGF to capsular fibroblast cultures, and intra-articularly in the knees of nude and Wistar rats. Significance Although adhesive capsulitis is a relatively common disease which causes prolonged pain and morbidity, very little is known about its underlying causes or pathogenesis. This study was designed to initiate a focused research effort to examine IAC from a biological perspective. By undertaking this project we hoped to draw clinical relevance to our laboratory investigations while extending discove ries regarding the cellular and molecular basis of IAC into t he formulation of novel therapeutic strategies for clinical application. Our laboratory hypothesis is that TGF1 is a primary cyto kine mediator of pathology in IAC. Our investigatory strat egy will be comprehensive in scope, enabling a fundamental characterization of the pathobiol ogy of this disease. We will develop an 29

PAGE 30

animal model of arthrofibrosis by lo cal delivery and overexpression of TGF1 cDNA in the knee joints of rats and characterize the molecular and cellular events leading to joint fibrosis. We will also exami ne the relationship between diabetes and fibrotic signaling in cell culture of synovial cells and in our model of IAC. Moreover, by further testing our hypothesis in an animal model of articular fi brosis we will draw parallels between the effects of intra-articular cytokine signaling and clinical pathology. It is intended that at the completion of this work we will have a greater understanding of IAC and be wellpositioned to develop effective strat egies for therapeutic intervention. 30

PAGE 31

Table 1-1. Stages of Idiopathic Adhesive Capsulitis Stage 1 Duration of symptoms: 0 to 3 months Pain with progressive loss of motion Hypertrophic vascular synovitis Rare inflammatory infiltrate Normal underlying capsule Stage 2 Duration of symptoms: 3 to 9 months Persistence of pain with loss of range of motion Loss of capsular volume with hypervascular synovitis No inflammatory infiltrate Stage 3 Duration of symptoms: >10 to 20 months Minimal pain with stiff shoulder Residual synovial thickening without hypervascularity Dense fibrotic capsule Stage 4 Duration of symptoms: 15 to 30 months Minimal pain Slow, steady recovery of range of motion 31

PAGE 32

Figure 1-1. TGFSmad signaling pathway. TGFbinds first to the constitutively active TGFtype II receptor (T RII) and then the ligand-receptor complex associates with the type I receptor (T R1). T RII phosphorylates T RI on specific serine and threonine residues in the GS domain. Activated T RI propagates the signal downstream by directly phosphorylating Smad2 and Smad3 that form heterotrimeric or dimeric complexes with Smad4 and translocate into the nucleus. Interaction of these complexes with transcription factors (TF) regulat es transcription. Adapted from Dijke et al., Trends in Biochemical Sciences, 2004. 32

PAGE 33

Figure 1-2. Immunohistochemistry analysis for CT GF expression in tissue from patients with adhesive capsulitis (A,B) or normal joints (C,D) visualized under bright field (A,C) and fluorescence (B,D) microscopy. Positive CT GF expression can be detected in tissue from patients with adhesive capsulitis (B) and not in normal joints (D). 33

PAGE 34

CHAPTER 2 MATERIALS AND METHODS Vector Production The adenoviral vectors (Ad.GFP, Ad.TGF1, Ad.CTGF) used in this study originated from replication-deficient type 5 adenovirus lacking E1 and E3 loci. The respective cDNAs were inserted in place of the E1 region, with expression driven by the human cytomegalovirus ear ly promoter/enhancer 79 High-titer suspensions of recombinant adenovirus were prepared by amplif ication in 293 cells, and purified using three consecutive CsCl gradients as previously described 80 Titers were determined by optical density at 260nm. Vesicular stomatitis virus G-protein ( VSV-G) pseudotyped lentiv iral vectors were produced by transient transfection, using Lipo fectamine (Invitrogen, Carlsbad, CA), of 293FT cells with the Virapower TM packaging plasmids, pLP1, pLP2 and pLP/VSVG, containing gag-pol, Rev, and VSV-G protein envelope (Invitrogen, Carlsbad, CA) with expression driven by the human cytomegalov irus early promoter/enhancer. The tansducing vector (pCDH1-GFP IRES NEO) wa s generated by insertion of the cDNA for GFP into the multiple cloning site on t he pCDH1-vector. At 48 and 72 hours, the conditioned media were harvested, filtered th rough a 0.45 m filter (Steri-cup; Millipore, Billerica, MA), and centrif uged at 20,000 rpm in a swinging bucket rotor for two hours. The viral pellets were resuspended in Opti -Mem (Invitrogen, Carlsbad, CA) and stored in -80C. Tissue Culture Early passage primary rat synovial fibroblas ts (RSF), obtained from normal male Wistar rats, were cultured at 37C in 5% CO 2 atmosphere in Dulbeccos Modified 34

PAGE 35

Medium (DMEM) (Invitrogen, Carlsbad, CA) supplemented with 10% fetal bovine serum and 1% penicillin/ streptomycin (Invitrogen, Carlsbad, CA). ASO Transfection Briefly, RSFs were grown to confluenc e and treated with either 100mM or 300mM of antisense oligonucleotide (ASO; ISIS Pharmeceutic als, Carlsbad, CA) (TGFASO, ISIS #105204; Scramble ASO, ISIS #110410; CTGF ASO, ISIS #124212) and 10 g/ml of Lipofectamine (Invitrogen, Carlsbad, CA) for 1 hour. Medium was added to each well and the cells were grown for 24 hrs. Cells were washed and grown in serum free medium for an additional 24 hrs before har vesting medium for specific ELISA. TGF1 ELISA RSFs were seeded into flat-bottom twelve well plates in complete medium containing 10% FBS. Upon confluency, cells were transduced with 4.0 x 10 vp 2.0 x 10 vp of Ad.TGF1. After 24 hrs of incubation, m edium was replaced with serum-free medium. Supernatants were co llected at 48 hrs and TGF1 levels were measured using a solid-phase ELISA with TGF1 ELISA kits for humans (R&D Systems) according to the manufacturers instructions. Four replicate wells were used to obtain all data points and the mean of all samples were calculated. 8 9 Cell Migration Assay Fibroblasts were seeded in 6-well dis hes with conditioned medium until cells reached visual confluence. To block prolif eration cells were treated with 1 ml of 5 g/ml of mitomycin C (Sigma, St. Louis, MO), in Optimem (Invitrogen, Carlsbad, CA) for one hour. After the cells were washed three times in phosphate buffered saline (PBS), the bottom of each well was vertically scratched us ing a one ml pipette tip. The cells were then treated with either 2 ml of medium supplemented with Dglucose (Sigma, St. Louis, 35

PAGE 36

MO) with or without varying amounts of either Ad.TGF1, Ad.CTGF, or Ad.GFP. For this, cells were placed in a minimal amount of media for one hour with the appropriate virus before supplementing with an additional 2 ml of media/well. Digital images were then taken at 24 and 48 hrs post mitomycin C treatment. Analysis was performed using computer software (GraphPad Prism; GraphPad Software Inc.) at individual time points using a one-way analysis of variance (ANOVA). Animal Experimentation Experiments were carried out on 6-7 week old male athymic nude rats and male Wistar rats weighing 150170g (Charles River Laboratorie s, Wilmington, MA) housed two per cage with free access to standard laboratory food and water. All animal procedures were approved by the Instituti onal Animal Care and Use Committee of the University of Florida. Gene delivery vectors were injected into the joint space of the knee through the infrapatellar ligam ent. At periodic intervals afte r intra-articular injection, animals were killed by CO 2 asphyxiation followed by thoracic puncture. The joint tissues were then harvested for analysis. For Anti-sense oligonucleotides (ASO) Experiments ASOs were co-injected into the kn ee joint at a concentration of 10 g/ l with Ad.GFP, Ad.CTGF, Ad.TGF1 or PBS. ASOs were a generous gift provided by Nick Dean of Excaliard Pharmaceuticals (C arlsbad, CA) and ISIS Pharmaceuticals (Carlsbad, CA). ISIS TGF1=105204; ISIS Scramble=110410; ISIS CTGF=124212. Induction of Diabetes in Animals Adult Wistar rats weighing 150-170g (Char les River Laboratories, Wilmington, MA) housed two per cage were used for inducing diabetes. The animals were fasted 4 hours 36

PAGE 37

before diabetes was induced usin g streptozotocin (STZ) (Sigma, St. Louis, MO.). Rats received two intraperitoneal injections of 45mg/kg of STZ, 72 hours apart. STZ was freshly prepared and dissolved in 0.05M citrat e buffer, pH 4.5 and injected within 15 minutes of preparation. Blood glucose m easurements were per formed on a weekly basis using blood obtained from the tails of non-fasting rats using OneTouch Ultra (LifeScan Inc, CA). Rats whose blood glucose levels were above 250 mg/dl were considered diabetic. Animals were killed periodically and tissues were harvested. Those animals who were diabetic for longer than 2 months received a single subcutaneous implantation of Linplant (Lin Shin Canada Inc, Canada) and every two months after as needed to prevent ketoacidosis. Preparation of Total RNA Total RNA was isolated from treated and control synovial and capsular tissues using the RNeasy mini kit (Qiagen, Valencia, CA). Briefly, tissues were harvested and stored in RNALater (Qiagen, Valencia, CA) until RNA extraction was performed, at which time the tissues were frozen in liqui d nitrogen and pulverized using a mortar and pestle. The pulverized tissue was added to l ysis Buffer RLT (Qiagen, Valencia, CA), homogenized using a 20-gauge needle, and RNA was harvested using RNeasy spin columns following manufacturers pr otocol (Qiagen, Valencia, CA). PCR Array The Extracellular Matrix and Adhesion Molecules PCR Array for rat (SABiosciences, Frederick, MD) was used to examine the expression of over 80 related genes. One g of RNA was DNase treated and reverse-transcribed using the RT 2 First Strand Kit following manufacturers protocol (SABiosciences, Frederick, MD). The resulting cDNA template was mixed with SYBR Green PCR master mix 37

PAGE 38

(SABiosciences, Frederick, MD) and 25l of t he mixture was equally aliquoted into each well of the 96-well plate already containing individual PCR primer sets. Differential analysis was performed using software prov ided by SABiosciences (SABiosciences, Frederick, MD). Briefly, the CT method was used for data analysis to determine foldincrease and decrease in ex pression between treated and cont rol tissues. The students t test was used to determine statistical signif icance. Each time poi nt represents n=3. Immunohistochemistry 5 m sections of formalin-fixed, decalc ified, paraffin-embedded blocks were cut and mounted on plus charged slides (Fisher Sc ientific, Pittsburgh, PA). Slides were deparaffinized and rehydrated through a series of xylenes and graded alcohols and blocked in 3% peroxide/methanol for 10 minutes at room temperature. Appropriate sections were stained with hemat oxylin and eosin (H&E) or to luidine blue. If required, heat mediated antigen retrieval was perform ed in Dako Target Retrieval Solution (DakoCytomation, Carpinteria, CA) for 20 minutes at 95C. Nonspecific binding was blocked in 15% normal serum matched to the secondary antibody species. Slides were incubated overnight at 4C with commercia lly available antibodies: mouse antiSMA at 1:1000 dilution (Sigma, St. Louis, MO) (no retrieval), goat anti-MMP-9 at 1:50 (Santa Cruz Biotechnology, Santa Cruz, CA), mouse anti-MMP-13 at 1:15 (Santa Cruz Biotechnology, Santa Cruz, CA). The appropriate biotin ylated secondary antibody (Vector Labs, Burlingame, CA) was applied to samples for 30 minutes at room temperature at a dilution of 1:200, followed by detection with an avidin-biotin-based peroxidase kit (ABC Elite; Vector Laboratories, Burlingame, CA). The antigen-antibody 38

PAGE 39

complex was observed by reaction with 3,3' -diaminobenzidine (DAB) and slides were counterstained with hematoxy lin and coverslipped. Detection of GFP 5 m sections collected from decalcified, paraffin embedded blocks were manually immunostained using rabbit anti-GFP (dilution 1:1000, ab290, Abcam; Cambridge, MA). Slides were heat retrieved in 10mM citra bu ffer (BioGenex, San Ramon, CA), pH 6.0 before blocking with normal serum and staining ov ernight at 4C. Positive signal was detected with anti-rabbit Alexafluor 488 (dilu tion 1:500; Molecular Probes, Eugene, OR). Controls consisting of isotype and concentration matched immunoglobulin were included for each section. Detection of Reporter ASO Intraarticular injection of 10 g/ l of reporter ASO (generous ly provided by Dr. Schultz and Dr. Dean) was performed. Tissues were harvested 10 days post injection, and paraffin embedded. 5 m paraffin sections from joints were stained for the reporter ASO using the following twostep horseradish peroxide (HRP) immunohistochemistry technique. The slides were pretreated with DAKO Blocking solution (Carpenteria, CA, USA), and DAKO Proteinase K solution followed by incubation with the primary antibody 2E1-B5 (Berkeley Antibody Company, Berkeley, CA, USA) which recognizes CG or TCG motif in phosphorothiate ASO. Secondar y antibody incubation was with Zymed Anti-IgG1 isospecific HRP conjugated secondary (San Francisco, CA, USA). Positive staining was detected with DAB as t he chromogen and revealing agent (DAKO Carpenteria, CA, USA). Sections were c ounterstained with haematoxylin and then photographed to document cellular localiz ation of reporter ASO expression. 39

PAGE 40

Microarray Sample Preparation Human tissue samples from patients wit h IAC and from control human shoulder were obtained post-surgery from Dr. Thomas Wright, University of Florida, and stored at -80 degrees in RNALater (Company, city). Fo r RNA extraction, the tissues were frozen in liquid nitrogen and pulverized using a mortar and pestle. Pulverized tissue was added to lysis Buffer RLT (Qiagen, Valenc ia, CA), homogenized using a 20-gauge needle, and RNA was harvested fo llowing manufacturers protocol using the RNAeasy Mini kit (Qiagen). RNA concentration wa s determined using the NanoDrop ND-1000 spectrophotometer (NanoDrop Technologies, Wilmington, DE). RNA integrity and quality were estimated on an Agilent 2100 Bioanal yzer (Agilent Technologies, Palo Alto, CA). RNA integrity number (RIN) index wa s calculated for each sample using the Agilent 2100 Expert software. RIN provides a numerical assessment of the integrity of RNA and facilitates the standardization of inte rpretation of RNA quality. To reduce experimental bias in data analysis due to poor RNA quality, only RNAs with RIN number >7.0 were further processed. RNA Amplification, Labeling and Array Hybridization For RNA amplification and labeling, 500 ng of total RNA were linearly amplified and labeled with Cy3-dCTP fo llowing the Agilent One-Co lor Microarray-Based Gene Expression Analysis protocol. To monitor the microarray analysis workflow a mixture of 10 different viral polyadenylated RNAs (Agil ent Spike-In Mix) was added to each RNA sample before amplificati on and labeling. Labeled cRNA was purified with the Qiagen RNAeasy Mini Kit (Qiagen, Valencia, CA), and sample concentration and specific activity (pmol Cy3/g cRNA) were measured in a NanoDrop ND-1000 40

PAGE 41

spectrophotometer. Hybridization was performed by the ICBR Core at the University of Florida. 41

PAGE 42

CHAPTER 3 GENERATION AND CHARACTERIZAT ION OF ADENOVIRAL VECTOR In vitro Expression of Adenoviral Vector To study and evaluate the role of CTGF in arthrofibrosis, we used adenovirus as a mechanism for gene delivery. Adenoviruses in fect both resting and dividing cells of many types and highly purified solutions of virus can easily be produced with high titres. Both the Ad.TGF1 and Ad.GFP vectors used in these studies were generated and characterized previously by our research group. The Ad.CTGF vector was generated for this project. For this, both the rat and human CTGF cDNAs were subcloned into the AdLox plasmid construct and packaged into adenovirus using Cre recombinase method described by Hardy et.al Briefly, 293 cells constitutively expressing bacteriophage Cre recombinase were co-transfected with re combinant AdLox vector plasmid carrying the CTGF cDNA and with 5 helper virus DNA. Following the formation of cytopathic effects, the cell lysates were used to infe ct cultures of 293 cells to amplify the recombinants. Vector particl es were purified using CsCl density gradient purification. 79 To determine the functionality of this vector, characterization of the adenovirus containing CTGF began with we stern blot analysis. Fibroblasts in culture were transduced with either Ad.CTGF or Ad.GFP at varying concentrations and cell lysates and conditioned media were run on 15% pre-cast SDS-polyacrylamide gels (BioRad). For detection of CTGF protein, a biotin anti-Human CTGF C-terminus antibody (R&D BAF 660) was used. It is known that the human CTGF protein resolves at multiple sizes including the full length peptide at 38KD, and partial proteolytic fragments between 1020KD that represent the degradation of the full length peptide by intracellular proteases within the ER or Golgi 82 As indicated by the arrow in Figure 3-1 the presence of a 38 42

PAGE 43

KD band is clearly visible in both Ad.hCTGF and Ad.rCTGF and absent in our control, confirming the presence of our proteins Additionally, an approximately 20KD band from the human CTGF protein is detectible, which is consist ent with the literature. The band at 45KD is most likely due to non-specific bi nding as it is seen in both our negative control and in our Ad.hCTGF and Ad.rCTGF lanes. Western blot of the supernatant and cell lysate revealed the same pattern of bands; however, a higher concentration of protein was found in the lysate (not shown). To further characterize this virus, a specific ELISA for human CTGF, using antibodies generously provided by Dr. Schultz was performed. For this, rabbit HIG-82 cells were transduced with equal amounts of Ad.h CTGF, Ad.rCTGF, or Ad.GFP control. As seen in Figure 3-2 Ad.hCTGF expression was confirm ed by specific ELISA, further confirming the functionality of our virus. Additional tests will need to be performed to verify the expression of the Ad.rCTGF virus. Lastly, a cell proliferation assay, us ing Ad.hCTGF was performed by Edith Sampson in Dr. Schultzs lab, using the CellTiter 96 AQ ueous One Solution Cell Proliferation Assay from Prom ega. For this, human corneal fibroblast (HCF) cells were seeded in a 96 well plate, gr own to 70-80% confluency and serum starved for 48 hours, after which time, virus was serially diluted in each well and assayed 48hours post infection. As shown in Figure 3-3 as increasing amounts of virus was added to cells, an increase in fibroblast proliferation occurr ed, which is characteristic of the CTGF gene. From these data the Ad.hCTGF, descri bed from here on as Ad.CTGF, appears to be functional and will be examined in vivo 43

PAGE 44

Overexpression of TGF1 In vivo Induces Fibrosis After characterizing and confirming the functionality of the Ad.CTGF vector in vitro we further characterized its activities in vivo to determine its role in fibrotic induction within the joint. Studies were carried out in the knee joints of Wistar rats to determine effects of CTGF, GFP, and TGF1 overexpression intra-articu larly. For this, 3.6 x 10 8 vp (viral particles) of Ad.GFP, Ad.hCTGF, or Ad.hTGF1 were injected into both knee joints of Wistar rats, four rats per group. Afterward, th e rats were analyzed daily for weight, knee diameter and vigor. Results in Figure 3-4 represent the percentage change in knee diameter relative to day ze ro. Ad.GFP animals did not have an increase in knee diameter and fluctuated within 7% of the value at day zero. A mild increase in knee diameter, about 20% relative to day zero, was observed immediately after Ad.CTGF injection and gradually leveled off to near normal levels. The greatest effect was observed in those rats receiving Ad.TGF1. These rats experienced a ~37% increase one day after injection and continued to increase to ~45% relative to day zero, maintaining an elevated knee diameter throughout the study. At 4 and 8 days post injection, animals we re sacrificed and tissues were harvested and processed for histology. H&E histological staining 4 days post injection, as shown in Figure 3-5 reveals a mild inflammatory respons e in Ad.GFP infected animals, most likely due to the immunogenic effects of adenovirus. Animals injected with Ad.CTGF revealed a modest increase in fibroblast prolif eration, as seen in Figure 3-5, with the image representing the most dramatic changes seen within this group. Additionally, histology from 8 days post injection showed little to no synovial fibroblast proliferation indicating the effects of Ad.CTG F were transient (not shown) Overall, while fibroblast proliferation seen in the Ad .CTGF injected animals was minimal and transient. The most 44

PAGE 45

striking response was observed in animals injected with Ad.TGF1. Histology from these animals demonstrated a severe hypercellu lar response, primarily of fibroblastic cells, resulting in synovial joint thickening. This synovial fibrotic expansion could be seen 8 days post injection (not shown). Fr om this experiment, it appears that while CTGF may play a role in other fibrotic c onditions or act as an important co-factor in fibrogenisis, its effects within the joint space are minimal and transient, and it does not appear to be a primary induc er of joint fibrosis. On the contrary, TGF1 induced a severe fibrotic response similar to that des cribed in stage 3 IAC. T herefore, for further experiments to model ar throfibrosis, Ad.TGF1 was delivered to the knee joints of rats. 45

PAGE 46

Figure 3-1. Western blot analysis of CTGF pr otein. HIG-fibroblast cells in culture were infected with Ad.hCTGF, Ad.rCTGF, and Ad.GFP control. Western blot analysis revealed the full length 38kDa protein in both Ad.rCTGF and Ad.hCTGF, as well as, a 20kDa proteolytically cleaved fragment in Ad.hCTGF and non-specific bands at 45kDa. 46

PAGE 47

Figure 3-2. hCTGF specific ELISA. Fibr oblasts in culture were infected with Ad.GFP control, Ad.rCTGF and Ad.hCTGF. An ELISA, using antibodies specific for hCTGF, confirmed its expression. 47

PAGE 48

Figure 3-3. Cell proliferation assay. Fibr oblasts in culture in a 96-well plate were infected with Ad.hCTGF and serially di luted among the wells. An MTS assay showed an increase in cell proliferation with increas ing amounts of virus. 48

PAGE 49

Figure 3-4. Percentage change in knee diameter relative to day zero of adenovirus injected Wistar rats. 49

PAGE 50

Figure 3-5. Histolgical H&E stain of Wistar rats knees, four days post intra-articular injection. Mild synovial inflamma tion could be seen following Ad.GFP injection, compared to normal control. Modest fibroblast proliferation was observed in Ad.CTGF injected animals. Ad.TGF1 animals displayed severe fibrotic expansion and proliferation. Images shown at 20x magnification. 50

PAGE 51

CHAPTER 4 GENE DELIVERY OF TGF-BETA1 INDUCES ARTHROFIBROSIS AND CHONDROMETAPLASIA OF SYNOVIUM IN VIVO Introduction Arthrofibrosis is a condition that arises from the development of excess fibrous tissue intra-articularly which leads to chr onic joint pain and loss of range of motion. It can occur in most joints and frequently onset s following injury, surgery, diabetes or immobilization; however, the precise etiology remains unclear 83, 84, 85, 86 A particularly common example is adhesive capsulitis of t he shoulder, also known as frozen shoulder syndrome. It is characterized by a painful fi brotic expansion of the synovium and joint capsule, which gradually results in the lo ss of active and passive motion of the joint. 87, 2 Although the disease is generally self-limiting, it can persi st for 2-3 years, leaving patients disabled with limited use of the affected arm 87, 1 Since the underlying causes remain unknown, no specific pharmacologic or non-surgical therapy has been shown to cure arthrofibrosis or provide significant long-term benefit. Pathologic fibrosis is a prominent feature of chronic disease in several organs, including the skin, liver, lung and kidney, and often begins with local injury and activation of normal repair mechanisms 88, 89 Following tissue damage, there is a need for local synthesis of reparative connective tissue, which involves the migration of fibroblasts to the wound site and their pro liferation. These cells then synthesize abundant levels of extracellular matrix prot eins, including collagens, proteoglycans and fibronectin. Many of these fi broblasts differentiate into myofibroblasts, which express high levels of -smooth muscle actin ( -SMA) that confers contractile activity to facilitate wound closure. In pathologic fibrosis, as the healing process loses its inflammatory component, there is no accompanying reduction in the myofibroblasts, as would occur 51

PAGE 52

during normal healing, and a contractile fibrot ic state persists. It is thought that mechanisms that continually stimulate myof ibroblast differentiation, or, conversely, those that specifica lly inhibit apoptosis or phenotypic reversion in these cells, are responsible for their persistence and that of the fibr otic condition 90 While the initiating causes of pathologic fibrosis are likely diverse, the protein factors that mediate wound healing and tissue repair are considered to play central roles. In this regard, transfo rming growth factor-beta 1 (TGF1) has been implicated as a participant in a majori ty of fibrotic conditions 17 This pleiotropic cytokine induces a broad array of biological activities, such as cellular proliferation, differentiation, regulation of inflammation and tumor progression 91, 92 TGF-1 is also a potent mediator of extracellular matrix (E CM) protein synthesis, and its expression is increased in numerous fibrotic conditions 21 Active TGF-1 binds to a heteromeric receptor complex consisting of TGFtype I and type II receptors, and signa ls intracellularly through transcriptional activators, Smads 2 and 3, as well as Smad independent pathways. Most of the profibrotic effects, fibroblast proliferation, myofibroblast di fferentiation, enhanced synthesis of matrix proteins and inhibition of collagen breakdown, are thought to be mediated through Smad signaling 93 Indeed, several ECM genes are di rect Smad targets, including collagen types I, III, V, VI, VII and X, and fibronectin 33-35 TGF-1 is also known to regulate expression of other proteins thought to drive fi brogenesis, including connective tissue growth factor (CCN2/CTGF), which is often considered to be a downstream mediator of TGF1-induced fibrosis 27 52

PAGE 53

Whereas other fibrotic conditions have re ceived considerable research attention, there are comparatively few reports regarding the pathogenesis of arthrofibrosis. This can be partially attributed to the scarcity of relevant animal models, as well as a lack of available human tissue from early stage dis ease for detailed analysis. Arthroscopy, due to its invasiveness, is not a diagnostic tool and is only used to remove fibrotic adhesions from late-stage immobilized joints that have pr oven resistant to other forms of therapy. Few studies have addressed the in volvement of specific growth factors in arthrofibrosis; however, it has been shown using immunohist ochemical staining and ELISA, that increased levels of TGF1 are found in synovial and capsular tissues and diffusely in the fibrotic ECM 1, 39, 41 We hypothesized that sustained over-production of TGF1 intra-articularly drives chronic arthrofibrosis. To test this, we used a recombinant adenovirus to deliver and locally overexpress the cDNA for human TGF1 in the joints of athymic nude rats and examined the effects of chronic stimulation on the local biology of the capsular tissues. Using this approach, we found TGF1 gene transfer rapidly induced a severe and persistent fibrotic condition that encased and immobilized the injected joint. Histologic examination, as well as focu sed expression arrays of the joint tissues, suggest that the developing fibrotic tissue possesses many of the molecular features of an aggressive tumor. Important findings r egarding the proliferative and pl astic nature of resident capsular fibroblasts were also revealed. 53

PAGE 54

Results Intra-articular Delivery of Ad.TGF1 Induces a Dose-Dependent Fibrotic Response: Prior to undertaking in vivo experiments, we first characterized the Ad.TGF1 vector for transgene expression in cultures of sy novial fibroblasts isolated from the joints of rats. The cells were incubated with increa sing doses of virus, and 48 hours later, the levels of human TGF1 in the conditioned media were determined by specific ELISA. As shown in Figure 4-1 TGF1 was expressed in a dose-dependent manner and achieved a maximum of ~1 g/ml at 2.0 x 10 9 viral particles (vp). Vector doses exceeding this were found to be toxic to the cells. After confirming the ability of the Ad.TGF1 construct to deliver and express the transgene in the target cells, we performed preliminary studies in vivo to determine a vector dose that would provide a signifi cant biological response without being detrimental to the health of the animals when expressed long term. We initially used Wistar rats, which are immunologically comp etent, but enable sustained expression of adenovirally delivered transgenes for about 7-10 days 94 For these experiments, groups of rats were injected in both knees with low (1.0 x 10 9 vp), medium (2.0 x 10 9 vp), or high (4.0 x 10 9 vp) doses of virus. Rats receiving 4.0 x 10 9 vp of Ad.GFP were used as controls for the pathological effects of vira l delivery. The experiment was scheduled for 7 days, at which time the animals would be sacrificed and the tissues harvested for analysis. Within three days, the knees receiving the Ad.TGF1 virus became visibly enlarged, increasing in size with the vector dose. At four days post injection the rats receiving the highest dose of Ad.TGF1 were sacrificed due to health concerns. They 54

PAGE 55

were lethargic, did not eat and showed labored respiration. Although the animals at the lower doses showed a prominent increase in joint diameter, they were otherwise healthy with normal appetite. Dissection of the joints from the animals killed at day 4, and those of the remaining groups sacrif iced at day 7, showed that the increase in joint size was due to the overgrowth of a dense scar-like tissue that encased the joints and effectively occluded joint motion. Sustained Overexpression of TGF1 in the Joints of Nude Rats Induces Severe Arthrofibrosis and Chondrometaplasia: After observing the initial fibrotic response to overexpression of Ad.TGF1 and identifying a working range of virus, we next wanted to determine the effects of sustained TGF1 over-expression for a prolonged per iod of time. We injected the medium dose (2.0 x 10 9 vp) of Ad.TGF1 into both knee joints of a group of nude rats. As a control for viral administration, we delivered a similar dose of Ad.GFP into a parallel group. Animals in each group were euthanized at 5, 10, and 30 days post injection and the joint tissues were harvested for analysis. As seen in Figure 4-2 by day 5 the knee joint diameter in the animals receiving Ad.TGF1 increased by about 40% relative to controls, and this increase was maintained throughout the cour se of the experiment. Following sacrifice and removal of the skin, gross inspection show ed that the knee joints were completely enveloped in fibrotic tissue such that the normal features of the joint were not discernable (similar to that shown in Figure 4-3C). Consistent with the pilot experiments, by day 5 the knee joints had become immobilized, and remained locked at 90 o flexion for the duration of the experiment. During dissection of the da y 5 animals, the fibrotic encasement was easily cut from the bone to expose the inta ct structures beneath. At each successive 55

PAGE 56

time point, however, the fibrotic tiss ue became progressively more dense and aggresssive, such that by day 30 it appeared to penetrate and fuse with the bony architecture of the knee joint, making it di fficult to distinguish pre-existing anatomy (Figure 4-3D). Histologic analysis of the harvested tiss ues showed that injection of Ad.GFP caused a slight increase in the number of leukocytes in the synovial intima at day 5, but otherwise the tissues appeared normal ( Figure 4-4 control). In stark contrast, injection of Ad.TGF1 stimulated a dramatic fi brotic response intra-articularly that, over time, changed and progressed into an aggressive fibrocartilaginous metaplasia. By day 5 there was an extensive pro liferation of reactive-appear ing stellate and spindled fibroblasts effacing much of the normal in traand periarticular connective tissues (Figure 4-4, day 5). The fibroblasts replaced the normal subsynovial adipose tissue, and radiated outward from the jo int space without altering the articular cartilage, which maintained its amphophilic appearance on tolu idine blue stains. The fibroblastic proliferation also extended into adjacent skeletal muscle, where it invaded between myocytes. There was little accompanyin g inflammatory component early and no metaplastic bone or cartilage was identified at 5 days. At 10 days, the fibroblastic proliferation had essentially replaced all of the normal anatomic structures in and surrounding the joint and began to fuse with articular cartilage, subperiosteal cortical bone, and ligamentous and capsular fibrous tissue (Figure 4-4, day 10). At these sites of fusion, the cells began to lose their once spindled morphology, instead appearing as rounded, chondrocyte-like cells. 56

PAGE 57

At thirty days, the normal microscopic anat omy of the joint was not recognizable, and the resident joint and capsular structures were displaced (Figure 4-4, day 30). The fibroblastic component was considerably less cellular and more densely fibrotic, and large areas were now composed of cartilagin ous tissue. The majority of the articular cartilage was replaced by metaplastic fibrocar tilaginous tissue, confirmed by the deeply amphophilic appearance of the tissue on toluidine blue stain. There was no clear delineation between the fibr ous component and the cartilaginous component. These cartilaginous masses penetrated the entire joint space leaving an abundant deposition of matrix enriched for proteoglycans. TGF1 Stimulates Expression of Genes for Matricellular Proteins, MMPs, Collagens and Adhesion Molecules In an attempt to determine the protein mediators responsible for driving the pathologic changes in the cells and tissues, RNA isolated from t he capsular/synovial tissues of control rats and those receiving Ad.TGF1 was analyzed by PCR-array for expression of 84 genes associat ed with cellular adhesion, extracellular matrix synthesis and remodeling. Differential analyses of the expression patterns, as seen in Table 4-1 strikingly illustrated the extent to which prolonged exposure to TGF1 activates the articular tissues. Numerous genes showed significant ly enhanced expression, many exceeding 100-fold, while only modest reductions in ex pression were observed sporadically in a handful of the genes. Analyzed over the course of the experiment, the most extensive and consistent changes in gene expression we re observed among the collagens, the matricellular proteins, and the proteolytic enzymes. 57

PAGE 58

Consistent with the early fi brotic expansion, collagen types I, III and V were elevated from 3 to 10 fold at days 5 and 10. As the hypercellu lar tissue began to differentiate and transition to a cartilagi nous phenotype at day 10, expression of collagen type II and hyaluronan and proteoglycan li nk protein 1 (HAPLN1/cartilage link protein) dramatically increased, with both genes showing >100 fold increase by day 30. Modest increases in expression of collagens VI and VIII were also ob served at different time points. Supporting the highly prolifer ative, expansive response of the capsular tissues, expression of the matricellu lar protein genes, as a group, was elevated at all time points. Those most profoundly induced were osteopontin (30-100 fold), followed by thrombospodins 1 and 2, tenascin-C and perios tin. Expression of other matricellular protein genes, such as CCN2/CTGF and secreted protein acidic and rich in cysteine (SPARC/osteonectin), was also modestly elevated throughout the study and only achieved statistical significance at certain time points. Characteristic of the expansive and invasi ve properties of the capsular tissues, expression of the MMPs was broadly induced at all time poi nts. Expression of MMP-12 (elastase) showed the greatest sustained induction ranging from ~100 to 200-fold over controls. This was followed by the colla genases, MMPs-13 and -8, which were also elevated throughout the experiment, but s howed >200-fold enhanced expression at days 10 and 30, respectively. MMPs-9 and -2 (the gelatinases), MMP-7 (matrilysin), MMP-11 (a member of the stromelysin family), and MMPs-14 and -16 (also termed membrane type; MT-MMPs-1 and -3 ) also showed significant increases in expression, ranging between ~3-80 fold induction thr oughout the course of the experiment. 58

PAGE 59

ADAMTS1 showed evidence of enhanced expressi on at each time point, but this was not statistically significant. A variety of cellular adhesion molecules also showed increased expression in the expanded capsular tissues. Of note, E-, Nand P-cadherins ( epithelial, neural, and placental, respectively) were each up-regul ated at day 5; however, only N-cadherin remained highly expressed for all 30 days. Interestingly, this was paralleled by heightened expression of neur al-cell adhesion molecule (NCAM). Both NCAM and Ncadherin are thought to coordinately inte ract during cellular condensation in chondrocytic differentiation as well as in tu mor cell proliferation. Several integrin receptor molecules showed increased expression during the ear ly stages of the experiment, including L and 3 subunits at day 5, and L M 2 and 3 at day 10. Increased expression of -catenin was also seen at day 10 and L-selectin showed ~16fold enhanced expression at day 30. Several other genes showed enhanced expression; the most notable being Emilin1 whose expression was increased >180-, 87and 36-fold at days 5, 10 and 30. Emilin1 is a connective tissue glycoprotein associ ated with elastic fibers, and is thought to contribute to cell motility, ti ssue differentiation and morphogenesis. 95 Certain laminins also showed increased expre ssion, particularly the laminin 1 chain at day 30 and laminin 1 at day 10. Fibronectin was also moderately enhanced at days 10 and 30. To solidify these data, we used immunohist ochemistry to examine the tissues for the presence of MMP-13 and MMP-9 at the protein level, as well as smooth muscle actin ( SMA), a myofibroblast marker ( Figure 4-5 ). Consistent with the expression patterns above, staining for MMP -13 was visible throughout t he fibrotic matrix at the 59

PAGE 60

later time points, with darker staining in the more cartilaginous tissues. Likewise, staining for MMP-9 was consistent with RNA analysis and appeared throughout the fibrotic synovium. The darkest staining wa s observed at day 10 at the borders between the newly expanded fibrotic tissues and ca rtilaginous regions. As expected, dark staining for SMA was detected at day 5 in the heavily fibrotic regions. 96 At days 10 and 30, as the cells appeared to differentiate toward a chondrocytic phenotype, SMA staining was significantly reduced. Resident Synovial/Capsular Fibroblasts Pr oliferate to form the Fibrotic Mass In fibrotic conditions of several tissues such as liver, lung and kidney, a level of uncertainty has surrounded the origin of the fibroblastic cells causing the pathology. In certain models, it remains unresolved as to whether the fibrosis arises from the local proliferation and differentiation of resident fibroblasts or from infiltrating progenitor cells originating from bone ma rrow, or elsewhere. To determine the origin of the cells re sponsible for generating the pathologic fibrotic and cartilaginous tissues in the joint, a lentiviral vector c ontaining the cDNA for GFP (LV-GFP), was injected into the knees of nude rats 48 hrs prior to injection of Ad.TGF1. We have shown previously that lentiv irus-based vectors, when delivered at sufficient titer, will transduce a significant proportion of the fibroblastic cells resident in the normal joint 97 Therefore, by using a lentivirus to deliver the cDNA for GFP we could stably mark the fibroblastic cells that were pre-existent in the capsular tissues; furthermore, since the lentiviral vector integrates its genetic payload into the genome of the transduced cell, any progeny that arise fr om the transduced cell will likewise contain the transgene and fluoresce green 78, 97 By pre-labeling the resident capsular cells in 60

PAGE 61

this manner, if the cells comprising the fi brotic mass arise from resident fibroblast populations, then a large percentage should fluoresce green. Alter natively, if the fibroblastic cells arise from circulating myofibroblasts or progenitor cells, then the bulk of the cell mass would be negative for GFP expression. Consistent with earlier expe riments, delivery of LV-GFP in normal control joints, generated a uniform layer of fluorescent cells across the entire expanse of the synovial lining and penetrated several cell layers deep ( Figure 4-6A ) 78, 97 In the joints of animals that subsequently received Ad.TGF1, examination at day 5 showed that the entire depth of the fibrotic synovial tissue was filled with elongated fluorescent fibroblastic cells. At day 10 the cells appeared more or ganized with a mixture of fluorescent elongated fibroblasts and more rounded chondro cytic cells (Figure 4-6C). By day 30, the cells in the cartilaginous tissues, as conf irmed by toluidine blue staining, were also predominantly GFP+ (Figure 4-6D). These re sults demonstrate that in arthrofibrosis, myofibroblasts arise from fibroblastic cells re sident in the connective tissues of the joint. These cells have a high proliferative capacit y and the ability to transdifferentiate into chondrocytic cells. Discussion Our investigations present a vivid demonstr ation of the stimul atory capacity of TGF1 and its potential as a pro-fibrotic cytokine in joint disease. Sustained overexpression of TGF1 in the knee joints of nude rats induced a severe fibrotic response arising from the rapid proliferation of synovial fibr oblasts, their differentiation into myofibroblasts and the synthesis of fibrillar collagenous matrix. Much like the adhesions associated with arthrofibrosis in humans, the fibrotic tissue progressively 61

PAGE 62

developed an aggressive phenotype and began to attach and fuse with the cartilaginous and bony surfaces. Fibrotic regions in cont act with articular cartilage began to undergo chondrometaplasia. Follo wing 30 days of TGF1 over-expression, sections of the expanded fibro-cartilaginous tissue had penetrated the articular cartilage and subchondral bone. Large portions of the hypertrophied exp anse had differentiated into hyaline-like cartilage, progressing in some regions toward osteogenesis. Although our data provide an extraordinarily severe representation of joint fibrosis, the pathologies observed are entirely consistent with those seen in clinical cases of arthrofibrosis that occur following joint surgery 98 Procedures involving the knee, such as anterior cruciate ligament re construction and high tibi al osteotomy, are particularly vulnerable to the dev elopment of this type of fibrosis 98 Histological examination of tissues recove red from fibrotic knee joints following surgical release frequently identifies fibrosis, vascular pr oliferation and synovia l chondrometaplasia 99 In many cases the fibro-chondrogenic ti ssues also contain endochondral bone formations 100-102 The similarities between the tissue phenotype of the rat TGF1 overexpression model shown here and human ar throfibrosis, indicate that the data generated in the rat knee have relevance to the human condition. As such, our results support the role of TGF 1 as a primary mediator of th e pathogenesis in arthrofibrosis and therefore, as a primary tar get in the prevention of fibrot ic conditions of the joint. Expression Profiling is Consistent with an Aggressive Fibrotic and Chondrometaplastic Phenotype The expression data from the PCR-arrays support the macroscopic and histologic findings, and provide insight into the molecular events through which TGF1 drives the fibrotic hypertrophy and cartilaginous morphogenes is of the tissues. Ov erall, the profile 62

PAGE 63

reflected in Table 4-1 is in accord with a highly mobile, aggressive tissue that produces large quantities of ECM protein. Similar to the histologic profile, the pattern of gene expression bears many similarities to that of an aggressive tumor. Consistent with the fibrotic phenotype obs erved at days 5 and 10, there was an accompanying increase in expression of vari ous laminins and ECM proteins, including interstitial collagens type I and III, and perice llular type V collagen. Further, as the tissue transitioned to hyaline-like cartilage, there was a dramat ic enhancement of expression of collagen type II, the predominan t structural protein of ar ticular cartilage. This was paralleled by an increase in transcription of cartilage link protein (HAPLN1), which serves to stabilize aggrecan and hyaluronan aggr egates in the articular cartilage matrix and contributes to chondrocyt e differentiation and maturation 103 Beyond cellular proliferati on and expression of structural proteins, the changes in tissue phenotype observed in response to TGF1 stimulation require extensive degradation of pre-existing ma trix as the emerging fibrotic tissue is generated and expands, and later as the fibrous mass becomes invasive and is replaced by cartilaginous tissue. In acco rdance with this, some of the more striking increases in gene expression were seen among the MMPs. Fibrosis has often been considered to be a process dominated by TIMPs, whereby increased inhibiti on of MMPs is thought to permit the accumulation of ECM protei ns, leading to fibrotic hypertrophy. 104, 105 Although TIMPs modulate MMP proteolysis, it is also known that TIMPs-1 and -2 facilitate the activation of certain MMPs 106 While we saw an early, but relatively modest, increase in TIMP-1 expression at day 5, dramatic increas es were seen in MMPs -2 -7 -8, -9, -11, 63

PAGE 64

12, -13, -14 and -16, which further increased at day 10. At various time points in the experiment, expression of certain MMPs was increased by 80 to 300-fold. MMPs are known to participate in numer ous diverse activities, including extracellular matrix remode ling, basement membrane disrupt ion, epithelial apoptosis, cell migration, and angiogenesis 107 Their roles in these processes occur either by direct matrix molecule cleavage or by generat ing bioactive mediators and biologic regulators 104, 107 Due to the complex and dynamic natur e of the articular tissues over the 30-day experiment, it is not possible to discern the contributions of individual MMPs in arthrofibrosis. However, our results s uggest that the induction and maintenance of arthrofibrosis does not hinge upon the ac tivity of any singl e MMP, but instead represents an orchestrated, interactive network between numerous MMPs and TIMPs. Also notable was the marked increase in ex pression of matricellular protein genes. These non-structural, secreted glycoproteins interact with cell surface receptors, the ECM and soluble extracellular factors (e.g. growth factor s and MMPs) to modulate cell function as well as regulate the activity or availability of proteins sequestered in the matrix 108 As a group, matricellular proteins are known to enhance cellular mobility, ECM synthesis, cellular differentiation and migration 109-113 They are expressed at high levels during development, but in healthy adults are typically only synthesized during active tissue remodeling, such as in wound repair and disease, particularly in cancer. Within this family, osteopontin showed t he greatest induction throughout. Consistent with our expression data, osteopontin is kn own to bind and activate MMPs-2 and -3 even in the presence of TIMPs 114-117 Somewhat surprisingly, despite the dramatic effects of TGF1 on the joint tissues, expression of CCN2/CTGF remained relatively 64

PAGE 65

unchanged. CCN2/CTGF is thought to be a major activator of TGFactivity and mediator of fibrosis in other tissues 21, 118, 119 Its low induction here and relatively poor response of rat joint tissue to Ad.CTGF suggests that CCN2/CTGF may not play a major role in arthrofibrosis, and that its absence may be compensated for by other members of this group. The increased expression of both N-c adherin and NCAM throughout the 30 day experiment is consistent with the developm ent of chondrometaplasia in the fibrotic tissues. These adhesion molecules are expressed during embryonal chondrogenesis, where they contribute to early mesenchymal cell condensation 120 Their enhanced expression is also in agreement with the aggressive phenoty pe of the expanded synovial tissues, as both molecules are known to contribute to enhanced cellular motility and migration and are associat ed with increased invasivene ss in several types of cancer 121, 122 Synovial and Capsular Fibroblasts have a Hi gh Proliferative Capacity and Innate Plasticity Following lentiviral-medi ated delivery of the cDNA for GFP to the knees of rats, a large percentage of fibroblasts resident in t he synovial lining, subsynovium and fibrous capsule were fluorescently tagged. In pr evious work we have shown that VSV-G pseudotyped lentiviral vectors primarily infect CD90+ and CD29+ fibroblasts in the synovium and capsular tissues After stimulation with TGF94 1, these GFP+ cells massively proliferated, and differentiated in to myofibroblasts, and later chondrocytic cells, such that the vast majority of cells in the expanded tissues fluoresced green. By stably pre-labeling the articular cells in this manner we showed that the immense fibrotic 65

PAGE 66

expansion of the synovium (and subsequent differentiation of the cells into chondrocytes) arises from cells resident to the joint space. For pathologic fibrotic conditions in organs such as the lung, kidney and liver, the fibroblast/myofibroblast populations responsi ble for generating the fibrotic mass are thought to originate variously from: (1) loca l, resident mesenchyma l cell populations, (2) epithelial cells that undergo epith elial to mesenchymal transition, and/or (3) circulating mesenchymal progenitors from bone ma rrow that home to sites of injury 123-127 With regard to arthrofibrosis, our data strongly support the first scenario, and largely exclude the other two as having a meaningful role Further, our data provide a striking demonstration of the prolif erative capacity, innate plasticity and chondro/osteogenic potential of synovial and capsular fibrobl asts. They also confirm the potency of TGF1 as an inducer of proliferation and chondr ogenic differentiation in these cells. The existence of mesenchymal progenitor cells in human synovium has been recognized since 2001, and the natural potent ial of these cells to spontaneously generate ectopic cartilaginous/o steogenic tissue is reflected in diseases such as chondromatosis and arthrofibrosis 128-130 The utility of these cells in therapies for cartilage and bone repair is currently being inve stigated, and several preclinical studies point to synovial-derived fibroblasts as being superior to bone marrow and adiposederived mesenchymal progenitors for these purposes 131, 132 The apparent facility with which articular fibroblasts differentiate along chondrogenic and osteogenic pathways suggest they are predisposed to these lineages and should be strongly considered as candidates for regenerative and tissue engineer ing strategies for connective tissue disorders. The application of TGF1 as a chondrogenic agent in vivo either as a 66

PAGE 67

recombinant protein or transgene product, however, should be more cautiously considered. Our results serve to emphasiz e the sensitivity of connective tissue fibroblasts to growth factor stimulation, and that protocols designed to induce cellular differentiation in cartilage and bone repair in vivo should be aware of the high capacity for toxic side effects in adjacent tissues. 67

PAGE 68

Table 4-1. Relative Expression Values of ECM and Associated Genes in the Joints of Nude Rats Receiving Ad.TGF1 Gene Day 5 Day 10 Day 30 Structural Proteins Collagen, type 1, alpha 1 *4.5 **9.7 *3.7 Collagen, type 2, alpha 1 6.4 **160.3 **223.3 Collagen, type 3, alpha 1 1.8 6.1 1.9 Collagen, type 4, alpha 1 1.1 2.3 *-2.1 Collagen, type 4. alpha 2 1.1 4.2 -1.1 Collagen, type 4, alpha 3 **-3.1 -1.7 -2.6 Collagen, type 5, alpha 1 **4.0 **5.8 2.6 Collagen, type 6, alpha 1 1.7 *4.8 *2.2 Collagen, type 8, alpha 1 *7.6 5.1 *3.3 Hapln 1 2.3 **33.8 **106.5 Versican **4.6 2.9 *4.5 Matricellular Proteins CCN2/CTGF 3.0 3.6 *4.4 Osteopontin 1 **32.8 **118.0 **130.0 Periostin ** 9.3 *7.3 **6.9 Sparc *2.8 *7.4 2.1 Spock 1 1.4 3.7 1.4 Tenascin-C **9.5 **8.9 **9.6 Thrombospondin 1 **26.1 *7.5 **20.0 Thrombospondin 2 **12.6 **8.8 **14.5 Cell Adhesion Proteins E-Cadherin **4.5 9.6 3.4 N-Cadherin **10.7 **29.8 **10.1 P-Cadherin **8.4 2.6 3.0 R-Cadherin *-5.4 3.8 -3.6 Catenin, alpha 1 1.2 3.7 -1.2 Catenin, alpha 2 1.7 1.9 1.4 Catenin, beta 1 2.0 *5.1 1.4 Contactin 1 *-3.2 2.0 -2.1 Icam-1 -1.0 3.4 2.0 Integrin, alpha 2 -1.1 2.6 1.0 Integrin, alpha 3 -1.8 1.8 -2.0 Integrin, alpha 4 1.2 5.6 1.8 Integrin, alpha 5 1.8 4.2 1.2 Integrin, alpha D 1.5 3.7 1.0 Integrin, alpha E 1.6 5.0 1.3 Integrin, alpha L *2.6 *9.9 1.6 Integrin, alpha M 1.5 *4.4 -1.6 Integrin, alpha V 1.4 4.4 -1.1 Integrin, beta 1 2.2 2.9 1.9 Integrin, beta 2 1.5 *5.8 -2.8 Integrin, beta 3 *2.3 *5.2 1.6 Integrin, beta 4 -1.5 3.6 -1.4 Ncam 1 ** 25.3 **15.0 **15.1 Ncam 2 1.3 2.1 1.6 Pecam -1.0 1.6 1.3 68

PAGE 69

Table 4-1. Continued Gene Day 5 Day 10 Day 30 E-Selectin 1.4 -1.7 1.8 L-Selectin 3.6 14.3 **15.83 P-Selectin 2.2 3.1 2.0 Vcam 1 -2.0 1.2 -1.7 Vitronectin -1.8 1.3 -2.3 Extracellular Matrix Proteins Fibronectin 1 1.3 *3.4 *2.7 Laminin alpha 1 *5.7 5.1 **25.0 Laminin alpha 2 -2.2 2.3 1.4 Laminin alpha 3 1.1 *3.5 -1.3 Laminin beta 2 -2.4 3.5 -1.4 Laminin beta 3 -1.5 *3.8 *2.9 Laminin gamma 1 21.6 *72.4 3.3 Tgfbi **6.9 *9.3 **3.9 Metaolloproteinases and Inhibitors Adamts 1 12.2 26.7 15.3 Adamts 2 (RGD1565950) 2.0 ** 4.3 *3.8 Adamts 5 -1.0 2.3 1.3 Adamts 8 *-4.4 -2.4 -3.9 MMP 1a 3.5 3.6 2.6 MMP 2 ** 5.7 **9.3 **6.3 MMP 3 -1.1 2.0 4.6 MMP 7 ** 13.2 **20.8 *9.0 MMP 8 *10.4 *229.9 *39.2 MMP 9 *11.0 *79.5 3.1 MMP 10 3.2 11.2 8.3 MMP 11 *2.9 **17.9 **4.4 MMP 12 ** 97.3 **182.9 **198.1 MMP 13 19.3 *58.9 *300.3 MMP 14 ** 7.0 **5.0 **6.4 MMP 15 -1.4 1.3 -2.1 MMP 16 *3.2 *9.6 **5.70 TIMP 1 **10.8 1.1 **6.8 TIMP 2 2.9 1.5 *2.5 TIMP 3 ** -4.1 3.1 -1.5 Elastic Fiber Proteins Emilin 1 *181.3 *87.0 *36.1 Fibulin 1 1.4 3.4 *3.1 Extracellular matrix protein 1 -1.5 3.8 1.5 Other Synaptotagmin I 3.1 3.0 9.2 CD44 Antigen -2.4 -1.5 -3.1 Sarcogycan 1.3 3.4 1.1 NTPDase-1 (CD39) 1.9 3.0 1.6 Values with an asterisk represent p<0.05. Values with two asterisks represent p<0.01. n=3 for each time point. 69

PAGE 70

Figure 4-1. Transgene expression following infection of ra t synovial fibroblasts with Ad.TGF1. Following isolation of fibroblasts from rat synovium, the cells were grown in monolayer in 12-well plates and infected with increasing doses of Ad.TGF1. At 24hrs post infection, the m edium was replaced by 0.5 ml of serum-free medium. At 48hrs post infection, the conditioned medium was harvested and TGF1 content measured by spec ific ELISA. Results are expressed in ng/ml as the mean of 3 replicates. Error bars represent + S.E.M. 70

PAGE 71

Figure 4-2. Intra-articular delivery of Ad.TGF1 induces joint swelling. The knees of nude rats were injected with 2.0 x 10 9 vp of Ad.TGF1, and the diameter of the joints was measured periodically wi th calipers. Injection of Ad.TGF1 rapidly induced joint thickening, and animals showed increased knee diameter throughout the experiment. For the Ad.TGF1 treated group, days 0 and 5 represent n=16 knees, days 5-10 represent n=8 knees, and days 1030 represent n=4 knees. For all time points normal control represents an n=8 knees. Values represent the mean of measurements for all knees in each group. Error bars represent + S.E.M. 71

PAGE 72

Figure 4-3. Local overexpression of Ad.TGF1 in the knee joint induce severe arthrofibrosis. Groups of rats were injected intra-articularly with 2.0 x 10 9 vp of Ad.TGF1 (panels A and B) or Ad.GFP (panels C and D) and were killed periodically thereafter, as described Images shown are representative of those sacrificed at day 30. (A and C) External views of knee joints after removal of the skin. (B and D) Internal vi ews of the joints following dissection. For joints receiving Ad.TGF1 there was a visible increase in joint size, and the knees became encased in a dense, scar-like tissue. Upon dissection, the fibrotic tissue was observed to override the joint space and displace preexisting structures. Arrows indica te the location of the patella. 72

PAGE 73

Figure 4-4. Capsular fibrosis and chondrogenesis induced by Ad.TGF1. Knees of nude rats receiving Ad.TGF1 were harvested at various time points, decalcified and processed for histology. Sections were stained with H&E or toluidine blue as indicated. Images at day 0 show the Ad.GFP treated joint with a thin synovial lining supported largely by adipose tissue. At 5 days post injection, an expansion of fibroblastic cells from the synovial lining generated the bulk of the fibrotic mass, occluding the adipos e layer. 10 days post injection, formation of chondrocyticic cells can be seen within the fibrotic tissue. By day 30, the majority of fibroblastic cells had differentiated into chondrocytes, as seen in the toluidine bl ue stain, invading neighboring tissues and displacing existing structures. Im ages in left two columns are at a magnification of 2.5x, and images in the right columns are at a magnification of 20x. 73

PAGE 74

Figure 4-5. Immunohistochemical staining for MMPs 9 and 13, and smooth muscle actin ( SMA). The knee joints of nude rats receiving Ad.GFP control and Ad.TGF1 were harvested at days 5, 10 and 30, paraffin embedded, sectioned and immunologically stained for the presence of MMP 9, MMP 13 or SMA as indicated. 74

PAGE 75

Figure 4-6. Arthrofibrosis and chondromet aplasia arise from resident synovial and capsular fibroblasts. A) Fibroblastic cells in the synovium and joint capsule of nude rats were fluorescently labeled following intra-articular injection of recombinant lentivirus containing the cDNA for GFP (LV.GFP). 48 hours after delivery of LV.GFP, Ad.TGF1 was injected into the joint, and the rats were killed periodically thereafte r. (B) Day 5 post Ad.TGF1 injection showed a fibrotic cell mass composed primarily of GFP+ cells. (C) At day 10 the GFP labeled cells began to acquire a chondrocytic appearance. (D) By day 30, the GFP+ cells had differentiated into a chondrocytic phenotype as evidenced by the rounded morphology. E) Joints receiving only Ad.TGF1 show no evidence of GFP expression. F) The cartilaginous phenotype of tissues in panel D is confirmed by metachromatic staining of proteoglycan with toluidine blue. 75

PAGE 76

CHAPTER 5 EXAMINING THE EFFECTS OF LONG TERM ARTHROFIBROSIS IN IMMUNOCOMPETENT ANIMALS Introduction IAC is characterized by a painful fibr otic expansion of the synovium and joint capsule, which gradually results in the loss of active and passive motion of the joint. It typically begins over several months and progresses in stages 87 The disease is generally self-limiting. Some individuals, however, mostly di abetics, never recover from the painful and stiff stages and remain pe rmanently disabled. Since the underlying causes of IAC remain unknown, no specific pharmacologic or non-surgical therapy has been shown to cure IAC or provide signific ant long-term benefit. A primary reason for the absence of literature r egarding the pathogenesis of IAC is the lack of available tissue from early stage disease for detail ed analysis. Few studies have addressed the involvement of specific growth factors in IAC. Pathologic fibrosis is a prominent feature of chronic disease in several organs, including the skin, liver, lung and kidney, and often begins with local injury and activation of repair pathways. While the physi cal causes of pathologic fibrosis vary, the protein factors that mediat e wound healing and tissue repai r are considered to play central roles. TGF1 has been implicated in a majori ty of fibrotic conditions 17 Previously, we chronically overexpressed TGF1 at high levels in the knee joints of immunocompromised nude rats, which led to very severe arthrofibrosis (see Chapter 4). The amount of virus delivered in these ex periments represented an extreme response to stimulation with TGF1. While relevant to various diseases of the joint, including 76

PAGE 77

synovial chondromatosis, chondrometaplasia, osteosarcoma and chondrosarcoma, it showed no signs of resolving. Therefore, in an effort to understand the biological processes that contribute to the development and resolution of arthrofibrosis and frozen shoulder, we wanted to establish an animal model that more closely reflected the level of disease in humans. As cells expressing nonhomologous transgenic proteins are immunologically cleared from the joints of immunocompetent Wistar rats within 21 days 94 we used this system to examine the effects of s hort term production of TGF1 intra-articularly, with the goal of inducing a fibrotic event that will more clos ely mirror the stages of IAC. By observing animals over the course of 120 days and delivering a low dose of Ad.TGF1, we hypothesized that the animals would undergo a remodeling process similar to stage 4 IAC. To follow the changes in gene expressi on over the course of disease in the capsular tissues, we used real-time PCR te chnology and histologic analysis. Using PCR arrays, we were able to establish the signaling patterns occurring throughout this process, which should be of use in future studies comparing these data to expression data obtained from hu mans with IAC. We found that TGF1 gene transfer very rapidly induced a fibrotic condition that completely encased and immobilized t he injected joint, induced a chondrogenic response, and over the course of the 120 day experiment, gradually resolved into a less aggressive, less fibrotic, and overall less cellular tissue. Histologic examination, as well as focused expression arrays of the joint tissues over time, suggested the developing fibrotic tissue adopted a phenotype similar to that of IAC and gradually remodeled with time. 77

PAGE 78

Results We hypothesized that over-production of TGF1 intra-articularly contributes to, and may be the cause of, chronic arthrofibrosis seen in patients with IAC. The goal of our experiments was to establish a less severe model of arthrofibrosis more similar in intensity to that of IAC. In doing so, we could more accurately model its onset and progression as well as the factors affecting its resolution. Immunocompetent Wistar rats received a single intra-articular injection of Ad.TGF1or Ad.GFP into both knee joints (5.0 x 10 7 vp). Groups of rats were killed periodically over the course of 120 days, and the tissues were harvested and analyzed. Within three days, rats developed a robust fibrotic re sponse. The knees became swollen and were slightly immobilized the joint. Although t he animals showed a prominent fibrotic response, they were otherwise healthy with normal appetite. Animals receiving Ad.TGF1 showed an increase in knee diameter at day 10 relative to controls, and the increased joint size persisted throughout the course of the experiment. Physical manipulation of the knee showed the joints were stiff, and while not as seen in the nude rat receiving a high er dose, range of motion was notably limited. While not quite as severe as previously seen ( Figure 4-3 ), gross inspection of the joint tissues at the times of sacrifice reveled that knee joints were encased in fibrotic tissue, obscuring the normal features of the join t. Also, similar to immunocompromised animals in Chapter 4, early in the time cour se this fibrotic encasement could easily be dissected away from the bone and structures beneath; however as the fibrosis developed, the tissue became more dense, making it difficult to distinguish pre-existing anatomical structures. 78

PAGE 79

Delivery of Ad.TGF1 to the Joints of Immunocompetent Rats Induces Arthrofibrosis and Chondrometapl asia that Resolves with Time The normal rat knee joint is surrounded by a thin layer of synovium, only a few cell layers thick, overlying a layer of adipose tissue. Following stimulation of the synovium with TGF1, there was a massive proliferation of elongated fibroblastic synovial cells surrounding the joint. These fi broblastic cells from the synov ium actively proliferate and are surrounded by collagen fibers, generating the bulk of the fibrotic mass replacing all of the adipose tissue seen in the normal joint ( Figure 5-1 Day 10). By day thirty, the hypertrophied fibrotic tissue extends outward to fill the entire joint space, displacing resident joint and capsular structures ( Figure 5-1 Day 30). The fibrotic tissue began to adhere to the articular cartilage displaying a blending together of elongated fibroblastic cells and rounded cartilage cells, so much so, t hat toluidine blue staining of the matrix composition and cell morphol ogy reveal the predominant phenotype of the cells overrunning the joint had changed to that of articular cartilage. Reminiscent of the resolution phase de scribed in human IAC, by day 90, the overall phenotype of the joint space of these animals was less severe and less cellular than that seen at day 30; however, much of the chondrogenic tissue remained ( Figure 5-1 Day 90). As opposed to the highly prolif erative state seen at days 10 and 30, by days 90 and 120, the joint appeared to resolve much of the fibrotic and chondrogenic mass that developed. At day 120, the majority of fibr otic overgrowth had been eliminated, there was an over all reduction in cellularity, and adipose tissue could be seen surrounding the joint space. While some fibrotic and cartilaginous tissue remained, the overall architecture of the joint began to once again resemble that of a normal joint (Figure 5-1, Day 120). 79

PAGE 80

Expression Profiles of Ge nes Stimulated by TGFDisplay Altering Patterns of Expression Consistent with the Phenotype of the Tissue Using this overall less severe model of IAC, we wanted to ex amine the expression profile over the course of the 120 day experiment, to observe changes in expression of ECM and adhesion molecules, using RNA isolated from the capsular/synovial tissues of control rats and those receiving Ad.TGF1. Differential analyses of the expression patterns can be seen in Table 5-1 Similar to our earlier experiment and consis tent with the early fibrotic expansion seen histologically, collagen types I, III and V we re elevated from 2 to 6 fold at day 10. As the hypercellular tissue transitioned in to a cartilaginous phenotype at days 10 and 30, expression of collagen type II and hy aluronan and proteoglycan link protein 1 (HAPLN1/cartilage link protei n) dramatically increased, with both genes showing >100 fold increase by day 10 and >45 and ~20-fold increase by day 30, respectively. Interestingly, at days 90 and 120 expression of all structural proteins decrease, including Collagen II and Hapln1, cons istent with the re modeling observed histologically. Supporting the highly prolifer ative, expansive response of the capsular tissues, expression of the matricellu lar protein genes, as a group, were highly elevated at the early time points, days 10 and 30. Those most profoundly induced were osteopontin (15-100 fold), followed by thrombospodins 1 and 2, tenascin-C and periostin. As a whole, the matricellular pr oteins gradually decreased in expression as the tissue continued to remodel through day 120 into a less aggressive and less fibrotic phenotype. The exception was Spock-1, also known as testican-1. While the exact functions of this matricellular protein ar e not fully understood, it is speculated to 80

PAGE 81

participate in the regulation of matrix turnover in cart ilage and inhibition of membranetype MMPs, consistent with the re modeling seen in this tissue. Characteristic of the expansive and invasi ve properties of the capsular tissues, expression of the MMPs was broadly induced at all time points. Interestingly, for many of MMPs, expression levels not only peaked at the induction of fibrosis at day 10, but also during the reduction phase seen at day 120. Expression of MMP-12 (elastase) showed the greatest levels of expression at days 10 and 30 ranging from >250 to >150fold over controls and decreased to a 45 fo ld induction by day 120. This was followed by the collagenases, MMPs-13 and -8, which we re elevated throughout the experiment, showed >100-fold and >60-fold enhanced ex pression at day 10, respectively. Additionally, both MMP-12 and -13 showed increases in ex pression at day 120, during remodeling. MMP-9 (gelatinase), MMP-7 (matrilysin), MMP-11 (a member of the stromelysin family), and MMPs-14 and 16 (also termed membr ane type; MT-MMPs-1 and -3) also showed significant increases in expression, ranging between ~2-50 fold induction throughout the course of the ex periment. Similarly, ADAMTS-2, -5, and -8 increased in expression as the tissue rem odeled at days 90 and 120, but not all points were statistically significant. A variety of cellular adhesion molecules also showed increases in expression in the expanded capsular tissues during fibr otic induction. Heightened expression of neural-cell adhesion molecule (NCAM) peaked at day 10 and remained increased from >20 to 2-fold, throughout the experiment. In terestingly, both NCAM and N-cadherin are thought to coordinately interact during cellular condensation in chondrocytic differentiation as well as in tumor cell proliferation; however decreases are seen in N81

PAGE 82

cadherin at days 10 and 30. Several integrin receptor molecules showed increased expression during the early stages of the experiment, including L E and 3 subunits at day 10 and 30. While most cell adhesion molecule s peak at onset of fibrosis integrin and E-selectin peak in expression during remodeling at day 120. Several other genes showed enhanced expre ssion at onset of fibrosis and an overall reduction during remodeling; the mo st notable being Emilin-1 whose expression peaked at day 10 and gradually diminished. Em ilin-1 is a connective tissue glycoprotein associated with elastic fibers, and is thought to contribute to cell motility, tissue differentiation and morphogenesis. 95 Certain laminins also showed increased expression at onset, part icularly the laminin 1 chain at day 30 and laminin 1 at days 10 and 30. Discussion Patients with IAC generally experience an extremely painful stage when the adhesions are developing within the joint ca psule, followed by a frozen/adhesive phase, during which time the pain subs ides but the glenohumeral join t remains stiff with limited rotation, and finally a thawing/resolution phas e. The spontaneous resolution can take up to 42 months and patients see an improvement in the range of motion as the adhesions resolve. Since not much is known about what causes the glenohumeral joints of patients to freeze or to thaw, we examined by histology and by gene expression analysis the differences that occur in a model of joint fibrosis in rodents. During the onset of fibrosis observed at day 10, there was an accompanying increase in expression of various ECM proteins, including inte rstitial collagens type I, II and III. As the tissue transitioned to hyalin e-like cartilage, dr amatic enhancement of 82

PAGE 83

expression of collagen type II, the predominant structural protein of articular cartilage along with an increase in transcription of ca rtilage link protein (HAPLN1) could be seen. As the tissue resolved over the 120 day experiment, there was an overall phenotypic decrease in the chondrometaplastic tissue and concurrent decreases in both collagen type II and cartilage link protein. While not examined in the pr esent study, apoptosis of fibrotic synovial cells may contribute to the resolution of fibrosis by acting as a mechanism for removing the cell population responsible fo r both producing the fibrotic ma trix and protecting the matrix from degradation via their production of TI MPs. Evidence suggests that a major mechanism mediating the loss of cells that are unwanted a fter a pathological process is apoptosis 133-136 However, the loss of activated sy novial fibroblasts is not in itself sufficient to allow a remodeling of the existing excess collagens, which requires matrix degradation to be upregulated. MMP s are key enzymes in the breakdown of the ECM. Collagenases, MMP-1, -8, -13, are members of the MMP family that degrade fibrillar collagens types I, II, and III MMP-13 is the major collagenase that degrades collagens in connective tissues in rats along with MM P-8, which is expressed in articular chondrocytes 137 synovial fibroblasts 138 osteoblasts and osteocytes 139 among other locations. Our results indicate that in the MMP8 expression peaks at the development of fibrosis showing >60 fold, when the tissue is expanding and differentiating into an articular cartilage-like phenotype, and tapers off throughout the experiment as the tissue remodels so that by day 120 its expressi on had decreased. MMP-13 mRNA expression levels peaked both at the onset of fibrosis and when the most remo deling of the joint 83

PAGE 84

space seemed to be occurring. At days 10 and 120 MMP-13 levels showed >100 fold increases and at days 30 and 90 expression le vels were >50 and >65, respectively. TIMPs-1 and -2 expression w ent from >25-70 fold increas es seen during the peak of fibrosis at days 10 and 30, to an approxim ate ~2-fold increase at days 90 and 120 consistent with the extensiv e remodeling and degradation of fibrotic tissue. While MMP activity was highly expressed throughout the experiment, several of the MMPs peaked during both the fibrotic induction at day 10, and the resolution at day 120. These data suggest important factors for promoting the matrix remodeling are the enhanced MMP synthesis and the removal of the inhibitory influences of the TIMPS on collagenase activity. Intra-articular delivery of a low dose of virus in immunocompetent Wistar rats observed over 120 days, resulted in a rapid and severe fibrotic induction that was less intense than previous experiments (see Chapt er 4), and remodeled into an overall less fibrotic tissue with time. The resulting pathol ogy more closely resembled that described in patients with IAC. An initial synovial fibr otic proliferation caus ed the joint to become stiff. However, areas of the synovium transformed into hyali ne-appearing cartilage, which is observed in various forms of arthrofibrosis within the knee joint, not necessarily in IAC. The formation of synovial chondromet aplasia could be due to the anatomical site of the intra-articular injections. This shortcoming is not easily remedied as it is not feasible to inject into the shoulder joint of rodents. This dense fibrotic chondrometaplastic tissue resolved to a large extent over the course of 120 days. While the cartilaginous tissues were not completely eliminated by the end of this study, normal joint features were once again visible, and one would predict that this tissue would 84

PAGE 85

completely resolve at a longer time poin t. Ultimately, these data will be compared to global expression patterns of RNA extracte d from tissue samples of human patients undergoing surgery for IAC. Comparing the expression profile s of what is seen in our model of arthrofibrosis to that in human pat ients, will provide a valuable insight into the signaling of this disease and, perhaps, hi ghlight potential tar gets for therapy. 85

PAGE 86

Table 5-1. Relative Expression Values of ECM and Associated Genes in the Joints of Wistar rats receiving 5.0*10 7 Ad.TGF1 Gene Day 10 Day 30 Day 90 Day 120 Structural Proteins Collagen, type 1, alpha 1 **6.3 *3.3 -1.0 1.5 Collagen, type 2, alpha 1 **106.3 48.2 **14.2 2.9 Collagen, type 3, alpha 1 **2.9 1.9 -3.0 -1.4 Collagen, type 4, alpha 1 1.0 *-2.0 -3.1 -2.3 Collagen, type 4. alpha 2 1.7 -1.5 -3.5 -2.5 Collagen, type 4, alpha 3 -1.6 -2.4 1.8 2.6 Collagen, type 5, alpha 1 **2.7 2.1 -2.3 -1.2 Collagen, type 6, alpha 1 **2.4 1.5 -1.2 1.5 Collagen, type 8, alpha 1 **7.6 *5.1 2.0 **2.4 Hapln 1 **165.4 *21.7 *21.9 6.0 Versican -1.2 *5.6 2.0 2.9 Matricellular Proteins CCN2 (CTGF) *3.0 2.0 2.1 2.0 Osteopontin 1 **100.6 *49.2 **80.4 15.9 Periostin **6.0 2.6 -4.4 -1.2 Sparc *2.1 1.7 -2.8 -1.6 Spock 1 -1.1 *6.3 7.7 8.1 Tenascin-C **7.9 1.9 1.8 2.8 Thrombospondin 1 **14.4 **11.9 3.1 2.5 Thrombospondin 2 **11.1 *7.6 2.5 3.2 Cell Adhesion Proteins E-Cadherin **2.9 **5.5 2.2 1.2 N-Cadherin *-2.8 *-6.6 2.8 1.7 P-Cadherin 1.9 3.0 1.1 1.8 R-Cadherin **-12.7 1.4 -1.4 -1.9 Catenin, alpha 1 -1.1 1.1 -2.3 -1.2 Catenin, alpha 2 -2.1 *19.6 12.6 3.4 Catenin, beta 1 **2.5 1.7 -1.3 1.3 Contactin 1 -2.1 -1.1 -1.3 2.2 Icam-1 -1.2 1.1 -1.0 1.7 Integrin, alpha 2 1.6 1.1 -1.2 1.2 Integrin, alpha 3 **-3.2 *-3.4 -1.2 1.5 Integrin, alpha 4 **2.7 1.7 -1.7 -1.1 Integrin, alpha 5 **1.6 1.1 -1.8 1.1 Integrin, alpha D -1.5 2.1 -1.6 1.3 Integrin, alpha E **31.5 **58.7 1.9 4.5 Integrin, alpha L *2.9 *3.1 -2.8 -1.4 Integrin, alpha M 1.7 1.7 -1.5 1.4 Integrin, alpha V 1.7 -1.1 -2.7 -1.3 Integrin, beta 1 **-3.1 **-4.8 *21.3 **18.1 Integrin, beta 2 1.2 1.1 -2.2 -1.2 Integrin, beta 3 *4.12 1.1 -2.1 -1.2 Integrin, beta 4 -1.6 -1.4 -1.5 1.1 Ncam 1 **24.3 *11.2 2.0 *4.7 Ncam 2 -3.0 4.9 2.9 1.5 86

PAGE 87

Table 5-1. Continued Gene Day 10 Day 30 Day 90 Day 120 Pecam 1.5 1.4 2.0 *3.0 E-Selectin -2.1 -2.9 *15.0 **20.7 L-Selectin **-9.8 -6.0 1.4 1.9 P-Selectin 1.7 1.6 -1.7 1.5 Vcam 1 -1.2 -1.5 1.3 1.5 Vitronectin *-4.4 -6.7 3.4 2.5 Extracellular Matrix Proteins Fibronectin 1 **-2.0 *-3.5 3.3 *3.7 Laminin alpha 1 3.1 *17.7 3.7 3.7 Laminin alpha 2 -1.6 1.3 -1.7 1.2 Laminin alpha 3 -1.0 -1.0 -1.8 -1.4 Laminin beta 2 -1.2 1.3 -2.1 -1.3 Laminin beta 3 **-9.2 **-6.8 1.1 1.5 Laminin gamma 1 10.0 18.4 1.8 1.9 Tgfbi **6.3 *4.1 -2.1 1.0 Metaolloproteinases and Inhibitors Adamts 1 *2.7 12.7 -1.5 1.4 Adamts 2 (RGD1565950) *1.7 1.6 2.7 **5.0 Adamts 5 -1.7 -1.2 4.8 *5.6 Adamts 8 -1.8 -2.6 6.0 8.9 Mmp 1a -1.9 21.8 7.2 2.3 Mmp 2 **1.7 1.3 -2.5 1.6 Mmp 3 *4.5 **39.4 3.1 3.8 Mmp 7 -1.5 *10.1 1.6 *16.8 Mmp 8 **61.8 *24.4 16.8 *9.2 Mmp 9 *51.9 11.5 9.3 14.0 Mmp 10 2.5 20.1 11.5 7.6 Mmp 11 **7.7 **5.8 1.3 *3.1 Mmp 12 **258.9 *157.3 6.9 *45.0 Mmp 13 105.5 51.1 *66.4 *112.2 Mmp 14 **7.6 *2.9 2.2 4.3 Mmp 15 **-3.8 -**5.3 1.3 2.4 Mmp 16 **4.6 *4.5 1.5 2.4 Timp 1 **35.3 **49.8 2.3 2.9 Timp 2 **74.5 **35.8 2.1 3.1 Timp 3 -1.9 -2.2 -2.2 -1.8 Elastic Fiber Proteins Emilin 1 *49.29 32.8 6.3 *10.4 Fibulin 1 1.4 2.8 -1.1 1.9 Extracellular matrix protein 1 2.1 1.3 1.3 2.1 Other Synaptotagmin I -2.0 5.7 3.2 1.7 CD44 Antigen *-2.5 -1.9 2.0 *2.8 Sarcogycan -1.4 1.1 -1.6 1.0 NTPDase-1 (CD39) *4.9 1.8 1.6 2.5 Values with an asterisk represent p<0.05. Values with two asterisks represent p<0.01. n=3 for each time point. 87

PAGE 88

Figure 5-1. Capsular fibrosis and chondrogenesis induced by Ad.TGF1 resolves with time. Knees of Wistar rats receiving Ad.TGF1 were harvested at various time points, decalcified and processed for histology. Sections were stained with H&E or toluidine blue. Control images, in the upper panel, show Ad.GFP treated joint with a thin synovial lining supported largely by adipos e tissue. 10 days post injection, an expansion of fibroblastic cells from the synovial lining generated the bulk of the fibrotic mass, occluding the adipose layer and formation of chondrocyticic cells can be seen within the fibrotic tissue. By day 30, the majority of fibroblastic cells had differentiated into chondrocytes, as seen in the toluidine blue stain, invading neighboring tissues and displacing existing structures. Day 90 histology shows signs of remodeling with a less aggressive tiss ue and reduction in cellularity. By day 120, much of the fibrotic and chondrogenic tissue has resolved. Images in left two columns are at a magnification of 2.5x, and images in the right columns are at a magnification of 20x. 88

PAGE 89

CHAPTER 6 EXAMINING THE RELATIONSHIP BETWEEN DIABETES AND ARTHROFIBROSIS Introduction According to statistics from the Centers for Disease C ontrol and Prevention, in 2007, 23.6 million people (7.8% if the population) in the United States were living with diagnosed or undiagnosed diabetes. Diabetes mellitus is a medical condition associated with abnormally high levels of gluc ose in the blood. Normally, blood glucose levels are controlled by insulin. In type I diabetes, destruction of beta cells, leads to hyperglycemia. Type I diabetes (insulin-dep endent), accounts for 10% of all cases and results from the autoimmune destruction of in sulin-producing beta cells in the pancreas by CD4+ and CD8+ T cells and macrophages infiltrating the islets 42 Afflicting over 150 million people worldwide, type II diabetes, (non-insu lin-dependent (NIDDM)), is an incurable metabolic disorder characterized by insulin resistance, decreased beta-cell function, and hyperglycemia 43 Approximately 20% of diabetic patients not only suffer with IAC, but have a more persistent and severe case. Their inability to properly regulate blood glucose may be responsible. Hyperglycemic conditions are k nown to stimulate elevated expression of TGF1 53, 60 Acute and chronic high glucos e exposure stimulates TGF1 transcription, leading to an increased pool of bioactive TGF140 High levels of ambient glucose have been shown to stimulate the production of extracellular matrix components, including collagens, via TGF1 and CTGF signaling 46 Diabetic animals displayed increased levels of TGF-1 53 mRNA and protein in kidney tissues, along with an activated Smad signaling pat hway, transducing the TGF1 signal 53, 54 89

PAGE 90

TGF1 has been shown to be elevated in plasma 49 secreted in the urine 50 elevated in the circulation 141 and found in kidney tissues 51, 52 of patients who suffer from diabetic nephropathy, a fibrotic condition that causes kidney function impairment, leading to end-stage renal disease. All three isoforms of TGF142 and TGF1 mRNA are markedly increased in renal biopsy s pecimens from patients with diabetic kidney disease 143 Data from tissue culture and animal models has demonstrated that elevated glucose levels stimulate signaling casc ades involved in cell proliferation and fibrogenesis. Compared with low glucose, high glucose environments decrease expression of MMPs, incr ease expression of TIMPs 144 and lead to accumulation of fibronectin, laminin, and types I and IV collagen 56 In several renal cell types, treatment with a TGF1 antagonist, such as a neutralizing monoclonal antibody 57 or antisense oligonucleotides 59 largely eliminated the rise in EC M expression due to high glucose. This pointed to TGF1 as a mediator of the profibro tic effects of high glucose on the kidney 53 Similar studies showed treatment with neutralizing monoclonal antibodies against TGF-1 prevented mRNA increases of TGF1, type IV collagen, and fibronectin in diabetic mice 60 and anti-TGF1 antibody therapy prevented mesangial matrix expansion and renal insufficiency 145 While the specific reason remains unknown, diabetics are prone to the development of IAC and often suffer fr om a more persistent, prolonged case 146 We hypothesize that this is due to increased levels of blood sugar that stimulate the release of TGFand other profibrotic cytokines that serve to exacerbate the fibr otic condition. 90

PAGE 91

To test this we exam ined the effects of TGF1 overexpression on the knee joints of rats with experimental diabetes induced by streptozotocin. Results D-glucose Stimulates Cell Migration Prior to undertaking experiments in our animal model of IAC, we examined the effects of elevated glucose on the function of TGF1 in RSF in vitro To determine if elevated glucose enhanced cellular migration, RSF cells were grown to confluence in 6well dishes prior to treatment with mito mycin C, which alkyl ates DNA and produces interstrand DNA cross-links, thereby i nhibiting DNA synthesis and fibroblast proliferation 147-151 The RSF cells were t hen transduced with 1.3 x 10 7 vp of the appropriate viral construct, either Ad.TGF1 or Ad.CTGF. Afterward, the medium in certain wells was supplemented with D-glucos e. Using a pipette tip, a scratch was made in the cellular monolayer to simulate a tissue lesion, and cell migration into the mock wound space was assayed at 48 hours post treatment. In normal medium, overexpression of CTGF induced cellular migration approximately 2 fold over t hat of control, while TGF1 increased motility approximately 3-fold. In the presence of elevated D-gl ucose, migration into the mock wound space was also enhanced approximately two-fold at each concentration tested: 8, 25 and 75 mM ( Figure 6-1 and Figure 6-2 ). An additive effect could be seen in the presence of elevated glucoses with TGF1 or CTGF where migration was increased approximately 3x and 4x that of control, respectively. Th is assay shows the potential for extracellular glucose to enhance the effects of TGF1 and CTGF-induced cellular migration in RSF cells. 91

PAGE 92

Streptozotocin Diabetic Induction Noting the effect elevated glucose had on stim ulating cellular migration of rat joint cells in culture, we wanted to examine the role that increased circulating glucose plays in our animal model. Based on these data, we hypothesized the presence of high glucose would enhance the effects of TGF1, leading to a more severe and prolonged case of arthrofibrosis. To test this, animals were rendered diabetic by injection streptozotocin (STZ). The STZ-induced diabetic rat is one of the most commonly employed animal models used to study diabetes and its associated complications. STZ is a glucosamine-nitrosourea compound that is particularly toxic to beta cells, the insulin-producing cells of the pancreas. The diabetes that results is similar to type I insulin-dependent diabetes mellitus in humans, and animals exhibit behavioral and biochemical signs consistent with the onset of diabetes. Male Wistar rats weighing approximat ely 175 g were fasted up to 6hrs prior to diabetic induction. Animals received two in tra-peritoneal injections of freshly prepared STZ, at a dose of 45 mg/k g, 72 hours apart. Animals were supplied 10% sucrose water for 48 hours post injection and body weights and blood glucose levels were measured 72 hours after second STZ injection, followed by measurement at weekly intervals, to identify onset and continued presence of diabet ic hyperglycemia. Diabetic animals were euthanized at 10, 30, 90 and 120 days after onset of diabet es, and joint tissues were harvested. As seen in Table 6-1 mean diabetic glucose was 480 mg/dL compared to 108 mg/dL in non-diabetic, normal animals. Blood glucose levels exceeding 250 mg/dL were considered diabetic. 92

PAGE 93

Diabetes Induced Alterations in Gene Expression in Synovial Tissue Diabetic and normal control rats were killed at days 10, 30, 90 and 120 and the joint tissues were harvested for examinatio n. No phenotypic differences were observed by gross visual inspection or in H&E stai ned tissue sections among the diabetic animals compared to non-diabetic animals (data not shown). To determine the effect of hyperglycemia on normal tissue homeostasis in the joint at the molecular level, RNA was isolated from the capsular tissues of diabetic induced animals and age-matched non-diabetic control rats was analyzed by PCRarray for differential expression of genes associated with cellular adhesion, extracellu lar matrix synthesis and remodeling. Numerous alterations were seen in gene expression patterns in the synovium of diabetic animals compared to non-diabetic anim als particularly at t he onset of diabetes, as shown in Table 6-2 Among the structural proteins, the majority of the collagens displayed decreased expression throughout the experiment; however, collagens 4, 6, 8 were increased at the later time points. C onsistent with these data, kidney tissues from animal models of diabetes and humans show an upregulation of Collagen IV 58, 144 Of the matricellular proteins, only Spock1 s howed enhanced expression at each of the four time points examined. Expression patterns reminiscent of that observed in the Ad.TGF1 fibrotic tissues in earlier chapters, were seen for several gr oups of genes, including expression TIMP-1 and -2, which were increased throughout the course of the experiment, along with proteases MMP 1a, 3, 7, 8, 10, and 11. Interestingly, ADAMTS -5 and -8 and MMP-3, 7, -10, and -12 all had increased expression levels at the onset of hyperglycemia and some were increased by the later time point s. Many cell adhesion proteins, including 93

PAGE 94

integrins 3 and 1, Eand L-selectins, N-cadher in, VCAM1 and vitronectin also displayed expression patterns similar to that seen in fibrotic ti ssues. Expression patterns of laminin 1, 1 and fibronection were also similar. Most of the extracellular matrix a nd adhesion genes examined, were either increased or decreased in diabetic joints compared to normal joints. Although no phenotypic differences were observed bet ween non-diabetic and diabetic animals, many of the alterations in gene expression ar e somewhat similar to animals receiving overexpression of TGF1 in Chapter 5, albeit to a lesse r extent. These results suggest that increased circulating levels of blood glucose lead to synovial expression patterns similar to that seen in fibrotic tissues. As these animals have altered levels of expression in many ECM and proteolytic ge nes, they could be more susceptible to development of fibrosis in the presenc e of an appropriate stimulus suggesting a possible explanation for the prevalenc e of IAC in diabetic patients. Diabetic Joints Receiving Ad.TGF1 Resulted in an Overall Less Severe Fibrosis After onset of diabetes, groups of diabet ic and non-diabetic Wistar rats were intra-articularly injected with 5.0x10 7 vp of Ad.TGF1 or Ad.GFP control in both knees. Animals were euthanized at 10, 30, 90 and 120 days post injection and joint tissues were harvested for analysis. After vector injection, there was an increase in knee diameter, and the joints showed modest decrease in ROM. Gross a ppearance and overall knee diameter did not differ between diabetic and non-diabetic. Tissues were examined histologically using H&E staining to determine differences in fi brotic phenotype and severity. As shown in Figure 6-3 no observable differences could be seen in the histology between normal 94

PAGE 95

and diabetic control joints. However, mild differences were observed between diabetic compared to non-diabetic animals receiving Ad.TGF1, for both groups. After 10 days, expansion of fibroblasts within the synovium replaced all adipose tiss ue, filling the entire joint space, and the fibrotic tissue began to fuse with articular cartilage. Interestingly, at these sites of fusion in non-diabetic animals the cells began to lose their spindled morphology, instead appearing as rounded, chondr ocyte-like cells. However, at these same junctions in diabetic animals, t here was no transition into a chondrocytic phenotype and cells retained their fibroblastic morphology. At thirty days, in both groups of animals the normal microsc opic anatomy of the joint was not recognizable and the resident joint and capsular structures were displaced. Large areas of neo-cartilaginous tissue were seen in non-diabetic animals, and the majority of the articular cartilage was replac ed by metaplastic fibrocartilaginous tissue, confirmed by the deeply ampho philic appearance of the tissue on toluidine blue stain (not shown). The tissue of diabetic anima ls was considerably less cellular, with a greater concentration of collagenous mate rial, and although cartilaginous formations were present in the 30 day animals they were noticeably reduced. At 90 days, the tissues in both groups of animals showed signs of resolving, and there was a reduction in overall cellularity compared to day 30 tissues. The diabetic tissue showed signs of increased chondrog enic formation, as seen in the rounded chondrocyte-like cells and toluidine blue stai ning. By day 120, both groups showed an overall reduction in cellularity, and the phenot ype of tissues resembled that of more normal tissue, with the reappearance of adipose ti ssue along the outer edge of the joint space. Overall, the tissues were less aggressive and appeared to switch from a 95

PAGE 96

proliferative phase to one of remodeling. Wh ile most of the cartilaginous developments were resolved in the non-diabetic animals, this process seem to lag in the diabetic joints as many of the cartilaginous masses remained within the joint. Decreases in Gene Expression at Onset of Fibrosis were seen between Diabetic and Non-Diabetic Animals Receiving Ad.TGF1 Counter to our expectations in the diabetic animals, several genes showed a much lower level of induction following delivery of Ad.TGF1 ( Table 6-3 and Table 6-4 ). The most notable changes were seen at day 10 in t he structural proteins, Collagen I, II and VIII, and Hapln1, all of which showed decreased expression from -2.6 to -326-fold. The decreased expression in Collagen II and H apln1 is consistent with the lack of cartilaginous formation seen histologically in diabetic compared to non-diabetic. Compared to non-diabetic animals, severa l matricellular proteins displayed decreased expression at day 10 including osteopontin, periostin and thrombospondins 1 and 2. Osteopontin and periostin also had decreased expression le vels at day 90 and 120 respectively. N-cadherins, integrin V and Ncam 1 were all slightly decreased at day 10 in diabetic animals. For the most part, MMPs displayed decreased expression levels compared to non-diabetic animals. At day 10, MMPs 8, 9, and 11-16 were all decreased -2 to -35-fold. Many of these remained decreased throughout the experiment. Discussion The experiments performed in the present study were designed to identify the effects of diabetes on arthrofibrosis. Diabetes has been identified as a risk factor for the development of IAC, and diabet ics with frozen shoulder typically have more severe and persistent disease. Diabetes was induced in immunocompetent male Wistar rats using 96

PAGE 97

the drug STZ. Studies have shown the resu lting experimental di abetes from STZ injections has many similarities to type I human diabetes, including hyperglycemia, glucosuria, hypoinsulinemia, hyperlipidemia and weight loss. Animals also exhibit many of the same complications seen in the human disease, such as retinopathy, alterations in angiogenesis, reduced bone formation and cardiovascular disease 158, 159 Using our model of arthrofibrosis and an animal model of diabetes, we examined the differences in the synovium of diabetic animals histologically and by alterations in expression patterns, and compared them to non-diabetic animals Histologically no differences were observed between diabetic and normal synovium. Expression profiles in diabetic joints displayed alterations in ECM and adhesion genes, with many genes showing an overall increase in expression, including the majority of metalloproteinases and many integrins over the course of the 1 20 day experiment. This is consistent with studies in other tissues, including the ki dney, that show increases in MMP and TIMP levels and various collagens, particularly collagen type IV. The elevated levels of expression of metalloproteinases and inhibito rs, collagens and integrins, in the joint synovium of diabetic animals compared to non-diabetic synovium, suggests a reason for diabetic joints to be prone to development of arthrofibrosis. The altered expression patterns that occur in the joint may pre-di spose diabetics to IAC and result in a longer lasting fibrotic response. Our initial hypothesis, that diabetic animals would suffer a more severe and longer lasting fibrotic response in our animal model of arthrofibrosis, proved inaccurate. Histologic comparisons of both diabetic-induced and non-di abetic animals in our TGF1 model, showed a robust fibrotic induction in both sets of animals with a few minor 97

PAGE 98

differences. Overall the fibrotic response ap peared to be very similar in both sets of animals. Non-diabetic animals appeared to have an overall more robust chondrometaplasia that onset earlier and reso lved later; however, diabetic animals had overall less chondrogenic formation thr oughout the experiment as evidenced by histology and expression patterns. Day 90 tiss ues are very similar in both groups of animals and histologic exam ination shows signs of re modeling with a decrease in overall cellularity. By day 120, tissues in both groups show resolution of much of the fibrotic adhesions and the appear ance of the fat pad can be seen on the outer layers of the joint space. While much of the chondr ogenic tissue appears to have resolved in the non-diabetic animals, it is less resolved in diabetic animals and can still be seen in the joint space. Expression data from PCRarrays show few differences between diabetic and nondiabetic animals receiving Ad.TGF1. Consistent with the histology data showing decreased in chondrogenesis in diabetic anim als, at day 10 both collagen type II and cartilage link protein (Hapln1) had decreased expression levels compared to nondiabetic synovium. Decreases in Osteopontin and MMPs -8-16 were also observed at day 10. Only a handful of changes in the ex pression profile are seen at days 30-90; however, by day 120 several genes, incl uding MMP-1, NCAM2, collagen type II and Hapln1 showed increased expression. The decreased cartilage formation observed in diabetic animals with fibrosis is consistent with current liter ature showing delayed fractu re healing and increases in cartilage resorption 47 These studies suggest ther e is an increase in chondrocyte apoptosis and osteoclastogenesis, accelerating the loss of cartilage during fracture 98

PAGE 99

repair 47 Others have suggested that hyperglyce mia leads to increased bone resorption and osteopnia 160 and decreased bone and cartilage formation 161 A study by Kayal et al. comparing diabetic and non-diabetic rats sh owed that while initial cartilage formation was similar, over time ther e was a decrease in bone volume, callus size and cartilage content corresponding with an increase in osteoclasts and an increase in cartilage resorption 47 They found that ADAMSTS 4 and 5, major aggrecanases that degrade cartilage, were higher in diabetic animals, but there were no differences between mRNA levels of MMP13 47 Type I diabetes in humans is associated with a decrease in skeletal mass and a delay in fracture healing 162, 163 Models of Type I diabetes have demonstrated reduced bone turnover and impaired fracture repair and STZ diabetic rats showed abnormal bone repai r to be insulin dependent 164, 165 Diminished expression of osteocalcin, collagen type I and transcripti on factors that regulate osteoblast differentiation were evident in an STZ model of diabetes 166 Altogether, altered expression patterns s een in diabetic joint synovium, combined with complications due to hyperglycemia, ma y predispose diabetics to arthrofibrotic development. Histology and expression data suggest diabetic animals may not necessarily suffer from a more severe case of IAC; however, t he altered signaling and expression profiles in these animals and their elevated circ ulating glucose may affect the resolution. These joints may take a l onger amount of time to resolve than nondiabetic joints. 99

PAGE 100

Table 6-1. Mean of blood glucose in diabetic and normal Wistar rats. Metabolic Parameter Normal Diabetic Blood Glucose (mg/dL) Day 10 102 480 Day 30 110 600 Day 90 110 360 Day 120 111 480 Body Weight (g) Day 10 290 365 Day 30 370 340 Day 90 385 540 Day 120 490 485 100

PAGE 101

Table 6-2. Relative signal values of ECM and adhesion genes in the joints of Diabetic Wistar rats compared to non-di abetic normal Wistar joints Gene Day 10 Day 30 Day 90 Day 120 Structural Proteins Collagen, type 1, alpha 1 -2.8 *-3.4 -1.9 -2.6 Collagen, type 2, alpha 1 1.4 1.1 1.8 *-13.1 Collagen, type 3, alpha 1 *1.7 -1.5 -2.1 -2.4 Collagen, type 4, alpha 1 -1.5 1.1 1.9 1.9 Collagen, type 4. alpha 2 -2.3 1.5 *2.1 2.1 Collagen, type 4, alpha 3 -1.8 -1.2 3.7 *5.4 Collagen, type 5, alpha 1 -1.8 -1.7 -1.2 -1.2 Collagen, type 6, alpha 1 *-3.9 -1.3 2.1 2.0 Collagen, type 8, alpha 1 *-3.2 *-3.3 **2.1 2.2 Hapln 1 1.2 1.1 5.9 -1.2 Versican -1.8 -1.3 2.1 1.8 Matricellular Proteins CCN2 (CTGF) 1.8 1.9 2.6 1.8 Osteopontin 1 1.3 2.4 2.4 -3.8 Periostin -1.5 -1.6 -1.7 -2.0 Sparc -1.1 -1.5 -1.1 -1.1 Spock 1 5.2 2.5 3.0 *6.1 Tenascin-C -4.3 -1.5 4.1 3.1 Thrombospondin 1 1.6 2.1 1.5 2.1 Thrombospondin 2 -1.1 1.2 2.6 2.0 Cell Adhesion Proteins E-Cadherin 3.8 -1.2 -1.1 1.6 N-Cadherin **-12.9 **-31.7 2.0 3.7 P-Cadherin 2.3 -1.0 -1.1 1.1 R-Cadherin *2.9 1.4 1.8 1.7 Catenin, alpha 1 1.6 *2.0 2.1 2.5 Catenin, alpha 2 9.7 1.0 2.0 *3.7 Catenin, beta 1 -1.6 1.6 *2.7 *3.1 Contactin 1 1.6 -1.3 *4.2 *4.3 Icam-1 -1.5 1.7 **3.6 **3.9 Integrin, alpha 2 1.5 -1.2 1.1 1.5 Integrin, alpha 3 **-2.7 1.2 **6.4 **5.0 Integrin, alpha 4 1.6 1.9 1.6 1.5 Integrin, alpha 5 *-1.9 1.0 1.9 1.9 Integrin, alpha D **4.1 1.7 2.0 3.7 Integrin, alpha E **26.7 **10.0 1.6 1.5 Integrin, alpha L 1.8 **4.0 1.9 *2.2 Integrin, alpha M 2.0 *2.9 2.7 3.2 Integrin, alpha V -1.4 1.9 1.5 1.7 Integrin, beta 1 **-61.5 **-13.7 **12.0 *24.4 Integrin, beta 2 1.7 *2.7 2.4 2.3 Integrin, beta 3 1.4 1.4 1.8 1.8 Integrin, beta 4 *-1.8 1.6 **3.0 **3.1 Ncam 1 2.2 1.2 1.1 -1.1 Ncam 2 3.2 1.0 2.0 3.7 101

PAGE 102

Table 6-2. Continu e d Gene Day10 Day 30 Day 90 Day 120 Pecam -1.3 2.1 *3.9 *5.1 E-Selectin **-13.3 **-6.8 *5.2 *15.5 L-Selectin *-7.9 **-17.4 2.0 *4.4 P-Selectin -1.3 1.4 1.4 1.7 Vcam 1 -4.5 1.2 **4.7 **5.6 Vitronectin *-6.4 -1.8 *6.4 **9.6 Extracellular Matrix Proteins Fibronectin 1 **-21.8 *-2.5 **9.2 *11.7 Laminin alpha 1 *17.7 2.9 2.0 3.7 Laminin alpha 2 -2.9 1.3 -1.0 1.1 Laminin alpha 3 1.6 2.0 1.9 1.9 Laminin beta 2 *5.8 3.0 1.5 1.4 Laminin beta 3 -1.2 2.7 1.9 **10.2 Laminin gamma 1 14.5 22.0 2.3 *4.2 Tgfbi 1.3 2.0 2.0 1.9 Metaolloproteinases and Inhibitors Adamts 1 7.0 7.6 1.5 *2.1 Adamts 2 (RGD1565950) **-5.8 *-3.1 *4.0 *4.0 Adamts 5 -1.2 1.5 **9.5 **20.8 Adamts 8 *-4.8 **-18.7 1.7 3.1 Mmp 1a 16.1 1.3 2.0 3.7 Mmp 2 1.3 1.1 2.6 2.1 Mmp 3 2.0 3.1 2.1 2.7 Mmp 7 *20.8 3.3 1.7 3.2 Mmp 8 *12.2 1.5 2.2 3.8 Mmp 9 1.6 -3.4 3.7 -1.2 Mmp 10 21.0 1.2 1.7 3.0 Mmp 11 1.9 1.6 *3.0 2.6 Mmp 12 *22.9 11.7 -1.6 -3.5 Mmp 13 -3.3 1.4 21.9 4.7 Mmp 14 *-3.8 -1.1 4.1 4.3 Mmp 15 **-10.9 *-2.9 **8.3 *9.9 Mmp 16 -1.3 -2.2 1.3 1.1 Timp 1 **13.5 **8.1 2.4 2.9 Timp 2 **10.1 **35.1 *5.7 *8.0 Timp 3 1.2 1.9 2.0 *2.6 Elastic Fiber Proteins Emilin 1 1.5 6.7 *5.6 *7.0 Fibulin 1 1.0 1.1 *3.6 *3.0 Extracellular matrix protein 1 3.1 2.1 3.0 2.7 Other Synaptotagmin I 3.4 1.4 2.1 3.7 CD44 Antigen *-3.3 -2.7 1.8 *3.1 Sarcogycan 1.4 1.7 2.8 *3.1 NTPDase-1 (CD39) -1.1 *2.0 *3.3 *3.5 Values with an asterisk represent p<0.05. Values with two asterisks represent p<0.01. n=3 for each time point. 102

PAGE 103

Table 6-3. Relative signal values of ECM and adhesion genes in the joints of Diabetic Wistar rats receiving Ad.TGF1 compared to untreated diabetic Wistar rats Gene Day 10 Day 30 Day 90 Day 120 Structural Proteins Collagen, type 1, alpha 1 **5.1 **9.7 **2.7 2.9 Collagen, type 2, alpha 1 -1.3 *57.7 -1.1 **171.3 Collagen, type 3, alpha 1 1.4 **3.4 1.4 *1.8 Collagen, type 4, alpha 1 -1.2 *-1.7 **-3.9 -*2.9 Collagen, type 4. alpha 2 1.8 -1.6 **-5.1 **-3.1 Collagen, type 4, alpha 3 -2.3 *-2.5 -1.3 1.0 Collagen, type 5, alpha 1 **2.5 **2.6 -1.5 1.2 Collagen, type 6, alpha 1 *5.4 6.9 -1.7 -1.0 Collagen, type 8, alpha 1 **12.7 **9.6 1.0 1.6 Hapln 1 1.5 **30.3 1.8 **25.5 Versican *4.6 **4.3 -1.0 *2.8 Matricellular Proteins CCN2 (CTGF) -1.1 -1.1 1.4 2.5 Osteopontin 1 *8.9 *36.9 9.2 *153.6 Periostin **2.6 *2.6 -1.6 -1.5 Sparc 1.1 2.0 *-1.9 -1.4 Spock 1 -3.5 1.3 4.6 3.0 Tenascin-C *14.0 **4.9 -1.7 1.6 Thrombospondin 1 **3.0 *3.7 1.4 *2.3 Thrombospondin 2 **4.6 **4.8 1.0 *2.1 Cell Adhesion Proteins E-Cadherin 1.0 **4.1 3.2 2.3 N-Cadherin *-2.6 *3.6 1.9 -1.3 P-Cadherin -5.3 1.1 2.3 3.5 R-Cadherin **-17.1 -2.3 -1.2 1.3 Catenin, alpha 1 -2.0 **-2.2 **-3.5 *-2.1 Catenin, alpha 2 *-11.2 **6.5 **15.3 5.7 Catenin, beta 1 1.7 -1.1 **-3.4 -1.9 Contactin 1 **-3.4 -1.1 -3.4 *-1.9 Icam-1 1.9 -1.3 -3.4 *-1.9 Integrin, alpha 2 -1.8 1.0 -1.0 2.1 Integrin, alpha 3 -1.2 **-3.4 *-6.1 -1.9 Integrin, alpha 4 1.4 1.0 *-2.9 -1.4 Integrin, alpha 5 1.9 1.1 *-2.7 -1.3 Integrin, alpha D **-6.2 *-1.9 **-2.9 -2.8 Integrin, alpha E 1.8 **6.7 2.7 4.3 Integrin, alpha L **3.0 -1.0 *-3.7 **-2.4 Integrin, alpha M 1.1 -1.1 -3.0 -2.2 Integrin, alpha V -1.3 *-2.1 *-2.6 *-1.6 Integrin, beta 1 7.1 **3.1 *2.8 -1.3 Integrin, beta 2 -1.3 -1.6 *-3.5 -1.8 Integrin, beta 3 -1.2 1.0 **-3.6 -1.4 Integrin, beta 4 1.4 -1.9 **-4.1 -1.5 103

PAGE 104

Table 6-3. Continued Gene Day 10 Day 30 Day 90 Day 120 Ncam 1 **2.9 **7.0 3.2 **4.1 Ncam 2 **-7.1 1.2 *4.3 2.3 Pecam 1.1 *-1.9 *-1.7 -1.6 E-Selectin 7.4 *3.2 3.0 1.4 L-Selectin 1.5 1.4 1.1 -2.2 P-Selectin 3.4 1.6 -2.6 -1.6 Vcam 1 2.9 -1.3 **-2.9 -3.4 Vitronectin -1.2 -3.3 -1.3 -2.2 Extracellular Matrix Proteins Fibronectin 1 5.9 1.2 -1.8 -1.1 Laminin alpha 1 *-8.6 2.9 2.6 2.1 Laminin alpha 2 5.3 *-1.6 -1.0 1.3 Laminin alpha 3 *-3.2 *-2.7 -2.2 -1.1 Laminin beta 2 **-7.9 **-2.9 *-1.9 1.2 Laminin beta 3 **-13.6 **-25.8 1.2 *-5.8 Laminin gamma 1 1.3 1.0 1.1 -2.8 Tgfbi **2.0 *1.6 *-2.5 -1.4 Metaolloproteinases and Inhibitors Adamts 1 1.8 **2.1 **-2.7 -1.4 Adamts 2 (RGD1565950) *5.1 **3.9 1.3 1.7 Adamts 5 -1.4 -1.5 -1.4 -2.1 Adamts 8 2.3 **21.7 5.4 3.9 Mmp 1a *-17.5 2.6 *8.2 4.2 Mmp 2 -1.4 -1.2 *-3.2 -1.4 Mmp 3 1.6 **5.3 *1.8 1.3 Mmp 7 *-12.4 -1.1 1.8 3.2 Mmp 8 -3.1 5.1 5.1 3.0 Mmp 9 1.7 *52.5 3.8 **14.3 Mmp 10 -18.3 *6.4 **14.5 5.1 Mmp 11 1.5 *2.9 -1.2 *2.4 Mmp 12 2.2 20.5 **5.7 **23.2 Mmp 13 **38.1 *40.4 2.1 **6.5 Mmp 14 *7.4 **3.0 -1.3 -1.1 Mmp 15 1.5 -1.6 *-5.2 *-2.1 Mmp 16 *2.0 **6.6 1.5 **3.8 Timp 1 1.4 **3.4 1.5 1.4 Timp 2 *3.0 1.1 -1.7 -1.6 Timp 3 -2.9 **-3.6 **-3.1 *-1.9 Elastic Fiber Proteins Emilin 1 **19.5 **4.6 2.4 1.9 Fibulin 1 19.5 4.6 2.4 1.9 Extracellular matrix protein 1 -2.4 -1.2 1.0 1.5 Values with an asterisk represent p<0.05. Values with two asterisks represent p<0.01. n=3 for each time point. 104

PAGE 105

Table 6-4. Relative signal values of ECM and adhesion genes in the joints of Diabetic Wistar rats receiving Ad.TGF1 compared to non-diabetic Wistar rats receiving Ad.TGF1 Gene Day 10 Day 30 Day 90 Day 120 Structural Proteins Collagen, type 1, alpha 1 *-2.6 -1.2 1.3 -1.9 Collagen, type 2, alpha 1 **-326.5 1.3 -14.2 3.2 Collagen, type 3, alpha 1 -1.1 1.2 1.8 -1.3 Collagen, type 4, alpha 1 -1.3 1.3 1.4 1.1 Collagen, type 4. alpha 2 -1.4 1.3 1.4 1.2 Collagen, type 4, alpha 3 -2.0 -1.2 -1.0 1.5 Collagen, type 5, alpha 1 **-1.7 -1.3 1.1 -1.2 Collagen, type 6, alpha 1 -1.7 3.5 1.3 -1.0 Collagen, type 8, alpha 1 *-3.6 -1.7 -1.2 1.0 Hapln 1 **-57.7 1.5 -3.0 *2.6 Versican -1.4 -1.8 -1.3 1.3 Matricellular Proteins CCN2 (CTGF) -1.3 -1.2 1.7 *1.7 Osteopontin 1 *-9.7 1.8 -4.7 1.9 Periostin **-3.0 -1.6 1.3 *-3.6 Sparc **-1.8 -1.2 1.2 -1.3 Spock 1 1.1 -1.9 1.1 1.6 Tenascin-C -1.5 *1.7 -1.0 1.3 Thrombospondin 1 *-2.6 -1.5 -1.9 1.4 Thrombospondin 2 **-2.4 -1.3 -1.1 -1.0 Cell Adhesion Proteins E-Cadherin -1.1 -1.7 -1.2 2.1 N-Cadherin **-10.5 -1.3 -1.0 1.3 P-Cadherin -4.1 -2.7 1.2 1.6 R-Cadherin 3.96 -2.4 1.2 1.6 Catenin, alpha 1 -1.1 -1.2 1.2 1.1 Catenin, alpha 2 2.2 -3.0 1.5 4.5 Catenin, beta 1 *-1.9 -1.2 -1.1 -1.0 Contactin 1 1.4 -1.3 1.6 -1.0 Icam-1 1.4 1.2 -1.1 -1.2 Integrin, alpha 2 -1.4 -1.3 -1.1 1.8 Integrin, alpha 3 1.2 1.2 -1.2 1.3 Integrin, alpha 4 1.1 1.1 -1.3 -1.2 Integrin, alpha 5 -1.4 1.0 1.0 -1.0 Integrin, alpha D 1.1 -2.4 1.1 -1.3 Integrin, alpha E 1.7 1.1 1.6 -1.0 Integrin, alpha L 1.7 1.2 1.1 -1.1 Integrin, alpha M 1.3 1.5 1.0 -1.4 Integrin, alpha V *-2.2 -1.1 1.2 1.0 Integrin, beta 1 -2.5 1.1 1.2 -1.3 Integrin, beta 2 1.4 1.4 1.2 1.1 Integrin, beta 3 -2.8 1.3 -1.2 1.1 Integrin, beta 4 1.0 1.1 -1.0 1.3 Ncam 1 **-2.8 -1.4 1.4 -1.7 105

PAGE 106

Table 6-4. Continued Gene Day 10 Day 30 Day 90 Day 120 Ncam 2 1.5 -3.8 1.8 4.0 Pecam -1.3 -1.3 1.1 -1.3 E-Selectin -1.3 -1.4 1.4 -1.7 L-Selectin 1.2 -3.8 1.8 4.0 P-Selectin 1.8 -1.3 1.1 *-1.3 Vcam 1 1.2 1.3 1.1 -1.2 Vitronectin -1.2 1.1 1.2 1.3 Extracellular Matrix Proteins Fibronectin 1 -1.6 1.7 1.4 *2.0 Laminin alpha 1 -1.1 -2.1 -1.3 1.5 Laminin alpha 2 1.6 -1.7 1.5 -1.2 Laminin alpha 3 -1.4 -1.3 1.1 1.8 Laminin beta 2 1.5 -1.3 1.6 1.5 Laminin beta 3 -1.5 -1.4 1.8 -1.1 Laminin gamma 1 -8.4 1.2 1.3 -1.7 Tgfbi **-2.3 -1.3 1.4 -1.0 Metaolloproteinases and Inhibitors Adamts 1 1.2 1.2 -1.4 -1.3 Adamts 2 (RGD1565950) -1.9 -1.3 1.5 -1.0 Adamts 5 1.6 1.3 1.3 1.3 Adamts 8 1.2 3.0 -1.1 -1.0 Mmp 1a 1.9 -6.1 1.4 5.0 Mmp 2 -1.3 -1.4 1.6 -1.4 Mmp 3 1.3 *-2.4 1.1 -1.5 Mmp 7 3.4 *-3.3 1.0 -2.3 Mmp 8 **-34.9 *-3.2 -2.4 -1.1 Mmp 9 **-24.4 1.3 -1.1 -1.6 Mmp 10 -1.4 -2.7 1.4 1.5 Mmp 11 *-3.4 -1.2 1.4 1.4 Mmp 12 *-9.0 1.5 *-2.4 *-9.4 Mmp 13 *-9.5 1.1 -2.3 *-5.0 Mmp 14 *-3.4 -1.1 1.2 -1.5 Mmp 15 -2.4 1.1 -1.2 1.4 Mmp 16 **-2.7 -1.5 1.0 1.3 Timp 1 -1.3 *-1.8 1.4 -1.0 Timp 2 *-2.5 1.1 1.4 1.2 Timp 3 -1.1 1.2 1.3 1.8 Elastic Fiber Proteins Emilin 1 -4.1 -1.1 1.8 -1.1 Fibulin 1 -1.3 -1.6 1.2 -1.3 Extracellular matrix protein 1 1.0 1.4 2.0 1.3 Other Synaptotagmin I 1.5 -2.7 -1.0 4.2 CD44 Antigen 2.0 -1.0 1.0 1.1 Sarcogycan -1.2 -1.3 1.4 1.2 NTPDase-1 (CD39) -1.3 *2.8 1.1 -1.1 Values with an asterisk represent p<0.05. Values with two asterisks represent p<0.01. n=3 for each time point. 106

PAGE 107

Table 6-5. Arbitrary scale measuring severity of fibrosis by a blinded observer. Scale represents 0 (Normal)-4 (severe fibrosis). Ad.TGF1 Treated Scale (0-4) Non-Diabetic 0.0 Control Diabetic 0.0 Non-Diabetic 2.5 Day 10 Diabetic 2.25 Non-Diabetic 4.0 Day 30 Diabetic 3.0 Non-Diabetic 2.75 Day 90 Diabetic 2.75 Non-Diabetic 2.25 Day 120 Diabetic 2.25 107

PAGE 108

Figure 6-1. Glucose stimulates rat synovial fi broblast migration. Migration into the mock wound space is seen in the presence of Ad.CTGF and Ad.TGF1 alone. Dglucose stimulates migration into the wound space on its own, with 25mM having the greatest effect. 108

PAGE 109

Figure 6-2. Glucose stimulates rat synovial fi broblast migration. Migration into the mock wound space is enhanced in the presence of Ad.CTGF and Ad.TGF1 alone. D-glucose stimulates migration into the wound space on its own, with 25mM having the greatest effect. Values are mean SEM. (O ne-way ANOVA, ** P < 0.01). 109

PAGE 110

Figure 6-3. Capsular fibrosis induced by Ad.TGF1 in diabetic animals. Knees of diabetic and non-diabetic Wistar rats receiving Ad.TGF1 were harvested at various time points, decalcified and pr ocessed for histology. Sections were stained with H&E. At 10 days post injecti on, an expansion of fibroblastic cells from the synovial lin ing generated the bulk of the fibrotic mass, occluding the adipose layer in both groups and the formation of chondrocytic cells can be seen within the fibrotic tissue in non-diabetic animals. By day 30, the majority of fibroblastic cells had differentiated into chondrocytes in non-diabetic animals. Day 90 tissues are similar in both sets of animals, showing signs of resolving. Both show signs of c hondrogenic formation. At day 120, both groups have resolved and display an ov erall decrease in cellularity and visualization of the fa tpad; however, diabetic tissues appear to be less resolved and much of the chondrogenic tissue remains. Images in left two columns are at a magnification of 2.5x and images in the right columns are at a magnification of 20x. 110

PAGE 111

CHAPTER 7 ANTISENSE-OLIGONUCLEOTIDES LESSEN SEVERITY OF JOINT FIBROSIS WHEN DELIEVERED AT ONSET OF DISEASE Introduction Anti-fibrotic strategies have increased importance and relevance in many tissues throughout the body, including the joint. Arthrofi brosis, in particular IAC, is a complex disease of unknown origin that has not been fully characterized. Although it is known to be a fibrotic condition in the synovium of the joint, the mechanisms of onset are not well defined. Treatment options are limited and range from physi cal therapy to massive surgical release. Given the prevalence of IAC, identification of an effective fibrotic modulator would be beneficial in treatment of a significant portion of the population. There is widespread evidence that enhanced synthesis of TGF1 leads to scarring and fibrosis in a variety of tissues throughout the body. In earlier chapters (Chapters 4-6) we demonstrated t he potency of overexpression TGF1 on the tissues of the joint and its ability to induce an arthrofibrotic event in rodents. TGFhas been observed in a variety of fibr otic disorders, thus, TGFhas been the target of many therapeutic approaches designed to treat fibrosis. However, due to its pleiotropic nature and many beneficial effects on cells, system ic treatments could have deleterious effects. Along with TGF, CTGF has also been observed in a variety of fibrotic disorders. Although not found to be highly expressed at the mRNA level in our model of arthrofibrosis (see Tables 4-1 and 51), it has been shown to modulate and enhance many of the effects of TGF, and could prove to be a useful and specific target for modulating fibrosis seen in the joint. 111

PAGE 112

Antisense oligonucleotides (ASO) coul d be a promising therapeutic strategy. ASOs have been used with some success in other diseases, including cancer and viral infections 167, 168 In most renal cell ty pes, treatment with a TGF1 antagonist ASO 59 reduces the rise in ECM expression due to hi gh glucose. ASOs are short, single strands of DNA or RNA that are complementar y to a particular sequence and upon specific hybridization with mRNA, inhibit gene expression. While ASOs were first described in 1978, modifications to their structure over th e years have led to improvements in their pharmacokinetics and pharmacodynamics 169 ASOs recognize and hybridize to target mRNA by Watson-Crick base pairing and trigger RNase Hmediated RNA destruction. First generat ion ASOs lack the nuclease resistance and tissue stability necessary for therapy. Second generation ASOs have been reengineered, improving potency, nuclease resistance, and tissue half-life. Second generation oligonuc leotides used in this study have been modified to contain 2O-Methoxy ethyl (MOE) and phosphorothioate, which has been shown to enhance stability in vivo Naturally occurring oligonucleotides are highly susceptible to rapid degradation by cellular nucleases. W hen the phoshodiester bond is modified by replacing one of the non-bridging oxygens with sulfur, there is a greater resistance to nuclease degradation. Because the 2-MOE modification does not allow recruitment of RNase H when bound to mRNA, only the termi nal residues of t he ASO sequence are 2MOE-modified, leaving the central r egion with only the phosphorothioate backbone modification, thereby coupling the resistanc e to degradation in tissues with improved potency while supporting activation of RNase H 170 112

PAGE 113

With the use our arthrofibrotic animal model, based on the overexpression of TGF1 in the joint space (Chapters 4-6), we eval uated the function of ASOs specific to TGF1 and CTGF in their ability to lim it joint fibrotic induction. Results TGFKnockdown by Specific ELISA As shown in Chapters 4-7, overexpression of Ad.TGF1 in the joints of rodents, leads to a pathology similar to that seen in human arthrofibrosis. Consequently, we hypothesized that treatment with ASOs specific to TGF1 and CTGF would reduce the severity of fibrosis seen in our ani mal model of IAC. Prior to undertaking in vivo experimentation we tested the effectiveness of TGF1 ASO in vitro Twelve-well plates of rat synovial fibroblasts (RSFs) were grow n to confluence prior to treatment with ASO. TGF1 ASO at 100nM (0.825 g/ml) and 300nM (2.47 g/ml) was transfected into the appropriate wells using lipof ectamine transfection reagent. Twenty-four hours post transfection cells were infected with Ad.TGF1 at a dose of 1.3 x 10 7 viral particles (vp). The culture medium was replaced with serum free medium 24 hours after transduction, and harvested forty-eight hours pos t-transduction for specific ELISA. As seen in Figure 7-1 knockdown of TGFexpression was observed after treatment with both the 100nM and 300nM concentrations of TGF1 ASO compared to Ad.TGF1 alone. Intra-articular Injection of Reporter ASO is Retained in the Joint Space After confirming the ability of the ASO to knockdown of TGFexpression in cultured cells, we worked to determine the fu nctionality of ASO delivery intra-articularly in the Ad.TGF1 fibrosis model in vivo We first examined their distribution within the 113

PAGE 114

joint following intra-articular injection of a reporter ASO generously provided by Dr. Gregory Schultz and Dr. Nick Dean. Approximately 10 g/ l of reporter ASO was diluted in 50 l of PBS and injected into the joint spac e of adult male Wist ar rats. 10 days post injection, the animals were killed and joint tissues were harvested and processed for further immunohistochemical analysis using ant ibodies specific to the reporter ASO. Antibody staining demonstrated the reporter ASO was retained within the joint space and its detection was seen throughout the fibrous synovium and fat pad and radiated outward from the joint space ( Figure 7-2 ). IHC staining was not observed outside the joint space. ASOs were Effective at Inhibiting Arthrofibrosis in Animal Model Knowing the relative location of the ASO s within the joint and their ability to knockdown TGFexpression in rat synovial fibrobl asts in culture, we wanted to determine their ability to inhibit fibrosis wit hin our animal model. For this, animals were divided into 6 categories: (1) normal controls receiving PBS, (2) animals receiving 5.0 x 10 7 vp of Ad.TGF1 alone, (3) animals receiving 5.0 x 10 7 vp of Ad.TGF1 and 10 g/ l TGF1 ASO, (4) animals receiving 5.0 x 10 7 vp of Ad.TGF1 and 10 g/ l CTGF ASO, (5) animals receiving 5.0 x 10 7 vp of Ad.TGF1, 10 g/ l TGF1 ASO and 10 g/ l CTGF ASO, (6) and finally those animals receiving 5.0 x 10 7 vp of Ad.TGF1 and 10 g/ l scramble ASO. At 5 days post injection, animals were killed and tissues were harvested and processed for histology. H&E staining differenc es in fibrotic severity between joints treated with ASOs and untreated joints. Normal control animals show a thin layer of synovial cells, overlying layer of adipose. In animals treated with Ad.TGF1 alone, the 114

PAGE 115

joint space was filled with a thick, hyper cellular tissue populated primarily by elongated fibroblastic cells expandi ng into the adipose layer ( Figure 7-3 ). Upon co-injection with the TGF1 ASO, and slightly less so with CTGF ASO, a reduction in the cellularity and overall fibrotic development was seen, result ing in less displacement of adipose tissue. When the oligos were co-injected together ( Figure 7-3 ), results were similar to TGF1 ASO alone. There was a reduction in the overall fibrotic induction, such that the fibrotic expansion was minor and did not over-ride the joint space. As expected, no differences were observed between the scramble ASO and Ad.TGFalone (Figure 7-3). Discussion The second generation ASOs selected for these studies appeared to effectively target TGF1 expression in vitro and reduce TGF1 mediated fibrosis in vivo Knockdown activity disappeared when a scram bled ASO was tested, indicating this inhibition occurs through a sequence spec ific hybridizationdependent mechanism. Using the reporter oligonucleotide we showed that these molecule s permeate the joint tissues and were largely contained within and along the joint space. Interestingly, the oligos were detected up to 10 days pos t injection. ASOs specific for TGFand CTGF resulted in a decrease in develop ment of fibrosis in our ar throfibrotic animal model. An antisense-based approach to arthrofibrosis, in particular IAC, is an intriguing alternative to current therapies, which are based on physical manipulation and surgical release of adhesions. Targeting mRNA using ASOs is an appeali ng approach to study cell signaling, since an inhibitor can be designed for any gene for which a partial sequence in available. Cell and tissue type must be optimized to ensure cellular uptake of ASO, as this can vary. Prior to bei ng considered a therapeut ic agent, important 115

PAGE 116

considerations regarding relevance of animal models for human disease must be considered, as well as characterization of potency, tissue distribution, administration and dosing, and clearance and metabolism al ong with various other parameters. Clinical experiences with second generation oligonucleotides, such as the ones used here, have thus far revealed little dose-limiting toxicity 171 Additionally, ASOs can be repeatedly administered, as neutralizing antibody responses are not mounted against ASOs. Clinical trials have been tested in patients with Crohns disease using an ASO targeting ICAM-1, which proved to be well tolerated and capabl e of producing longlasting disease remissions 168 However promising these init ial experiments in our animal model of arthrofibrosis appear, many more rigorous tests need to be performed to fully examine the success of ASOs in our model. Preliminary data shown here examines inhibition of fibrotic induction by simultaneous injection of each ASO and Ad.TGF1 into the knee joints of Wistar rats. TGF1 blockade was effective at ame liorating the induction of acute fibrosis, as was CTGF blockade although to a lesser extent. However, since patients are typically not seen by physicians until they are experiencing pain and stiffness (Stages 2 or 3 of IAC), adhesions and scar ti ssue have already formed within their joint capsule. Therefore it woul d be prudent to check the effectiveness of ASOs at varying levels of fibrotic severity, including progre ssion and resolution. For example, it would be necessary validate ASOs in our animal model at onset of fibrosis (similar to Stage 1 in humans), at a frozen state (similar to Stage 3 and 4) and even determine if it enhances the rate of resolution of joint fibrosis While simultaneous injection of either TGFor CTGF ASO with our fibrotic stimulator (Ad.TGF1) appears to lessen the severity of 116

PAGE 117

fibrosis, more replicates and further exper iments need to be performed to tighten this preliminary data and determine fibrotic knockdown at each stage of disease. 117

PAGE 118

Table 7-1. Arbitrary scale in which a bli nded observer measured severity of fibrosis. Scale represents 0 (normal)-4 (severe fibrosis). Substance Intra-articularl y Injected Scale (0-4) Ad.TGF1 Alone 3.5 Ad.TGF1 with TGF1 ASO 1.5 Ad.TGF1 with CTGF ASO 2.5 Ad.TGF1 with both TGF1 ASO and CTGF ASO 1.5 Ad.TGF1 with scramble ASO 3.0 Normal Control 0.0 118

PAGE 119

Figure 7-1. Transgene expression following transduction of rat synovial fibroblasts with Ad.TGF1 and/or transfection with an antisens e oligonucleotide specific for TGF. Following isolation of fibroblasts from rat synovium, the cells were grown in monolayer in 12-well plates and transfected with either 100nM or 300nM TGF1 ASO using Lipofectamine. At 24hrs post transfection, cells were transduced with 1.3 x 10 7 vp of Ad.TGF1. The next day, medium was replaced by 0.5 ml of serum-free m edium. At 48hrs post transduction, the conditioned medium was harvested and TGF1 content measured by specific ELISA. Results are expressed as the mean of 3 replicates. Error bars represent + S.E.M. (One-way ANOVA, ** P < 0.01). 119

PAGE 120

Figure 7-2. Immunohistochemical staining of reporter anti-s ense oligonucleotide. Knee joints of Wistar rats intr a-articularly injected with 10 g/ l of reporter ASO were harvested at day 10, paraffin embedded, sectioned and immunologically stained for the presence of reporter ASO. 120

PAGE 121

Figure 7-3. Capsular fibrosis induced by Ad.TGF1 in Wistar rats is inhibited by cointra-articular injection of ASOs. Knees of Wistar rats receiving Ad.TGF1 alone (a), Ad.TGF1 and TGF1 ASO (b), Ad.TGF1 and CTGF ASO (c), Ad.TGF1, CTGF ASO and TGF1 ASO (d), Ad.TGF1 and scramble ASO (e), or PBS control (f) were harvested 5 days post injection, decalcified and processed for histology. Sections were stained with H&E. Images are at a magnification of 20x. 121

PAGE 122

CHAPTER 8 SUMMARY AND FUTURE DIRECTIONS Arthrofibrosis is a condition that arises from the development of excess fibrous tissue intra-articularly, which leads to chr onic joint pain and loss of range of motion. The prevalence of arthrofibrosis has been sited to occur in up to 35% of the population, and has been described in joints throughout the body 83, 84, 85, 86 In an effort to establish an animal model of IAC, an arthro fibrotic condition affecting ~3% of the population, and develop an understanding of the cellular and molecular events contributing to arthrofibrosis, we used an adenovirus to deliver and over-express TGF1 cDNA (Ad.TGF1) in the knee joints of athymic nude rats and immunocompetent Wistar rats. We were able to both model fibrotic induction and resolution in the joints of Wistar rats and obtain an expression profile of ECM genes and their altered signaling patterns. We initially hypothesized t hat prolonged release of TGF1 was responsible for initiating and maintaining t he fibrotic phenotype; however, our data do not necessarily support this. Rather, it appears that acute overexpression of TGF1 is sufficient to induce a very severe fibrosis. Its maintenan ce is not reliant on the sustained presence of TGF1, but appears to arise from a lack of mechanisms that remove the collagenous ECM that has been deposited. Indeed, over time, we observe a reduction in the cellularity of the fibrotic mass, but the fibrous matrix still rema ined. Without a resident or infiltrating cell population to effect degradat ion of the fibrotic ECM, there is no mechanism in place for its remova l and this is destined to persist. In this regard, treatment options would appear most effective when administered prophylactically to high risk pati ents prior to the synthesis of constricting fibrotic matrix. This scenario would be applicable in both the fibrosis arising post knee surgery or from 122

PAGE 123

IAC. Preliminary experiments examining th e effects of ASO technology in our animal model of fibrosis proved to be modestly successful. Co-injection of ASOs and Ad.TGF1 resulted in knock down of fibrotic induction compared to Ad.TGF1 alone. However, more experiments need to be performed in or der to determine the usefulness of ASOs as a potential therapy for IAC. Examining their function when injected into the joint at varying levels and stages of fibrosis w ould be ideal, as patients with IAC will not typically see their physician until they are in the middle to late stag es of IAC. Currently the only therapy available to patients with IAC is physical manipulation under anesthesia and surgical release by arthrosc opy. This technology, if successful, could provide a much needed alternat ive treatment for IAC. Concerning the molecules that contribute to fibrosis, a widely held opinion is that the activities of TIMPs become dominant over proteolytic enzymes, and this enables the build-up of excess ECM associated with fibros is. Our data, however, strongly contradict this idea. Rather it appears t hat fibrosis is very heavily dependent upon the activities of the MMPs. These molecules appear necessary to allow the dramatic expansion, motility and overgrowth of fibroblasts, as well as des truction of pre-existi ng matrix. It appears that the MMPs, in conjunction with matricellu lar proteins, are the primary mediators in the induction of fibrosis and that these mole cules offer strong targets for intervention. Indeed these molecules are expressed onl y during tissue growth and remodeling, allowing drugs to selectively target their ac tivities without risk to normal adult tissues. Somewhat surprisingly, despite the dramatic effects of TGF1 on the joint tissues, expression of CCN2/CTGF remained relati vely unchanged. CCN2/CTGF is thought to be a major activator of TGFactivity and mediator of fibrosis in other tissues 21, 118, 119 123

PAGE 124

Initially, we believed that CCN2/CTGF would pl ay a larger role in the development of joint fibrosis. However injection of Ad.CTG F resulted in little change to the joint space and expression analysis showed little difference in our model. Its relatively low induction here suggests that CCN2/CTGF may not play a ma jor role in arthrofibrosis, and that its absence may be compensated for by other mem bers of the matrice llular proteins. Although our data provide a severe representation of joint fibrosis, the pathologies observed are consistent with those seen clinically 98 Procedures involving the knee, are particularly vulnerable to the dev elopment of this type of fibrosis 98 Histological examination of tissues recove red from fibrotic knee joints following surgical release frequently identifies fibrosis, vascular pr oliferation and synovia l chondrometaplasia 99 In primary synovial chondr omatosis, metaplastic hyaline cartilage expands the subsynovial connective tissue. While histologically pr imary synovial chondromat osis appears similar to chondrosarcoma, it is a benign condition that is typically self-limiting and true malignant transformation is rare. In many ca ses the fibro-chondrogenic tissues also contain endochondral bone formations 100-102 The similarities between the tissue phenotype of the rat TGF1 overexpression model shown here and human arthrofibrosis, indicate that the data generat ed in the rat knee have relevance to human conditions. Articular cartilage has a poor ability to self-regenerate and is frequently damaged as a result of trauma or pat hological situations, such as osteoarthritis and rheumatoid arthritis; therefore, a method of repairing articular cartilage is of clinical relevance. Since articular cartilage has little capacity for self repair, a variety of cell based approaches are being investigated. Currently, chondrocytes isolated from healthy 124

PAGE 125

cartilage and mesenchymal stem cells from bone marrow, fat, and other tissues are being tested in animal models. Our data in Chapter 4 demonstrate the enormous proliferative potential of fibr oblastic cells resident in the joint capsule and synovium. Moreover, these cells appear predisposed to differentiation along a chondrogenic lineage. These cells appear to offer dist inct advantages for r egenerative applications, relative to progenitor cells isolated from bone marrow or adipose tissue that may be fated toward osteoblastic or adipogenic pathw ays respectively or to chondrocytes isolated from a very limited supply of healthy non-weight bearing cartilage. Our findings further demonstrate that delivery of the growth factor TGF1 stimulates local prolifer ation and induces chondrogenesis. Our observations are consistent with data obtained from cultures of both embryoni c and adult fibroblasts that are able to undergo differentiation into a chondrogenic phenotype in the presence of chondrogenic inducers 172 and other results which show that TGF1 is able to induce chondrogenic differentiation in mesenchymal progenitors 173, 174 A phase I clinical trial is being presented using an ex vivo therapeutic approach to initiate cartilage repair by infecting primary chondrocytes with a virus modified to express TGF1 and then injecting these TGF1 expressing chondrocytes into the knee joint (TissueGene, Gaithersburg, MD). From the data we have presented, the application of TGF1 as a chondrogenic agent in vivo either as a recombinant protein or transgene product, should be explored with extreme c aution. Our results serve dramatize the sensitivity of connective tissue fibroblasts to growth factor stimulation, and that protocols designed to induce cellular differentiati on in cartilage and bone repair in vivo should be aware of the high capacity for toxic side effects of thes e pleiotropic agents in adjacent tissues. 125

PAGE 126

While the animal model of joint fibrosis described in Chapter 5 resembles the phenotype of IAC and shows signs of resolvi ng, comparing the expression profiles of our model of arthrofibrosis to what occurs in human patients should provide a valuable insight into the signaling of this diseas e and, perhaps, highlight potential targets for therapy. To enable a comprehensive description of the biology of IAC at the molecular level, we will use microarray analyses to determine the differences in global transcription patterns between normal capsular tissues and those from patients with IAC. Capsular tissues from affected join ts have been recovered fr om patients with IAC and to allow for comparative controls, tissue from the glenohumoral capsule was similarly collected from surgical patients wit hout IAC. These tissue samples collected from patients with and without IAC, are currently being analyz ed using Agilent one-color microarrays. By determining the genes and gene classes that are over and under expressed in IAC and comparing those with the biological analyses performed here, a clearer understanding of the pathogenesis and progression of the disease should emerge with respect to inflammation, fi brosis, fibrotic signaling, and TGF1. It would be of interest to measure endogenous TGFexpression in the long term fibrotic model to determine its levels of activity du ring the remodeling process and compare to expression data obtained in hum ans. These studies should provide direct information about the profibrotic condition in IAC and perhaps allow insight into its relationship to other fibrotic, autoimmune or hyperplas tic conditions, such as scleroderma. 126

PAGE 127

LIST OF REFERENCES 1. Hannafin, J. A. & Chiaia, T. A. A dhesive capsulitis. A treatment approach. Clin Orthop Relat Res, 95-109 (2000). 2. Dias, R., Cutts, S. & Ma ssoud, S. Frozen shoulder. Bmj 331 1453-6 (2005). 3. Siegel, L. B., Cohen, N. J. & Gall, E. P. Adhesive c apsulitis: a sticky issue. Am Fam Physician 59, 1843-52 (1999). 4. van Royen, B. J. & Pavlov, P. W. Treat ment of frozen shoulder by distension and manipulation under local anaesthesia. Int Orthop 20, 207-10 (1996). 5. Binder, A. I., Bulgen, D. Y., Hazleman, B. L. & Rober ts, S. Frozen shoulder: a long-term prospective study. Ann Rheum Dis 43, 361-4 (1984). 6. Bulgen, D. Y., Binder, A. I., Hazleman, B. L., Dutt on, J. & Roberts, S. Frozen shoulder: prospective clinical study with an evaluation of three treatment regimens. Ann Rheum Dis 43, 353-60 (1984). 7. Placzek, J. D. et al. Long-term effectiveness of tr anslational manipulation for adhesive capsulitis. Clin Orthop Relat Res 181-91 (1998). 8. Helbig, B., Wagner, P. & Dohler, R. M obilization of frozen shoulder under general anaesthesia. Acta Orthop Belg 49, 267-74 (1983). 9. Rhind, V., Downie, W. W. Bird, H. A., Wright, V. & Engler, C. Naproxen and indomethacin in periarthr itis of the shoulder. Rheumatol Rehabil 21 51-3 (1982). 10. Binder, A., Hazleman, B. L., Parr, G. & Roberts, S. A controlled study of oral prednisolone in frozen shoulder. Br J Rheumatol 25, 288-92 (1986). 11. de Jong, B. A., Dahmen, R., Hogeweg, J. A. & Marti, R. K. Intra-articular triamcinolone acetonide injection in patients with capsulitis of the shoulder: a comparative study of two dose regimens. Clin Rehabil 12 211-5 (1998). 12. Quigley, T. B. Indications for Manipul ation and Corticosteroids in the Treatment of Stiff Shoulders. Surg Clin North Am 43, 1715-20 (1963). 13. Steinbrocker, O. & Argyros, T. G. Froz en shoulder: treatment by local injections of depot corticosteroids. Arch Phys Med Rehabil 55, 209-13 (1974). 14. Franklin, T. J. Therapeutic approaches to organ fibrosis. Int J Biochem Cell Biol 29, 79-89 (1997). 15. Clark, R. A. Cutaneous tissue repair: basic biologic considerations. I. J Am Acad Dermatol 13 701-25 (1985). 127

PAGE 128

16. Desmouliere, A., Darby, I. A. & Gabbi ani, G. Normal and pathologic soft tissue remodeling: role of the myofib roblast, with special emphasis on liver and kidney fibrosis. Lab Invest 83, 1689-707 (2003). 17. Border, W. A. & Noble, N. A. Transforming growth factor beta in tissue fibrosis. N Engl J Med 331 1286-92 (1994). 18. Czaja, M. J. et al. In vitro and in vivo association of transforming growth factorbeta 1 with hepatic fibrosis. J Cell Biol 108 2477-82 (1989). 19. Isaka, Y. et al. Gene therapy by skeletal muscle expression of decorin prevents fibrotic disease in rat kidney. Nat Med 2 418-23 (1996). 20. Johnson, R. J., Floege, J., Couser, W. G. & Alpers, C. E. Role of platelet-derived growth factor in glomerular disease. J Am Soc Nephrol 4 119-28 (1993). 21. Ihn, H. Pathogenesis of fibr osis: role of TGF-beta and CTGF. Curr Opin Rheumatol 14, 681-5 (2002). 22. Miyazono, K., Kusanagi, K. & Inoue, H. Divergence and convergence of TGFbeta/BMP signaling. J Cell Physiol 187 265-76 (2001). 23. Piek, E., Heldin, C. H. & Ten Dijke, P. Specificity, diversity, and regulation in TGF-beta superfamily signaling. Faseb J 13, 2105-24 (1999). 24. Nunes, I., Gleizes, P. E ., Metz, C. N. & Rifkin, D. B. Latent transforming growth factor-beta binding protei n domains involved in activation and transglutaminasedependent cross-linking of latent transforming growth factor-beta. J Cell Biol 136 115163 (1997). 25. Saitoh, M. et al. Identification of important regions in the cytoplasmic juxtamembrane domain of type I receptor that separ ate signaling pathways of transforming growth factor-beta. J Biol Chem 271 2769-75 (1996). 26. Wrana, J. L., Attisano, L ., Wieser, R., Ventura, F. & Massague, J. Mechanism of activation of the TGF-beta receptor. Nature 370 341-7 (1994). 27. Holmes, A. et al. CTGF and SMADs, maintenanc e of scleroderma phenotype is independent of SMAD signaling. J Biol Chem 276 10594-601 (2001). 28. Dennler, S. et al. Direct binding of Smad3 and Sm ad4 to critical TGF betainducible elements in the promoter of human plasminogen activa tor inhibitor-type 1 gene. Embo J 17, 3091-100 (1998). 29. Jonk, L. J., Itoh, S., He ldin, C. H., ten Dijke, P. & Kruijer, W. Identification and functional characterization of a Smad binding element (SBE) in the JunB promoter that acts as a transforming growth factor-beta, activin, and bone morphogenetic proteininducible enhancer. J Biol Chem 273 21145-52 (1998). 128

PAGE 129

30. Nagarajan, R. P., Zhang, J., Li, W. & Chen, Y. Regul ation of Smad7 promoter by direct association with Smad3 and Smad4. J Biol Chem 274 33412-8 (1999). 31. Taylor, L. M. & Khachigi an, L. M. Induction of platelet-derived growth factor Bchain expression by transforming growth fact or-beta involves transactivation by Smads. J Biol Chem 275 16709-16 (2000). 32. Lai, C. F. et al. Transforming growth factor-beta up-regulates the beta 5 integrin subunit expression via Sp1 and Smad signaling. J Biol Chem 275 36400-6 (2000). 33. Takagawa, S. et al. Sustained activation of fibroblast transforming growth factorbeta/Smad signaling in a mu rine model of scleroderma. J Invest Dermatol 121 41-50 (2003). 34. Verrecchia, F., Chu, M. L. & Mauviel, A. Identification of novel TGF-beta /Smad gene targets in dermal fibroblasts using a combined cDNA microarray/promoter transactivation approach. J Biol Chem 276 17058-62 (2001). 35. Verrecchia, F. & Mauviel, A. Transformi ng growth factor-beta signaling through the Smad pathway: role in extracellular matrix gene expression and regulation. J Invest Dermatol 118 211-5 (2002). 36. Postlethwaite, A. E., Keski-Oja, J., Moses, H. L. & Kang, A. H. Stimulation of the chemotactic migration of human fibroblasts by transforming growth factor beta. J Exp Med 165 251-6 (1987). 37. Jester, J. V., Huang, J., Petroll, W. M. & Cavanagh, H. D. TGFbeta induced myofibroblast differentiation of rabbit kerato cytes requires synergistic TGFbeta, PDGF and integrin signaling. Exp Eye Res 75, 645-57 (2002). 38. Zhang, H. Y. & Phan, S. H. Inhibition of myofibroblast apoptosis by transforming growth factor beta(1). Am J Respir Cell Mol Biol 21, 658-65 (1999). 39. Rodeo, S. A., Hannafin, J. A., Tom, J. Warren, R. F. & Wickiewicz, T. L. Immunolocalization of cytokines and their receptors in adhesive capsulitis of the shoulder. J Orthop Res 15, 427-36 (1997). 40. Bunker, T. D., Reilly, J., Baird, K. S. & Hamblen, D. L. Expression of growth factors, cytokines and matrix meta lloproteinases in frozen shoulder. J Bone Joint Surg Br 82, 768-73 (2000). 41. Ryu, J. D. et al. Expression of vascular endothelial growth factor and angiogenesis in the diabetic frozen shoulder. J Shoulder Elbow Surg 15, 679-85 (2006). 42. Foulis, A. K., McGill, M. & Farquharson, M. A. Insulitis in type 1 (insulindependent) diabetes mellitus in man--ma crophages, lymphocytes, and interferongamma containing cells. J Pathol 165 97-103 (1991). 129

PAGE 130

43. Matthaei, S., Stumvoll, M., Kellerer, M. & Haring, H. U. Pathophysiology and pharmacological treatment of insulin resistance. Endocr Rev 21, 585-618 (2000). 44. Galkowska, H., Wojewodzka, U. & Olszewski, W. L. Chemokines, cytokines, and growth factors in keratinocytes and dermal endothelial cells in the margin of chronic diabetic foot ulcers. Wound Repair Regen 14 558-65 (2006). 45. Maruyama, K. et al. Decreased macrophage number and activation lead to reduced lymphatic vessel formation and contri bute to impaired diabetic wound healing. Am J Pathol 170 1178-91 (2007). 46. Twigg, S. M. et al. Advanced glycosylation end products up-regulate connective tissue growth factor (insulin-like growth factor-binding protein-related protein 2) in human fibroblasts: a potential mechanism for expansion of extrace llular matrix in diabetes mellitus. Endocrinology 142 1760-9 (2001). 47. Kayal, R. A. et al. Diminished bone formation during diabetic fracture healing is related to the premature resorption of cartilage associated with increased osteoclast activity. J Bone Miner Res 22, 560-8 (2007). 48. Lobmann, R. et al. Expression of matrix-metalloproteinases and their inhibitors in the wounds of diabetic and non-diabetic patients. Diabetologia 45, 1011-6 (2002). 49. Roestenberg, P. et al. Connective tissue growth factor is increased in plasma of type 1 diabetic patients with nephropathy. Diabetes Care 27, 1164-70 (2004). 50. Gilbert, R. E. et al. Urinary connective tissue growth factor excretion in patients with type 1 diabetes and nephropathy. Diabetes Care 26, 2632-6 (2003). 51. Riser, B. L. et al. Regulation of connective tissue grow th factor activity in cultured rat mesangial cells and its expression in experimental diabetic glomerulosclerosis. J Am Soc Nephrol 11, 25-38 (2000). 52. Ito, Y. et al. Expression of connective tissue growth factor in human renal fibrosis. Kidney Int 53, 853-61 (1998). 53. Ziyadeh, F. N. Mediators of diabetic renal disease: the case for tgf-Beta as the major mediator. J Am Soc Nephrol 15 Suppl 1 S55-7 (2004). 54. Isono, M., Chen, S., Hong, S. W., Iglesias-de la Cruz, M. C. & Ziyadeh, F. N. Smad pathway is activated in the diabet ic mouse kidney and Sm ad3 mediates TGFbeta-induced fibronectin in mesangial cells. Biochem Biophys Res Commun 296 135665 (2002). 55. McLennan, S. V. et al. Effects of glucose on ma trix metalloproteinase and plasmin activities in mesangial cells : possible role in diabetic nephropathy. Kidney Int Suppl 77, S81-7 (2000). 130

PAGE 131

56. Ayo, S. H., Radnik, R., Ga roni, J. A., Troyer, D. A. & Kreisberg, J. I. High glucose increases diacylglycerol mass and activates prot ein kinase C in mesangial cell cultures. Am J Physiol 261 F571-7 (1991). 57. Ziyadeh, F. N., Sharma, K., Ericksen, M. & Wolf, G. Stimulation of collagen gene expression and protein synthesis in murine mes angial cells by high gl ucose is mediated by autocrine activation of trans forming growth factor-beta. J Clin Invest 93, 536-42 (1994). 58. Han, D. C., Isono, M., Ho ffman, B. B. & Ziyadeh, F. N. High glucose stimulates proliferation and collagen type I synthesis in renal cortical fibroblasts: mediation by autocrine activation of TGF-beta. J Am Soc Nephrol 10 1891-9 (1999). 59. Han, D. C., Hoffman, B. B., Hong, S. W., Guo, J. & Ziyadeh, F. N. Therapy with antisense TGF-beta1 olig odeoxynucleotides reduces kidney weight and matrix mRNAs in diabetic mice. Am J Physiol Renal Physiol 278 F628-34 (2000). 60. Sharma, K., Jin, Y., Guo, J. & Ziyadeh, F. N. Neutra lization of TGF-beta by antiTGF-beta antibody attenuates kidney hypertr ophy and the enhanced extracellular matrix gene expression in STZ-induced diabetic mice. Diabetes 45, 522-30 (1996). 61. Zhang, J., Liu, Z., Liu, H., Li, Y. & Li, L. [Regulation of the expression and function of glucose transporte r-1 by TGF-beta 1 and high glucose in mesangial cells]. Chin Med J (Engl) 113 508-13 (2000). 62. Gnudi, L., Thomas, S. M. & Viberti, G. Mechanical forces in diabetic kidney disease: a trigger for impaired glucose metabolism. J Am Soc Nephrol 18, 2226-32 (2007). 63. Ziyadeh, F. N., Han, D. C., Cohen, J. A., Guo, J. & Cohen, M. P. Glycated albumin stimulates fibronectin gene expression in glomerular mesangial cells: involvement of the transformi ng growth factor-beta system. Kidney Int 53, 631-8 (1998). 64. Isono, M., Mogyorosi, A., Han, D. C., Ho ffman, B. B. & Ziyadeh F. N. Stimulation of TGF-beta type II receptor by high glucos e in mouse mesangial cells and in diabetic kidney. Am J Physiol Renal Physiol 278 F830-8 (2000). 65. Dyer, D. G. et al. Accumulation of Maillard reaction products in skin collagen in diabetes and aging. J Clin Invest 91, 2463-9 (1993). 66. Vlassara, H., Bucala, R. & Stri ker, L. Pathogenic effects of advanced glycosylation: biochemical, biologic, and clin ical implications for diabetes and aging. Lab Invest 70, 138-51 (1994). 67. Vlassara, H., Brownlee, M. & Cera mi, A. Nonenzymatic glycosylation of peripheral nerve protein in diabetes mellitus. Proc Natl Acad Sci U S A 78, 5190-2 (1981). 131

PAGE 132

68. Kreisberg, J. I., Garoni, J. A., Radni k, R. & Ayo, S. H. High glucose and TGF beta 1 stimulate fibronectin gene expres sion through a cAMP response element. Kidney Int 46, 1019-24 (1994). 69. Oh, J. H., Ha, H., Yu, M. R. & Lee, H. B. Sequential effects of high glucose on mesangial cell transforming growth factor-beta 1 and fibronectin synthesis. Kidney Int 54, 1872-8 (1998). 70. Ha, H., Yu, M. R. & Lee, H. B. Hig h glucose-induced PKC activation mediates TGF-beta 1 and fibronectin synthesis by peritoneal mesothelial cells. Kidney Int 59, 46370 (2001). 71. Yung, S., Liu, Z. H., Lai, K. N., Li L. S. & Chan, T. M. Emodin ameliorates glucose-induced morphologic abnormalities and synthesis of transformi ng growth factor beta1 and fibronectin by human per itoneal mesothelial cells. Perit Dial Int 21 Suppl 3 S41-7 (2001). 72. Li, J. H. et al. Advanced glycation end products activate Smad signaling via TGFbeta-dependent and independent me chanisms: implications for diabetic renal and vascular disease. Faseb J 18, 176-8 (2004). 73. Evans, C. H. et al. Gene therapy for rheumatic diseases. Arthritis Rheum 42, 116 (1999). 74. Ghivizzani, S. C., Oligino, T. J., Glor ioso, J. C., Robbins, P. D. & Evans, C. H. Direct gene delivery strategies for the treatment of rheum atoid arthritis. Drug Discov Today 6 259-267 (2001). 75. Bandara, G. et al. Intraarticular expression of biologically active interleukin 1receptor-antagonist protein by ex vivo gene transfer. Proc Natl Acad Sci U S A 90, 10764-8 (1993). 76. Otani, K. et al. Suppression of antigen-induced arthriti s in rabbits by ex vivo gene therapy. J Immunol 156 3558-62 (1996). 77. Ghivizzani, S. C. et al. Constitutive intra-articular expression of human IL-1 beta following gene transfer to rabbit synovium produces all major pathologies of human rheumatoid arthritis. J Immunol 159 3604-12 (1997). 78. Gouze, E. et al. Lentiviral-mediated gene delivery to synovium: potent intraarticular expression with amp lification by inflammation. Mol Ther 7 460-6 (2003). 79. Hardy, S., Kitamura, M., Harris-Stansil, T., Dai, Y. & Phipps, M. L. Construction of adenovirus vectors through Cre-lox recombination. J Virol 71, 1842-9 (1997). 80. Ng, P. & Graham, F. L. Construction of first-generation adenoviral vectors. Methods Mol Med 69, 389-414 (2002). 132

PAGE 133

81. Heng, E. C., Huang, Y., Black, S. A ., Jr. & Trackman, P. C. CCN2, connective tissue growth factor, stimulates collagen deposition by gingival fibroblasts via module 3 and alpha6and beta1 integrins. J Cell Biochem 98, 409-20 (2006). 82. Moussad, E. E. & Brigst ock, D. R. Connective tissue growth factor: what's in a name? Mol Genet Metab 71, 276-92 (2000). 83. Gholve, P. A. et al. Arthrofibrosis of the knee after ti bial spine fracture in children: a report of two complicated cases. Hss J 4 14-9 (2008). 84. Lee, S. K., Gargano, F. & Hausman, M. R. Wrist arthrofibrosis. Hand Clin 22 529-38; abstract vii (2006). 85. Schiavone Panni, A., Cerc iello, S., Vasso, M. & Tartar one, M. Stiffness in total knee arthroplasty. J Orthop Traumatol 10, 111-8 (2009). 86. Boldt, J. G., Munzinger, U. K., Zanetti, M. & Hodler, J. Arthrofibrosis associated with total knee arthroplasty: gray-scale and power Doppler sonographic findings. AJR Am J Roentgenol 182 337-40 (2004). 87. Neviaser, R. J. & Neviaser, T. J. The frozen shoulder. Diagnosis and management. Clin Orthop Relat Res 59-64 (1987). 88. Wynn, T. A. Cellular and mole cular mechanisms of fibrosis. J Pathol 214 199210 (2008). 89. Pohlers, D. et al. TGF-beta and fibrosis in differ ent organs molecular pathway imprints. Biochim Biophys Acta 1792 746-56 (2009). 90. Gressner, O. A., Rizk, M. S., Kovalenko, E., Weiskirchen, R. & Gressner, A. M. Changing the pathogenetic roadmap of liver fibrosis? Where did it start; where will it go? J Gastroenterol Hepatol 23, 1024-35 (2008). 91. Rahimi, R. A. & Leof, E. B. TGF-bet a signaling: a tale of two responses. J Cell Biochem 102 593-608 (2007). 92. Leivonen, S. K. & Kahar i, V. M. Transforming growth factor-beta signaling in cancer invasion and metastasis. Int J Cancer 121 2119-24 (2007). 93. Bonniaud, P. et al. TGF-beta and Smad3 signaling link inflammation to chronic fibrogenesis. J Immunol 175 5390-5 (2005). 94. Gouze, E. et al. Transgene persistence and cell turnover in the diarthrodial joint: implications for gene therapy of chronic joint diseases. Mol Ther 15, 1114-20 (2007). 95. Spessotto, P. et al. beta 1 Integrin-dependent cell adhesion to EMILIN-1 is mediated by the gC1q domain. J Biol Chem 278 6160-7 (2003). 133

PAGE 134

96. Zhang, H. Y., Gharaee-Kermani, M., Z hang, K., Karmiol, S. & Phan, S. H. Lung fibroblast alpha-smooth muscle actin ex pression and contractile phenotype in bleomycin-induced pulmonary fibrosis. Am J Pathol 148 527-37 (1996). 97. Gouze, E. et al. In vivo gene delivery to synovium by lentiviral vectors. Mol Ther 5 397-404 (2002). 98. Magit, D., Wolff, A., Sutton, K. & Medve cky, M. J. Arthrofibrosis of the knee. J Am Acad Orthop Surg 15, 682-94 (2007). 99. Murphey, M. D., Vidal, J. A., Fanburg-Smith, J. C. & Gajewski, D. A. Imaging of synovial chondromatosis with radi ologic-pathologic correlation. Radiographics 27, 146588 (2007). 100. Vargas-Gonzalez, R. & Sanchez-Sosa, S. Fibrocartilaginous dysplasia (fibrous dysplasia with extensive ca rtilaginous differentiation). Pathol Oncol Res 12, 111-4 (2006). 101. Zoricic, S., Maric, I., Bobinac, D. & Vukicevic, S. Expression of bone morphogenetic proteins and cartilage-der ived morphogenetic proteins during osteophyte formation in humans. J Anat 202 269-77 (2003). 102. Matsui, Y. et al. Intrapatellar tendon lipoma with chondro-osseous differentiation: detection of HMGA2-LPP fusion gene transcript. J Clin Pathol 59, 434-6 (2006). 103. Wirrig, E. E. et al. Cartilage link protein 1 (Crt l1), an extracellular matrix component playing an important role in heart development. Dev Biol 310 291-303 (2007). 104. Gueders, M. M., Foidar t, J. M., Noel, A. & Ca taldo, D. D. Matrix metalloproteinases (MMPs) and tissue inhibito rs of MMPs in the respiratory tract: potential implications in asthma and other lung diseases. Eur J Pharmacol 533 133-44 (2006). 105. Hemmann, S., Graf, J., Roderfeld, M. & Roeb, E. Expression of MMPs and TIMPs in liver fibrosis a systematic re view with special emphasis on anti-fibrotic strategies. J Hepatol 46 955-75 (2007). 106. English, J. L. et al. Individual Timp deficiencies differentially impact pro-MMP-2 activation. J Biol Chem 281 10337-46 (2006). 107. Pardo, A. & Selman, M. Matrix meta lloproteases in aberrant fibrotic tissue remodeling. Proc Am Thorac Soc 3 383-8 (2006). 108. Bornstein, P. & Sage, E. H. Matricellu lar proteins: extracellu lar modulators of cell function. Curr Opin Cell Biol 14, 608-16 (2002). 134

PAGE 135

109. Alford, A. I. & Hankenson, K. D. Matricellular proteins : Extracellular modulators of bone development, remodelin g, and regeneration. Bone 38, 749-57 (2006). 110. Mo, F. E. & Lau, L. F. The matricellu lar protein CCN1 is essential for cardiac development. Circ Res 99, 961-9 (2006). 111. Moura, R., Tjwa, M., Vandervoort, P., Cludts, K. & Hoylaerts, M. F. Thrombospondin-1 activates medial smoot h muscle cells and triggers neointima formation upon mouse carotid artery ligation. Arterioscler Thromb Vasc Biol 27 2163-9 (2007). 112. Bradshaw, A. D. & Sage, E. H. SPARC, a matricellular protein that functions in cellular differentiation and tissue response to injury. J Clin Invest 107 1049-54 (2001). 113. Bornstein, P. Matricellular proteins: an overview. Matrix Biol 19, 555-6 (2000). 114. Agnihotri, R. et al. Osteopontin, a novel substrate fo r matrix metalloproteinase-3 (stromelysin-1) and matrix me talloproteinase-7 (matrilysin). J Biol Chem 276 28261-7 (2001). 115. Pardo, A. et al. Up-regulation and profibrotic role of osteopontin in human idiopathic pulmonary fibrosis. PLoS Med 2 e251 (2005). 116. Castellano, G. et al. Activation of the osteopontin /matrix metalloproteinase-9 pathway correlates with pros tate cancer progression. Clin Cancer Res 14, 7470-80 (2008). 117. Gao, Y. A., Agnihotri, R., Vary, C. P. & Liaw, L. Expression and characterization of recombinant osteopontin peptides represent ing matrix metalloproteinase proteolytic fragments. Matrix Biol 23, 457-66 (2004). 118. Kobayashi, T. et al. Connective tissue growth fact or mediates the profibrotic effects of transforming growth factor-bet a produced by tubular epithelial cells in response to high glucose. Clin Exp Nephrol 9 114-21 (2005). 119. Leask, A., Parapuram, S. K., Shi-Wen, X. & Abraham, D. J. Connective tissue growth factor (CTGF, CCN 2) gene regulation: a potent clinical bio-marker of fibroproliferative disease? J Cell Commun Signal 3 89-94 (2009). 120. Woods, A., Wang, G. & Beier, F. Regulation of chondr ocyte differentiation by the actin cytoskeleton and adhesive interactions. J Cell Physiol 213 1-8 (2007). 121. Lehembre, F. et al. NCAM-induced focal adhesion assembly: a functional switch upon loss of E-cadherin. Embo J 27, 2603-15 (2008). 122. Frame, M. C. & In man, G. J. NCAM is at the heart of reciproc al regulation of Ecadherinand integrin-mediated adhes ions via signalin g modulation. Dev Cell 15, 494-6 (2008). 135

PAGE 136

123. Shintani, N. & Hunziker, E. B. Chondr ogenic differentiation of bovine synovium: bone morphogenetic proteins 2 and 7 and transforming growth factor beta1 induce the formation of different types of cartilaginous tissue. Arthritis Rheum 56, 1869-79 (2007). 124. Fiorito, S., Magrini, L. Adrey, J., Mailhe, D. & Br outy-Boye, D. Inflammatory status and cartilage regenerativ e potential of synovial fibr oblasts from patients with osteoarthritis and chondropathy. Rheumatology (Oxford) 44, 164-71 (2005). 125. Mochizuki, T. et al. Higher chondrogenic potential of fibrous synoviumand adipose synovium-derived cells compared with subcutaneous fat-derived cells: distinguishing properties of mese nchymal stem cells in humans. Arthritis Rheum 54 843-53 (2006). 126. Shirasawa, S. et al. In vitro chondrogenesis of hum an synovium-derived mesenchymal stem cells: optimal conditi on and comparison with bone marrow-derived cells. J Cell Biochem 97 84-97 (2006). 127. Jiang, X. et al. [Study on the directed induci ng process of cartilage cells differentiated from human ma rrow mesenchymal stem cells]. Zhonghua Er Bi Yan Hou Ke Za Zhi 37, 137-9 (2002). 128. De Bari, C., Dell'Accio, F., Tylzanow ski, P. & Luyten, F. P. Multipotent mesenchymal stem cells from adult human synovial membrane. Arthritis Rheum 44, 1928-42 (2001). 129. Karystinou, A. et al. Distinct mesenchymal progenitor cell subsets in the adult human synovium. Rheumatology (Oxford) 48 1057-64 (2009). 130. Crawford, A., Frazer, A., Lippitt, J. M., Buttle, D. J. & Smith, T. A case of chondromatosis indicates a sy novial stem cell aetiology. Rheumatology (Oxford) 45, 1529-33 (2006). 131. Krampera, M., Pizzolo, G. Aprili, G. & Franchini, M. Mesenchymal stem cells for bone, cartilage, tendon and skeletal muscle repair. Bone 39, 678-83 (2006). 132. Ahmed, N., Stanf ord, W. L. & Kandel, R. A. Mesenchymal stem and progenitor cells for cartilage repair. Skeletal Radiol 36, 909-12 (2007). 133. Patel, T. & Gores, G. J. Apoptosis and hepatobiliary disease. Hepatology 21, 1725-41 (1995). 134. Vaux, D. L. & Strasser, A. The molecular biology of apoptosis. Proc Natl Acad Sci U S A 93 2239-44 (1996). 135. Evan, G. I., Brown, L., Whyte, M. & Harrington, E. Apoptosis and the cell cycle. Curr Opin Cell Biol 7 825-34 (1995). 136. Fraser, A. & Evan, G. A license to kill. Cell 85 781-4 (1996). 136

PAGE 137

137. Cole, A. A. et al. Chondrocyte matrix metallopr oteinase-8. Human articular chondrocytes express neutrophil collagenase. J Biol Chem 271 11023-6 (1996). 138. Hanemaaijer, R. et al. Matrix metalloproteinase-8 is expressed in rheumatoid synovial fibroblasts and endothelial cells. Regul ation by tumor necrosis factor-alpha and doxycycline. J Biol Chem 272 31504-9 (1997). 139. Sasano, Y. et al. Gene expression of MMP8 and MMP13 during embryonic development of bone and cartilage in the rat mandible and hind limb. J Histochem Cytochem 50, 325-32 (2002). 140. Poczatek, M. H., Hugo, C., Darley-Usm ar, V. & Murphy-Ullric h, J. E. Glucose stimulation of transforming grow th factor-beta bioactivity in mesangial cells is mediated by thrombospondin-1. Am J Pathol 157 1353-63 (2000). 141. Sharma, K. et al. Increased renal production of tr ansforming growth factor-beta1 in patients with type II diabetes. Diabetes 46, 854-9 (1997). 142. Yamamoto, T. et al. Expression of transforming growth factor-beta isoforms in human glomerular diseases. Kidney Int 49, 461-9 (1996). 143. Iwano, M. et al. Quantification of glomerular TGF-beta 1 mRNA in patients with diabetes mellitus. Kidney Int 49, 1120-6 (1996). 144. McLennan, S. V. et al. Decreased matrix degradati on in diabetic nephropathy: effects of ACE inhibition on the expression a nd activities of matrix metalloproteinases. Diabetologia 45, 268-75 (2002). 145. Ziyadeh, F. N. et al. Long-term prevention of renal insufficiency, excess matrix gene expression, and glomerular mesangial matrix expansion by treatment with monoclonal antitransforming growth factor -beta antibody in db/db diabetic mice. Proc Natl Acad Sci U S A 97 8015-20 (2000). 146. Milgrom, C. et al. Risk factors for idiopathic frozen shoulder. Isr Med Assoc J 10, 361-4 (2008). 147. Lee, J. S., Oum, B. S. & Lee, S. H. Mitomycin c infl uence on inhibition of cellular proliferation and subsequent synthesis of type I collagen and laminin in primary and recurrent pterygia. Ophthalmic Res 33, 140-6 (2001). 148. Nieto, A. et al. Effect of mitomycin-C on human foreskin fibroblasts used as feeders in human embryonic stem cells: immunocytochemistry MIB1 score and DNA ploidy and apoptosis evaluated by flow cytometry. Cell Biol Int 31, 269-78 (2007). 149. Heilmann, C., Schonfeld, P., Schluter, T., Bohnensack, R. & Behrens-Baumann, W. Effect of the cytostatic agent idarubici n on fibroblasts of t he human Tenon's capsule compared with mitomycin C. Br J Ophthalmol 83, 961-6 (1999). 137

PAGE 138

150. Blacker, K. L., Williams, M. L. & Gold yne, M. Mitomycin Ctreated 3T3 fibroblasts used as feeder layers for human keratinocyt e culture retain the capacity to generate eicosanoids. J Invest Dermatol 89 536-9 (1987). 151. Chen, T. C. & Chang, S. W. Mito mycin C Modulates IL-1R Expression on Corneal Fibroblasts and Modifies Corneal Ep ithelial Cell Migrati on via IL-1 Related Hepatocyte Growth Factor. Invest Ophthalmol Vis Sci (2009). 152. Ha, T. S. et al. Regulation of renal laminin in mice with type II diabetes. J Am Soc Nephrol 10 1931-9 (1999). 153. van Det, N. F. et al. Regulation of glomerular epithelial cell production of fibronectin and transforming growth factor-beta by high glucose, not by angiotensin II. Diabetes 46, 834-40 (1997). 154. Suzuki, D. et al. In situ hybridization studies of matrix metalloproteinase-3, tissue inhibitor of metalloprot einase-1 and type IV colla gen in diabetic nephropathy. Kidney Int 52, 111-9 (1997). 155. Del Prete, D. et al. Down-regulation of glomerular matrix metalloproteinase-2 gene in human NIDDM. Diabetologia 40, 1449-54 (1997). 156. Shankland, S. J., Ly, H., Thai, K. & Scholey, J. W. Glomer ular expression of tissue inhibitor of metalloproteinase (T IMP-1) in normal and diabetic rats. J Am Soc Nephrol 7 97-104 (1996). 157. Knoll, K. E., Pietrusz, J. L. & Liang, M. Tissue-specific transcriptome responses in rats with early strept ozotocin-induced diabetes. Physiol Genomics 21, 222-9 (2005). 158. Mosci, P., Vecchiarelli, A., Cenci, E., Puliti, M. & Bistoni, F. Low-dose streptozotocin-induced diabetes in mice. I. Course of C andida albicans infection. Cell Immunol 150 27-35 (1993). 159. Kunjathoor, V. V., Wilson, D. L. & LeBoeuf, R. C. Increased atherosclerosis in streptozotocin-induced diabetic mice. J Clin Invest 97, 1767-73 (1996). 160. Krakauer, J. C. et al. Bone loss and bone turnover in diabetes. Diabetes 44, 77582 (1995). 161. Gooch, H. L., Hale, J. E ., Fujioka, H., Balian, G. & Hurwitz, S. R. Alterations of cartilage and collagen expressi on during fracture healing in experimental diabetes. Connect Tissue Res 41, 81-91 (2000). 162. Herskind, A. M., Christensen, K., No rgaard-Andersen, K. & Andersen, J. F. Diabetes mellitus and healing of closed fractures. Diabete Metab 18 63-4 (1992). 163. Soejima, K. & Landing, B. H. Osteoporosis in juvenile-onset diabetes mellitus: morphometric and comparative studies. Pediatr Pathol 6 289-99 (1986). 138

PAGE 139

164. Goodman, W. G. & Hori, M. T. Diminished bone formation in experimental diabetes. Relationship to osteoi d maturation and mineralization. Diabetes 33, 825-31 (1984). 165. Funk, J. R., Hale, J. E., Carmines, D., Gooch, H. L. & Hurwitz, S. R. Biomechanical evaluation of early fractu re healing in normal and diabetic rats. J Orthop Res 18, 126-32 (2000). 166. Lu, H., Kraut, D., Gerstenf eld, L. C. & Graves, D. T. Diabetes interferes with the bone formation by affecting the expression of transcription factors that regulate osteoblast differentiation. Endocrinology 144 346-52 (2003). 167. Tamm, I., Dorken, B. & Hartmann, G. Antisense therapy in oncology: new hope for an old idea? Lancet 358 489-97 (2001). 168. Yacyshyn, B. R. et al. A placebo-controlled tr ial of ICAM-1 antisense oligonucleotide in the treat ment of Crohn's disease. Gastroenterology 114 1133-42 (1998). 169. Stephenson, M. L. & Zamecnik, P. C. Inhibition of Rous sarcoma viral RNA translation by a specific oligodeoxyribonucleotide. Proc Natl Acad Sci U S A 75 285-8 (1978). 170. Dean, N. M. & Griffey, R. H. Identification and characterization of secondgeneration antisense o ligonucleotides. Antisense Nucleic Acid Drug Dev 7 229-33 (1997). 171. Henry, S. et al. Chemically modified oligonucleot ides exhibit decreased immune stimulation in mice. J Pharmacol Exp Ther 292 468-79 (2000). 172. Lengner, C. J. et al. Primary mouse embryonic fibroblasts: a model of mesenchymal cartilage formation. J Cell Physiol 200 327-33 (2004). 173. Tuli, R. et al. Transforming growth factorbeta-mediated chondrogenesis of human mesenchymal progenitor cells involves N-cadherin and mitogen-activated protein kinase and Wnt signaling cross-talk. J Biol Chem 278 41227-36 (2003). 174. Barry, F., Boynton, R. E ., Liu, B. & Murphy, J. M. Chondrogenic differentiation of mesenchymal stem cells from bone marro w: differentiation-dependent gene expression of matrix components. Exp Cell Res 268 189-200 (2001). 175. Brittberg, M. et al. Treatment of deep cartilage defects in the knee with autologous chondrocyte transplantation. N Engl J Med 331 889-95 (1994). 139

PAGE 140

BIOGRAPHICAL SKETCH Rachael Susan Watson was raised in Bainbridge Township, Ohio. She graduated summa cum laude of Kenston Hi gh Schools class of 1999. Her undergraduate studies began the following fall at the University of Mount Union, Alliance, Ohio, where she majored in biology and minored in chemistr y. During the summers of her junior and senior years, Rachael began working as a research assistant at the NASA Glenn Research Center in Cleveland, Ohio in the lab of Dr. Greg Zimmerli, where she contributed to the understandi ng of loss of bone density in microgravity. After graduating magna cum laude and re ceiving her Bachelor of Science in May 2003, she began work for Dr. Yung Huang at Case Western Reserve University in Cleveland, Ohio developing a cell based vaccine for influenza, until joining the Interdisciplinary Program in Biomedical Sciences (IDP) at the Universi ty of Florida in August of 2004. In May of 2005, Rachael began her doctoral study under t he guidance Dr. Steven C. Ghivizzani, in the Department of Or thopaedics and Rehabilitation w here she investigated the idiopathic disease, Adhesive Capsulitis. During the course of her graduate studies she received extensive training in the use of tech niques to study arthrofibrosis, including the development viral vectors for gene transfer, techniques to study in vivo gene signaling and the development of animal models. She received a graduate student research grant from Clinical and Translation Science Institut e at the University of Florida. Rachael presented her research at a talk for a Keystone Symposia on TGF-beta and received a scholarship from the Keystone Symposia for her work. She also presented her research at a poster presentation fo r the American Society of Gene and Cell Therapy. Her graduate research characterizing the role of TG F-beta in arthrofibrosis is currently in submission. Rachael received her Ph.D. in the spring of 2010. 140