"Medicine, Faculty of"@en . "Medicine, Department of"@en . "Experimental Medicine, Division of"@en . "DSpace"@en . "UBCV"@en . "Khani-Hanjani, Abbas"@en . "2009-07-15T15:57:37Z"@en . "2000"@en . "Doctor of Philosophy - PhD"@en . "University of British Columbia"@en . "Rheumatoid arthritis (RA) is a chronic, debilitating disease characterized by inflammatory\r\nthickening of articular soft tissue, with extension of the synovium over the articular\r\ncartilage, cartilaginous destruction and subchondral erosion leading to progressive joint\r\ndeformity and immobility. RA currently affects over half a million subjects in Canada.\r\nMore than 1,300 patients die annually from gastric bleeding secondary to the use of\r\nNSAIDs. The clinical manifestations of RA encompass a spectrum ranging from mild to\r\nsevere disease, with irreversible joint damage and disability. Our current understanding\r\ndoes not allow for the accurate prediction of the severity in RA. Yet, increasing evidence\r\nsuggests that early intervention with immunosuppressive drugs may mitigate the relentless\r\nprogression of this disease and prevent irreversible joint damage. The main objective of this\r\nstudy was to explore selected immune response genes potentially involved in disease\r\nseverity using patient populations stratified according to disease severity. PCR-RFLP and\r\nfragment length analyses of microsatellites were used to study the effect of genetic\r\npolymorphisms of nine genes on chromosomes 2, 4, 6, 12 and 22, which included HLA-DR\u00CE\u00B21,\r\nHLA-DQ\u00CE\u00B21, HLA-DP\u00CE\u00B21, TNF, IL-1\u00CE\u00B2, IL-IRN, IL-2, IL-2R\u00CE\u00B2, and IFN\u00CE\u00B3. HLA-DR\u00CE\u00B21\r\nalleles containing the QKRAA or QRRAA epitope, HLA-DQ\u00CE\u00B21*302, HLA-DP\u00CE\u00B21\r\n*401 and the 126bp allele of IFN\u00CE\u00B3 were identified as high risk alleles, which\r\ndemonstrate a significant association with the severe RA. In contrast the HLA-DR\u00CE\u00B21\r\nalleles containing the DERRA epitope and HLDR\u00CE\u00B21*8, HLA-DQ\u00CE\u00B2 1*402/502/603/604,\r\nHLA-DP\u00CE\u00B21*201/501/1001 and the 122bp allele of IFN\u00CE\u00B3 were significantly decreased in\r\nfrequency in the population with severe RA compared to those with mild disease,\r\nsuggesting a potential protective effect of these alleles on the progression of the disease.\r\nThe distribution of genetic polymorphism for all other gene markers were either similar\r\nbetween populations or had a statistically insignificant value in the prediction of the\r\nseverity of RA. The combination of HLA class II and IFN\u00CE\u00B3 high-risk alleles yielded a more\r\nspecific prediction of disease severity. A logistic regression model was designed which\r\nincorporated demographic and clinical data, HLA class II and IFN\u00CE\u00B3 gene polymorphisms in\r\norder to predict the severity of disease. The model was able to discriminate between mild\r\nand severe RA patients with an odds ratio of greater than 447, Chi-square=28.83, p<0.0001\r\nand an overall accuracy of over 90%."@en . "https://circle.library.ubc.ca/rest/handle/2429/10814?expand=metadata"@en . "7388966 bytes"@en . "application/pdf"@en . "GENETIC FACTORS ASSOCIATED WITH DISEASE SEVERITY IN RHEUMATOID ARTHRITIS by ABBAS KHANI-HANJANI B.Sc, The University of British Columbia, 1995 A THESIS SUBMITTED IN PARTIAL FULFILMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY in THE FACULTY OF GRADUATE STUDIES Faculty of Medicine Department of Experimental Medicine We accept this thesis as confirming to the required standard THE UNIVERSITY OF BRITISH COLUMBIA March 2000 \u00C2\u00A9 Abbas Khani-Hanjani, 2000 In presenting this thesis in partial fulfilment of the requirements for an advanced degree at the University of British Columbia, I agree that the Library shall make it freely available for reference and study. I further agree that permission for extensive copying of this thesis for scholarly purposes may be granted by the head of my department or by his or her representatives. It is understood that copying or publication of this thesis for financial gain shall not be allowed without my written permission. Department of tf, cw tb> V hK^Jbd The University of British Columbia Vancouver, Canada DE-6 (2/88) ABSTRACT Rheumatoid arthritis (RA) is a chronic, debilitating disease characterized by inflammatory thickening of articular soft tissue, with extension of the synovium over the articular cartilage, cartilaginous destruction and subchondral erosion leading to progressive joint deformity and immobility. RA currently affects over half a million subjects in Canada. More than 1,300 patients die annually from gastric bleeding secondary to the use of NSAIDs. The clinical manifestations of R A encompass a spectrum ranging from mild to severe disease, with irreversible joint damage and disability. Our current understanding does not allow for the accurate prediction of the severity in RA. Yet, increasing evidence suggests that early intervention with immunosuppressive drugs may mitigate the relentless progression of this disease and prevent irreversible joint damage. The main objective of this study was to explore selected immune response genes potentially involved in disease severity using patient populations stratified according to disease severity. PCR-RPLP and fragment length analyses of microsatellites were used to study the effect of genetic polymorphisms of nine genes on chromosomes 2, 4, 6, 12 and 22, which included H L A -D R p l , HLA-DQP 1, H L A - D P p l , TNF, IL-ip, IL-1RN, IL-2, IL-2RP, and IFNy. H L A -DRpi alleles containing the Q K R A A or Q R R A A epitope, HLA-DQP 1*302, H L A -DPpl *401 and the 126bp allele of IFNy were identified as high risk alleles, which demonstrate a significant association with the severe RA. In contrast the H L A - D R p i alleles containing the DERRA epitope and HLDRpi*8, HLA-DQP 1*402/502/603/604, HLA-DPpl*201/501/1001 and the 122bp allele of IFNy were significantly decreased in frequency in the population with severe R A compared to those with mild disease, suggesting a potential protective effect of these alleles on the progression of the disease. ii The distribution of genetic polymorphism for all other gene markers were either similar between populations or had a statistically insignificant value in the prediction of the severity of RA. The combination of H L A class II and IFNy high-risk alleles yielded a more specific prediction of disease severity. A logistic regression model was designed which incorporated demographic and clinical data, H L A class II and IFNy gene polymorphisms in order to predict the severity of disease. The model was able to discriminate between mild and severe R A patients with an odds ratio of greater than 447, Chi-square=28.83, p<0.0001 and an overall accuracy of over 90%. i i i TABLE OF CONTENTS Abstract ii List of Tables viii List of Figures xi List of Symbols and Abbreviations xii Acknowledgements xv Chapter One Introduction 1 1.1 Hypothesis 1 1.2 Background 2 1.2.1 Epidemiology 2 1.2.2 Clinical Manifestations 3 1.2.3 Diagnosis 4 1.2.4 Treatment and Management 5 1.2.5 Risk Stratification 9 Chapter Two Immunopathogenesis of Rheumatoid Arthritis 10 2.1 Rheumatoid Arthritis: An Autoimmune Disease 10 2.2 Loss of Tolerance to Autologous Antigens 11 2.3 Mediation of Joint Injury in Rheumatoid Arthritis 15 Chapter Three The Genetics of Rheumatoid Arthritis 18 3.1 Introduction 18 3.2 Human Leukocyte Antigens 19 3.3 T-Cell Receptors 24 iv 3.4 Cytokine Genes 25 3.4.1 T N F Family of Genes 25 3.4.2 IL-1 Family of Genes 27 3.4.3 IL-2 Family of Genes 29 3.4.4 IFNy Family o f Genes 31 3.5 Other Genes 32 Chapter Four Methods 34 4.1 Rationale 34 4.2 Subjects and Selection Criteria 34 4.3 Candidate Genes 35 4.4 Materials and Solutions 36 4.5 Deoxyribonucleic A c i d Extraction and Quantification Methods 40 4.5.1 D N A Quantification: Using Gene Quant 41 4.6 Polymerase Chain Reactions 42 4.6.1 Primer Designs 42 4.6.1.1 H L A Class II Primers 42 4.6.1.2 T N F Microsatellite Primer Sets 43 4.6.1.3 T N F 2 (-308) Primers 44 4.6.1.4 Interleukin 1 Beta (IL-1P) Primers 44 4.6.1.5 Interleukin 1 Receptor Antagonist ( IL-1RN) Primers 44 4.6.1.6 Interleukin 2 (IL-2) Primers 44 4.6.1.7 Interleukin 2 Receptor Beta (IL-2Rp) Primers 44 4.6.1.8 Interferon Gamma (IFNy) Primers 44 4.6.2 Polymerase Chain Reaction Mixes 45 4.6.3 Cyc l ing File Parameters 45 4.7 Restriction Fragment Length Polymorphism 47 4.7.1 H L A Class II Enzyme Digests 47 4.7.1.1 H L A - D P p i Enzyme Digests 47 4.7.1.2 H L A - D Q p l Enzyme Digests 50 4.7.1.3 H L A - D R p i Enzyme Digests 52 4.7.2 T N F 2 Enzyme Digests 55 4.7.3 I L - l p Enzyme Digests 55 4.8 Electrophoresis 56 4.8.1 Polyacrylamide Gel Electrophoresis 57 4.8.2 Agarose Gel Electrophoresis 57 v 4.9 Microsatellite Fragment Analysis 57 4.9.1 Procedures 57 4.10 Statistical Methods 58 Chapter Five Results 61 5.1 Introduction 61 5.1.1 Influence of Demographic and Clinical Factors on the Odds of Severe versus Mild Disease 63 5.2 Class II Human Leukocyte Antigens 65 5.2.1 HLA-DRpl 65 5.2.1.1 The Effect of HLA-DRpl Gene Dose 67 5.2.1.2 Allele Group Frequency 68 5.2.1.3 Influence of HLA-DRpi on the Odds of Severe versus Mild Disease, Adjusted for Demographic and Clinical Variables 69 5.2.1.4 Electrostatic Charges 70 5.2.2 HLA-DQpl 72 5.2.2.1 Allele Group Frequency 73 5.2.2.2 Influence of HLA-DQP1 on the Odds of Severe versus Mild Disease, Adjusted for Demographic and Clinical Variables 74 5.2.3 HLA-DPpl 75 5.2.3.1 Allele Group Frequency 76 5.2.3.2 Influence of HLA-DPpi on the Odds of Severe versus Mild Disease, Adjusted for Demographic and Clinical Variables 77 5.2.4. Combined Effect of HLA-DRpl and DQP1 78 5.2.5 Combined Effect of HLA-DRpi, DQP1 and DPpl 81 5.2.5.1 Combined Influence of HLA-DRpl, DQpl and DPpi on the Odds of Severe versus Mild Disease, Adjusted for Demographic and Clinical Variables 82 5.3 Tumor Necrosis Factor 84 5.3.1 TNF2 (-308) Polymorphism 84 5.3.2 TNF Microsatellite Polymorphism 84 5.4 Interleukin-1 Genes 89 5.4.1 IL-1 Alpha 89 5.4.2 IL-1 Beta 89 5.4.3 IL-1 Receptor Antagonist 91 5.5 Interleukin-2 Genes 92 5.5.1 Interleukin-2 Alpha Microsatellite Polymorphism 92 5.5.2 IL-2 Receptor Beta Microsatellite Polymorphism 93 v i 5.6 Interferon Gamma Gene 94 5.6.1 IFNy Microsatellite Polymorphism 94 5.6.2 Influence of IFNy on the Odds o f Severe versus M i l d Disease, Adjusted for Demographic and Clinical Variables 96 5.6.3 Combined Influence of H L A - D R B 1 , D Q B 1 , D P p i and IFNy on the Odds of Severe versus M i l d Disease, Adjusted for Demographic and Cl inical Variables 98 5.6.4 Goodness o f Fit Analysis for Combined Influence of H L A - D R B 1 , D Q P 1 , D P p i a n d IFNy on the Odds of Severe versus M i l d Disease, Adjusted for Demographic and Clinical Variables 99 5.6.5 A Reduced Model Combining Only the Effects of H L A - D R p i and IFNy 99 5.6.6 Combined Influence of Only H L A - D R p l a n d IFNy on the Odds of Severe versus M i l d Disease, Adjusted for Demographic and Clinical Variables 100 5.6.7 Goodness of Fit Analysis for Combined Influence of Only H L A - D R p l a n d IFNy on the Odds of Severe versus M i l d Disease, Adjusted for Demographic and Clinical Variables 103 5.7 Confirmation of H L A - D R p i and IFNy Results 104 5.7.1 Combined Influence of H L A - D R p i a n d IFNy Pooled Data on the Odds of Severe versus M i l d Disease, Adjusted for Demographic and Clinical Variables 107 Chapter Six Discussion 109 Chapter Seven Summary 129 References 130 Appendix \5M vii LIST OF TABLES Table Page 4.1 Sty I enzyme digest patterns. 55 4.2 Fnu 4HI and TaqI enzyme digest patterns. 56 5.1 Demographic and clinical characteristics of patients with severe and mild Rheumatoid Arthritis. 62 5.2 Logistic regression analysis using combination of four demographic and clinical data to predict the outcome of RA. 63 5.3 Detailed results of logistic regression analysis examining the influence of demographic and clinical parameters on the outcome of RA. 64 5.4 Frequencies of patients and controls expressing HLA-DRpi alleles. 66 5.5 Frequencies of patients and controls expressing HLA-DRpl alleles when individuals with HLA-DRpl *0401/0404 alleles were excluded. 67 5.6 Frequencies of patients and controls expressing high risk alleles, low risk alleles, both or neither at HLA-DRpl locus. 69 5.7 Logistic regression analysis using demographic, clinical data and HLA-DRpi markers to predict the outcome of RA. 69 5.8 Frequencies of patients and controls according to the amino acid charges at positions 70 and 71. 71 5.9 Frequencies of patients and controls expressing HLA-DQP1 alleles. 72 5.10 Frequencies of patients and controls expressing high risk alleles, low risk alleles, both or neither at HLA-DQP1 locus. 73 5.11 Logistic regression analysis using demographic, clinical data and HLA-DQP1 markers to predict the outcome of RA. 74 5.12 Frequencies of patients and controls expressing HLA-DPpi alleles. 75 5.13 Frequencies of patients and controls expressing high risk alleles, low risk alleles, both or neither at HLA-DPpi locus. 76 5.14 Logistic regression analysis using demographic, clinical data and HLA-DPpl markers to predict the outcome of RA. 77 v i i i 5.15 Frequencies of patients and controls expressing HLA-DR.pi/DQPl haplotypes. 79 5.16 Frequencies of patients and controls expressing HLA-DR.pi/DQp/DPpi allele combinations. 81 5.17 Logistic regression analysis using demographic, clinical data and all HLA class II gene markers to predict the outcome of RA. 83 5.18 TNF (-308) polymorphism frequencies of patients with RA and controls. 84 5.19 TNFa microsatellite polymorphism frequencies of patients with RA and controls. . 85 5.20 TNFb microsatellite polymorphism frequencies of patients with RA and controls. 86 5.21 TNFc microsatellite polymorphism frequencies of patients with RA and controls. 87 5.22 TNFd microsatellite polymorphism frequencies of patients with RA and controls. 87 5.23 TNFe microsatellite polymorphism frequencies of patients with RA and controls. 88 5.24 IL-1P polymorphism frequencies of patients with RA and controls. 90 5.25 IL-1 receptor antagonist polymorphism frequencies of patients with RA and controls. 91 5.26 IL-2cc microsatellite polymorphism frequencies of patients with RA and controls. 92 5.27 IL-2 receptor beta microsatellite polymorphism frequencies of patients with RA and controls. 93 5.28 IFNy microsatellite polymorphism frequencies of patients with RA and controls. 95 5.29 Frequencies of patients and controls expressing high risk alleles, low risk alleles, both or neither alleles at IFNy locus. 96 5.30 Logistic regression analysis using demographic, clinical data and IFNy markers to predict the outcome of RA. 97 5.31 Logistic regression analysis using demographic, clinical data, all HLA class II and IFNy gene markers to predict the outcome of RA. 98 ix 5.32 Goodness of fit analysis showing the discriminatory power of the model in section 5.6.3. 99 5.33 Frequencies of patients and controls expressing high risk alleles, low risk alleles, or others at HLA-DRpl and IFNy loci. 100 5.34 Logistic regression analysis using demographic, clinical data, HLA-DRpi and IFNy gene markers to predict the outcome of RA. 101 5.35 Detailed results of logistic regression analysis using demographic, clinical data, HLA-DRpi and IFNy gene markers to predict the outcome of RA. 102 5.36 Goodness of fit analysis showing the discriminatory power of the model in section 5.6.6. 103 5.37 IFNy microsatellite polymorphism frequencies of patients with RA and controls (pooled data). 105 5.38 Frequencies of patients and controls expressing high risk alleles, low risk alleles, both or neither at IFNy locus (pooled data). 106 5.39 Logistic regression analysis using demographic, clinical data, HLA-DRpi and IFNy gene markers to predict the outcome of RA (pooled data). 107 5.40 Detailed results of logistic regression analysis using demographic, clinical data, HLA-DRpl and IFNy gene markers to predict the outcome of RA (pooled data). 108 x LIST OF FIGURES Figure Page 4.1 HLA-DPpl RFLP: A l primer set. 48 4.2 HLA-DPpl RFLP: A2 primer set. 48 4.3 HLA-DPpl RFLP: A3 primer set. 49 4.4 HLA-DPpl RFLP: A4 primer set. 49 4.5 HLA-DQP 1 RFLP: DQR1 amplification. 50 4.6 HLA-DQpl RFLP: DQR2 amplification. 51 4.7 HLA-DRpl RFLP: DR1 amplification. 52 4.8 HLA-DRpl RFLP: DR2 amplification. 52 4.9 HLA-DRpl RFLP: DR52A amplification. 53 4.10 HLA-DRpl RFLP: DR4 amplification. 54 4.11 HLA-DRpi RFLP: DR7, 9, 10 amplifications. 54 5.1 Electropherogram showing five alleles (122-130bp) at the polymorphic region in the first intron of the IFNy gene. 94 5.2 IFNy allele frequencies in Caucasian patients with severe and mild RA compared to controls. 104 xi L I S T O F S Y M B O L S A N D A B B R E V I A T I O N S Symbols DP H L A class II marker DQ H L A class II marker DQL3 5' generic HLA-DR primer DQR1 3' group specific HLA-DR primer DQR2 3' group specific HLA-DR primer DR H L A class II marker IFNf interferon forward primer IFNr interferon reverse primer IL-1 pf interleukin-1 beta forward primer IL-1 Br interleukin-1 beta reverse primer IL-lRaf interleukin-1 receptor antagonist forward primer IL-lRar interleukin-1 receptor antagonist reverse primer IL-2f interleukin-2 forward primer IL-2r interleukin-2 reverse primer IL-2RBf interleukin-2 receptor beta forward primer IL-2Rpr interleukin-2 receptor beta reverse primer IL-ip interleukin-1 beta IL-2 interleukin-2 IL-2RP interleukin-2 receptor beta IR4f TNFa forward primer IR2r TNFa reverse primer IRlf TNFb forward primer IR5r TNFb reverse primer IR6f TNFc forward primer IR7r TNFc reverse primer IR1 If TNFd forward primer IR12r TNFd reverse primer IR13f TNFe forward primer IR14r TNFe reverse primer MgCb magnesium chloride NaHC03 sodium bicarbonates NaOH sodium hydroxide Na2EDTA sodium ethylenediaminetetraacetic acid NH4 ammonium NH4CI ammonium-chloride xii Abbreviations AA amino acid Ab antibody ABI applied biosystems ACD Acid citric dextrose ACR American College of Rheumatology Ags antigens AG adenine-guanine AIDS acquired immunodeficiency syndrome APC antigen presenting cell B-cell bone marrow derived lymphocyte B.M bone marrow bp base pair(s) C complement CA cytosine-adenine C A M cell adhesion molecules CAT computerized axial tomography CD Crohn's disease CD4 cluster of differentiation antigen CD8 cluster of differentiation antigen dATP deoxy-adenosine 5' triphosphate dCTP deoxy-cytosine 5' triphosphate DE aspartic acid-glutamic acid DERAA aspartic acid-glutamic acid-arginine-alanine-alanine dGTP deoxy-guanine 5' triphosphate DHFR dihydrofolate reductase DMARD disease modifying antirheumatic drug DNA deoxyribose nucleic acid DNTR dinocleotide tandem repeats dTTP deoxy-thymidine 5' triphosphate EAF extra articular features EBV Epstein-barr virus EDTA ethylene diamine tetraacetic acid ESR erythrocyte sedimentation rate Fc fragment crystalline GG guanine-guanine GM-CSF granulocyte-macrophage colony-stimulating factor gP glycoprotein GT guanine-thymine HLA Human leukocyte antigen HPLC high-pressure liquid chromatography IBD inflammatory bowel disease ICAM intercellular adhesion molecule IDDM insulin-dependent diabetes mellitus IFN interferon IgG immunoglobulin-G li invariant peptide LPS lipopolysaccharide LT lymphotoxin MCP-1 monocyte chemotactic protein-1 x i i i M R I magnetic resonance imaging N S statistically non-significant N S A I D nonsteroidal anti-inflammatory drug O R odds ratio P C R polymerase chain reaction Q K R A A glutamine-lysine-arginine-alanine-alanine Q R R A A glutamine-arginine-arginine-alanine-alanine QS quantum sufficiat R A Rheumatoid arthritis R B C red blood cell r h I L - I R N recombinant human interleukin-1 receptor antagonist R N receptor antagonist R N A ribonucleic acid R F Rheumatoid factor R F L P restriction fragment length polymorphism S A R D slow acting antirheumatic drug SDS sodium dodecyl sulphate S E shared epitope S L E systemic lupus erythromatosus SSOP sequence specific oligonucleotide probe s T N F R soluble tumor necrosis factor receptor T A P transporter in antigen processing T-cell Thymus-derived lymphocyte T C R T-cell antigen receptor T E M E D N , N , N-N-tetramethylethylenediamine T G F transforming growth factor Th helper T-lymphocyte T N F tumor necrosis factor U C ulcerative colitis U V ultra-violet V variable V N T R variable nucleotide tandem repeat W B C white blood cell xiv ACKNOWLEDGEMENTS I would like to acknowledge all o f the people who have assisted me with my work presented in this thesis. M y deepest gratitude goes to my supervisor, Dr. Paul Keown, who has given me his utmost support and guidance throughout this project. In addition, my sincere thanks goes out to all of the patients that agreed to participate in this study, without whom this study would not have been possible. I would also like to thank Dr . A . Chalmers, D . Lacaille, C . Beattie, who helped to recruit and assess the patients in the study. I would also like to express my appreciation to Dr. D . Hoar, Dr. D . Horsman and Dr. F. Jirik who have guided and assisted me with the many steps throughout this thesis. I would also like to acknowledge C . Home and all the staff of the Immunology Laboratory at Vancouver General Hospital for their technical insight and cooperation. K . Rangno and M . Cummings did a terrific job in their recruitment o f patients and in blood collection. S. Adamat and M . Anderson were of great assistance in the microsatellite poymorphism work. I would also like to thank student colleagues for the intellectual discussions we shared over the past few years. I would also like to thank my parents for their encouragement, optimism and complete support. Finally, I would like to express my sincere gratitude to my wife for her patience, encouragement and ongoing support. xv C H A P T E R O N E I N T R O D U C T I O N 1.1 Hypothesis Our current understanding of rheumatoid arthritis (RA) suggests a sequential mechanism of molecular pathogenesis in which three immunological steps may be postulated: (a) exposure to exogenous antigen(s) provides the stimulus to break peripheral tolerance to endogenous auto-antigen(s) due to molecular mimicry between relevant epitope(s) (b) expression of a restricted set of human leukocyte antigen ( H L A ) and T-cell antigen receptor (TCR) genes capable of presenting and recognizing these exogenous antigen(s) enables the construction of a tri-molecular complex which expands appropriate arthritogenic T-cell clones (c) inheritance of defined polymorphisms within relevant cytokine genes leads to altered cytokine gene expression and potentiation of immunologic activity, functional helper T-lymphocyte 1 (Thl ) predominance within the joints, and thus to their irreversible destruction. The objective of this study is to identify specific polymorphisms: within the region of class II H L A , tumor necrosis factor (TNF), interleukin-1 beta (IL-1P), interleukin-1 receptor antagonist ( IL-1RN) , interleukin-2 (IL-2), interleukin-2 receptor beta (IL-2RP) and interferon gamma (IFNy) gene loci which are closely associated with disease severity in R A . This may enable the identification of patients who are at high or low risk of disease progression, and thus aid in the selection of optimal therapy at an early stage in the disease course. I 1.2 Background 1.2.1 Epidemiology Rheumatoid arthritis is a chronic, debilitating disease characterised by inflammatory thickening of articular soft tissue, with extension of the synovium over the articular cartilage, cartilaginous destruction and subchondral erosion leading to progressive joint deformity and immobility (Jacobson, DL. et al. 1997 and Krane, SM. et al. 1986). RA affects about 0.8 percent in the world's population with a prevalence ranging from 0.3 to 2.1 percent. All races and ages are affected although the incidence and severity of the disease differ widely throughout the world. Residents of North America are at an increased risk of acquiring RA compared to residents of the Caribbean and Africa (Silman, AJ. et al 1994). RA currently affects over half a million people in Canada, imposing an immense economic burden of $668 CAN., monthly per patient (The North American Rheumatoid Arthritis Disease Management Study Group, 1998). Similarly, the total costs of RA in the United States have been estimated up to $14 billion U.S., per year (Callahan, L. 1998). Treatment of RA consumes 14% of all drug prescription costs and is attended by significant clinical risks (The North American Rheumatoid Arthritis Disease Management Study Group, 1998). RA affects three times as many women as men, but this difference diminishes with increasing age. RA generally develops between the ages of 40 and 50 and is six times more frequent in women in their 60's compared to women in their 20's (Felson, DT. 1992). Hormones and neurotransmitters may be important factors in the onset and/or progression of RA. Manifestations of RA subside in 70% of women during pregnancy and reoccur in the early postpartum period, perhaps due to changes in the bioactivity of prolactin (Anisman, H. et al. 1996). Education and socio-economic status are closely associated with the onset and progression of RA. Individuals with less education and a lower socio-economic status are at higher risk of acquiring RA compared to those with higher education (Pincus, T. et al.1988). Individuals with history of 2 cigarette smoking may be at an increased risk of developing R A (Silman, A J . et al 1996). Stress has also been reported to be a factor in the progression of R A . R A occurs more frequently in the winter than in the summer (Felson, T D . 1992). 1.2.2 Clinical Manifestations R A presents in 60% of patients with fatigue, anemia, generalized weakness, anorexia and non-specific musculoskeletal symptoms before the appearance of synovitis. In about 10% of patients the onset o f R A is acute and aggressive with swelling and tenderness of the joints o f the hands, wrists, knees and feet and other extra-articular features. The pattern of the joint involvement in R A is symmetric. Cl in ical symptoms include swelling, morning stiffness, pain, limitation o f movement, warmth and tenderness of the joint capsule and sensitivity o f the joints to stretch and/or distention (Lipsky, PE . , 1998 and Wolfe, F . 1997). The clinical manifestations of R A encompass a spectrum of disease activity (Wolfe, F. 1997). A t one end o f this spectrum, progression of the disease is slow and self-limiting with minimal or no systemic effects, the patient is generally fully mobile and the disease is controlled with minimal therapy. A t the other end of the spectrum, occurring in about one out of six patients with R A , the disease progresses inexorably to a systemic debilitating disorder. Disease activity in these patients is progressive with both articular and extra articular features ( E A F ) . These latter include rheumatoid nodules, pleuropulmonary manifestations, rheumatoid vasculitis, neurological manifestations, Sicca syndrome and Felty's syndrome. Joint destruction in this group o f patients generally leads to immobility and handicap within a few years after the onset o f the R A (Lipsky, PE. , 1998 and Morgan, GJ . et al. 1993). 3 1.2.3 Diagnosis RA is a complex disease with no single diagnostic indicator. Clinical, laboratory, and imaging techniques are therefore employed in combination, and criteria have been established to promote a consistent and universal diagnosis of RA. Presently, RA is most commonly defined by the 1987 American College of Rheumatology criteria (1987 ACR criteria) for purposes of study. These include 1. Morning stiffness, 2. Arthritis of three or more joint areas, 3. Arthritis of hand joints, 4. Symmetric arthritis, 5. Rheumatoid nodules, 6. Serum Rheumatoid factor (RF), and, 7. Radiographic changes. Patients who have at least four of these seven criteria are considered to have definite RA (Arnett, FC. et al. 1988). Due to the nonspecific symptoms of the disease in the early stages, the diagnosis of RA may be difficult and often there is a substantial time lag from the onset of the disease to its diagnosis (Chan, KA. et al 1994). A detailed history and physical examination are critical to verify the presence or absence of symptoms and their symmetrical or non-symmetrical pattern. Tender and swollen joint counts indicate the degree and extent of inflammatory activity in the joints of patients with RA. The recent data shows progression of the disease is rapid (Lipsky, PE. 1998). Yet none of the currently available prognostic markers are capable of accurately identifying the outcome of the disease at an early stage in the individual patient. Laboratory measures may be contributory, but are seldom diagnostic. Sixty percent of patients with RA have a high titer of autoantibodies to the Fc (fragment crystalline) region of immunoglobulin G (IgG), known as RF, in their blood. Studies have shown that there is a difference in the avidity of RF in patients with RA. High-avidity RF is directly correlated to disease activity and patients with high-avidity RFs have higher tender joint count scores (Newkirk, MM. et al. 1995). In general, the high titer of RF in these patients is associated with more complex and systemic disease with EAF and rapid progression (Kalsi, J. et al. 1993). However, the presence of RF is not specific to RA 4 since RF in high titers is also seen in other diseases like cryoglobulinemia (Gorevic, PD. et al. 1980) and a low titer of RF may be seen in 5% of healthy individuals. Patients with RA frequently suffer from the normochormic and normocytic anemia typical of many inflammatory diseases. This is thought to be the result of ineffective erythropoiesis and directly associated with activity of the disease. In almost all RA patients there is an increase in erythrocyte sedimentation rate (ESR) which directly correlates with disease activity. Joint aspiration is often employed to assist in diagnosis. The synovial fluid of patients with inflammatory arthritis is generally more turbid, less viscous and higher in protein concentration than normal synovial fluid. An increase in the white cell count, especially of polymorphonuclear leukocytes, in the synovial fluid also indicates the presence of an inflammatory arthritis, although this is not specific to RA (Lipsky, PE. 1998). X-ray, computerized tomography or magnetic resonance imaging (MRI) scans and other forms of imaging can show damage to articular surfaces, and are used to measure the degree of destruction of the joints and articular cartilage and inflammation (Lipsky, PE. 1998). However, the usefulness of these techniques is limited to the later stages of the disease, when the progression of RA has already led to the destruction of the joints. Therefore, a primary reason for conducting this study was to identify reliable predictive marker(s) capable of determining the outcome of RA at an early stage. 1.2.4 Treatment and Management There is no specific cure for RA. Treatment of this disease therefore involves a variety of supportive, pharmacological, physical and surgical approaches directed to reducing the inflammation and destruction of the joints and to limiting the systemic features of the disease. RA can have an enormous impact on the physical well being of an individual and can ultimately limit functional capability. Education is important to help patients and their families adjust to the 5 disease, reduce unnecessary stress on the joints, and improve overall outcome. Psychotherapy and counseling may be helpful in this adjustment. Rest, relaxation and splinting are important in reducing joint inflammation while cooling or heating of the affected joints is often helpful in relieving pain. Proper exercise and nutrition, which help to maintain muscle strength and mobility, round out the treatment program. None of the currently available pharmaceutical agents is capable of curing RA; however, all share a similar goal of reducing the symptoms and/or delaying the progression of the disease. Patients may be treated with one or a combination of drugs to control the pain and progression of RA. Nonsteroidal anti-inflammatory drugs (NSAIDs) are generally used as the first step of management. NSAIDs block the activity of cyclooxygenase in the production of prostaglandins, prostacyclin and thromboxanes and thus act as analgesic and anti-inflammatory drugs. Side effects from this group of drugs are serious and include gastric bleeding, hemorrhage and even death (Simons, LS. et al. 1996 and Brooks, PM. et al. 1991). Cyclooxygenase-2 (COX-2) inhibitors are a newer and more specific class of anti-inflammatory drugs used in the management of RA (Simon, LS. et al. 1999, Choy, EHS. et al. 1997). These drugs are specifically designed to inhibit COX-2 and have no effect on COX-l activity, and therefore potentially fewer adverse effects on the gastrointestinal tract. The results from the clinical trial of Celecoxib (a COX-2 inhibitor) in the management of RA are promising, although the gastrointestinal tract adverse effects are not completely eliminated (Simon, LS. et al. 1999). The long-term side effects, if any, of COX-2 inhibitors remain to be identified. Disease modifying antirheumatic drugs (DMARDs), also known as slow acting antirheumatic drugs (SARDs), generally have no direct anti-inflammatory or analgesic effect and are usually prescribed simultaneously with NSAIDs. This group of drugs includes gold compounds, 6 antimalarials, Sulfasalazine, D-penicillamine, Azathioprine, Methotrexate and Leflunomide ( O ' D e l l , JR. 1999, Smolen, JS. et al. 1999 and Tett, SE . 1993). These agents mitigate the rate o f progression and course of the disease although their mechanisms of action are different. In brief, gold compounds decrease T-cell proliferation, T N F a , lysosomal enzyme release, and neutrophil chemotaxis and random migration. Antimalarial drugs function in the acidic environment within the Golg i apparatus, lysosomes and endosomes, affecting multiple aspects of immune regulation including antigen (Ag) processing and R F production. They also decrease T-lymphocyte proliferation, IL-1 production and activity of enzymes capable of joint destruction. Sulfasalazine inhibits various prostanoids like leukotriene B 4 and scavenges reactive oxygen molecules. D -penicillamine inhibits collagen cross-links, therefore slowing the deposition of collagen in fibrotic processes. Azathioprine is a purine analogue, which inhibits deoxyribose nucleic acid ( D N A ) synthesis and cell proliferation. Finally Methotrexate inhibits dihydrofolate reductase ( D H F R ) leading to inhibition o f adenosine production and cell proliferation. D M A R D s are also associated with considerable unwanted side effects including renal impairment and liver toxicity, thus requiring close patient monitoring (Conaghan, P G . et al. 1996, Kremer, J M . et al. 1994 and Kavanaugh, A F . 1992). Glucocorticoids may be administered by oral, intravenous or intra-articular routes. A dose o f less than 7.5 mg/day o f prednisone often helps to control R A symptoms, although monthly pulses o f high-dose glucocorticoid therapy used in conjunction with D M A R D s seem to have a more potent effect in suppressing R A symptoms. Glucocorticoids specifically inhibit macrophage and T-cell function by inhibiting the production of interleukin-1 (IL-1), T N F , IL-2 , interlukin-4 (IL-4) and interlukin-6 (IL-6), and exert a non-specific immunosuppressive and anti-inflammatory activity through the redistribution of lymphocytes from the vascular components to the lymphoid tissue. They also inhibit migration of monocytes to the site of inflammation by blocking the synthesis and the release o f chemotactants. Cushingoid changes, weight gain, fatigue and osteoporosis are 7 some of the negative side effects associated with use of glucocorticoids (Choy, EHS. et al. 1997, Kirwan, JR. 1995 and Cash, JM. et al.1994). Recently, more specific immunosuppressive drugs have been used in the management of RA. Cyclosporin A is capable of suppressing T-cell proliferation and activation via down-regulation of IL-2 gene transcription, which in turn reduces disease activity and bone erosion in RA (Tugwell, P. et al 1995 and 1990). Unfortunately, it may have toxicity and side effects including nephrotoxicity and increased blood pressure (Ludwin, D. et al. 1993). Leflunomide is one the most recent immunomodulatory drugs which inhibits pyrimidine synthesis through inhibition of dihydro-orotate dehydrogenase (Smolen, JS. et al. 1999). The result of clinical trail for Leflunomide are promising and indicate less adverse side effects and better health-related quality of life associated with the use of Leflunomide compared to Sulphasalazine and Methotrexate (Smolen, JS. et al. 1999 and Strand, V. et al. 1999). These drugs are generally reserved until patients are unresponsive to all other medications, and are then often used in combination with other DMARDs. Finally, a new group of biological therapies that are specifically designed for RA is now emerging (Sander, O. et al. 1998). Preliminary data from use of anti-tumor necrosis factor monoclonal antibodies or soluble TNF receptors indicates impressive success in reducing disease activity and offers promise in reducing the resultant joint injury (Tak, PP. et al.1997, 1995, Paleolog, EM. et al. 1996, Elliott, MJ. et al 1994, and Feldmann, M. 1992). The potential of treating RA using anti-CD4 monoclonal antibodies and recombinant human interleukin-1 receptor antagonist (rhIL-IRN) has also been studied (Odeh, M. 1997, Drevlow, BE. et al. 1996, Tak, PP. et al 1995 and Horneff, G. 1993). Results from these studies on anti-CD4 are conflicting and show no significant benefit; however, results of the effect of rhIL-IRN show a clear benefit for RA patients from this therapy. Gene therapy, which involves the cloning and transfection of 8 genes like soluble T N F receptor, human interleukin-1 receptor antagonist, interleukin-2 receptor and anti-inflammatory cytokine genes including interleukin-10 (IL-10) are considered as an exciting future means o f management (Evans, C H . et al. 1996). Most of these new therapies are still limited to clinical trials, and the long-term side effects are not known. 1.2.5 Risk Stratification Increasing evidence suggests that treatment with powerful and specific immunomodulatory agents in the early stages o f R A can delay the progression of the disease and the destruction o f the joints (Tak, PP. et al. 1995 and Tett, SE . 1993). Early intervention with gold, Cyclosporin A or Azathioprine may mitigate the relentless progression of this disease and prevent irreversible joint damage (Lipsky, P E . 1998). To optimize this approach, however, it is essential to understand the mechanisms responsible for disease progression and to define those patients who are at greatest risk o f severe destructive arthropathy. Since the manifestation and outcome o f the disease differs among R A patients the availability of predictive marker(s) that can distinguish accurately between patients who are destined to develop severe or mild R A is essential for a successful management plan. Accurate stratification of risk would make it possible to individualize therapy; to offer appropriate, powerful (and often expensive) therapy to those at highest risk, while shielding those with mild disease from the toxic side effects of unnecessary anti-rheumatic drugs. Many attempts have been made to identify means of predicting disease progression in the early stages of R A . Although various markers, including R F titer and H L A - D R p l t y p i n g , have been used in the past, none has shown sufficient sensitivity, specificity and consistency to predict the severity o f R A . 9 C H A P T E R T W O I M M U N O P A T H O G E N E S I S O F R H E U M A T O I D ARTHRITIS 2.1 Rheumatoid Arthritis: An Autoimmune Disease The exact cause o f R A is not clear. However, it is believed that R A is triggered by a nonspecific inflammatory response to a yet unknown exogenous antigen which results in the breakdown of tolerance to specific self antigens. In the joints, the activation of T-cells (primarily CD4+) and the production of proinflammatory and anti-inflammatory cytokines follow this initial inflammation (Panayi, G S . et al. 1998 and Urbina-Joiro, H . et al 1998). The latter act to control the immune response and the inflammation. The former lead to further activation of other mononuclear and polymorphonuclear leukocytes (Feldmann, M et al. 1996 and Salmon, M . et al. 1995). Production and release o f major proinflammatory cytokines like IL-1 , IL-2, IFNy, granulocyte-macrophage colony-stimulating factor ( G M - C S F ) , tumor necrosis factor alpha (TNFoc), transforming growth factor beta (TGFP) and macrophage-CSF orchestrate a much larger and systemic inflammatory response that eventually leads to rheumatoid synovitis and extra articular injury. Clinical and histological data suggest that R A is an immune-mediated disease, whose onset and progression is closely associated with activity of the immune system in genetically susceptible hosts (Odeh, M . 1997, Moreland, L W . et al 1993 and Panayi, GS . et al 1992). The fact that a large percentage of R A patients have high titers of R F in their blood and a large number of CD4+ T-cells infiltrating their joints suggests the importance of B - and T-cell activity in R A . T-cells play an essential role in the onset and/or progression of chronic and severe R A (Firestein, G S . et al. 1990). Investigators who believe that a certain type of T-cell clone is involved in the progression of R A , 10 hypothesize that a superantigenic insult activates T-cells bearing specific classes of TCR and results in the breakdown of self-tolerance. This is followed by clonal expansion of auto-reactive T-cells, which are specific for an arthritogenic antigen (Kotzin, BL. et al. 1993). Macrophage activity, cytokine profile, inflammation of the joints and further progression of the disease to a systemic condition in severe RA patients are further evidence of immune-associated responses in the progression of the disease. Many antirheumatic drugs exert their effects by regulating one or another of these mechanisms involved in the normal process of immune response. Progression of RA is closely associated with both genetic and environmental factors (Harris, E. Jr. 1990). Among the former, HLA-DRpi alleles appear to play an important role. Infection is considered to be one of the major environmental factors involved in the etiology of rheumatic diseases (Krause, A. et al 1996), and could represent a source of the superantigen alluded to above. Yet, lack of evidence for a consistent foreign organism or antigen in the joints or blood of RA patients and the continuous self-destructing characteristic of the severe form of RA, in particular, argue against this being due to infection alone. Rather, RA is accepted to belong to the family of autoimmune diseases in which the immune system initiates a cell-mediated or a humoral immune response against the self-antigen(s) leading to tissue or organ injury following the breakdown of mechanisms that normally control self-tolerance. Over 40 autoimmune diseases have been identified, including rheumatoid arthritis, insulin-dependent diabetes mellitus (IDDM), inflammatory bowel disease (IBD), Crohn's disease (CD), systemic lupus erythromatosus (SLE) and others (Hang, LM. et al. 1997), which affect 5 to 10 percent of the general population (Leslie, RD. et al. 1994 and Hang, L M . et al. 1997). 11 2.2 Loss of Tolerance to Autologous Antigens Tolerance to self antigens is one of the most unique and important characteristics of the immune system. Self-tolerance is mediated by mechanisms that are acquired during the maturation of lymphocytes in the generative lymphoid organs (thymus for T-cells and bone marrow (B .M. ) for B -cells) and actively maintained in the peripheral lymphoid organs such as lymph nodes and spleen. Loss o f self-tolerance in the immune system may permit an immune reaction to occur against self-peptides under appropriate conditions (Abbas, A . 1994). While the mechanisms underlying this are varied and are not always clear, it has been reported that there is a genetic relationship between the breakdown of self-tolerance in some autoimmune diseases like R A and certain inherited and/or mutated genes (Brinkman, B M N . et al. 1996 and Gregersen, P K . et al 1987). Other studies have reported an association between the onset of autoimmune diseases, including R A and infections, tissue and humoral factor injuries, and factors causing immunodeficiency diseases (Albani, S. et al 1996, 1995). The principal reasons normally cited for loss of self-tolerance and subsequent progression of autoimmune diseases include: 1) Failure to delete self-reactive clones; 2) Exposure to exogenous antigen(s) leading to the breakdown of peripheral tolerance; and 3) Abnormal regulation of lymphocyte response(s) resulting in the potentiation of immunologic activity. Autoimmunity may result from a failure of deletion of either auto-reactive B - and/or T-cells, although the latter appear more important because of their critical role as regulators of key immune responses. This is important in autoimmune diseases that are primarily T-cell driven, like R A (Salmon, M . et al. 1995). T-cell education in the thymus occurs via both positive and negative selection, in which H L A molecules play a key role for presentation of self-antigens. H L A genes thus influence thymic selection and the residual lymphocyte repertoire (Goronzy, JJ. et al 1993), 12 and may lead to the presence of auto-reactive T-lymphocytes in the periphery (Hughes, A L . et al 1996). This could be a potential cause of R A i f the peripheral tolerance process does not deactivate these T-lymphocytes. The demonstration of certain T-cell clones in the rheumatic joints and specific H L A molecules associated with the progression of R A , supports the concept that this may result from the persistence of auto-reactive T-cell clones (Salmon, M . et al. 1995 and Chaouni, I. et al. 1990) responsible for local production of proinflammatory cytokines (Katsikis, P D . et al 1994 and Yanni , G . et al.1993). Exogenous antigen(s) may trigger the breakdown of self-tolerance and the onset of autoimmune diseases by at least three different mechanisms which include molecular mimicry between exogenous antigen(s) and self-peptide(s), superantigenic stimulation of auto-reactive cells, and/or non-specific stimulation of the immune system leading to enhanced production of IL-2. Structural similarities between two antigenic peptides can result in the phenomenon o f molecular mimicry. These similarities are often seen in the evolutionary conserved peptide sequences among a vast number of species. Molecular mimicry between foreign and endogenous (self) Ags may result in an immune response, with lymphocyte proliferation and/or antibody (Ab) release against both foreign and self-antigen(s) leading to autoimmune injuries (Davies, J M . 1997). A n example of this is rheumatic fever which occurs after streptococcal infection and is characterized by the production of anti-streptococcal Abs, and acute endo-myocarditis (Hang, L M . et al. 1997). Studies have shown similar findings of molecular mimicry between specific subtypes o f heat shock proteins in primitive bacteria and the shared epitope (SE) of H L A - D R . p i in R A patients (Albani, S. et al 1995 and 1992). Moreover, the similarity of glycoprotein 110 of Epstein-Barr virus ( E B V ) to the SE and its presence in the rheumatic joints also suggest a cross-reaction between foreign and self-antigens may be involved in the onset and/or progression of R A (Hang, L M . et al. and Albani, S. et al 1996). 13 Polyclonal activators and superantigens are capable of stimulating T- and B - self-reactive lymphocytes that have escaped the negative selection of central tolerance, irrespective o f their antigenic specificity (Abbas, A K . et al 1994). Superantigens like lipopolysaccharide (LPS) and staphylococcal enterotoxin are capable of activating B - and T-lymphocytes, respectively, by interacting with specific surface molecules, through a receptor-independent pathway, and lead to polyclonal activation and expansion. L i , S. et al. (1996) have reported the presence o f certain types of foreign superantigens like staphylococcal enterotoxin A and B , and streptococcal M and their involvement in activating auto-reactive T-cells in R A patients, although these results are not consistent (Albani, S. 1996). Polyclonal lymphocyte activation could also lead to systemic, rather than organ-specific, autoimmune injury, and this mechanism may be partially responsible for the progression of systemic R A in the severe form of the disease. In the maintenance of peripheral tolerance, self reactive T-cells normally encounter co-stimulator deficient antigen presenting cells (APCs) in the periphery and remain anergized. However, this peripheral tolerance may be overcome i f the A P C s that present self-Ag(s) to auto-reactive T-lymphocytes do so in the presence of infectious foreign Ags. Under these circumstances the A P C s may express co-stimulator molecules and activate T-lymphocytes. Supporting studies (Grassi, W . et al. 1998, Lipsky, P E . 1998, Kraft, S M . et al. 1985 and Svartz, N . 1975) in R A suggest that a fever of 38\u00C2\u00B0C-40\u00C2\u00B0C in some R A patients may indicate the presence of a simultaneous infection. Furthermore, infectious agents like hepatitis C virus and parvovirus B19 have been reported in the peripheral blood, inflammatory tissues and synovial fluid of patients with R A (Caramschi, P. et al. 1996 and Schnitzer, TJ . et al. 1996) although these results have not been consistently reproducible. Studies conducted on experimental mice (Abbas, A K . et al. 1994) suggest that over-expression of IL-2 could also overcome peripheral tolerance, and may be one of the mechanisms responsible for the breakdown of tolerance in R A patients. However, the available data on the presence and/or 14 over-expression o f IL-2 in the joints of patients with R A is inconsistent and contradictory (Buchan, G.et al 1988 and Firestein, G S . et al 1988). Other studies suggest that the B-lymphocytes of ( N Z B x N Z W ) F I mice, which develop a disease resembling human S L E , are hyper-responsive to Ags. The amount o f antibodies produced by these B-cells upon recognition of self- or foreign-antigens is greater than in normal B-cells (Abbas, A . et al. 1994). Yet, there is no data supporting a higher production o f antibodies by the B-lymphocytes of R A patients. Abnormally high or low production of certain regulatory cytokines may lead to abnormal stimulation of auto-reactive lymphocytes resulting in potentiation o f autoimmune injury. Changes in cytokine production can also alter the sub-populations of T-cells (Th2 versus T h l ) . Proinflammatory cytokines IL-2, IFNy and interlukin-12 (IL-12) promote the differentiation of ThO C D 4 + T-cells to T h l subsets, which may in turn suppress the number of regulatory or suppressor Th2 T-lymphocytes leading to the onset and/or progression of autoimmune disorders (Abbas, A . et al. 1994). Abnormally high or low cytokine production has been associated with genetic polymorphisms in humans and could be one of the mechanisms responsible for the onset and/or progression of autoimmune diseases like R A (Mu, H . et al. 1999 and Mulcahy, B . et al. 1996). 2.3 Mediation of Joint Injury in Rheumatoid Arthritis The synovial joints are the most important and primary sites o f R A activity. Histological findings from the inflamed joints of patients with R A indicate that the major cell types in the synovial membrane include a mixture of primarily CD4+ T-cells, CD8+ T-cells and macrophages. Mino r cell types include fibroblasts, plasma cells, endothelium and dendritic cells. This cell mixture is different from that in the synovial fluid in which the major cell type is the neutrophil (Kingsley, G . et al. 1997 and Feldmann, M . et al. 1996). 15 Studies on RA joint cells suggest that among the proinflammatory cytokines there is a hierarchy in which TNFoc is predominant and is the driving force for the production of other inflammatory cytokines (Brennan, FM. et al. 1989). The presence of chemokines in high physiological concentration in the joints results in the increased expression of cell adhesion molecules (CAMs) at the cell surface. This leads to infdtration of the synovium, inflammation of synovial tissue, proliferation and thickening of synovial tissue and finally cartilage and bone destruction (Veale, DJ. et al. 1996). Activated CD4+ T-cells promote B-cell proliferation and production of autoantibodies including RF. The presentation of antibody/antigen complexes to the T-cells by antigen presenting cells in the joints acts as a positive feedback mechanism to further activate the T-cells (Brennan, FM. et al.1998, Firestein, GS. 1992 and Cush, J. et al. 1991). There is a cascade of pro-inflammatory and anti-inflammatory cytokines in RA, in which TNFa stimulates production of IL-1, anti-inflammatory cytokines (IL-10, IL-1RN, soluble TNF receptor (sTNFR) and pro-inflammatory cytokines (IL-6, IL-8, GM-CSF and others). After stimulation by TNFa, IL-1 exerts similar effects on the stimulation of anti-inflammatory and/or pro-inflammatory cytokines (Brennan, FM. et al.1998, Butler, D. et al. 1995 and Brennan, FM. et al.1989). Further, TNFa and IL-1 play a major role in stimulating the cells of the pannus (articular cartilage) to produce large amounts of degradative enzymes including collagenase and other proteases, which are involved in breaking down collagen and causing further injury to the joints and vessels (Bresnihan, B. 1998, Muller-Ladner, U. et al. 1998 and Zvaifler, NJ. et al.1994). These injuries lead to the exposure of T-cells to modified autoantigens like collagen, immunoglobulin or heat shock proteins that are homologous to foreign antigens, causing proliferation of uncommitted CD4+ T-cells and cytokine release. The cytokine profile and the histological features of the joint injury in RA supports the understanding that the inflammatory reactions are mostly Thl driven (Dudler, J. et al. 1998). T-cell cloning studies from the joints in 16 R A patients suggest the presence of T h l like cells producing IL-2 and IFNy in the joints (Quayle, A J . et al 1993 and Miltenburg, A J . et al. 1992). The importance of IL-2 in T-cell proliferation and growth makes this cytokine an important one in cell-mediated, and especially T-cell driven, autoimmune diseases. The presence of IL-2, especially in the early stages of an autoimmune response, is crucial. However, the presence of IL-2 might not be necessary for the progression o f the disease in the later stages of R A . On the contrary, production of IL-4 and IL-10 by T h l and Th2 subset o f T-lymphocytes, is uncommon (Del Prete, G . et al. 1993 and Fiorentino, D F et al 1989). The ongoing tissue and joint inflammation at the final stage of severe R A resembles a delayed type hypersensitivity reaction, which seems to be independent o f any foreign antigen (Feldmann, M . et al 1996). Therefore, in the progressed form of R A , outcomes are similar among individuals with severe disease. 17 C H A P T E R T H R E E T H E G E N E T I C S OF R H E U M A T O I D A R T H R I T I S 3.1 Introduction Up to one third of patients with RA report a positive family history of RA in first-degree relatives; suggesting that this disease is influenced by genetic factors (Reveille, JD. 1998, Rigby, AS. et al. 1992 and Nunez, G. et al. 1984). The frequency of RA is higher in monozygotic twins (15%) compared to dizygotic and non-twin relatives (2-3%) (Silman, AJ. et al. 1993, 1992). The prevalence of RA in monozygotic and dizygotic twin studies indicates that the onset and development of RA is not a random event but modulated by the genetic susceptibility of the host (Lawrence, JS. et al. 1970 and Bellamy, N. et al. 1992). Despite this, no single gene has been found to completely explain the susceptibility or severity of RA (Weyand, CM. et al. 1995), indicating that RA is a complex disease, in which multiple genes are responsible for onset and severity. Vertical transmission of RA in family and twin studies (2-15%) is less than 100% and the prevalence of RA is not the same throughout the world. Thus, in addition to genetic predisposition, other factors like hormones and the environment, which may include viral infections, blood transfusions, obesity and smoking, may also be involved in the onset and/or progression of RA in humans (Symmons, DP. et al. 1997). It is therefore believed that RA develops as result of the interaction between complex multigenic predisposing factors and environmental stimuli. The genetic basis of RA has been studied extensively. Investigators have engaged primarily in the study of HLA genes and, secondarily, in certain non-HLA genes like TCR, cytokine, transporter in 18 antigen processing (TAP) and invariant peptide (li) genes. All of these have been suggested to be associated with the onset and/or progression of RA (Singal, DP. et al. 1994, Stastny, P. et al. 1988, Janossy, G. et al. 1981, Danis, VA. et al. 1994, and Krischmann, DA. et al. 1995). A dominant mode of inheritance of RA was rejected by Deighton, CM. et al (1993), who have suggested that HLA haplotype sharing in multiple families is only important for severe RA and not for mild RA. Furthermore, studies of families with multiple affected members have shown no strict linkage of HLA haplotypes and disease onset, thus indicating that HLA genes may be involved in RA severity rather than initiation (Pritchard, MH. 1994). 3.2 Human Leukocyte Antigens The HLA class I gene region is located on the short arm of chromosome six (6p21.31) telomeric to the HLA class II region (Beck, S. et al. 1999). It spans over approximately 1900 kilobases and consists of three principal gene clusters A, B, and C with A component being the most telomeric. HLA class I peptides consists of an a chain with a l , a2, a3 domains, and are associated with p2 microglobulin on the cell surface. Studies of extended HLA haplotypes in RA have associated certain HLA class I antigens with RA (Mulcahy, B. et al. 1996). However, these associations are mostly due to linkage disequilibrium between the two HLA class I and II loci, which are separated by a distance of one centimorgan (Klitz, W. et al. 1995 and Puttick, AH. et al. 1990) and there is little data showing an independent association between HLA-class I antigens and either susceptibility to or progression of RA. The HLA class II region is also located on the short arm of chromosome six and spans a distance of approximately 800 kilobases. The HLA class II contains three major and independent gene clusters designated DP, DQ and DR. The arrangement of these genes is such that the DP cluster is the most centromeric, DR the most telomeric and DQ is located between the two. These genes 19 give rise to 3 different groups of cell surface peptides: designated D P , D Q and D R . H L A class II peptides consist o f two similar a and p chains. The alpha chain is a 32-34 k D protein and is larger than the beta chain, which is 29-32 k D in weight. Both polypeptide chains have N-l inked oligosaccharide groups. The extra-cellular region of H L A class II molecules consists of a peptide-binding region and an immunoglobulin-like region. The peptide binding region, which consists of a l and p i chains provides a cleft for the H L A class II molecule. The sides of this cleft are supported by two helices open at both ends which can accept peptide fragments of size 10-40 amino acids (AAs) . The floor of the cleft consists of 8 beta-pleated strands, four of which are derived from the alpha-1 and four from the beta-1 chain. The peptide binding regions of H L A class II are the polymorphic regions o f the molecule, while the immunoglobulin-like regions are non-polymorphic. H L A class II antigens are expressed primarily on bone marrow derived cells which serve as professional antigen presenting cells. Synthesis and expression of the H L A molecules are highly regulated at the gene transcription level by cell type specific factors, inflammatory factors and cytokines such as interferon gamma (IFNy) which can up-regulate the expression o f both H L A class I and II molecules. H L A class II molecules bind exogenous peptide Ags which are accumulated in the endosomes. Upon the fusion of endosomes and exocytic vesicles (carrying uncharged class II molecules) the low p H inside the endosome activates the enzymes which break the invariant chain and expose the binding site of H L A class II molecules. H L A class II molecules then load peptide Ags in phagocytic lysosomes and are subsequently transferred to the plasma membrane o f the A P C s where these Ags are presented to C D 4 + T-cells (T-helper cells). Because C D 4 + T-cells recognize peptide Ags in the context o f H L A class II molecules, the selection and expansion o f lymphocyte clones and hence the specificity o f the consequent immune response is critically dependent on the fine molecular structure of this bi-molecular complex. 20 Studies of the effect of HLA-class II in RA have mostly addressed the question of susceptibility, rather than progression and severity. In these studies there has generally been no distinction among RA patients on the basis of disease severity. The expression of HLA class II molecules on most cell types in joints in the later stages of RA is both a hallmark of activation and indicates the importance of these molecules in the onset and progression of the disease (Klareskog, L. et al. 1982 and Janossy, G. et al. 1981). Epidemiologic studies have suggested that HLA class II genes are non-randomly associated with the expression of RA (Perdriger, A. et al 1996 and Nepom, GT. et al. 1992, 1989). However, these findings are not consistent (Teller, K. et al. 1996), and differences exist between racial-ethnic groups in the specific HLA alleles involved. Overall, these studies suggest that certain subtypes of HLA-DRlpl*01(101,102) and HLA-DRpl*04 (401,404,405,408) are associated with RA in Caucasians and Orientals, respectively, while HLA-DRpi*03 (1402) and HLA-DRpl*10 (1001) are the predominant RA-associated HLA alleles in Arabs, Tlingit Indians, and Tamil and Hindi Indians, respectively (Mody, GM. et al. 1994 and Satter, MA. et al. 1990). In addition, there is a low frequency of HLA-DR4 in RA patients from northern Italy while Greek and Israeli Jewish RA patients commonly type as HLA-DR10 and HLA-DR1, respectively (Salvarani, C. et al. 1992 and Gao, X.etal. 1983). A more comprehensive study (Gregersen, PK. et al. 1987) described an association between a shared sequence motif in the third hypervariable region of the HLA-DRpi and the onset and progression of RA, irrespective of ethnic background. This motif, denoted QKRAA or QRRAA at the amino acid level (Q=glutamine, K=lysine, R=arginine, A=alanine) is found at the third hypervariable region of HLA-DRpi (Gregersen, PK. et al 1987). The difference between the 21 QKRAA and QRRAA motifs is due to the substitution of an arginine for a lysine at position 71 in the HLA-DRpl gene. The substitution of arginine and lysine at position 71 is considered conservative, particularly since they both have basic side chains with a positive charge. At the nucleic acid level this reflects the substitution of a Guanine for an Adenine nucleotide. Crystallographic studies have shown that a specific binding site known as the rheumatoid pocket on the a helix and the P-pleated sheet, which forms the floor of the HLA class II molecule, may influence antigen binding in RA patients. This \"rheumatoid pocket\" encompasses the amino acid positions 13, 70, 71, 74 and 78 of the P chain and position 9 on the a chain. Similarities between the RA-associated shared epitope (SE) motif, which stretches from amino acids 70-74 of the HLADRpl peptide, and the rheumatoid pocket also support the importance of this latter in the expression of RA (Stern, LJ. et al. 1994 and Weyand, CM. et al 1995). In normal individuals the pocket allows the binding of large aliphatic side chains in contrast to which the rheumatoid pockets incorporating RA-associated sequences disfavors positively charged AAs due to the electrostatic interaction of arginine in position 71 of the a helix. There is evidence that other distinct pockets may be important in different HLA-DR-associated diseases like polymyalgia rheumatica and giant cell arteritis (Weyand, CM. et al. 1994, 1992). The importance of the rheumatoid pocket is consistent with the finding that only certain allelic variants of HLA-DR4 are relevant for RA. The RA-linked alleles *0401, *0404, *0405 and *0408 can be distinguished from the non-associated variants *0402 and *0403 through substitution of different AAs in positions 70, 71 and 74 (Winchester, R. 1994). For example, the RA-linked alleles contain QK or QR at positions 70 and 71 while the non-associated RA-linked alleles contain DE (D=aspartic acid and E=glutamic acid) at the same positions, and carry different charges. However, the rheumatoid pocket alone cannot explain the severity and progression of RA. 22 Vehe, R. et al (1994) reported that certain HLA-DR alleles are directly associated with the presence or absence of RF. Furthermore, Weyand, CM. et al (1995) suggested that even though seropositive and seronegative variants of RA may carry the SE, the genotype of the two disease variants could be clearly distinguished. Patients who are rheumatoid factor seropositive have a high frequency of the QKRAA motif whereas seronegative patients are associated with the high frequency of QRRAA. It seems that the substitution of lysine in position 71 results in the activation of T-helper cells, which can activate rheumatoid factor-secreting B-cells (Weyand, CM. et al 1995). The QKRAA sequence stretch is found mainly in HLA-DRP 1*401, whereas the QRRAA variant is associated with HLA-DRpi*0101, *404, *405 and *408. Seropositive patients develop a more severe form of the disease than do seronegative patients. Patients with HLA-DR4+ who were seronegative developed erosive disease in the early stages and needed to be treated more aggressively, while HLA-DR4\" seronegative patients developed erosions later in the disease course and could be treated with nonsteroidal anti-inflammatory agents or hydroxychloroquine (Weyand, CM.etal. 1995). Most of these studies focused on the association of HLA markers with susceptibility to RA, however none of the associations were strong enough to predict the severity and outcome of RA. Studies on the association of HLA-DQP1 with RA have also reported an increase in frequency of HLA-DQpl*301/302 in patients with RA (Zanelli, E. et al. 1998 and Taneja, V. et al. 1996). However, the strong linkage disequilbrium between HLA-DRpi and DQpl makes it difficult to distinguish the independent effect of each allele and has produced conflicting results (Taneja, V. et al. 1996 and Singal, DP. et al. 1987). Thus, the frequencies of the alleles in both loci are often reported together as an extended HLA-DR-DQ haplotype (Zanelli, E. et al. 1998). 23 A possible contribution of H L A - D P p i to susceptibility of R A has been reported (Stephens, H A F . et al 1989 and Gao, X . et al 1991) although this association is not undisputed due to the presence o f a moderate linkage disequilibrium between H L A - D R and D P (Imanishi, T. et al 1991). The combination o f H L A - D P p l * 0 4 0 1 with the subtypes of H L A - D R p i that contain arginine at position D R p l - 7 1 has been reported with increased frequency in R A patients (Perdriger, A . et al 1996). 3.3 T-Cell Receptors Two different types of T C R s are found in humans. The most common of the two, which is found on the majority of T-cells, consists of a heterodimeric polypeptide (a and p) and is involved in the recognition of A g - H L A complexes. The second subset of T-cells carry T C R s composed of y and 8 polypeptide chains and their mode of antigen recognition is unknown. In addition, the T C R complex also contains five other polypeptides known as the CD3 complex (y, 5, 8, and r\) which are involved in signal transduction. Unlike H L A genes, T C R complex genes are unlinked and located on chromosomes 7, 11 and 14 in humans. Studies examining the influence of T C R genes on the synovial lesion in R A have generally been conducted on small study populations and the evidence is conflicting and inconclusive. Stastny, P. et al (1988) found an increased expression of a T C R V p 8 restriction fragment length polymorphism (RFLP) in patients with R A but this data could not be confirmed by other studies. A n increase in VP3+, V p l 4 + and V p l 7 + T-cells has been reported in studies of the T-cells repertoire of R A patients (Goronzy, JJ. et al. 1995), while other studies have failed to confirm these findings, but have shown an increase in VP2+, VP6+ and VP15+ T-cells (Jenkins, R N . et al. 1993, Maruyama, T. et al. 1993 and Cooper, S M . et al. 1994). A n increase in y8 T C R has been suggested to imply the presence o f microbial superantigens (Abbas, A . et al. 1994). Studies conducted by Chaouni et al. 1990, reported an increase in yS T C R bearing cells (5% in normal versus 15% in R A patients) in the joints of R A patients, but these results have not been duplicated. 24 3.4 Cytokine Genes In addition to H L A , T C R and immunoglobulin genes, which affect R A susceptibility, cytokine genes could contribute to the immune responsiveness of the host. The presence of pro-inflammatory and immune-regulatory cytokines in high concentrations in the joints of patients with R A supports the important role of the cytokines in R A . Certain proinflammatory cytokines like T N F , IL-1 , IL-2, IFNy, and their receptors play a critical role in the inflammatory injury of immune-mediated diseases. These cytokines are often found in high levels in the joints of R A patients and are important factors in the onset and maintenance of chronic inflammation of R A (Lipsky, P E . et al.1989). The level of production o f these cytokines may; however, vary between subjects according to their H L A and/or other gene profiles. Studies have reported an association between cytokine production and H L A class II genes (Petrovsky, N , et al. 1997, Brinkman, B M N . et al. 1994 and Jacob, C O . et al. 1990). Jacob, and his colleagues (1990) showed an association between inheritance of H L A - D R 3 or H L A - D R 4 and high production of T N F while H L A - D R 2 was associated with low production of T N F . Certain cytokine gene polymorphisms has been associated with a higher production of cytokines, and a higher frequency of these polymorphisms have been reported to be associated with some autoimmune diseases including I D D M , Crohn's disease, Felty's syndrome and R A (Awata, T. et al. 1994, Bioque, G . et al. 1995, Brinkman, B M N . et al. 1994, Cantagrel, A . et al. 1999 and Danis, V A . et al. 1994). 3.4.1 TNF Family of Genes T N F is a 17kD cytokine produced and secreted mainly by mononuclear phagocytes and T-cells in response to infections. The secreted or soluble form of T N F is a homotrimer of 51kD proteins which initiates its action by binding to T N F cell surface receptors (55kD and 75kD). T N F a is one of the most crucial cytokines in the pathogenesis of R A and acts to potentiate inflammatory injuries 2 5 by stimulating the production of other cytokines like IL-1, IL-6 and IL-8 (Feldmann, M. 1996, Maini, RN. 1996, and Brennan, FM. et al. 1989) and increasing expression of cell adhesion molecules leading to infdtration of leukocytes at sites of inflammation. Also, TNFa is one of the major cytokines that induces HLA class II expression on the surface of the cells in the joints of RA patients. A bi-allelic TNFa polymorphism, TNF1 and TNF2, has been reported at position -308 in the promoter region, of the TNFa gene (Wilson, AG. et al. 1993). This polymorphism has been associated with the progression of cerebral malaria, SLE, IDDM, IBD and Felty's syndrome (McGuire, W. et al 1994, Wilson, AG. et al 1994, Pociot, F. et al. 1993, Bouma, G. et al. 1995 and Brinkman, BMN. et al. 1994). The TNF2 allele is thought to be associated with the increased TNFa production, although these results are controversial (Wilson, AG. et al 1993 and Brinkman, BMN. et al. 1996). Furthermore, Danis, VA. et al (1995) have reported a strong association and increased frequency of an uncommon TNF2 allele in patients with RA and increased production of IL- la in vitro. Five more polymorphic regions have been described within the TNF gene. A total of thirty-three variable nucleotide tandem repeat (VNTR) polymorphisms have been reported at these five loci (a-e) (Udalova, IA. et al. 1993). Thirteen of these are at the a (al-al3) locus, seven at b (bl-b7), two at c (cl-c2), seven at d (dl-d7) and four at e (el-e4). Loci a and b are in the upstream region and loci c is in the first intron of the human TNFp (lymphotoxin) gene while, loci e and d are located downstream of the human TNFa gene. Some of these microsatellite polymorphisms have been associated with susceptibility to RA and other autoimmune diseases like IDDM (Mulcahy, B. et al. 1996). High frequencies of TNFal/b5, in association with HLA-DR3-B18, have been reported in IDDM patients (Monos, DS. et al. 1995). More recently TNFa6, b5, c l , d3 and e3 have been 2 6 associated with the susceptibility to R A (Mulcahy, B . et al. 1996). The 10.5-kb allele of N c o l R F L P o f T N F beta (TNFP) gene has also been associated with better prognosis of gastric and lung cancers and increased susceptibility to I D D M and non- IDDM (adult onset) and monokine responses (Shimura, T. et al 994, Pociot, F . et al. 1991,1993 and Vendrell, J. 1995). 3.4.2 IL-1 Family of Genes The IL-1 family o f cytokines consists o f three members: IL-1 alpha ( I L - l a ) , IL-1 beta (IL-1P) and IL-1 receptor antagonist ( IL-1RN). Interleukin-1 a and P are pro-inflammatory cytokine produced mainly by activated mononuclear phagocytes in response to bacterial products, T N F , IL-1 and contact with C D 4 + T-cells (Abbas, A . et al. 1994). Both I L - l a and I L - i p are approximately 17kD proteins with identical physiological activity which bind to the same cell surface receptor, yet, they have different isoelectric points and show less than 30% structural homology. IL-1 has similar physiological importance to T N F in R A (Feldmann, M . 1994). IL-1 is directly involved as a mediator of local inflammation in the joints and, by its actions on mononuclear phagocytes and vascular endothelium, increases the expression of surface molecules leading to leukocyte adhesion. Further, IL-1 potentiates inflammatory responses by increasing the production of chemokines and expression of intercellular adhesion molecules ( I C A M - 1 and E -selectin) leading to recruitment and infiltration of other leukocytes at the site. IL-1 also indirectly stimulates the production of other pro- and anti-inflammatory cytokines like IL-1 , IL-6, IL-8, G M C -SF, M C P - 1 and R A N T E S (Feldmann, M . 1996 and Main i , R N . 1995). IL-1 has been reported as a potential mediator in the pathogenesis of R A by stimulating rheumatoid synovial cell collagenase and cartilage matrix degradation (Maini , R N . 1995) and thus is of particular interest in this study. I L - 1 R N is a biologically inactive protein produced by human mononuclear phagocytes, which acts as a naturally occurring inhibitor that controls the pro-inflammatory effect o f I L - 1 . I L - 1 R N is 27 structurally homologous to IL-1 , and thus is capable of binding to IL-1 receptors and acts as a natural competitive inhibitor for IL-1 . Recombinant I L - 1 R N has been tested for its therapeutic potential in the treatment of R A with benefit (Drevlow, B E . et al. 1996 and Dinarello, C A . 1991). Considering the crucial function of IL-1 in the inflammatory responses, interleukin-1 gene polymorphims may affect the production of IL-1 and thus play an important role in the progression of R A , both through a direct and/or indirect effect of IL-1 . The genes for IL-1 a , IL-i p and I L - R N are separate, and are all on chromosome 2 (Steinkasserer, A . et al. 1992). The polymorphisms o f the IL-1 family of genes have been previously documented and outlined below (Bai l ly , S. et al. 1996, Pociot, F . et al. 1992 and Tarlow, J K . et al. 1993). Seven polymorphic tandem repeats (A1-A7) have been reported within intron 6 o f the IL-1 a gene with allele sizes ranging from 620bp to 1220bp (Bailly, et al. 1993, 1996). N o relationship was found between these polymorphisms and the onset and/or progression o f R A (Bai l ly , S. et al. 1995). Certain alleles of these polymorphisms have been associated with higher production o f IL-1 and disease progression. For example, an inverse relationship between numbers o f 46bp tandem repeats and secretion o f I L - l a , was reported by Bai l ly , S. et al. in 1996. However, these differences did not reach statistical significance. Four R F L P s have been described for the IL-1 P gene by D i Giovine et al (1992). Fnu 4HI IL-1P R F L P and TaqI IL-1 P R F L P polymorphisms have been reported within exon 5 o f the IL-1 P gene (Guasch, JF. 1996 and D i Giovine, FS . 1992). Al le le 2 o f TaqI R F L P in human I L - l p has been associated with a higher production of physiologically active I L - l p and is increased in frequency in patients with I D D M (Pociot, F. et al. 1992). The potential role of the allele 2 of TaqI R F L P of 2 8 IL-1 P gene polymorphism in the pathogenesis of other autoimmune diseases like IBD, Crohn's disease (CD), and ulcerative colitis (UC) (Bioque, G. et al. 1995) has been studied. Five 86-bp tandem repeats (alleles A1-A5) have been reported in intron 2 of the IL-1RN gene by Lennard, A. et al. (1992) and Tarlow, JK. et al. (1993). Combined frequencies of A1-A3 were reported to be 98.6% with A4 and A5 frequencies of 1.4% (0.007 each). Each repeat has three potential protein binding sites which could be functionally important. These sites are an alpha-interferon silencer A, a beta-interferon silencer B and an acute phase response element. The effect of this polymorphism has been studied in association with Graves' disease, systemic lupus erythematosus and nephropathy in human diabetes mellitus (Blakemore, AI. et al. 1996, 1995, and 1994). A high frequency of IL-1 RN allele 2 has also been reported to be associated with UC, CD, IDDM and lichen sclerosis (Bioque, G. et al. 1995 and Blakemore, A. et al. 1995). 3.4.3 IL-2 Family of Genes The IL-2 family of cytokines consists of four members: Interleukin-2 (IL-2), IL-2 receptor alpha (IL-2Ra), IL-2 receptor beta (IL-2RP) and IL-2 receptor gamma (IL-2Ry). IL-2 is a 14 to 17 kD glycoprotein secreted primarily by CD4+ T-cells in response to antigen activation, and induces its action through IL-2 receptor (Abbas, A. et al. 1994). The IL-2 receptor consists of three chains, a 55kD polypeptide or a chain, a 70 to 75kD polypeptide or P chain, and a 64kD polypeptide or y chain. The p and y chains are expressed coordinately and form a complex known as IL-2RPy. IL-2a has no detectable biological activity and acts as a third anchor point which together with IL-2Rpy complex allows the binding of IL-2 to its tri-peptide receptor. The IL-2RPy complex is involved in signal transduction and has much higher affinity than IL-2Ra for IL-2. Interleukin-2 is an essential cytokine for growth of T-lymphocytes, and for facilitating progression from the G l to S phase of the cell cycle. In addition to its function as the major autocrine growth factor, IL-2 29 enhances the cytolytic function of the natural killer cells and promotes human B-cells for antibody synthesis. Data on IL-2 cytokine concentration at the initiation and during the progression of autoimmune diseases are not consistent, although it is clear that suppression of I L -2 production via Cyclosporin A is beneficial in autoimmune diseases like R A . Interleukin-2 is a single gene on chromosome 4 in humans, which contains four exons, three introns and a promoter region homologous to that of the interferon gamma gene. A sequence homologous to the viral enhancer element has also been reported in the second intron. Fifteen dinucleotide repeat polymorphisms known as alleles A 1 - A 1 5 have been reported in the 3' flanking region o f the IL-2 gene (Epplen, C. et al. 1994). The amplified D N A size of these polymorphisms ranges from 115bp to 147bp. Genetic polymorphisms that may play a role in the expression of this cytokine could have an important impact in the pathogenesis of autoimmune diseases including R A , since reports have indicated that concentration of IL-2 is increased at the site o f synovial tissue injury in the joints of patients with R A . The promoter region o f IL -2RP gene contains no T A T A box elements but instead has a G T (n) (guanine-thymine) dinucleotide repeat whose polymorphisms may play an important regulatory role in the expression of I L - 2 R P (Gnarra, JR. et al. 1990). Eight G T dinucleotide repeat polymorphisms have been reported 215bp upstream from the first transcription initiation site within the promoter region of IL -2Rp gene (Brewster, E S . et al.1991 and Gnarra, JR. et al.1990). I L - 2 R P microsatellite gene polymorphisms have been associated with the progression of some putative autoimmune diseases like schizophrenia (Ganguli, JR. et al. 1989,1994 and Weeks, D E . et al 1988). Considering the important role of IL-2 in the inflammatory processes o f autoimmune diseases, these polymorphisms could have an important implication in progression of an inflammatory autoimmune disease like R A . 30 3.4.4 IFNy Family of Genes IFNy is a potent proinflammatory homodimeric glycoprotein of approximately 21 to 24kD subunits which is an important component in the progression of autoimmune diseases. It is produced by both ThO and T h l CD4+ helper T-cells and all CD8+ T-cells. IFNy transcription is initiated directly by antigen activation and is induced by IL-2 and IL-12. IFNy acts as a potent activator o f mononuclear phagocytes, increases expression of H L A class I and II molecules by almost all the cells of the joints, promotes T- and B-cel l differentiation, increases the cytolytic activity o f natural killer cells and activates vascular endothelial cells leading to lymphocyte extravasation. IFNy has been reported as one of the important cytokines involved in the pathogenesis and progression o f R A (Hopkins, SJ. et al. 1988). Pathological studies, however, have reported inconclusive and contradictory data regarding the amounts of IFNy in the synovial fluid and synovium of patients with R A (Feldmann, M . et al. 1996). IFNy is coded for by a single gene which contains four exons with three intervening regions and is mapped to chromosome 12 in humans. Six dinucleotide ( C A ) repeat polymorphisms known as alleles A 1 - A 6 have been reported in the first intron of the human IFNy (A.Ruiz-Linares, 1993). These microsatellite polymorphisms have been associated with progression o f I D D M and Graves' disease. A high frequency of IFNy alleles A 3 and A 6 has been associated with the early and/or abrupt onset of I D D M (Awata, T. et al. 1994) while a high frequency o f allele 5 and a low frequency of allele 2 has also been reported in patients with Graves' disease (Siegmund, T. et al. 1998). Thus, it has been suggested that the microsatellite polymorphisms of IFNy may have predictive value for I D D M and Graves' disease (Awata, T. et al. 1994 and Siegmund, T. et al. 1998). The physiological importance of these polymorphisms is not known. However, it is thought that they may affect regulation of IFNy gene expression especially since the presence of a DNasel -hypersensitive site and an enhancer-like element has been reported in the first intron of IFNy 31 (Young, HA. et al 1990). Recent in vitro studies have indicated that the microsatellite polymorphisms in the first intron of IFNy gene in humans have an important effect on the production and expression of physiologically active IFNy cytokine (Pravica, V. et al. 1999). The polymorphisms of IFNy have not been studied in RA patients. 3.5 Other Genes Transporter in antigen processing (TAP) proteins have an important role in the assembly of antigen(s) which are loaded on to HLA class I molecules. TAP genes are located on the short arm of chromosome 6 within the class II region of the HLA complex. Two PCR-SSOP (Sequence Specific Oligonucleotide Probe) polymorphismic sites have been described for TAP I and three for TAP II genes (Kelly, A. et al. 1990, Colonna, M. et al. 1992 and Vandevyver, C. et al. 1995). Studies have shown a variable association between the TAP I and II gene polymorphisms and the onset and development of RA. However, these results are not consistent. Singal, DP. et al (1994) reported a significant increase in the frequency of TAP IP0101-1693A allele in RA patients. However, Marsal, S. et al (1994) reported no difference in the frequency of this and other TAP II polymorphisms in RA patients with HLA-DR4+ by comparison with HLA-DR4+ controls. Other studies on TAP I and Tap II gene polymorphisms have also reported conflicting results demonstrating the lack of an independent influence of TAP I and TAP II loci on susceptibility to RA (Vandevyver, C. et al. 1995). Invariant peptide (li) has also been considered to play an important role in the onset and/or progression of RA (Krischmann, DA. et al. 1995 and Bodmer, HS. et al. 1995). Kirschmann, DA. et al (1995) reported that the binding and origin of naturally processed peptides, bound to RA-associated HLA-DR alleles (HLA-DRP1*0401, 0404) was different than those in non-RA-associated HLA-DR (HLA-DRpl*0402) alleles. Using capillary liquid chromatography, they also 32 showed that the amount of li that bound to RA-associated HLA-DRs was significantly lower than for non-RA associated HLA-DRs. They concluded that a low concentration of li allows the exposure of binding sites of HLA class II molecules and may contribute to the loading of self-Ags to HLA class II. These self-Ags are then presented to the T-cell by HLA class II of APCs leading to autoimmunity in RA patients. 33 C H A P T E R F O U R M E T H O D S 4.1 Rationale It was postulated that the demonstration of differential frequency in individual gene polymorphisms between the three study groups would: \u00E2\u0080\u00A2 permit the reliable identification of patients at high or low risk of progression to severe destructive disease, \u00E2\u0080\u00A2 facilitate individualised therapy in patients with severe RA to ultimately reduce deformity, debility and treatment costs, while simultaneously avoiding the use of toxic and expensive agents in individuals at low risk of disease progression, and \u00E2\u0080\u00A2 identify novel targets for potential intervention in RA leading to new treatments for this disease. 4.2 Subjects and Selection Criteria Three groups of subjects consented to take part in the study: Group A consisting of patients with severe RA, Group B consisting of patients with mild RA and Group C, consisting of normal controls. For subjects with severe disease, all patients with adult-onset RA attending the Cyclosporin clinic at the Mary Pack Arthritis Centre were invited to participate. This clinic supervises the management of all RA patients receiving Cyclosporin in the Province of British Columbia. To receive this therapy, patients must have failed treatment with at least 3 conventional DMARDs or have severe RA associated eye disease. For subjects with mild disease, all patients with adult-onset RA attending the antimalarial treatment follow-up eye program at the Vancouver General Hospital Eye Care Centre were invited to participate. Subjects were included in the study if they had received antimalarials for at least 3 years, without any current or prior use 34 of other DMARDs and if their disease was well controlled (< 4 swollen joints and no severe deformities). A control group of Caucasian subjects randomly selected from the donor list of the British Columbia provincial bone marrow transplant registry served as healthy controls. Al l patients had to be Caucasian and were evaluated by a rheumatologist (Dr. D. Lacille and/or Dr. A. Chalmers) for clinical assessment of their RA and to verify eligibility for the study. Data collected included age, gender, duration of RA, previous and current therapy, American College of Rheumatology classification of functional status (Hochberg, MC. et al 1992), presence or history of extra-articular features, active joint count, swollen joint count and the presence of RA joint deformities. Medical records were reviewed to obtain the results of RF analyses, for which a titre of greater than 1:80 was considered positive. ESR was measured in all patients with mild RA. 4.3 Candidate Genes Genetic polymorphisms of five groups of genes (HLA, IL-2, TNF, IL-1, and IFNy) were examined in patients with severe and mild RA and were compared to healthy controls. These genetic markers were selected because of their role in the pathogenesis of RA. HLA class II loci genes are involved in the presentation of the foreign and/or self-antigens to the T-lymphocytes, and their polymorphism could affect the Ags that are presented and the consequent clonal expansion of the T-cell repertoire. IFNy was selected because of its physiological effects in increasing the expression of HLA molecules, activating mononuclear phagocytes and promoting differentiation of B- and T-lymphocytes (ThO, Thl and Th2). The IL-2 family of genes (IL-2 and IL-2RP) was selected since these genes are critical to the initial immune responses and are responsible for the growth and clonal expansion of T-lymphocytes. The TNF and the IL-1 family of genes (IL-la, IL-1P, and IL-1RN) have similar immunological effects and were selected for their importance and key role in the 35 inflammatory processes of RA. The effect of IL-1 and TNF on endothelial and surrounding cells of the joints includes increased expression of adhesion molecules (E-selectin and ICAM-1) and a higher production of other proinflammatory cytokines and chemokines such as IL-6, IL-8, GM-CSF, MCP-1 and RANTES. 4.4 Materials and Solutions RBC Lysis Buffer 0.144M N H 4 C I 0.001MNaHCO3 - added dH 20 to 2.0L - stored at 4\u00C2\u00B0C Nuclei Lysis Buffer 10.OmM Tris base 400.0 mMNaCl 2.0mMNa2 EDTA added dH 20 to 500ml stored at 4\u00C2\u00B0C 15.2g 0.17g 0.61g H.7g 0.37g or 2.0 ml of 0.5M Na 2EDTA stock Protease K Stock Buffer 2.0mM Na2EDTA - Stored at 4\u00C2\u00B0C or stored frozen in 50.0 ml aliquots l x T E 4.7 mM Tris HC1 l.OmM Na2EDTA - added dH 20 to 400 ml - adjusted pH to 8.0 with 10N NaOH - added dH 20 to 500ml - autoclaved or filter sterilized - stored at room temperature 2.0ml (0.5M stock) 1.0% SDS (10% SDS stock) 50.0 ml dH20 448.0ml 0.37g 1.0ml (0.5M stock) 36 Saturated Ammonium Acetate 19.2M NrL acetate* 740.0g - added dH20 to 500 ml -stored at room temperature -* saturated solution - added additional dH 20 with a dropper until it went into solution without heating 10 x Loading Buffer 0.25% Bromophenol Blue 0.25g 25.0% Ficoll type 400 25.0g 0.1% SDS (10% SDS stock) 1.0ml -QS to 100ml with dH 20 - adjusted pH to 7.5 - stored at room temperature - Bromophenol Blue runs at approximately 400 base pairs 5x Loading Buffer Bromophenol Blue 50mM Tris Base 50mM Na2EDTA SDS Sucrose - dissolved Tris and EDTA in 50ml dH 20 - adjusted pH to 7.6 - dissolved SDS, Bromophenol Blue, sucrose in Tris-EDTA - adjusted final volume to 100ml with dH 20 - stored at room temperature 1% Ethidium Bromide Ethidium Bromide lO.Omg -added dH 20 to 1.0ml - stored at 4\u00C2\u00B0C 0.10g 0.60g l-68g 0.50g 40.0g 0.5M Na2EDTA Na2EDTA 186.1g dH 20 800.0ml - mixed until dissolved (sped up by mixing at 50\u00C2\u00B0C for one hour and added 40ml 10N NaOH) - adjusted pH to 8.0 with NaOH - added dH 20 to adjust for total volume of 1.0 L and autoclaved or filter sterilized - aliquoted and stored frozen in 40ml quantities 37 10% SDS Stock SDS 50.0/100.0g - added d H 2 0 to 5 0 0 m l / l L - heated to 65\u00C2\u00B0C to dissolve - stored at 4\u00C2\u00B0C lOx TBE (Gel Buffer) -added d H 2 0 to 2.0L - stored at 4\u00C2\u00B0C 5x TBE Tris base Boric acid Stock N a 2 E D T A ( 0 . 5 M ) - added d H 2 0 to 1.0/2.0L - adjusted p H to 8.3 - stored at 4\u00C2\u00B0C 0.1M NaOH N a O H 4.0g - added 1.0L d H 2 0 and dissolved - stored at room temperature 0.1M HC1 concentrated HC1 8.3ml - added 800 d H 2 0 , mixed and QS to 1.0L with d H 2 0 1.5% Agarose Gel Agarose 0 . 5 x T B E - boiled, stirring constantly - cooled to 50\u00C2\u00B0C and added 2.0pl ethidium bromide - poured into a casting form, inserted slot former (1mm) and cooled for 30 minutes - poured in 0.5X T B E to cover gel by 2mm, applied samples and ran 0 .89M Tris Base 215.6g 0 .89M Boric acid HO.Og 0.02 N a 2 E D T A 80.0ml (from 0 .5M stock) 54.0/108.0g 27.0/54.0g 20.0/40.0 ml 1.2gr 80ml 12.5% Polyacrylamide Gel 30% Acrylamide/bis (29:1) 4.0ml 5x T B E 2.0ml d H 2 0 3.9ml 38 - mixed together and degassed for at least five minutes - added : 10% ammonium persulfate 70pl T E M E D 7pl - mixed by swirling gently - loaded the gel immediately - let set for at least one hour - assembled the gel, added l x T B E to the chambers, applied samples and ran 10% Ammonium Persulfate ammonium persulfate 1 OOmg d H 2 0 1.0ml - made fresh weekly - stored at 4\u00C2\u00B0C 30% Acrylamide/bis acrylamide/bis(29:l) 3Og bottle d H 2 0 73ml - protected the mix from light (covered with aluminum foil) - stored at 4\u00C2\u00B0C Sequencing Gel Ultra pure Urea 18.00gr (dissolved in 17ml of d H 2 0 ) Long Ranger (gel solution) 5ml 5x T B E 10ml d H 2 0 Added to adjust total volume to 50ml - mixed and filtered with 0.45Micron filter, and degassed under vacuum - added 250jul of 10% ammonium sulfate - added 3 5pi o f T E M E D , mixed gently and loaded Deionized Formamide - laboratory standard formamide was stirred with mixed bead resin, filtered, aliquoted in small quantities, and kept at -20\u00C2\u00B0C Standard Size Markers - O X 174 D N A / H a e III marker (Promega) was used for polymerase chain reaction ( P C R ) - R F L P - lOObp D N A ladder (Pharmacia) was used for I L - 1 R N size marker - Gene Scan 500 R O X or T A M R A size markers (API) were used for identification of dinucleotide tandem repeats ( D N T R ) 39 Polymerase Chain Reaction Reagents 5uM primers - primers were received in a freeze dried form and were then diluted to a 5uM working solution and kept at -20\u00C2\u00B0C lOx dNTP's - supplied in a kit from Pharmacia as lOOmM of dATP, dCTP, dGTP and dTTP. - diluted lOpl of each nucleotide with 960ul of dH20 (total volume=1000u.l) and kept stock vials and working mix at -20\u00C2\u00B0C Taq D N A polymerase - supplied from Roche and stored at -20\u00C2\u00B0C lOx P C R buffer and lOx M g - supplied from Promega and kept at -20\u00C2\u00B0C. lOx Mg was supplied as 25 mM MgC12. A 1000u.l 15mM working solution was made by diluting 600ul of supplied Mg C12 and 400u.l of dH20 and kept at 4\u00C2\u00B0C P C R mixes - prepared prior to PCR set up and in volumes that were sufficient for 33 reactions. These mixes were prepared as follows and kept at -20\u00C2\u00B0C: * specific primer set for each reaction (please refer to the primer sets) 4.5 Deoxyribosnucleic Acid Extraction and Quantification Methods An ACD tube of whole blood was collected from each individual for DNA extraction. Tubes were spun for 20 minutes at 2000 rpm and plasma was discarded using sterile pipettes. The buffy coat (all nucleated cells) and approximately 1ml of underlying red cells were transferred to a labeled 15ml conical polypropylene centrifuge tube (total of 2ml approximately). Cold red blood cell (RBC) lysis buffer was added to the 9ml mark on the tube and mixed well by inverting the tube. The tubes were then left at room temperature for 15 minutes, and were inverted several times during this incubation to ensure even lysis of RBC. The tubes were spun again for 20 -lOx PCR buffer lOOul ioom lOOpl ioom 470ul -1 Ox dNTP - 5' Primer* - 3' Primer* -dH2Q 40 minutes at 2000 rpm; supernatants were discarded. Extra cold RBC lysis buffer was used to wash leftover RBC and cell ghosts without disturbing the pellet. 3ml of cold nuclei lysis buffer was added to the pellet in the tube and mixed well to totally resuspended the cells. 200pl of 10% SDS (a detergent which lyses nuclear and white blood cell (WBC) membranes by degrading their lipid content) and 600ul of Protease K solution (mixture of 2mg Protease K per 1ml of Protease K buffer that digests endonucleases and histoproteins) was added to each sample. Samples were mixed by inverting the tube gently, and incubated overnight at 37\u00C2\u00B0C or for two hours at 65\u00C2\u00B0C. 1ml of saturated ammonium acetate was added to each tube, which was then shaken vigorously for at least 1 minute and left at room temperature for 15 minutes. Tubes were spun for 15 minutes at 2500 rpm and the supernatant was poured into another sterile 15ml polypropylene tube. Exactly two volumes of absolute ethanol (99%) at room temperature were added and mixed gently by inverting the tube until the DNA precipitated. In cases where the recovered DNA concentration was not high enough to precipitate and be visible to the naked eye when inverted, the tube was centrifuged at 15000 rpm for 5 minutes, decanted, blotted and air dried slightly before adding lx TE. Precipitated DNA was gently spooled from the tube with a 1 pi inoculating loop, and air-dried briefly for 1 minute. The DNA was then transferred to a sterile screw cap Sarstedt tube (2ml) containing 500ul lx TE buffer, incubated at 65\u00C2\u00B0C for at least one hour, cooled to room temperature and then stored at 4\u00C2\u00B0C. All DNA extraction was done in the laminar flow hood using sterile techniques, and the pipette tips used for extraction were plugged to avoid cross contamination. 4.5.1 DNA Quantification: Using Gene Quant DNA samples were incubated at 65\u00C2\u00B0C for 15 minutes, vortexed briefly and 30pl of each sample was transferred to a separate EZ tube. lOul of the 30ul sample was transferred to another EZ tube with 990ul deionized water and vortexed thoroughly (total volume=lml). The spectrophotometre 41 reference reading was set at 260nm with deionized water and the absorbance for each sample was measured at A260nm and A280nm. The A260/A280 ratio was measured and had to be grater than 1.6. D N A concentration for each sample was then calculated [A260 (sample) x 5= u.g/p.1]. The 20pl D N A sample in the original E Z tube was diluted to a concentration o f 100ng/u.l for P C R setup. 4.6 Polymerase Chain Reaction 4.6.1 Primer Designs The following primer sets were used throughout the experimental procedures in this study. The fluorescently tagged primers were made by Applied Biosystems (ABI) and all others were made by University o f Calgary, University Core Laboratory. 4.6.1.1 HLA Class II Primers - H L A - D P B 1 Primers: 5' Primer (PB1): 5' Primer (PB2): 3 ' Primer (PB3): 3 ' Primer (PB4): (Fukumori, Y . e ta l . 1992) - H L A - D Q B 1 primers: 5' Generic Primer (DQL3) 3 ' Group Specific Primers: (DQR1): (DQR2): (Mitsunaga, S. et al 1995) 42 5 ' - C A G A G A A T T A C G T G T A C C A G 5 ' - C C C G C A G A G A A T T A C C T T T T 5 ' - C A G G G T C A C G G C C T C G T 5 ' - C T G C A G G G T C A T G G G C C 5 ' - A T C C C C G C A G A G G A T T T C G T G 5 ' - T C C C G C G G T A C G C C A C C T C 5 ' - G T C G T G C G G A G C T C C A A C T G - H L A - D R p l primers: 5' Group Specific Primers: (DR1): (DR2): (DR52A): (DR4): (DR7): (DR9): (DR10): 3' Generic Primer: (3' Gen): (Home, C. et al. 1993, 1993) 5'-CGT TTC T T G TGG C A G C T T A A G T T 5'-TTC C T G T G G C A G C C T A A G A G G 5'-CCA CGT T T C T T G G A G T A C T C T A C 5'- A C G . T T T CTT G G A G C A GGT T A A T C A 5'-CCA C G T T T C C T G T G G C A G G G 5'-GTG C G G T A T C T G C A C A G A G 5'-GTT T C T T G G A G G A G G T T A A G T T 5'-TCG C C G C T G C A C T G T G A A G C T 4.6.1.2 TNF Microsatellite Primer Sets Set 1 (TNFa): - 5' Primer (IR4f): -3' Primer (IR2r): Set 2(TNFb): -5'Primer (IRlf): - 3' Primer (IR5r): Set 3 (TNFc): - 5' Primer (IR6f): - 3' Primer (IR7r): Set 4 (TNFd): -5'Primer (IRllf): -3'Primer (IR12r): Set 5 (TNFe): - 5' Primer (IR13f): -3'Primer (IR14r): 5'-NED-CCT C T C T C C C C T G C A A C A C A C A 5'-GCC T C T A G A TTT C A T C C A G C C A C A 5'-GCA C T C C A G C C T A G G C C A C A G A 5'-TET-GTG TGT GTT G C A G G G G A G A G A G 5'- 6FAM-GGT T T C T C T G A C T G C A T C T T G T C C 5'-TCA T G G G G A G A A C C T G C A G A G A A 5'-AGA T C C T T C C C T G T G A G T T C T G C T 5'- 6FAM-CAT A G T G G G A C T C T G T C T C C A A A G 5'- TET-GTG C C T GGT T C T G G A G C C T C T C 5'-TGA G A C A G A G G A T A G G A G A G A C A G (Udalova,IA. et al. 1993) 43 4.6.1.3 TNF2 (-308) Primers * - 5' Primer (PI): 5'- AAG GAA ACA GAC CAC AGA CCT G - 3' Primer (P2): 5'- ACA CAC AAG CAT CAA GGA TAC C (Nedospasov, SA. etal. 1986) 4.6.1.4 Interleukin-1 Beta (IL-lp) Primers * - 5' Primer (IL-1 pf): 5'- CGT ATA TGC TCA GGT GTC CTC C - 3' Primer (IL-1 pr): 5'- ACA TGG AGA ATT AGC AAG CTG CC (Guasch JF. et al. 1996, Bioque G. et al. 1995, Pociot F. et al. 1992, and Clark BD. et al. 1986) 4.6.1.5 Interleukin-1 Receptor Antagonist (IL-1RN) Primers * - 5' Primer (IL-lRNf): 5'- TCA GCA ACA CTC CTA TTG ACC TGG - 3' Primer (IL-lRNr): 5'- GTC TCA TCT TCC TGG TCT GCA GG (Tarlow KJ. et al. 1993, and Lennard A. et al. 1992) 4.6.1.6 Interleukin-2 (IL-2) Primers * - 5' Primer (IL-2f): 5'- 6FAM-AAA GAG ACC TGC TAA CAC ACA CAC - 3' Primer (IL-2r): 5'- CCT ATG TTG GAG ATG TTT ATT GTT TC (Epplen C. et al. 1994, and Shows T. et al. 1984) 4.6.1.7 Interleukin-2 Receptor Beta (IL-2Rp) Primers * - 5' Primer (IL-2Rpf): 5'-7/ffiX-GAG AGG GAG GGC CTG CGT TC - 3' Primer (IL-2RPr): 5'- CAC CCA GGG CCA GAT A A A GAT CT (Brewster ES. et al 1991, and Gnarra JR. et al. 1990) 4.6.1.8 Interferon Gamma (IFNy) Primers * - 5' Primer (IFNyf): 5'- 6FAM-AGA CAT TCA CAA TTG ATT TTA TTC TTA C - 3' Primer (IFNyr): 5'- CCT TCC TGT AGG GTA TTA TTA TAC G (Ruiz-Linares 1993 and Gray PW. et al. 1982) * Denotes primers which were altered in length or independently designed (from the published sequences) in order to produce more sensitive, specific and robust primers with higher melting points than the published primers. 44 4.6.2 Polymerase Chain Reaction Mixes A general set of conditions and reagents were used for PCR amplification of all the samples. A Perkin Elmer Thermal Cycler was used for all PCR reactions. The total reaction volume for each sample was 30pl, for all amplifications, which consisted of: A master mix of 29uj/sample (all reagents except DNA) was made and aliquoted to each tube. One microliter of sample DNA (lOOpg/ul) was then added to each tube and overlaid with one drop of mineral oil. A positive (previously typed) and a negative (without DNA) control were included in each run. Samples and reagents were kept chilled at all times until transferred to a preheated 95\u00C2\u00B0C thermocycler block. After samples were transferred to the thermocycler, all samples were kept at 95\u00C2\u00B0C for a period of 5 minutes (hot start) before their appropriate cycling file was applied. Depending on the locus being amplified, each amplification had a specific cycling parameter which was stored in the memory of the thermocycler and was applied after the \"hot start\" period as follows. 4.6.3 Cycling File Parameters For all HLA class II, IL-1P, IL-2 and IL-2RP amplifications: -Five minutes \"hot start\" at 95\u00C2\u00B0C followed by: - PCR mix: -MgC12(15mM) -Taq DNA Polymerase -Genomic DNA(100u.g/ml) 26uJ 3 pi 0.15ixl lul 32 cycles of 94\u00C2\u00B0C 64\u00C2\u00B0C 45 seconds 60 seconds For all TNF microsatellite amplifications: -Five minutes \"hot start\" at 95\u00C2\u00B0C followed by: 31 cycles of 94\u00C2\u00B0C 68\u00C2\u00B0C 45 seconds 60 seconds 45 cycle of 94\u00C2\u00B0C 68\u00C2\u00B0C For all TNF2 (-308) amplification: -Five minutes \"hot start\" at 95\u00C2\u00B0C followed by: 33 cycles of 1 cycle of 94\u00C2\u00B0C 60\u00C2\u00B0C 72\u00C2\u00B0C 94\u00C2\u00B0C 60\u00C2\u00B0C 72\u00C2\u00B0C -For IFNy amplification: -Five minutes \"hot start\" at 95\u00C2\u00B0C followed by: 31 cycles of 94\u00C2\u00B0C 62\u00C2\u00B0C 1 cycle of 94\u00C2\u00B0C 62\u00C2\u00B0C 45 seconds 5 minutes 30 seconds 30 seconds 45 seconds 30 seconds 30 seconds 10 minutes 45 seconds 60 seconds 45 seconds 5 minutes -For IL-1RN amplification: -Five minutes \"hot start\" at 95\u00C2\u00B0C followed by: 31 cycles of 1 cycle of 94\u00C2\u00B0C 60\u00C2\u00B0C 72\u00C2\u00B0C 94\u00C2\u00B0C 60\u00C2\u00B0C 72\u00C2\u00B0C 1 minute 1 minute 3 minutes 1 minute 1 minute lOminutes Upon completion of the PCR cycles all samples were checked for positive amplification. Four microliters of each sample was mixed with 6pl of 5x loading buffer and applied to a 1.5% agrose gel. The gel was then electrophoresed for 30 minutes at 115volts, stained with ethidium bromide and photographed for positive amplification results under ultra-violet (UV) light. 46 4.7 Restriction Fragment Length Polymorphism Restriction enzyme digests were used for high resolution typing of HLA class II, TNF2 and IL-1 p. Restriction enzymes were used to cut the amplified DNA products at different sites resulting in different fragment sizes so that the digested DNA produced different fragment sizes upon electrophoresis. A total of 4pl of the each amplicon was digested with restriction enzymes in a total volume of lOui (dH20, enzyme buffer(s) and enzyme 1 unit/digest) and incubated in a water bath according to the temprature specifications for each enzyme. The following flow-charts were used for typing of the alleles in the different experiments. 4.7.1 H L A Class II Enzyme Digests 4.7.1.1 HLA-DPpi Enzyme Digests \" For typing of HLA-DPpi alleles, 4 different PCR amplifications were performed (A1-A4) each using a different set of primers. After amplification each amplicon was digested by restriction enzymes. The primer sets used in these amplifications were as follows: A1 =PB 1 and PB3 A2=PB 1 and PB4 A3=PB2 and PB3 A4=PB2 and PB4 Set A l primers amplified a group of alleles which were then digested according to Figure 4.1. 47 Figure 4.1: H L A - D P p i R F L P : A l primer set. 101,301,601,901,1001,1101,1301,1401 and 1701 301,601,1101,1401 BstuI 1101 301,601,1401 I Fok 1 Ava II 301,1401 101,901,1001,1301,1701 BstuI I 901,1001,1701 1 Fok 1 1 r 601 901,1001 Rsa 1 or Sea 1 901 1001 101,1301 I E c o N l \u00E2\u0080\u0094I I I 1701 1301 101 Set A2 primers amplified a group of alleles which were then digested according to Figure 4.2. Figure 4.2: H L A - D P p l R F L P : A2 primer set. 1501,1801 I E c o N l 1501 1801 Set A3 primers amplified a group of alleles which were then digested according to Figure 4.3. 48 Figure 4.3: H L A - D P p l R F L P : A3 primer set. DPB7 501.DPB7, 801,1601,1901 Bstu 501,801,1601,1901 Ddel 501,1901 I Fok 1 801,1601 I Fok 1 I I 501 1901 801 1601 Set A4 primers amplified a group of alleles which were then digested according to Figure 4.4. Figure 4.4: H L A - D P p l R F L P : A4 primer set. 201,202 201,202,401,402 EcoNl 401,402 Sac I 201 202 Bstu I 401 402 49 4.7.1.2 HLA-DQ(31 Enzyme Digests For typing of HLA-DQP1 alleles, 2 different PCR amplifications were performed (DQR1-DQR2). For DQR1 amplification, DQR1 and DQL3 primer set was used; whereas, DQR2 and DQL3 was used for DQR2 amplification. After amplification each amplicon was digested by restriction enzymes and typed according to Figures 4.5 and 4.6. Figure 4.5: H L A - D Q p l R F L P : DQR1 amplification. 501,502,503,504,601,602,603,604 Rsal 501,502,504,603,604 503 601 602 Cfo 501 502,504 603 604 S I N I 502 504 50 Figure 4.6: HLA-DQB1 R F L P : DQR2 amplification. 201,301,302,303,304,305,401,402 RSAI 201 301,304 BsaHI 302,303,305 BsaHI 401,402 Cfol 301 304 302,303 305 401 402 Cfol 302 305 51 4.7.1.3 HLA-DRpi Enzyme Digests To type for HLA-DR.pi alleles, 7 different PCR amplifications were performed, DR1, DR2, DR52A, DR4, DR7, DR9 and DR10. All amplifications were different in their 5' primers but the same consensus 3' primer was used for all 7 amplifications. Each amplicon was then digested by restriction enzymes for high resolution typing according to Figures 4.7 to 4.11. Figure 4.7: H L A - D R p i R F L P : DR1 amplification. 101,102,103 SIN I 101,102 103 HphI 101 102 Figure 4.8: H L A - D R p i R F L P : DR2 amplification. 1501,1502,1601,1602 Rsal 1501,1502 1601,1602 Hphl Asp 700 1501 1502 1601 1602 52 Figure 4.9: H L A - D R p l R F L P : DR52A amplification. 301,302,801,802,8031,8032,1101,1102,1103,1104,1201,1202, 1301,1302,1303,1304,1305,1401,1402,1403,1404,1405 Rsal 301 1101 1102 1103 1104 1304 1201 1202 Rsal 1301 1302 1305 Rsal Rsal 1201 1202 1401 1404 HphI 302 8031 1402 1403 1405 1401 1404 M n l l 801 802 8032 1303 Rsal 1101 1102 1103 1304 1104 1301 1305 1302 HphI HphI HphI I T 8031 302 1402 1403 1301 1302 1102 1304 801 802 8032 1303 | Hph I Ksp I 1405 8032 1303 801 802 1101 1103 1104 Bsrbl 1103 1104 302 1403 53 Figure 4.10: HLA-DRpi RFLP: DR4 amplification. 401,402,403,404,405,406,407,408,409,410, 411,412,413,414,415,416,1410 M n l l 401,409 413,416 HphI 402,412 414 HphI 401,409 413 402,412 414 416 403,406 407,411 1410 HphI 404,405 410 HphI 107 404^ 1410 403,406 4 ,410 405,408 411 415 Hae II Hae II Rsa I Hae II Hae II 401,416* 409 402 412 403,411 406 404 410 Hae II I I 403 411 405 408 unable to subtype further Figure 4.11: HLA-DRpl RFLP: DR7,9,10 amplifications. 701,901,1001,DRP5 Rsal 701 901 1001 DRP5 54 4.7.2 TNF2 Enzyme Digests Polymerase chain reaction-restriction fragment length polymorphism (PCR-RFLP) techniques were used to distinguish between the mutant alleles with an adenine at position -308 versus a guanine nucleotide in normal or wild type alleles. Samples were amplified with PI and P2 primer sets and digested with \"Sty F enzyme at 37\u00C2\u00B0C overnight. The digested amplicons were then run on 12.5% acrylamide gel to detect the cutting pattern of the enzyme. Samples were stained with ethidium bromide. Sty 1 restriction enzyme recognizes the amplified DNA sequence only if the guanine nucleotide is at position -308 and cuts the amplicon. Homozygous mutants were not cut so they were observed as a single band under UV light. Homozygous wild types were completely cut and were observed as two smaller bands under UV light. A standard DNA size marker, (j)X174 DNA/Hae III (Promega), was used for sizing of the fragments. Heterozygous individuals had a pattern that contained both cut and uncut alleles. Different allele sizes are presented in Table 4.1. Table 4.1 Sty I enzyme digest patterns. Allele Type DNA Fragment Size Homozygous mutant 143 bp Homozygous wild type 123 bp and 20bp Heterozygous 143bp, 123bp and20bp 4.7.3 IL-ip Enzyme Digests Samples were amplified with IL-1 pf and IL-1 Br primer sets and digested with \"Fnu 4HI and or TaqF enzymes. Each amplicon was first digested individually with both enzymes according to the enzyme protocol. Second, a double digest of both enzymes was repeated to determine if both 55 mutations were on the same chromosome and were transferred simultaneously. In the later case 4u.l of each amplicon was first digested with Fnu 4HI enzyme at 37\u00C2\u00B0C for one hour and then TaqI was added to the sample and the digest was continued at 65\u00C2\u00B0C for another hour. The digested amplicons were then stained with ethidium bromide and run on 12.5% acrylamide gel and were observed under UV light to detect the cutting pattern of the enzyme. A standard DNA size marker, <))X174 DNA/ Hae III (Promega), was used for sizing of the fragments. The allele sizes and pattern of the cuts are presented in Table 4.2. Table 4.2 Fnu 4HI and TaqI enzyme digest patterns. Cut Patterns DNA Fragment Size - 327bp ~ TaqI 21 Ibp and 116bp Fnu 4HI - 288bp and 39bp Fnu4HI TaqI 211bp,77bp and39bp * * Denotes both cut sites were on the same chromosome 4.8 Electrophoresis Electrophoresis was used to separate different lengths of DNA fragments. All samples were stained with ethidium bromide and a standard DNA size marker, X174 DNA/Hae III (Promega) or lOObp DNA lader (Pharmacia), was used for sizing of the fragments. For enzyme digested DNA, 12.5% polyacrylamide gel was used. For DNA repeats that differed by more than 50bp in length a 1.5%) agarose gel was used to distinguish different alleles. 56 4.8.1 Polyacrylamide Gel Electrophoresis Protean II electrophoresis apparatus with a 0.75mm spacer was used for electrophoresis of enzyme digested DNA. Ten milliliters (for two gels) of 12.5% polyacrylamide gel was prepared and used for electrophoresis. One time TBE buffer was used to run the polyacrylamide gels. Two microliters of lOx loading buffer was added to the enzyme digested amplicon, spun briefly and loaded into the slots. Gels were run at 115 volts until the blue dye had run out of the bottom of the gel (approximately 1.5 hours). Gels were then stained separately in 20ml of 0.5ug/ml ethidium bromide for 5 minutes on top of a gentle shaker; and photographed under UV light. 4.8.2 Agarose Gel Electrophoresis Agarose minigel electrophoresis (1.5%) was used to identify different alleles of the IL-1RN, 85bp repeat polymorphism. 4.9 Microsatellite Fragment Analysis An ABI 377 DNA sequencer, ABI Prism 377 Collection, Genescan (version 2.1) and Genotyper (version 2.1) software were used to identify microsatellite and dinucleotide repeat polymorphisms. 4.9.1 Procedures Approximately 0.5-1 pi of each amplified DNA sample was added to a premix that contained 2pl of deionized formamide, 0.5ul loading dye and 0.5pl Genescan 500 size-standard, TAMRA or ROX (PE Applied Biosystems). Samples were then heated at 95\u00C2\u00B0C for 2.5 minutes and transferred quickly to ice. Samples were loaded to the pre-warmed sequencing gel (50\u00C2\u00B0C) and run at 3000 volts, 60.0 mA for 2 hours. Filter set C or D was used to detect different fluorescent colors and the laser scan was set at 2400 scans per hour for optimal resolution. Data was gathered in gel files 57 by the 377 Collection software. A size-standard file was created manually by assigning appropriate sizes, according to the manufacturer, to the electropherogram peaks of each size-standard marker (see Appendix Figures 1 and 2). This size-standard file was then applied to all the samples in that particular run to assign the exact fragment size to each sample. Data was then transferred to Genotyper 2.1 for further analysis and allele assignment. Genotyper was set to filter out nonspecific and out-of-range alleles and then assign the appropriate allele type to each sample by using the defined ranges. Electropherograms generated by Genescan were then manually double-checked to make sure that Genotyper had not missed an allele. 4.10 Statistical Methods A two-sided t-test was used to compare continuous variables including age, disease duration, age of onset, active and swollen joint count and duration of therapy among individuals in both patient populations. Standard deviation and p values were calculated and a p value of less than 0.05 was considered statistically significant. The proportion of patients expressing individual alleles of all candidate genes was compared in both study groups and controls using the chi-square test with Yate's correction. Binomial analyses of the data were performed using Graphpad Prism statistical software. Statistical tests have been adjusted for the effect of multiple tests involving the IFNy locus by multiplying the observed significance values (i.e. p values) obtained by the number of tests performed (i.e. 12 tests, number of alleles present n=6, number of comparisons, q=2). All other analyses were considered as secondary and hypotheses-generating, and adjustment of the statistical significance level was therefore not made for these analyses. 58 Individuals were classified into different risk groups according to alleles most closely associated with disease status at each gene locus. HLA-DR.pi alleles which had an odds ratio difference of more than 2 fold between mild and severe groups were generally considered high risk alleles (OR>2) or low risk alleles (OR<0.5) assuming neither of the probabilities was zero. Similarly, for all other loci, alleles with an odds ratio greater than 2 between the severe and mild groups, were considered to confer high risk for progression of severe RA, while those with an odds ratio of less than 0.5 were considered to confer low risk. Individuals with high and low alleles were grouped as \"both\" and those with neither high nor low risk alleles were grouped as \"neither\". For analysis of combined frequencies of HLA class II loci (DR, DQ and DP), alleles were grouped according to their expression of high risk (triple positive high risk), low risk (triple positive low risk), or other alleles at all three loci. The risk of severe versus mild disease associated with the presence of each group was estimated by calculating the chi-square and p value. For analyses of combined frequencies of HLA-DRpi and IFNy, alleles were grouped according to the expression of high risk alleles for both genes (double positive high risk), low risk alleles for both genes (double positive low risk) or other alleles at the two loci. The risk of severe versus mild disease associated with the presence of each group was estimated by calculating the chi-square and p value. Multiple logistic regression was performed to calculate the OR of severe versus mild disease associated with the presence of each allele group after controlling for other prognostic markers of RA. Covariates evaluated were gender, age at onset, disease duration and rheumatoid factor positivity. A forward stepwise inclusion method using the likelihood ratio was employed. Only covariates found to be significant predictors of disease severity were included in the final model used to evaluate the variable of interest (high or low risk alleles). P values for the variable of interest were calculated using the Likelihood ratio test, which is superior for small sample sizes. 59 Statistical analysis was performed using S-Plus 2000 statistical software. The significance level for the primary analysis, in which logistic regression was used to evaluate the influence of presence or absence of each epitope on disease severity, was set at 0.05. P values were adjusted for multiple testing by multiplying them by the number of comparisons made (i.e. 2 comparisons, Severe vs Control and Severe vs Mild). For FILA-DRpi and IFNy loci, alleles were grouped according to their expression of the high risk, low risk, both or neither alleles for pooled analysis. The risk of severe versus mild disease associated with the presence of each allele or group of alleles was estimated by the calculation of chi-square and p value. 60 C H A P T E R F I V E R E S U L T S 5.1 Introduction Seventy patients were identified in the Cyclosporin Clinic, of whom 12 were non-Caucasian and were thus eliminated. Of the remaining 58 patients, only 48 met the study criteria and agreed to participate. These patients all had severe RA characterized by conventional ACR criteria, joint deformity and erosions, who required Cyclosporin-A therapy for disease control. Invitations to participate in the study were sent to 369 patients in the Hydroxychloroquine Clinic. Fifty patients responded of whom 39 met the selection criteria and agreed to participate. These patients all had mild non-erosive disease requiring no more than anti-malarial (Hydroxychloroquine) treatment. The study was therefore conducted using a final total of 87 Caucasian patients divided into two discrete groups having distinct grades of RA. Both groups had a minimum disease history of 3 years to ensure comparable risk exposure. A third group, a demographically matched population of 105 normal subjects, Group C, served as the control group for comparison. The demographic and clinical features of the RA patients are shown in Table 5.1. There was no significant difference in the mean age and gender of patients with severe or mild disease, although other factors of prognostic importance differed significantly between the two groups. The proportion of females was lower (67 vs 77%), the age at onset was lower (39+14 vs 49+13 years) and duration of disease was longer (19+12 vs 12\u00C2\u00B17 years) in subjects with severe disease than with mild disease. Of the 45 patients with severe RA for whom RF results were available, 39 (86.7%o) were RF positive, while in the 28 patients with mild RA for whom RF results were 61 available, 10 (35.7%) were RF positive. As expected, patients in the severe group had more active and more severe disease, as demonstrated by a higher number of active and swollen joint counts, greater frequency of deformities and extra-articular features, and poorer functional classification (Table 5.1). Patients with severe RA had been on Cyclosporin for an average of 26 months, while those with mild disease had been on anti-malarials for an average of 90 months. Table 5.1 Demographic and clinical characteristics of patients with severe and mild Rheumatoid Arthritis. Parameters Control Severe RA\" Mild R A a Age (years) 41 \u00C2\u00B12 58 \u00C2\u00B1 12 61 \u00C2\u00B1 13 Gender (females) 48 (46%) 32 (67%) 30 (77%) Age of onset (years) NA 39\u00C2\u00B1 14 49 \u00C2\u00B1 13 Disease duration (years) NA 19 \u00C2\u00B1 12 12 + 7 Rheumatoid Factor positive NA 39/45 (87%) 10/28 (36%) Active joint count NA 14 \u00C2\u00B1 12 5 \u00C2\u00B1 5 Swollen joint count NA 8 \u00C2\u00B1 8 1 \u00C2\u00B12 Joint deformities NA 48(100%) 10(26%) Extra-articular featuresb NA 36 (75%) 14 (36%) Functional class: I NA 0 18/39(46%) II NA 21/41 (51%) 19/39 (49%) III NA 16/41 (39%) 2/39 (5%) IV NA 4/41 (10%) 0 Duration of therapy (months) NA 26 \u00C2\u00B122 90 \u00C2\u00B147 Values represent mean +/- Standard Deviation or number of subjects (%). Where specified, denominator represents the number of subjects with available data. a All differences between the two groups are significant (p value < 0.05), except age and gender. b Presence of an extra-articular feature other than sicca syndrome. NA not applicable. 62 5.1.1 Influence of Demographic and Clinical Factors on the Odds of Severe versus Mild Disease. The four variables, age at onset, disease duration, RF, which differed significantly between the groups with severe or mild disease by univariable analysis, and gender were incorporated in a forward step-wise logistic regression analysis to determine the effect of these parameters on the outcome of RA assuming all other variables were held constant. An incremental predictive value for each of the clinical parameters in the logistic regression analysis is shown in Tables 5.2 and 5.3. Two of these clinical parameters proved to have a statistically significant (p<0.05) predictive value on the outcome of the disease when used in the logistic regression model (Table 5.2). This analysis indicated that gender and the presence of RF were the two important clinical factors that predicted disease severity with odds ratios of 3.68 (p=0.0487) and 10.19 (p<0.0001), respectively (Tables 5.2 and 5.3). Duration of the disease and age at onset had a minimal effect on the outcome of the disease with insignificant p values of 0.1450 and 0.0827 and odds ratios of 1.58 and 0.40, respectively (Tables 5.2 and 5.3). Table 5.2 Logistic regression analysis using combination of four demographic and clinical data to predict the outcome of R A . Factors Chi-Square d.f P value Age at onset 101 I 0.0827 (NS) Gender 3.88 1 0.0487 Duration 2.12 1 0.1450 (NS) RF 21.34 2 O.0001 63 Table 5.3 Detailed results of logistic regression analysis examining the influence of demographic and clinical parameters on the outcome of R A . Factors Low High Diff. Effect S.E. Lower 95% Upper 95% OR Age at onset 33 55 22 -0.92 0.55 -2.00 0.15 0.40 OR details 33 55 22 0.40 NA 0.14 1.16 Duration 9 18 9 0.46 0.32 -0.18 1.09 1.58 OR details 9 18 9 1.58 NA 0.84 2.98 Gender m:f 1 2 NA 1.30 0.69 -0.06 2.66 3.68 OR details 1 2 NA 3.68 NA 0.94 14.31 RFp:n 1 2 NA 2.32 0.64 1.07 3.58 10.19 OR details 1 2 NA 10.19 NA 2.91 35.71 RF unk:n 1 3 NA -0.38 0.95 -2.24 1.49 0.69 OR details 1 3 NA 0.69 NA 0.11 4.43 Patients with severe RA (n = 48), patients with mild RA (n = 39). Odds ratios (OR) and chi-square statistics are marginal (i.e. have been adjusted for all other factors). Odds ratios here reflect the distribution of patients observed rather than underlying prevalence of mild and severe RA. RF: Because results not available for all patients, subjects were divided into three categories consisting of those with a positive RF (p), negative RF (n) and those for whom this result was unknown (unk). Subjects with missing RF data were not significantly different from those with negative (n) results. Effect of each parameter= natural logarithm of OR m:f male versus female S.E. standard error 64 5.2 Class II Human Leukocyte Antigens 5.2.1 HLA-DRpl The proportion of patients expressing each HLA-DRpi allele in the three study groups is shown in Table 5.4. Eighty-eight percent of the patients (42/48) with severe RA had one or more of the disease-associated alleles HLA-DRpi*101, 102, 401, 404, 405, 408, or 1001 expressing the QKRAA/QRRAA epitope, compared with 54% (21/39) of those with mild disease (OR=6.00; p=0.0011) and 41% (43/105) of the controls (OR=10.93; pO.OOOl). This effect was principally due to the elevated prevalence of HLA-DRpi*401 and 404 in patients with severe disease. Fifty-two percent of the patients (25/48) with severe RA were positive for HLA-DRpl*0401, compared with 23% (9/39) with mild disease (OR=3.6; p=0.0112) and 16% (17/105) of controls (OR=5.6; pO.OOOl). HLA-DRpl*404 was present in 21% of patients (10/48) with severe RA compared to 13% (5/39) with mild RA (OR=1.8; p=NS) and 6% (6/105) of controls (OR=4.3; p=0.0107). There was no significant difference in the other alleles expressing the QKRAA/QRRAA epitope. HLA-DRpi*405 or 408 were indifferent between the three study groups, each being present in less than 3% of patients with severe RA, mild RA or controls. HLA-DRpi*101 and/or 102 were also present with comparable frequency in all groups: 23% (11/48) of patients with severe RA, 21% (8/39) with mild RA (OR=l.l; p=NS) and 18% of controls. In contrast, only 6% of patients (3/48) with severe RA had one of the alleles HLA-DRpl* 103/402/1102/1103/1301/1302 which contained the DERAA (D=Aspartic acid, E=glutamic acid, R=arginine, A=alanine) motif, compared with 31% (12/39) in the mild RA group (OR=0.15; p=0.0064) and 21% (22/105) of the controls (OR=0.25; p=0.0407). This difference was principally due to the different frequencies of HLA-DRpi* 1301/02, which were expressed by only 4% of 65 patients with severe RA compared with 21% of those with mild RA (OR=0.16; p=0.0414) and 19% of controls (OR=0.81; p=0.0288). The antigens HLA-DRpT* 0103/0402/1102/1103 which also exhibit the sequence motif DERAA at positions 70-74 of the third hypervariable region are infrequent in normal controls, and did not show any difference between the three study groups. Table 5.4 Frequencies of patients and controls expressing H L A - D R p i alleles. Controls Severe RA Mild RA DRpl % % ORa Pa % ORa Pa OR b Pb 101 15 21 1.5 NS 18 1.2 NS 1.2 NS 102 3 6 2.3 NS 3 0.90 NS 2.5 NS 103 1 0 0.72 NS 8 8.7 NS 0.11 NS 301 32 21 0.66 NS 21 0.54 NS 1.0 NS 401 16 52 5.6 O.0001 23 1.5 NS 3.6 0.0112 402 1 2 2.2 NS 3 2.7 NS 0.81 NS 403/06 3 0 0.30 NS 0 0.37 NS - -404 6 21 4.3 0.0107 13 2.4 NS 1.8 NS 405 3 2 0.72 NS 0 0.37 NS 2.5 NS 407 6 2 0.36 NS 0 0.19 NS 2.5 NS 408 2 0 0.43 NS 0 0.52 NS - -701 19 21 1.1 NS 21 1.1 NS 1.0 NS 801/02/03 8 2 0.30 NS 8 1.2 NS 0.25 NS 901 1 2 2.2 NS 0 0.88 NS 2.5 NS 1001 1 2 2.2 NS 5 5.62 NS 0.41 NS 1102/1103 1 0 0.72 NS 0 0.88 NS - -1101/04 10 10 0.99 NS 10 0.98 NS 1.02 NS 1201 4 0 0.23 NS 0 0.29 NS - -1301/02 19 4 0.18 0.0288 20 1.1 NS 0.16 0.0414 1303 2 0 0.43 NS 3 1.3 NS 0.27 NS 1401 8 0 0.12 NS 8 1.0 NS 0.11 NS 1501/02/03 30 19 0.53 NS 26 0.79 NS 0.67 NS 1601/02/03 5 4 0.87 NS 3 0.52 NS 1.6 NS a compared with controls b severe compared with mild 66 DRp 1*101 was significantly more common among patients with severe RA than in controls (53% vs 17%), O.R=5.5; p=0.006) only when the patients with DRpi*401and 404 were excluded from analysis (Table 5.5). HLA-DRB1* 1001 showed no association with disease or severity, being present in 2%, 5% and 1%, respectively, of patients with severe or mild RA, and normal controls (p = NS) Table 5.5 Frequencies of patients and controls expressing HLA-DRR1 alleles when individuals with HLA-DRpi*401/404 alleles were excluded. Controls Severe RA Mild RA (n=82) (n=15) (n=28) DRpi % % ORa Pa % ORa Pa ORb Pb 101 17 53 5.5 0.0060 23 1.5 NS 3.8 NS 102 4 7 1.9 NS 4 1.0 NS 1.8 NS 103 1 0 1.7 NS 7 6.7 NS 0.32 NS 405 2 7 2.9 NS 0 0.61 NS 5.5 NS 408 1 0 1.7 NS 0 1.0 - - -1001 1 0 1.7 NS 4 3.2 NS 0.55 NS compared with controls b severe compared with mild 5.2.1.1 The Effect of HLA-DRpi Gene Dose Twenty-three percent of patients (11/48) with severe disease expressed two disease-associated DRpl alleles, compared with 13% (5/39) with mild disease (OR=2.02; p=NS), and 5% (5/105) of controls (OR=5.95; p=0.0018). Eight percent of patients (4/48) with severe disease were homozygous for DRpl*0401, compared with 3% (1/39) with mild disease (OR=3.45; p=NS); this combination was not observed amongst the controls. No patients or control subjects expressed two alleles containing the DERAA motif, nor were they homozygous for these alleles. Four percent of patients with severe RA (2/48) expressed alleles with the QKRAA/QRRAA and DERAA motifs, 6 7 compared with 10% (4/39) with mild RA and 6% (6/105) of controls. Ten percent of patients with severe RA (5/48), compared with 26% (10/39) with mild RA and 44% (46/105) of controls expressed neither epitope. 5.2.1.2 Allele Group Frequency The HLA-DRpi alleles containing the SE (QKRAA or QRRAA) appeared to be associated with a high risk for severity of RA. In contrast, HLA-DRpi alleles containing the DERAA sequence appeared to be associated with a low risk of severe RA. Patients and controls were therefore divided into four categories according to the risk of alleles they were carrying, high risk or SE+, low risk or DE+/DR8+, both and neither. Chi-square analysis showed that the frequencies of these four risk groups differed significantly between patients with mild and severe RA, Chi-square=16.91 and p=0.0007. Further comparison of only high and low risk alleles of HLA-DRpi, among patients with both mild and severe RA indicated that there is an odds ratio difference of approximately 22 fold between the high and low risk groups of alleles among the two patient populations. The proportion of patients expressing each of these alleles in the three study groups is shown in Table 5.6. Eighty-one percent of patients (39/48) with severe RA were positive for high risk alleles, compared with 41% (16/39) with mild disease (OR=6.23; p=0.0003) and 34% (36/105) of controls (OR=8.31; pO.OOOl). In contrast, only 2% of patients (1/48) with severe RA had at least a single low risk allele compared with 23% (9/39) in the mild RA group (OR=0.07; p=0.0066) and 22% (23/105) of the controls (OR=0.08; p=0.0039). There was no significant difference in the frequency of patients with both, high and low risk, alleles or neither alleles between severe and the mild groups or between the mild RA and control groups. However the difference in the frequency of high risk, low risk and neither alleles remained highly significant between the severe and control groups. 68 Table 5.6 Frequencies of patients and controls expressing high risk alleles, low risk alleles, both or neither at HLA-DR(31 locus. Categories Controls Severe RA MildRA ORb Pb % % ORa pa % OR a Pa High (SE+) 34 81 8.31 O.0001 41 1.33 NS 6.23 0.0003 Both 7 6 0.93 NS 13 2.09 NS 0.45 NS Neither 37 10 0.20 0.0014 23 0.51 NS 0.39 NS Low (DE+/DR8+) 22 2 0.08 0.0039 23 1.07 NS 0.07 0.0066 compared with controls b severe compared with mild 5.2.1.3 Influence of HLA-DRpi on the Odds of Severe versus Mild Disease, Adjusted for Demographic and Clinical Variables HLA-DRpi typing results were added to the four previously defined demographic and clinical parameters in a forward step-wise logistic regression analysis to determine if additional information on the outcome of RA could be gained by typing for HLA-DRpi. Patients were divided into four groups according to the alleles expressed (high risk or SE+, low risk or DE+, both SE+/DE+ and neither as previously described) for analysis. The results of the logistic regression analysis are shown in Table 5.7. This analysis indicated that information on the typing of HLA-DRpl has statistically significant incremental value in predicting the outcome of the disease after considering all clinical parameters, Chi-squre=17.83 and p=0.0005. Table 5.7 Logistic regression analysis using demographic, clinical data and H L A -D R p i markers to predict the outcome of R A . Factors Chi-Square d.f P value Age at onset 5.92 1 0.0150 Gender 3.46 1 0.0628 (NS) Duration 0.76 1 0.3819 (NS) RF 13.74 2 0.0010 HLA-DR-pl 17.83 3 0.0005 69 5.2.1.4 Electrostatic Charges Comparison of the sequence of the putative high risk (QKRAA or QRRAA) and low risk (DERAA) epitope indicates that there is a difference of two amino acids, carrying different charges, between these motifs. The high risk epitopes consist of neutral and positively charged amino acids at positions 70 and 71, respectively, while the low risk epitope has negatively charged amino acids at the same positions. To determine whether the electrostatic charges on amino acids 70 and 71 within the HLA-DRB1 peptide groove influence the risk of disease by determining the binding pattern of antigenic peptides, patients were divided into eight different groups according to the amino acid charges at these positions. The proportion of patients expressing combinations of these alleles in the three study groups is shown in Table 5.8. Thirty-seven percent of patients (18/48) with severe RA had a neutral AA at position 70 and a positive AA at position 71 on both alleles, compared with 13% (5/39) with mild disease (OR=4.08; p=0.0187) and 21% (22/105) of controls (OR=2.26; p=0.0496). In contrast, none of the patients (0/48) with severe RA had either allele with negatively charged AAs at position 70 and 71 alone, compared with 15% (6/39) in the mild RA group (OR=0.05; p=0.0168) and 8% (8/105) of the controls (OR=0.12; p=NS). There was no significant difference in the frequency of patients with other alleles among the three groups. 70 Table 5.8 Frequencies of patients and controls according to the amino acid charges at positions 70 and 71. Alleles A1/A2 Controls Severe RA MildRA ORb Pb \u00E2\u0080\u00A2% % ORa Pa % ORa Pa N/N 0 0 - - 0 - - - -P/P 0 0 - - 0 - - - -O/O 21 37 2.26 0.0496 13 0.55 NS 4.08 0.0187 N/O 10 6 0.57 NS 18 1.87 NS 0.30 NS N/P 3 0 0.30 NS 3 0.90 NS 0.27 NS N/X 8 0 0.12 NS 15 2.21 NS 0.05 0.0168 P/0 4 4 1.10 NS 8 2.10 NS 0.52 NS Others 54 53 0.91 NS 44 0.65 NS 1.41 NS N denotes having a negatively charged AA in positions70 and 71. P denotes having a positively charged AA in positions 70 and 71. O denotes having neutral AA at position 70 and a positively charged AA at position 71. X denotes any other allele. a compared with controls b severe compared with mild 71 5.2.2 HLA - D Q P 1 The proportion of patients expressing each HLA-DQpi allele in the three study groups is shown in Table 5.9. Fifty percent of patients (24/48) with severe RA were positive for HLA-DQpl*0302, compared with 23% (9/39) with mild disease (OR=3.3; p=0.019) and 20% (21/105) of controls (OR=4.0; p=0.0003). In contrast, 0% of patients (0/48) with severe RA had HLA-DRpi* 0603 compared with 13% (5/39) in the mild RA group (OR=0.065; p=0.036) and 16% (17/105) of the controls (OR=0.05; p=0.0074). The frequency of the rest of the alleles were not statistically different between the study groups (p<0.05) and/or did not have an odds ratio which varied significantly from OR=l. Table 5.9 Frequencies of patients and controls expressing H L A - D Q p i alleles. Controls Severe RA MildRA DQP1 % % ORa Pa % ORa Pa ORb Pb 201 45 33 0.62 NS 38 0.71 NS 0.80 NS 301 31 27 0.85 NS 28 0.84 NS 0.94 NS 302 20 50 4.0 0.0003 23 1.1 NS 3.33 0.019 303 7 6 0.93 NS 0 0.35 NS 6.08 NS 402 8 2 0.26 NS 8 0.96 NS 0.26 NS 501 19 27 1.6 NS 26 1.6 NS 1.08 NS 502 6 4 0.72 NS 10 1.8 NS 0.38 NS 503 7 0 0.13 NS 8 1.1 NS 0.11 NS 601 2 4 2.2 NS 5 2.6 NS 0.80 NS 602 29 15 0.43 NS 13 0.43 NS 1.16 NS 603 16 0 0.05 0.0074 13 0.72 NS 0.065 0.036 604 5 2 0.43 NS 8 1.6 NS 0.25 NS Others 4 0 0.23 NS 0 0.27 NS - -a compared with controls b severe compared with mild 72 5.2.2.1 Allele Group Frequency HLA-DQpi alleles were grouped based on the differential odds ratios between mild and severe RA, as previously described in the statistical methods. The HLA-DQP1*302 allele was considered to have high risk for severity of RA. In contrast, HLA-DQP 1* 402/502/603/604 alleles were considered to have low risk for severe RA. Patients and controls were therefore divided into four categories according to the risk of alleles they were carrying, high risk or DQ 3 02+, low risk or DQ 402+/502+/ 603+/604+, both and neither. The proportion of patients expressing each of these alleles in the three study groups is shown in Table 5.10. Forty-six percent of patients (22/48) with severe RA were positive for high risk alleles, compared with 18% (7/39) with mild disease (OR=3.87; p=0.0T19) and 15% (16/105) of controls (OR=4.71; p=0.000T). In contrast, only 4% of patients (2/48) with severe RA had low risk alleles compared with 33% (13/39) in the mild RA group (OR=0.09; p=0.0010) and 32% (34/105) of the controls (OR=0.09; p=0.0003). There was no significant difference in the frequency of patients with both alleles or neither allele between severe, mild and the control groups. However the difference in the frequency of high and low risk alleles remained highly significant between the severe and control groups. Table 5.10 Frequencies of patients and controls expressing high risk alleles, low risk alleles, both or neither at H L A - D Q p i locus. Controls Severe RA MildRA ORb Pb Categories % % ORa Pa % ORa Pa High (302+) 15 46 4.71 0.0001 18 1.22 NS 3.87 0.0119 Both 5 4 0.87 NS 5 1.08 NS 0.80 NS Neither 48 46 0.93 NS 44 0.85 NS 1.09 NS Low (402+/502+/ 32 4 0.09 0.0003 33 1.04 \u00E2\u0080\u00A2 NS 0.09 0.0010 603+/604+) a compared with controls b severe compared with mild 73 5.2.2.2 Influence of HLA-DQpi on the Odds of Severe versus Mild Disease, Adjusted for Demographic and Clinical Variables HLA-DQpi typing results were added to the four demographic and clinical parameters in a forward step-wise logistic regression analysis to determine if additional information on the outcome of RA could be gained by typing for HLA-DQPI. Patients were divided into four groups according to the alleles expressed (high risk or DQ 302+, low risk or DQ 402+/502+/ 603+/604+, both and neither based on their differential odds ratios, as previously described in the statistical methods) for the analysis. The results of the logistic regression analysis are shown in Table 5.11. This analysis indicated that information on the typing of HLA-DQPI has no statistically significant (p<0.05) incremental value in predicting the outcome of the disease after considering for the defined demographic and clinical parameters, Chi-square=7.27 and p=NS. Table 5.11 Logistic regression analysis using demographic, clinical data and H L A -D Q p i markers to predict the outcome of R A . Factors Chi-Square d.f P value Age at onset L71 I 0.1911 (NS) Gender 1.42 1 0.2336 (NS) Duration 1.24 1 0.2651 (NS) RF 16.85 2 0.0002 HLA-DQ-pl 7.27 3 0.0639 (NS) 74 5.2.3 HLA-DPpl The proportion of patients expressing each HLA-DPpi allele in the three study groups is shown in Table 5.12. Seventy-nine percent of patients (38/48) with severe RA were positive for HLA-DPpl *0401, compared with 54% (21/39) with mild disease (OR=3.26; p=0.022) and 62% (31/50) of controls (OR=2.3; p=NS). In contrast, 2% of patients (1/48) with severe RA had HLA-DPpl* 0501 compared with 15% (6/39) in the mild RA group (OR=0.12; p=NS) and 6% (3/50) of the controls (OR=0.33; p=NS). The frequency of the rest of the alleles did not reach a statistically significant level (p<0.05) and/or did not have an odd ratio which varied significantly from OR=l. Table 5.12 Frequencies of patients and controls expressing H L A - D P p i alleles. Controls Severe RA MildRA DPpl % % OR a Pa % ORa Pa OR b Pb 101 8 6 0.77 NS 5 0.59 NS 1.23 NS 201 20 23 1.2 NS 38 2.3 NS 0.48 NS 202 0 0 - NS 3 3.7 NS 0.27 NS 301/1401 36 23 0.53 NS 23 0.65 NS 0.99 NS 401 62 79 2.3 NS 54 0.71 NS 3.26 0.0224 402 26 27 1.1 NS 18 0.59 NS 1.70 NS 501 6 2 0.33 NS 15 2.7 NS 0.12 NS 601 0 4 5.4 NS 0 - NS 4.25 NS 801 0 0 - NS 0 - NS - NS 901 6 2 0.33 NS 0 0.16 NS 2.50 NS 1001 4 2 0.51 NS 8 2.6 NS 0.26 NS 1101 2 4 2.1 NS 5 2.5 NS 0.80 NS 1301 0 0 - NS 0 - NS - -1501 0 0 - NS 3 3.7 NS 0.27 NS 1601 2 0 0.34 NS 0 0.40 NS - -1701 0 2 3.19 NS 3 3.7 NS 0.81 NS 1801 0 0 - NS 0 - NS - -1901 2 0 0.34 NS 0 0.40 NS - -a compared with controls b severe compared with mild 75 5.2.3.1 Allele Group Frequency HLA-DPB1 alleles were grouped based on the differential odds ratios between mild and severe RA, as previously described in the statistical methods. The HLA-DPpi*401 allele was considered to have high risk for severity of RA. In contrast, HLA-DPpi* 201/501/1001 alleles were considered to have low risk for severe RA. Patients and controls were therefore divided into four categories of risk according to the alleles they were carrying, high risk or DP 401+, low risk or DP 201+/501+/1001+, both and neither. The proportion of patients expressing each of these alleles in the three study groups is shown in Table 5.13. Fifty-eight percent of patients (28/48) with severe RA were positive for high risk alleles, compared with 36% (14/39) with mild disease (OR=2.50; p=NS) and 54% (27/50) of controls (OR=4.04; p=0.0002). In contrast, only 6% of patients (3/48) with severe RA had low risk alleles compared with 38% (15/39) in the mild RA group (OR=0.11; p=0.0006) and 20% (10/50) of the controls (OR=0.63; p=NS). There was no significant difference in the frequency of patients with high risk alleles, both alleles or neither allele between severe and the mild groups. Also there was no significant difference in the frequency of patients with low risk alleles and neither allele between severe and the control groups. However there was a significant difference in the frequency of patients with high risk alleles and both alleles between severe and the control groups. Table 5.13 Frequencies of patients and controls expressing high risk alleles, low risk alleles, both or neither at H L A - D P p l locus. Controls Severe RA MildRA Categories % % OR a Pa % OR a Pa ORb Pb High(401+) 54 58 4.04 0.0002 36 1.62 NS 2.50 NS Both 8 21 6.64 0.0020 18 5.52 0.0129 1.20 NS Neither 18 15 1.82 NS 8 0.89 NS 2.05 NS Low (201+/ 20 6 0.63 NS 38 5.94 0.0001 0.11 0.0006 501+/1001+) a compared with controls b severe compared with mild 76 5.2.3.2 Influence of HLA-DPpi on the Odds of Severe versus Mild Disease, Adjusted for Demographic and Clinical Variables HLA-DPpi typing results were added to the four previously defined demographic and clinical parameters in a forward step-wise logistic regression analysis to determine if additional information on the outcome of RA could be gained by typing for HLA-DPpi. Patients were divided into four groups according to the alleles expressed (high risk or DP 401+, low risk or DP 201+/501+/1001+, both and neither based on their differential odds ratios, as previously described in the statistical methods) for the analysis. The results of the logistic regression analysis are shown in Table 5.14. This analysis indicated that information on the typing of HLA-DPpl has statistically incremental value in predicting the outcome of the disease after considering for demographic and clinical parameters, Chi-square= 17.28 and p=0.0006. Table 5.14 Logistic regression analysis using demographic, clinical data and H L A -D P p i markers to predict the outcome of R A . Factors Chi-Square d.f P value Age at onset 7.12 1 0.0076 Gender 1.78 1 0.1822 (NS) Duration 2.68 1 0.1018 (NS) RF 16.47 2 0.0003 HLA-DP-pl 17.28 3 0.0006 77 5.2.4 Combined Effects of HLA-DRpl and DQpl HLA-DRpl/DQpl frequencies in the three study groups are shown in Table 5.15. Nineteen percent of patients (9/48) with severe RA were positive for each of the combinations HLA-DRpl/DQpl*040i/0301 or HLA-DRpl/DQpl*0404/0302, compared with 13% (5/39) with mild disease (OR=1.7; p=NS) and 5% (5/105) of controls (OR=4.62; p=0.013). Thirty-three percent of patients (16/48) with severe RA were positive for HLA-DRpi/DQpi*0401/0302 compared with 10% (4/39) with mild disease (OR=4.37; p=0.022) and 11% (12/105) of controls (OR=4.27; p=0.00T3). In contrast, 0% of patients (0/48) with severe RA had HLA-DRpi/DQpi*1301/0603 compared with 8% (3/39) in the mild RA group (OR=0.11; p=NS) and 13% (14/105) of the controls (OR=0.065; p=0.019). The frequency of the rest of the allele combinations was not significantly (p<0.05) different from the mild group and/or did not have an odd ratio which varied significantly from OR=l. 78 Table 5.15 Frequencies of patients and controls expressing H L A - D R R l / D Q p i haplotypes. Controls Severe RA Mild RA DRBl/DQpl % % ORa Pa % ORa Pa ORb Pb 101-501 14 21 NS 20 NS NS 102-501 3 6 NS . 5 NS NS 103-301 0 1 NS 0 - NS 103-501 1 0 NS 5 NS NS 1501-502 1 0 NS 5 NS NS 1501-602 27 15 NS 15 NS NS 1501-601 0 0 - 2 NS NS 1501-603 2 0 NS 5 NS NS 1502-601 1 4 NS 2 NS NS 1601-502 5 4 NS 2 NS NS 401-301 5 19 4.62 0.013 13 2.94 NS 1.57 NS 401-302 11 33 4.27 0.0013 10 0.98 NS 4.37 0.0221 402-302 1 2 NS 2 NS NS 403-302 2 0 NS 0 NS -404-301 1 2 NS 0 NS NS 404-302 5 19 4.62 0.013 13 2.94 NS 1.57 NS 405-201 0 1 NS 0 - NS 405-301 1 0 NS 0 NS -405-302 2 0 NS 0 NS -407-301 6 0 NS 0 NS -407-302 0 2 NS 0 - NS 408-301 1 0 NS 0 NS -408-402 1 0 NS 0 NS -301-201 31 19 NS 20 NS NS 301-301 0 1 NS 0 - NS 301-501 0 1 NS 0 - NS 301-603 1 0 NS 0 NS -1101-301 7 6 NS 5 NS NS 1101-302 0 1 NS 0 - NS 1104-301 3 2 NS 5 NS NS 1201-301 4 0 NS 0 NS -1301-302 0 1 NS 0 - NS 1301-602 1 0 NS 0 NS -1301-603 13 0 0.065 0.019 8 0.54 NS 0.11 NS 1301-604 0 0 - 1 NS NS 1302-603 0 0 - 1 NS NS 1302-301 0 0 - 1 NS NS 1302-604 5 2 NS 5 NS NS 79 Table 5.15 Continued Controls Severe RA M i l d R A DRpl /DQpl % % OR a pa % OR a pa OR b pb 1303-201 1 0 NS 0 NS -1303-301 0 0 - 2 NS NS 1305-301 1 0 NS 0 NS -1401-503 6 0 NS 7 NS NS 1402-301 1 0 NS 0 NS -1404-503 1 0 NS 0 NS -701-201 13 17 NS 20 NS NS 701-303 6 4 NS 2 NS NS 701-603 1 0 NS 0 NS -801-301 0 2 NS 0 - NS 801-402 4 2 NS 7 NS NS 802-402 1 0 NS 0 NS -8032-301 3 0 NS 0 NS -811-402 1 0 NS 0 NS -901-303 1 2 NS 0 NS NS 1001-501 1 2 NS 2 NS NS 1001-502 0 0 _ 1 NS NS a compared with controls b severe compared with mild 80 5.2.5 Combined Effects of HLA-DRpi , DQpl and DPpl The combinations with greatest difference in their frequencies among the three groups were chosen for analysis of the combined effect of HLA class II antigens on RA. These results are shown in Table 5.16. Twenty-seven percent of patients (13/48) with severe RA were positive for HLA-DRpi *0401/ DQpi*302/DPpl*401 combination, compared with 5% (2/39) with mild disease (OR=6.87; p=0.016) and 14% (7/50) of controls (OR=2.28; p=NS). In contrast, 0% of patients (0/48) with severe RA had the HLA-DRpl*1301, 1302/DQpi*603 and DPpl*201, 501, 1001 combination, compared with 13% (5/39) in the mild RA group (OR=0.07; p=0.0364) and 4% (2/50) of the controls (OR=0.20; p=NS). The frequency of the rest of the allele combinations did not reach a statistically significant level (p<0.05). Table 5.16 Frequencies of patients and controls expressing H L A - D R p i / D Q p i / D P p i allele combinations. Controls Severe RA Mild RA DRpi/DQpl/DPpl % % OR a Pa % OR a Pa OR b Pb 401/302/401 14 27 2.28 NS 5 0.33 NS 6.87 0.0159 401/301/401 2 19 11.3 0.016 5 2.65 NS 4.27 NS 404/302/401 2 17 9.80 0.030 8 4.08 NS 2.40 NS 101,2/501/402 2 10 5.67 NS 5 2.65 NS 2.15 NS 401/302/402 4 6 1.60 NS 5 1.30 NS 1.23 NS 101,2/501/301,1401 10 6 0.60 NS 8 0.75 NS 0.80 NS 301/201/201,501 10 6 0.60 NS 15 1.64 NS 0.37 NS 1301,02/603/201,501,1001 4 0 0.20 NS 13 3.53 NS 0.07 0.0364 a compared with controls b severe compared with mild 81 The combined data from HLA-DR, DQ and DP analysis was categorised into three groups of high risk allele (at all three loci as previously defined), low risk allele (at all three loci as previously defined) and others (without high or low risk alleles at all three loci). Chi-square analysis showed that the frequencies of these three risk groups differed significantly between patients with mild and severe RA, Chi-square=l 1.19 and p=0.0037. Further comparison of only high and low risk triple positive alleles of HLA class II, among patients with both mild and severe RA, indicated that there is an odds ratio difference of approximately 31 fold between the high and low risk groups of alleles among the two patient populations. 5.2.5.1 Combined Influence of HLA-DRpi, DQpl, and DPpl on the Odds of Severe versus Mild Disease, Adjusted for Demographic and Clinical Variables HLA-DRpi, HLA-DQPI and HLA-DPpi typing results were added to the four previously defined demographic and clinical parameters in a forward step-wise logistic regression analysis to determine if additional information on the outcome of RA could be gained by typing for all of the HLA class II genes. Patients were divided into four groups (high risk, low risk, both and neither as previously described) for each of the HLA class II genes for analysis. The results of the logistic regression analysis are shown in Table 5.17. This analysis indicated that information on the typing of HLA class II has statistically significant incremental value in predicting the disease outcome even after considering the demographic and clinical parameters, Chi-square=31.08 and p=0.0003. 82 Table 5.17 Logistic regression analysis using demographic, clinical and all H L A class II genes markers to predict the outcome of R A . Factors Chi-Square d.f P value Age at onset 4.56 1 0.0326 Gender 0.98 1 0.3221 (NS) Duration 0.89 1 0.3466 (NS) RF 10.96 2 0.0042 HLA-DRpi, DQ81 &DPpl 31.08 9 0.0003 83 5.3 Tumor Necrosis Factor Genes 5.3.1 TNF2 (-308) Polymorphism The frequencies of individuals expressing TNF (-308) polymorphism are shown in Table 5.18. None of the patients with either severe or mild RA were homozygous for the mutant allele AA (adenine- adenine) compared with 2% (1/50) of controls (OR=0.34, p=NS). Sixty-nine percent of patients (33/48) with severe RA were homozygous for the wild type allele GG (guanine-guanine) compared with 77% (30/39) of mild RA patients and 66% (33/50) of controls (OR=0.66 and 1.13, respectively, p=NS). Thirty-one percent of patients (15/48) with severe RA were heterozygous for both alleles AG (adenine-guanine) compared with 23% (9/39) of mild RA patients and 32% (16/50) of controls (OR=0.97 and 1.51, respectively, p=NS). This data did not indicate a significant difference in frequency of these three alleles between the three study groups. Table 5.18 T N F (-308) polymorphism frequencies of patients with R A and controls. TNFc Controls Severe RA MildRA OR b Pb % % OR a Pa % ORa Pa A/A 2 0 0.34 NS 0 0.42 NS - NS G/G 66 69 1.13 NS 77 1.72 NS 0.66 NS A/G 32 31 0.97 NS 23 0.64 NS 1.51 NS a compared with controls b severe compared with mild 5.3.2 TNF Microsatellite Polymorphism The frequencies of TNFa microsatellite polymorphisms in each of the three study groups are shown in Table 5.19. In addition to the thirteen dinucleotide repeats (al-al3), one new polymorphism (al4) with 122bp was detected in our study. Nineteen percent of patients (9/48) with severe RA were positive for TNFa2 compared with 5% (2/39) with mild disease (OR=4.3; p=0.0024) and 34% (17/50) of controls (OR=0.44; p=NS). Forty percent of patients (19/48) with severe RA and 38% (15/39) of mild RA patients were positive for TNFa6 compared with 20% (10/50) of controls 84 (OR=2.5; p=NS or OR=2.57; p=0.035 for total patient population compared to the control). In contrast, 27% of patients (13/48) with severe RA had TNFal 1 compared with 51% (20/39) in the mild RA group (OR=0.35; p=0.036) and 32% (16/50) of the controls (OR=0.79; p=NS). The difference in the frequency of the rest of the alleles did not reach a statistically significant level (p<0.05) and/or did not have an odd ratio which varied significantly from OR=l. Table 5.19 TNFa microsatellite polymorphism frequencies of patients with R A and controls. Controls Severe RA Mild RA TNFa % % OR a Pa % ORa Pa ORb Pb al 6 2 0.33 NS 0 0.17 NS 2.50 NS a2 34 19 0.44 NS 5 0.10 0.0024 4.27 NS a3 24 35 1.74 NS 38 1.98 NS 0.88 NS a4 8 10 1.34 NS 8 0.96 NS 1.39 NS a5 10 4 0.39 NS 5 0.49 NS 0.80 NS a6 20 40 2.62 NS 41 2.78 NS 0.94 NS a7 12 15 1.25 NS 8 0.61 NS 2.05 NS a8 2 4 2.13 NS 5 2.65 NS 0.80 NS a9 0 4 5.43 NS 3 3.84 NS 1.65 NS alO 26 19 0.66 NS 15 0.52 NS 1.27 NS a l l 32 27 0.79 NS 51 2.24 NS 0.35 0.036 al2 0 2 3.19 NS 3 3.93 NS 0.81 NS al3 6 6 1.04 NS 8 1.31 NS 0.80 NS a!4* 0 2 3.19 NS 0 - NS 2.50 NS a compared with controls b severe compared with mild * denotes new alleles Two new mononucleotide repeats (t>8-t>9) were detected with 13 lbp and 132bp respectively, in addition to the seven previously reported (bl-b7) TNFb polymorphisms. The frequencies of these polymorphisms are shown in Table 5.20. At the TNFb locus, 23% percent of patients (11/48) with severe RA were positive for TNFb4 compared with 8% (3/39) with mild disease (OR=3.6; p=NS) 85 and 10% (5/50) of controls (OR=2.68; p=NS). Two percent of patients (1/48) with severe RA were positive for the novel allele, TNFb8, compared with 13% (5/39) with mild disease (OR=0.14; p=NS) and 0% (0/50) of controls (OR=16.10; p=0.032). The difference in the frequency of the rest of the alleles did not reach a statistically significant level (p<0.05) and/or did not have an odds ratio which varied significantly from OR=l. Table 5.20 TNFb microsatellite polymorphism frequencies of patients with R A and controls. Controls Severe RA MildRA TNFb % % ORa % ORa OR b Pb bl 0 0 - - 0 - - - -b2 0 0 - - 0 - - - -b3 26 12 0.41 NS 13 0.42 NS 0.97 NS b4 10 23 2.68 NS 8 0.75 NS 3.57 NS b5 24 19 0.73 NS 13 0.47 NS 1.57 NS b6 62 62 1.02 NS 69 1.38 NS 0.74 NS bl 46 58 1.64 NS 49 1.11 NS 1.47 NS b8* 0 2 3.19 NS 13 16.10 0.032 0.14 NS b9* 4 0 0.20 NS 3 0.63 NS 0.27 NS a compared with controls b severe compared with mild * denotes new alleles Two dinucleotide repeats (cl-c2) have been previously reported for TNFc polymorphism. The frequencies of these polymorphisms are shown in Table 5.21. At the TNFc locus, 85%) percent of patients (41/48) with severe RA were positive for TNFcl compared with 95% (37/39) with mild disease (OR=0.32; p=NS) and 90% (45/50) of controls (OR=0.65; p=NS). The difference in the frequency of the rest of the alleles did not reach a statistically significant level (p<0.05) and/or did not have an odds ratio which varied significantly from OR=l. 86 Table 5.21 TNFc microsatellite polymorphism frequencies of patients with R A and controls. Controls Severe RA MildRA TNFc % % ORa Pa % ORa Pa OR b Pb cl 90 85 0.65 NS 95 2.06 NS 0.32 NS c2 58 50 0.72 NS 44 0.56 NS 1.29 NS a compared with controls b severe compared with mild Seven dinucleotide repeats (dl, d7) have been previously reported for TNFd polymorphisms. The frequencies of these polymorphisms are shown in Table 5.22. Twenty-one percent of patients (10/48) with severe RA were positive for TNFd5 compared with 10% (4/39) with mild disease (OR=2.3; p=NS) and 8% (4/50) of controls (OR=3.0; p=NS). The difference in the frequency of the rest of the alleles did not reach a statistically significant level (p<0.05) and/or did not have an odds ratio which varied significantly from OR=l. Table 5.22 T N F d microsatellite polymorphism frequencies of patients with R A and controls. TNFd Controls Severe RA MildRA ORb Pb % % ORa Pa % ORa Pa dl 24 25 1.06 NS 18 0.69 NS 1.52 NS d2 2 6 3.27 NS 5 2.65 NS 1.23 NS d3 68 77 1.58 NS 85 2.59 NS 0.61 NS d4 58 42 0.52 NS 38 0.45 NS 1.14 NS d5 8 21 3.03 NS 10 1.31 NS 2.30 NS d6 2 2 1.04 NS 3 1.29 NS 0.81 NS d7 0 0 _ NS 5 6.73 NS 0.15 NS a compared with controls b severe compared with mild 87 Four dinucleotide repeats (el, e4) have been previously reported for TNFe polymorphisms. The frequencies of these polymorphisms are shown in Table 5.23 below. Ninety-four percent of patients (45/48) with severe RA were positive for TNFe3 compared with 97% (38/39) with mild disease (OR=0.24; p=NS) and 98% (49/50) of controls (OR=0.18; p=NS). The difference in the frequency of the rest of the alleles did not reach a statistically significant level (p<0.05) and/or did not have an odds ratio which varied significantly from OR=l. Table 5.23 TNFe microsatellite polymorphism frequencies of patients with R A and controls. TNFe Controls Severe RA MildRA OR b Pb % % ORa Pa % OR a Pa el 36 29 0.73 NS 23 0.53 NS 1.37 NS e2 6 6 1.04 NS 5 0.85 NS 1.23 NS e3 98 94 0.18 NS 97 0.77 NS 0.24 NS e4 0 0 - NS 3 0.63 NS 0.27 NS 8 compared with controls b severe compared with mild 88 5.4 Interleukin-1 Genes 5.4.1 IL-1 Alpha Although several attempts were made to type this locus according to the protocol and primer sequences in the published papers, consistent and specific amplifications were not obtained. PCR conditions (temperature, cycling time, enzyme, magnesium, etc) were changed several times and Perfect Match polymerase enhancer (Stratagene) was used, but none of the observed amplicons were similar to published findings or specific and reproducible. 5.4.2 IL-1 Beta In addition to the two reported human IL-ip polymorphisms, a third polymorphism was identified by carrying out a double digest of the amplified DNA. Amplicons were simultaneously digested with restriction enzymes, TaqI and Fnu 4HJ, to determine whether the two polymorphisms were carried on the same chromosome (cis) or not. Determining the phase (cis or trans) provided additional information to further evaluate the effects of the IL-lp genetic polymorphism on the physiological activity of IL-1 p. The proportion of patients expressing each IL-1 beta allele in the three study groups is shown in Table 5.24. Six percent of patients (3/48) with severe RA were homozygous for allele 2 of IL-lp (TaqI -/- allele), compared with 3% (1/39) with mild disease (OR=2.53; p=NS) and 2% (1/50) of controls (OR=3.27; p=NS). In contrast, 54% of patients (26/48) with severe RA were homozygous for allele 1 of IL-1 P (TaqI +/+) compared with 64% (225/39) in the mild RA group (OR=0.66; p=NS) and 64% (32/50) of the controls (OR=0.67; p=NS). However, the difference in the frequencies of these two alleles did not reach a statistically significant level between all three groups. The frequencies of the rest of the alleles were not significantly (p\u00C2\u00A30.05) different from the mild group and/or did not have an odds ratio which varied significantly from OR=l. 89 Table 5.24 I L - l p polymorphism frequencies of patients with R A and controls. Controls Severe RA MildRA IL-lp % % ORa Pa % OR a Pa ORb Pb Fnu 4HI +/+ 44 42 0.91 NS 46 1.09 NS 0.83 NS Fnu 4HI+/- 46 50 1.17 NS 41 0.82 NS 1.44 NS Fnu 4HI-/- 10 8 0.82 NS 13 1.32 NS 0.62 NS Taql+/+ 64 54 0.67 NS 64 1.00 NS 0.66 NS Taql+/- 34 40 1.27 NS 33 0.97 NS 1.31 NS Taql-/- 2 6 3.27 NS 3 1.29 NS 2.53 NS Fnu 4HI/TaqI* 76 60 0.48 NS 72 0.80 NS 0.60 NS a compared with controls b severe compared with mild + enzyme cut and - uncut denotes new alleles 90 5.4.3 IL-1 Receptor Antagonist The frequencies of IL-1RN alleles are shown in Table 5.25. Forty-two percent of patients (20/48) with severe RA were positive for A2, compared with 56% (22/39) with mild disease (OR=0.55; p=NS) and 64% (32/50) of controls (OR=0.4; p=0.044). There was no significant difference in frequencies of A l and A3 alleles among all groups. Table 5.25 IL-1 receptor antagonist polymorphism frequencies of patients with R A and controls. IL-1RN Controls Severe RA MildRA ORb Pb % % ORa Pa % ORa Pa A l 88 88 0.95 NS 90 1.19 NS 0.80 NS A2 64 42 0.40 0.044 56 0.73 NS 0.55 NS A3 4 4 1.04 NS 5 1.30 NS 0.80 NS A4 0 0 - - 0 - - - -A5 0 0 - - 0 - - - -a compared with controls b severe compared with mild 91 5.5 Interleukin-2 Genes 5.5.1 Interleukin-2 Alpha Microsatellite Polymorphism The results of the IL-2 typing are shown in Table 5.26. In addition to the 15 known polymorphisms, we have detected one new polymorphism with a 122bp allele. Table 5.18 shows this new polymorphism, in bold, along with the proportion of patients expressing each IL-2 allele in the three study groups. Twenty-three percent of patients (11/48) with severe RA were positive for A15, compared with 13% (5/39) with mild disease (OR=2.02; p=NS) and 8% (4/50) of controls (OR=3.42; p=NS). In contrast, 0% of patients (0/48) with severe RA had A8 compared with 10% (4/39) in the mild RA group (OR=0.08; p=NS) and 6% (3/50) of the controls (OR=0.14; p=NS). Table 5.26 IL-2a microsatellite polymorphism frequencies of patients with R A and controls. Controls Severe RA MildRA ORb Pb IL-2 % % ORa Pa % OR a A l (148bp) 0 2 3.19 NS 0 - - 2.50 NS A2(146bp) 0 2 3.19 NS 5 6.73 NS 0.39 NS A3(144bp) 2 4 2.13 NS 8 4.08 NS 0.52 NS A4 (142bp) 10 8 0.82 NS 5 0.49 NS 1.68 NS A5 (140bp) 12 17 1.47 NS 15 1.33 NS 1.10 NS A6(138bp) 22 27 1.32 NS 18 0.78 NS 1.70 NS A7(136bp) 28 25 0.86 NS 28 1.01 NS 0.85 NS A8(134bp) 6 0 0.14 NS 10 1.79 NS 0.08 NS A9(132bp) 28 25 0.86 NS 38 1.61 NS 0.53 NS A10(130bp) 20 12 0.57 NS 10 0.46 NS 1.25 NS A l l (128bp) 24 19 0.73 NS 23 0.95 NS 0.77 NS A12(126bp) 6 6 1.04 NS 5 0.85 NS 1.23 NS A13 (124bp) 4 0 0.20 NS 5 1.30 NS 0.15 NS A* (122bp) 2 0 0.34 NS 5 2.65 NS 0.15 NS A14(118bp) 0 0 - - 0 - - - -A15 (114bp) 8 23 3.42 NS 13 1.69 NS 2.02 NS a compared with controls b severe compared with mild * denotes new alleles 92 5.5.2 IL-2 Receptor Beta Microsatellite Polymorphism In addition to the eight reported polymorphisms we detected three new polymorphisms with 147bp, 165bp and 169bp, respectively, which are shown in bold. The frequency of patients expressing each of the IL-2Rp alleles in the three study groups is shown in Table 5.27. Twenty-five percent of patients (12/48) with severe RA were positive for allele 1, compared with 15% (6/39) with mild disease (OR=1.83; p=NS) and 36% (18/50) of controls (OR=0.59; p=NS). Also, thirty-seven percent of patients (18/48) with severe RA were positive for allele 2, compared with 28% (11/39) with mild disease (OR=1.53; p=NS) and 22% (11/50) of controls (OR=2.13; p=0.NS). In contrast, allele 5 was present in 2% (1/48) of patients with severe disease compared to 8% (3/39) of mild RA patient (OR=0.25; p=NS) and 12% (6/50) of control (OR=0.16; p=NS). The frequency of the rest of the alleles was not significantly (p<0.05) different from the mild group and/or did not have an odd ratio which varied significantly from OR=l. Table 5.27 IL-2 receptor beta microsatellite polymorphism frequencies of patients with R A and controls. Controls Severe RA MildRA IL-2Rp % % ORa Pa % ORa ORb Pb -2(169bp)* 0 2 3.19 NS 0 - - 2.49 NS 0 (165bp)* 6 8 1.42 NS 10 1.79 NS 0.79 NS 1 (163bp) 36 25 0.59 NS 15 0.32 NS 1.83 NS 2(161bp) 22 37 2.13 NS 28 1.39 NS 1.53 NS 3 (159bp) 34 37 1.16 NS 38 1.21 NS 0.96 NS 4(157bp) 28 29 1.06 NS 36 1.44 NS 0.73 NS 5 (155bp) 12 2 0.16 NS 8 0.61 NS 0.25 NS 6(153bp) 0 0 - - 0 - - - -7(151bp) 16 12 0.75 NS 20 1.35 NS 0.55 NS 8(149bp) 34 29 0.80 NS 33 0.97 NS 0.82 NS 9 (147bp)* 0 0 - 3 3.93 NS 0.27 NS a compared with controls b severe compared with mild * denotes new alleles 93 5.6 Interferon Gamma Gene 5.6.1 IFNy Microsatellite Polymorphism A total of 6 alleles were documented ranging from 120 to 130bp (figure 5.1), which were shown to contain a sequence ranging from 11 (120bp allele) to 16 (130bp allele) C A dinucleotide repeats. Figure 5.1: Electropherogram showing five alleles (122-130bp) at the polymorphic region in the first intron of the IFNy gene. l O O 1 1 O 1 2 0 1 3 O U O 1 5 0 UHl I FTgel | 1 O O b p | | 1 3 9 b p j | 1 5 Q b p | | 1 S O b p | LTHJ I Haol | 1 O O b p l | 1 3 9 b p | | 1 5 Q b p | 11 6 Q b p | 11 O O b p l | 1 3 9 b p | | 1 5 Q b p | | 1 S O b p | 94 The proportion of patients expressing each IFNy allele in the three study groups is shown in Table 5.28. The frequency of the polymorphisms in normal subjects ranged from 0% for the 130bp and 120bp alleles, to 68% for the 122bp and 124bp alleles. The genotype frequencies did not deviate from the expected value by Hardy-Weinberg equilibrium. Seventy-three percent of patients (35/48) with severe RA were positive for the 126bp allele, compared with 21% (8/39) with mild disease (OR=10.43; pO.OOOl) and 12% (6/50) of controls (OR=19.74; pO.OOOl). In contrast, only 6% of patients (3/48) with severe RA had the 122bp allele compared with 64% (25/39) in the mild RA group (OR=0.037; pO.OOOl) and 68% (34/50) of the controls (OR=0.019; pO.OOOl). The frequency of the rest of the alleles was not significantly (p<0.05) different from the mild group and/or did not have an odds ratio which varied significantly from OR=l. Table 5.28 IFNy microsatellite polymorphism frequencies of patients with R A and controls. IFNy No. ofCA Controls Severe RA MildRA Severe vs Mild RA Alleles repeats % % OR3 P^ Corr. % OR a OR\" b P Corr. 130bp 16 0 6 7.77 NS 0 - - 6.08 NS 128bp 15 16 15 0.90 NS 15 0.95 NS 0.94 NS 126bp 14 12 73 19.74 O.0001 21 1.89 NS 10.43 O.0001 124bp 13 68 77 1.58 NS 67 0.94 NS 1.68 NS 122bp 12 68 6 0.019 O.0001 64 0.50 NS 0.037 O.0001 120bp 11 0 0 - 3 3.93 NS 0.27 NS a compared with controls b severe compared with mild P con . Corrected P value To examine the individual and combined effects of the 126bp and 122bp polymorphisms, patients and controls were divided into four categories according to the expression of IFNy alleles associated with a high risk (126bp) or low risk (122bp) of severe RA, both (126bp and 122bp) or neither. The proportion of patients expressing each of these alleles or combinations in the three study groups is shown in Table 5.29. Seventy-three percent of patients (35/48) with severe RA were positive for the 95 126bp allele, compared with 10% (4/39) with mild disease (OR=23.56; pO.OOOl) and 4% (2/50) of controls (OR=64.62; pO.OOOl). In contrast, only 6% of patients (3/48) with severe RA had the 122bp allele compared with 54% (21/39) in the mild RA group (OR=0.057; pO.OOOl) and 70% (35/50) of the controls (OR=0.029; pO.OOOl). Table 5.29 Frequencies of patients and controls expressing high risk alleles, low risk alleles, both or neither at IFNy locus. Controls Severe RA Mild RA Categories % % ORa pa % OR a P a OR\" pb High (126bp) 4 73 64.62 O.0001 10 2.74 NS 23.56 O.0001 Low(122bp) 70 6 0.029 O.0001 54 0.50 NS 0.057 O.0001 Both 8 0 0.11 NS 10 1.31 NS 0.081 NS Neither 18 21 1.20 NS 26 1.57 NS 0.76 NS compared with controls b severe compared with mild Chi-square analysis showed that the frequencies of these four risk groups differed significantly between patients with mild and severe RA, Chi-square=41.66 and pO.OOOl. Further comparison of only high and low risk alleles of IFNy, among patients with both mild and severe RA, indicated that there is an odds ratio difference of approximately 61 fold between the high and low risk groups of alleles among the two patient populations. 5.6.2 Influence of IFNy on the Odds of Severe versus Mild Disease, Adjusted for Demographic and Clinical Variables IFNy typing results were added to the four previously defined demographic and clinical parameters in a forward step-wise logistic regression analysis to determine if additional information on the outcome of RA could be gained by typing for IFNy. Patients were divided into four groups according to the alleles expressed (high risk or 126bp+, low risk or 122bp+, both or 126bp+/122bp+ and neither as previously described) for analysis. The results of the Logistic 96 Regression analysis are shown in Table 5.30. This analysis indicated that information on the typing of IFNy has a statistically significant incremental value in predicting the outcome of the disease after considering previously defined demographic and clinical parameters, Chi-square=34.48 and pO.OOOl. Table 5.30 Logistic regression analysis using demographic, clinical data and IFNy markers to predict the outcome of RA. Factors Chi-Square d.f P COT. value Age at onset 0.30 1 0.5812 (NS) Gender 4.15 1 0.0416 Duration 5.44 1 0.0197 RF 14.32 2 0.0008 IFNy 34.48 3 O.0001 P corr. Corrected P value 97 5.6.3 Combined Influence of HLA-DRpi , DQpl, DPpi and IFNy on the Odds of Severe versus Mild Disease, Adjusted for Demographic and Clinical Variables To examine if additional information on the outcome of RA could be gained by adding the typing of all HLA class II genes and IFNy, HLA- DR, DQ, DP and IFNy typing results were added to the four previously defined demographic and clinical parameters in a forward step-wise logistic regression analysis. Patients were divided into four groups according to the alleles expressed (high risk, low risk, both or neither as previously described) for each of the three HLA class II and IFNy genes for analysis. The results of the logistic regression analysis are shown in Table 5.31. This analysis indicated that information on the typing of HLA class II and IFNy has a statistically significant incremental value in predicting the outcome of the disease after considering for demographic and clinical variables, Chi-square=29.57 and p=0.0005 for the combined effect of all HLA class II genes. Furthermore, all HLA class II genes individually have a statistically significant incremental predictive value, which is shown in Table 5.31. Table 5.31 Logistic regression analysis using demographic, clinical data, all H L A class II and IFNy gene markers to predict the outcome of R A . Factors Chi-Square d.f p con-, value Age at onset 0.36 1 0.5464 (NS) Gender 8.60 1 0.0034 Duration 0.70 1 0.4011 (NS) RF 7.72 2 0.0210 IFNy 32.97 3 O.0001 HLA-DRpi ,DQpl ,DPpl 29.57 9 0.0005 HLA-DQpl,DPpl 18.19 6 0.0058 HLA-DRpi 10.59 3 0.0142 HLA-DQpi 11.57 3 0.0090 HLA-DPpl 9.65 3 0.0218 P Corr. Corrected P value 98 5.6.4 Goodness of Fit Analysis for Combined Influence of HLA-DRpi, DQP1, DPpi and IFNy on the Odds of Severe versus Mild Disease, Adjusted for Demographic and Clinical Variables To identify the accuracy of the model presented in Table 5.31 in discriminating between patients with mild and severe RA, the predicted outcome was calculated for all RA patients and compared with the known disease status (Table 5.32). Some of the clinical parameters were absent in three patients with severe RA; thus, only 84 patients had all required variables for the analysis. The overall accuracy of this model proved to be 96.4% (81/84). Ninety-five percent (37/39) of the patients predicted to have mild RA indeed had the mild form of the disease and 98% of those patients predicted to have severe disease were also correctly identified. Characteristics of the three individuals who were misclassified by this model are shown in Table 1 of the Appendix. Table 5.32 Goodness of fit analysis showing the discriminatory power of the model in section 5.6.3. Categories True Mild True Severe Patients predicted to have Mild disease 37 1 Patients predicted to have Severe disease 2 44 5.6.5 A Reduced Model Combining Only the Effects of HLA-DRpi and IFNy The combined effects of HLA-DRpi and IFNy markers on the odds of severe versus mild disease were analysed to determine whether this provides a more efficient and parsimonious model than did typing of IFNy and all three HLA class II genes. Patients were categorised into three groups according to the HLA DRpi and IFNy alleles expressed (high risk alleles or SE+/126bp+, low risk or DE+, HLA-DRpl*08+/122bp+ alleles and others). The results are shown in Table 5.33. Fifty-eight percent of patients (28/48) with severe RA were positive for both the SE and IFNy 126bp alleles, compared with 3% (1/39) with mild disease (OR=53.20; pO.OOOl) and 2% (1/50) of controls (OR=68.60; pO.OOOl). In contrast, none of the patients (0/48) with severe RA were 99 positive for the DERAA epitope or HLA-DRpl *08 and IFNy 122bp alleles compared with 13% (5/39) in the mild RA group (OR=0.065; p=0.036) and 4% (2/50) of the controls (OR=0.20; p=NS). Chi-square analysis showed that the frequencies of these three risk groups differed significantly between patients with mild and severe RA, Chi-square=32.75 and pO.OOOl. Comparison of only patients with both high risk or both low risk HLA-DRpl and IFNy alleles indicated an odds ratio of 209 fold between groups with mild and severe RA. Table 5.33 Frequencies of patients and controls expressing high risk alleles, low risk alleles, or others at H L A - D R p l and IFNy loci. Controls Severe RA MildRA ORb Pb Categories % % ORa pa % ORa pa Double positive high 2 58 68.60 O.0001 3 1.29 NS 53.20 O.0001 Double positive low 4 0 0.20 NS 13 3.52 NS 0.065 0.036 Others 94 42 0.05 O.0001 85 0.35 NS 0.13 0.0001 a compared with controls b severe compared with mild 5.6.6 Combined Influence of only H L A - D R p i and IFNy on the Odds of Severe versus Mild Disease, Adjusted for Demographic and Clinical Variables To examine the extent of information gained on the outcome of RA by using only the HLA-DRpi (instead of all three HLA-class II genes) and IFNy data, HLA-DRpi and IFNy typing results were added to the four previously defined demographic and clinical parameters in a forward step-wise logistic regression analysis. Patients were divided into four groups according to the alleles expressed (high risk, low risk, both or neither as previously described) for HLA-DRpi and IFNy genes for analysis. The results of the logistic regression analysis are shown in Tables 5.34 and 5.35. This analysis indicates that HLA-DRpiand IFNy each have a statistically significant incremental value in predicting the outcome of the disease after adjusting for demographic and clinical variables, Chi-square=l 1.83 and p=0.0098. However, a comparison of the two analyses in Tables 100 5.31 and 5.34 indicates that HLA-DQ and DP have an additional statistically significant incremental predictive value on the outcome of the disease even after adjusted for clinical data, HLA-DR, and IFNy gene polymorphisms (Chi-square=18.19; p=0.0058). Table 5.34 Logistic regression analysis using demographic, clinical data, H L A -DRB1 and IFNy gene markers to predict the outcome of R A . Factors Chi-Square d.f p corr. value Age at onset 0.58 1 0.4443 (NS) Gender 4.52 1 0.0334 Duration 3.14 1 0.0762 (NS) RF 9.23 2 0.0099 IFNy 28.03 3 <0.0001 HLA-DRpi 11.38 3 0.0098 P com Corrected P value 101 Table 5.35 Detailed results of logistic regression analysis using demographic, clinical data, H L A - D R p i and IFNy gene markers to predict the outcome of R A . Factors Low High Diff. Effect S.E. Lower 95% Upper 95% OR Age at onset 33 55 22 -0.67 0.90 -2.43 1.08 0.51 OR details 33 55 22 0.51 NA 0.09 2.95 Duration 9 18 9 1.09 0.64 -0.17 2.35 2.98 OR details 9 18 9 2.98 NA 0.85 10.49 Gender m:f 1 2 NA 2.56 1.36 -0.12 5.23 12.93 OR details 1 2 NA 12.93 NA 0.89 187.60 RF p:n 1 2 NA 3.11 1.30 0.56 5.67 22.47 OR details 1 2 NA 22.47 NA 1.74 289.65 RF unk:n 1 3 NA 0.79 1.74 -2.62 4.20 2.21 OR details 1 3 NA 2.21 NA 0.07 66.96 IFNy H:L 4 1 NA 5.52 1.73 2.14 8.91 250.57 OR details 4 1 NA 250.57 NA 8.47 7417.19 IFNy B:L 4 2 NA -7.24 24.50 -55.26 40.78 0.00 OR details 4 2 NA 0.00 NA 0.00 5.11xl017 IFNy N:L 4 3 NA 2.20 1.34 -0.43 4.83 9.04 OR details 4 3 NA 9.04 NA 0.65 125.78 HLA-DRpi H:L 4 1 NA 4.17 2.04 0.17 8.16 64.49 OR details 4 1 NA 64.49 NA 1.19 3499.00 HLA-BRpl B:L 4 2 NA -0.31 2.41 -5.03 4.40 0.73 OR details 4 2 NA 0.73 NA 0.01 81.72 HLA-DRpl N:L 4 3 NA 1.87 2.76 -3.55 7.28 6.49 OR details 4 3 NA 6.49 NA 0.03 1458.43 Controls (n = 50), patients with severe RA (n = 48), patients with mild RA (n = 39). Odds ratios (OR) and chi-square statistics are marginal (i.e. have been adjusted for all other factors). Odds ratios here reflect the distribution of patients observed rather than underlying prevalence of mild and severe RA. HLA-DRpi: H = high risk alleles, L = low risk alleles, B = both high and low risk alleles, N = neither IFN-y: H = high risk allele, L = low risk allele, B = both high and low risk alleles, N = neither. RF: Because results not available for all patients, subjects were divided into three categories consisting of those with a positive RF (p), negative RF (n) and those for whom this result was unknown (unk). Subjects with missing RF data were not significantly different from those with negative (n) results. Effect of each parameter= natural logarithm of OR m:f male versus female S.E. standard error NA not applicable 102 5.6.7 Goodness of Fit Analysis for Combined Influence of only HLA-DRpi and IFNy on the Odds of Severe versus Mild Disease, Adjusted for Demographic and Clinical Variables To identify the accuracy of the model presented in Table 5.34 in discriminating between patients with mild and severe RA, the predicted outcome was calculated for all RA patients and compared with the known disease status (Table 5.36). Some of the clinical parameters were absent in three patients with severe RA; thus, only 84 patients had all required variables for the analysis. The overall accuracy of this model proved to be 90.5% (76/84). Eighty-seven percent (34/39) of the patients predicted to have mild RA indeed had the mild form of the disease and 93% (42/45) of those patients predicted to have severe disease were also correctly identified. Characteristics of the eight individuals who were misclassified by this model are shown in Table 2 of the Appendix. Table 5.36 Goodness of fit analysis showing the discriminatory power of the model in section 5.6.6 Categories True Mild True Severe Patients predicted to have Mild disease 34 3 Patients predicted to have Severe disease 5 42 103 5.7 Confirmation of HLA-DRB1 and IFNy Results The observations were confirmed on an independent sample of 27 subjects, comprising 12 patients with severe RA and 15 controls. Figure 5.2 shows that the allele frequencies in the original subjects with severe RA differed substantially from those in both controls and patients with mild RA, and that these observations were identical when repeated in independent groups. The results of these groups were therefore pooled (control n=65, severe RA n=60 and mild RA population n=39) and reported in Table 5.37. Figure 5.2: IFNy allele frequencies in Caucasian patients with severe and mild RA compared to controls. 100 90 120bp 122bp 124bp 126bp 128bp 130bp IFNg Alleles Subjects in phase 1 are indicated by solid lines and icons: Severe R A (n=48) \u00E2\u0080\u0094 \u00E2\u0080\u00A2 \u00E2\u0080\u0094 , Controls (n=50) \u00E2\u0080\u0094 \u00E2\u0080\u00A2 \u00E2\u0080\u0094 , M i l d R A (n=39) \u00E2\u0080\u0094 \u00E2\u0080\u00A2 \u00E2\u0080\u0094 . Subjects in phase 2 are indicated by open lines and icons: Severe R A (n=12) \u00E2\u0080\u0094 O \u00E2\u0080\u0094 , Controls (n=15) \u00E2\u0080\u0094 \u00E2\u0080\u00A2 \u00E2\u0080\u0094 . 104 The proportion of patients expressing each IFNy allele in the three study groups is shown in Table 5.37. Patients with severe RA differed significantly from normals and patients with mild disease in the frequency of two of these alleles. The 126bp allele was present in 73% of patients with severe RA compared with 21% of patients with mild RA (OR: 10.66, pO.OOOl) and 12% of normal subjects (OR: 19.59, pO.OOOl). In contrast, the 122bp allele was detected in only 7% of patients with severe RA compared with 64% of patients with mild disease (OR: 0.04, pO.OOOl) and 80% of normal subjects (OR: 0.018, pO.OOOl). There was no significant difference in the frequencies of the other microsatellite polymorphisms between the three study groups. Table 5.37 IFNy microsatellite polymorphism frequencies of patients with RA and controls (pooled data). IFNy Alleles No. of CA repeats Controls Severe RA MildRA Severe vs Mild RA % % ORa p a C m . % ORa Pa ORD P Corr. 130bp 16 0 7 10.43 NS 0 - - 6.29 NS 128bp 15 12 17 1.42 NS 15 1.29 NS 1.10 NS 126bp 14 12 73 19.59 O.0001 21 1.84 NS 10.66 O.0001 124bp 13 68 73 1.31 NS 67 0.95 NS 1.37 NS 122bp 12 80 7 0.018 O.0001 64 0.45 NS 0.04 O.0001 120bp 11 0 0 - 3 5.1 NS 0.21 NS Controls (n = 65), patients with severe RA (n = 60), patients with mild RA (n = 39). a compared with controls b severe compared with mild P corr. Corrected P value To examine potential interactions between the alleles, subjects in each study group were allocated to one of four categories depending upon whether they expressed the 126bp allele, the 122bp allele, both or neither. The results are shown in Table 5.38. Almost three quarters (72%) of patients with severe RA expressed the 126bp allele without the 122bp allele compared with 10%) of patients with mild RA (OR: 22.13, pO.OOOl) and 5% of normal subjects (OR: 52.27, pO.OOOl). In contrast, only 5% of patients with severe RA expressed the 122bp allele without 105 the 126bp allele compared with 54% of patients with mild R A (OR: 0.045, pO.OOOl) and 72% of normal subjects (OR: 0.020, pO.OOOl) . The 126 and 122bp alleles were conjointly expressed by only one patients with severe disease compared with 10% of patients with mild disease (OR: 0.148, p = ns) and 8% of normal subjects (OR: 0.20, p = ns). There was no difference in the frequencies of subjects who expressed neither o f these alleles between the three study groups. Although the sample size o f the study does not enable us to determine this with confidence, it appears that homozygosity for these alleles influences the propensity to develop severe disease by comparison with heterozygosity. None of the patients with severe R A were homozygous for the 122bp allele, compared with 8% (3/39) of those with mild disease and 17% (11/65) of normal controls. In contrast, 10% (6/60) o f patients with severe R A were homozygous for the 126bp allele, compared with 3% (1/39) o f those with mild disease and 0% (0/65) o f normal controls. Table 5.38 Frequencies of patients and controls expressing high risk alleles, low risk alleles, both or neither at IFNy locus (pooled data). Controls Severe R A M i l d R A Severe vs M i l d R A Categories N (%) N (%) O R a P a N (%) O Ra P a O Rb P b 126bp allele 3 (5) 43 (72) 52.27 O.0001 4 (10) 2.36 N S 22.13 O.0001 122bp allele 47 (72) 3 (5) 0.020 O.0001 21 (54) 0.45 N S 0.04 O.0001 Both 5 (8) 1 (2) 0.20 N S 4 (10) 1.37 N S 0.15 N S Neither 10 (15) 13 (22) 1.52 N S 10 (26) 1.90 N S 0.80 N S Controls (n = 65), patients with severe R A (n = 60), patients with mild RA (n = 39). a compared with controls b severe compared with mild R A 106 5.7.1 Combined Influence of HLA-DRB1 and IFNy Pooled Data on the Odds of Severe versus Mild Disease, Adjusted for Demographic and Clinical Variables The results of the logistic regression analysis on pooled data are shown in Table 5.39 and 5.40. IFNy typing results, clinical parameters and HLA-DRpi typing were used in a forward step-wise logistic regression analysis to determine if the previously observed information revealed by the combination of these markers, on the outcome of RA, holds true. Results of logistic regression analysis indicates that by combining the clinical factor information with the typing of IFNy and HLA-DRpi, the prediction of disease outcome will be more accurate than using any of these three factors alone. The odds ratio for IFNy (high risk versus low risk alleles) when combined with HLA-DRpi and clinical factors is over 477 between severe and mild RA patients. Table 5.39 Logistic regression analysis using demographic, clinical data, H L A -D R p i and IFNy gene markers to predict the outcome of R A (pooled data). Factors Chi-Square d.f p C O T . value Age at onset 3.84 1 0.0500 Gender 6.61 1 0.0101 Duration 1.68 1 0.1949 (NS) RF 7.27 2 0.0264 IFNy 28.83 3 O.0001 HLA-DRpi 15.66 3 0.0013 P COIT . Corrected P value 107 Table 5.40 Detailed results of logistic regression analysis using demographic, clinical data, H L A - D R p i and IFNy gene markers to predict the outcome of R A (pooled data). Factors Low High Diff. Effect S.E. Lower 95% Upper 95% OR Age at onset 33 55 22 -1.52 0.85 -3.18 0.14 0.22 OR details 33 55 22 0.22 NA 0.04 1.15 Duration 9 18 9 0.92 0.73 -0.50 2.35 2.52 OR details 9 18 9 2.52 NA 0.60 10.48 Gender m:f 1 2 NA 2.78 1.26 0.30 5.25 16.08 OR details 1 2 NA 16.08 NA 1.35 191.07 RF p:n 1 2 NA 2.94 1.27 0.46 5.42 18.94 OR details 1 2 NA 18.94 NA 1.58 226.73 RF unk:n 1 3 NA 2.26 1.48 -0.64 5.16 9.56 OR details 1 3 NA 9.56 NA 0.53 173.30 IFNy H:L 4 1 NA 6.10 1.82 2.53 9.68 447.35 OR details 4 1 NA 447.35 NA 12.56 15930.00 IFNy B:L 4 2 NA -0.11 1.81 -3.66 3.44 0.90 OR details 4 2 NA 0.90 NA 0.03 31.05 IFNy N:L 4 3 NA 2.75 1.38 0.05 5.45 15.63 OR details 4 3 NA 15.63 NA 1.05 233.00 HLA-DRpi H:L 4 1 NA 4.35 2.12 0.20 8.50 77.57 OR details 4 1 NA 77.57 NA 1.22 4912.91 HLA-BRpi B:L 4 2 NA -1.28 2.30 -5.78 3.23 0.28 OR details 4 2 NA 0.28 NA 0.00 25.35 HLA-DRpi N:L 4 3 NA 1.57 2.61 -3.54 6.69 4.83 OR details 4 3 NA 4.83 NA 0.03 800.91 Controls (n = 50), patients with severe RA (n = 48), patients with mild RA (n = 39). Odds ratios (OR) and chi-square statistics are marginal (i.e. have been adjusted for all other factors). Odds ratios here reflect the distribution of patients observed rather than underlying prevalence of mild and severe RA. HLA-DR|31: H = high risk alleles, L = low risk alleles, B = both high and low risk alleles, N = neither IFNy: H = high risk allele, L = low risk allele, B = both high and low risk alleles, N = neither. RF: Because results not available for all patients, subjects were divided into three categories consisting of those with a positive RF (p), negative RF (n) and those for whom this result was unknown (unk). Subjects with missing RF data were not significantly different from those with negative (n) results. Effect of each parameter= natural logarithm of OR m:f male versus female S.E. standard error N A not applicable 108 C H A P T E R SIX DISCUSSION While the etiology of RA remains incompletely understood, its devastating consequences affect millions of people around the world. Intense research is currently focused on understanding the etiology and pathology of RA, developing new medications to reduce pain, increase the quality of life, and slow the rate of progression of the disease, and elucidating new and reliable prognostic markers to distinguish patients who will progress to the severe form of RA in the early stages of disease. The principal objective of this study was to examine the relationship between HLA and cytokine gene polymorphisms and RA to try to identify alleles related to disease severity, and also to combine these with known risk factors in order to more precisely predict outcome in this disorder. Genetic typing was employed to examine nucleic acid polymorphisms within the target genes in individuals with mild and severe disease. This offers the potential to provide a sensitive and reliable marker(s) of outcome at the early stages of the disease before joint destruction occurs. The latter is especially important due to the current lack of reliable and sensitive methods for early prediction of the outcome and severity in RA. This study deliberately selected patient populations at extremes of the RA spectrum in order to maximize the differences in the severity and outcomes of RA. A control group was used to identify the differences in the frequencies of the relevant genetic markers in RA patients from the norm. This is one of the few studies, to our knowledge, which has been designed in this manner to specifically address the role of genetic markers in the prediction of severity in RA. Most other studies which have tried to answer this question have compared a heterogeneous group of RA 109 patients to normal control subjects (Wagner, U. et al 1997). However, this design has disadvantages. The data in this study only represents patients at either end of the RA spectrum, and lacks representatives of RA patients with disease of intermediate severity. It is therefore not possible to measure the sensitivity and/or specificity of polymorphisms identified in this study across the full spectrum of clinical disease, since a continuum of RA patients was not used. In recent years, studies of genetic predisposition to complex multigenic autoimmune diseases have increasingly attempted to address the whole human genome, rather than focusing on a specific area. These types of studies are of special interest for gene hunting in multigenic autoimmune diseases including RA. Recent technological developments in the area of molecular biology have made genome wide scans of patients with complex diseases ever more popular. Genome-wide linkage studies conducted on Caucasian sib pairs with RA have shown significant association between 14 candidate regions and susceptibility to RA (Cornelis, F. et al 1998). An alternative to this popular, yet expensive and labor intensive method of gene scanning is to focus on key genes that may be involved in the progression of the disease of interest. Genetic markers within, or in proximity to, these key genes may provide a possible explanation of the effects of genetic differences, among individuals, on disease outcome. This method is not only less expensive and time consuming, but has been demonstrated to be as effective as conducting a genome wide scan if the markers are chosen carefully and are sufficiently polymorphic (Kruglyak,L 1997). The candidate genes for this study included HLA-DRpi, DQpl, DPpl and TNF (all on chromosome 6), IL-1 receptor and IL-1 receptor antagonist (on chromosome 2), IL-2 and IL-2 receptor (on chromosomes 4 and 22, respectively) and IFNy (on chromosome 12). The chromosomal locations of the candidate genes in this study are consistent with those reported in 110 the genome-wide linkage study by Cornelis et al (1998), and the products of these genes have all been shown to have a crucial effect on the severity and/or pathogenesis of RA. These genes all have highly polymorphic regions and are present on five different chromosomes. This increased the probability of success of the study in that other important genes affecting the severity of RA which were in proximity to and in linkage disequilibrium with these markers might be simultaneously detected. HLA class II DR.pi alleles have been shown to have prognostic value in RA, mostly in determining susceptibility to the disease rather than severity and progression (Nepom, GT. et al. 1992). While the implicated alleles vary depending on the ethnic background of the subjects examined, patients who carry the alleles with the SE appear to be at higher risk of developing RA than those without this cassette. However, while it has been suggested that there is a significant association between SE alleles and susceptibility to RA, this same association has not been shown consistently for severity and outcome of RA (Suarez-Almazor, ME. et al. 1995). Nepom, GT. et al. (1996) have shown that HLA class II DR.pi may have prognostic implications in patients with erosive and progressive disease when used in combination with other clinical and laboratory assessment techniques. However, they have also reported that combining the multiple alleles which express the SE into one cluster reduces the specificity of this test even though sensitivity may increase. The data from the current study confirmed this observation and showed that 88% (42/48) of patients with severe RA carried at least one of the alleles with the SE compared to 54% (21/39) of mild patients (OR=6.00, p=0.0011) and only 41% (43/105) of controls (OR=10.93, p<0.0001). The odds ratio between the severe and mild groups suggests that the chance of developing severe RA is about six times greater when an individual with RA has an allele which includes the SE. Our data indicates that, while there is a strong association between 111 RA and SE bearing alleles, there is also a hierarchy among these alleles. HLA-DRpi* 401/404 seem to be more closely associated with the severe form of RA, while HLA-DRpl*101/102 or 1001 alone do not seem to have the same effect, and their effect may depend on the second co-dominant allele for the severity of the disease. Fifty-two percent (25/48) of patients with severe RA carried HLA-DRf31*401, compared to 23% (9/39) of those with mild RA (OR=3.6, p=0.0112) and 16% (17/105) of the controls. Studies have shown seropositive patients carrying HLA-DRpi*401 are at higher risk of progressing to erosive disease than those with other subtypes of SE bearing alleles (Gough, A. et al. 1995, 1994). On the other hand, the frequency of HLA-DRpl*101/102 or 1001 was similar among all groups. Therefore, this data supports the use of individual alleles rather than SE containing allele clusters. In addition to alleles that are associated with a high risk of susceptibility to and/or severity of RA, more recent studies, including this one, have reported a second group of alleles whose frequencies are decreased in patients with severe RA (Zanelli, E. et al. 1998 and Khani-Hanjani, A. et al. 1998). These alleles, which include HLA-DRpi*0103/0402/l 102/1103/1301 and 1302, have been shown to reduce the chance of progression of the severe RA in animal models and thus appear to have a protective effect (Zanelli, E. et al. 1995). The effect of these alleles has been suggested to be due to the DERAA motif at positions 70-74 of the third hypervariable region of HLA-DRpl. Only 6% of patients with severe RA had one of the DERAA alleles compared to 31% of the mild (OR=0.15; p=0.0064) and 21% of the controls (OR=0.25; p=0.0407). From these analyses, the probability of having severe RA is six times less in individuals with one of the alleles containing the DERAA motif compared to those without. Again, the data reported here indicate that there is a hierarchy of effect. Only HLA-DRpi*1301/1302 showed a statistically significant reduction in frequency in the severe patient 112 group. Four percent of patients with severe RA were positive for HLA-DRpi*1301/1302 compared to 20% of patients with mild RA (OR=0.16; p=0.0414). Stratification of RA patients according to high risk (QKRAA/QRRAA containing) or low risk (DERAA containing or DR8) alleles produced a statistically significant difference between mild and severe RA patients with a chi-square of 16.91. Further, results of multiple logistic regression analysis show an additional predictive value for HLA-DRpi markers in disease severity, after controlling for other clinical prognostic factors (Chi-square= 17.83; p=0.0005). FILA-DRpi markers alone have the highest single incremental predictive value in disease severity, among other HLA class II genes, when incorporated to the logistic regression model. The DERAA motif includes negatively charged amino acid DE at positions 70 and 71, which could affect the binding of Ag to HLA and influence TCR recognition and binding (Wucherpfennig, KW. et al 1995). These AAs may come into direct contact with the TCR due to their physical position on the DRpi chain and so affect the T-cell repertoire and consequent immune responses (Olson, RR. et al. 1992 and Stern, LJ. et al 1994). This is especially important in the study of an autoimmune disease like RA since a number of association studies in humans have reported that the HLA genes could have a crucial effect on, and do influence, the construction of human TCR VP repertoire in RA (Williams, WV. et al 1993, Walser-Kuntz, DP. et. al 1995 and Gulawani-Akolkar, B. et al. 1995). The frequencies of AAs with different charges at positions 70 and 71 in the two patient groups are shown in Table 5.8. There is a significant increase in the frequency of patients carrying alleles with neutral amino acid at position DRpi*70 and positively charged amino acid at position DRpl*71 among the patients with severe RA compared to patients with mild disease (OR=4.08; 113 p=0.0187). In contrast, the frequency of severe patients who have at least one allele with the negatively charged AAs at these positions was significantly reduced compared to the patients with mild RA (OR=0.05; p=0.0168). This data supports the hypothesis that HLA alleles with negatively charged AAs at positions 70 and 71 might have a protective effect against the severe form of the disease in humans, consistent with the results of animal studies (Zanelli, E., et al. 1995). Differences at the HLA level affect the antigen binding and T-cell repertoire and may be one of the reasons why patients with mild RA did not progress to severe disease. It may be that the antigens responsible for the onset of RA are negatively charged and thus cannot be efficiently presented to the T-cell by HLA molecules with DERAA motif (Cerna, M . et al 1995). Some HPLC (high-pressure liquid chromatography) studies have recently shown that RA-associated HLA-DRpi molecules bind more avidly to negatively charged AAs like aspartic acid and glutamic acid and do not allow binding of positively charged AAs like arginine and lysine (Friede, T. et al. 1996). Therefore, it is probable that the amino acid charges at position 70 and/or 71 of the SE and DERRA epitopes play a crucial role in the progression and or severity of autoimmune diseases and RA. The analysis of HLA-DRpi frequencies indicated that polymorphisms at this locus alone are not sufficiently specific to accurately predict the severity of RA. Although 88% of patients with severe RA carried one of the SE alleles, over 50% of patients with mild RA also had the SE containing alleles. Further, 12% of patients with severe RA did not carry any of the SE alleles. Thus, the SE alone is unlikely to have a high predictive value in clinical settings. 114 Studies have reported that the shared epitope motif, QKRAA/QRPvAA, of HLA-DRpl is similar to certain viral proteins (EBV gpl 10) and heat shock proteins of bacteria like Escherichia coli, Lactobacilus lactis and Brucella ovis DnaJ heat shock protein (Albani, S. et al. 1996). These similarities could be one explanation for the development of an autoimmune disease like RA following a viral or bacterial infection. Upon infection with the viral or bacterial agents that carry an Ag homologous to the shared epitope, these Ags are processed by APCs and presented to T-cells. Following this, activated T-cells clonally expand to destroy foreign Ag(s). However, since the foreign Ag(s) is homologous to the DRpi SE, the clonally expanded T-cells do not recognize them as being native Ag(s) and the tolerance mechanisms breaks down due to molecular mimicry. The SE motifs are ingested in the process of recycling surface antigens, like any other surface proteins, and then processed and presented at the cell surface to the T-cells (Albani, S. et al 1996). This process initiates an immune response against self-Ag (SE) in patients with RA. Investigators have suggested that HLADQpl is responsible for the presentation of the SE cassette of the HLA-DRpl, which resembles the foreign antigens, after the onset of autoimmune diseases by the exogenous antigen(s). They also indicate that the abundant expression of DQ in the joints is one of the reasons for the localization of the inflammatory responses in the joints of RA patients (Bradley, DS. et al. 1997, Albani S. et al 1995, Nabozny GH. et al. 1996, Zanelli, E. et al. 1995, and Anderson LC. et al. 1991). Therefore, polymorphism of the third hypervariable region of the HLA-DQPI locus is an important factor in determining the type of antigen that is presented and consequently in the progression of an autoimmune disease like RA. There is a strong linkage disequilibrium between HLA-DRpi and DQpi, however, and it is important to determine the effects of each locus on the disease. Recent studies of other autoimmune diseases like IDDM and celiac disease have demonstrated independent effects of DRpi and DQPI in the progression of 115 these diseases (Johansen, BH. et al 1996, Thorsby, E. et al. 1996, Cucca, F. et al. 1995 and Erlich, HA. etal. 1993). The analysis of HLA-DQpi gene frequencies reported here showed a strong association between DQP1 and disease severity in RA. Further, there was a difference between high and low risk alleles, with DQpl*302 and DQP1*603 respectively having the most significant effects. The allele DQP1*302 was significantly increased in frequency in the severe group compared to the mild and control groups (OR=3.33; p=0.GT9 and OR=4.0; p=0.0003), respectively. These results are similar to the frequency of DRpi*401 among patients with severe and mild RA (OR=3.6; p=0.0112), yet these two loci in combination may produce more informative results. Analysis of the data also indicates that certain HLA-DQPI alleles are less frequent in the severe group and might protect against the development of severe RA. In particular, the frequency of HLA-DQP1*603 is significantly decreased in the severe group compared to the mild and control groups (OR=0.065; p=0.036 and OR=0.05; p=0.0074), respectively. However, results of multiple logistic regression analysis show that no additional information is gained by adding HLA-DQpi markers to the logistic regression model (Chi-square=7.27; p=NS). The linkage disequilibrium between HLA-DRpi and DPpl (global disequilibrium=0.35) is not as strong as that between HLA-DRpi and DQpl (global disequilibrium 0.92) due to the larger distance and the presence of a recombination hot spot between the HLA-DRpi and HLA-DPpl loci (Klitz, W. et al 1995 and Cullen, M. et al. 1997). A number of studies have reported a positive association between certain DPpl subtypes especially HLA-DPpi*0201, 0301, 0401, and the susceptibility to and clinical manifestation of RA (Carthy, D. et al. 1995, Perdriger, A. et al. 1992, Gao, X. et al 1991 and Stephens, HAF. et al. 1989). Analysis of our data also indicates an increase 116 in the frequency of HLA-DPpi*0401 allele in patients with severe RA compared to the mild group (OR=3.26; p=0.0224). The frequency of patients with HLA-DPpl*0201, 0501 and 1001 was decreased in the severe group compared to the mild group. However, none of these values reached a statistically significant value perhaps due to the small patient numbers and/or mild effect of these alleles. Further, results of multiple logistic regression analysis show an additional predictive value for HLA-DPpi markers in disease severity, after controlling for other clinical prognostic factors (Chi-square=17.28; p=0.0006). Even though the odds ratio of HLA-DPpi*0401 (OR=3.26) in patients with severe versus mild disease is similar to that of HLA-DRpl*0401 (OR=3.6) and DQpl*0302 (OR=3.33), the combination of high risk alleles in these loci yields a larger odds ratio. The frequency of patients carrying both HLA-DRpl*401 and DQpl*302 was significantly higher in severe RA patients compared to the mild (OR=4.37; p=0.0221), a slight increase in the odds ratio when combining the two loci together, by comparison with either locus alone. The small degree of change in the odds ratio is not unexpected considering the tight linkage disequilibrium between HLA-DRpi and HLA-DQpi. Similarly, the frequency of patients carrying HLA-DRpl*1301 and DQpl*603 was lower in the patients with severe RA compared to those with mild disease (OR=0.11; p=NS), although this did not reach a significant level due to the small size of the patient populations. This odds ratio was not substantially different from that of HLA-DRpl*1301 alone (OR= 0.16 versus mild disease). The frequency of patients who carry all of the high risk alleles for the three loci (HLA-DRpi*0401, DQ*0302 and DP*0401) is much higher in the severe group compared to the mild group (OR=6.87; p=0.0159). This represents almost a two-fold increase in the odds ratio compared to each individual locus, which suggests that these loci have independent and additive 117 effects, thus, when combined can predict the outcome and severity of the disease more accurately than if employed alone. In contrast, the frequency of the low risk alleles HLA-DRpl* 1301, 1302/ DQpl*0603 and DPpl*0201, 501, 1001 from the three loci was statistically significantly decreased in patients with severe RA compared to those with mild RA (OR=0.07; p=0.0364); a large decrease in the odds ratio compared with low risk DRpi or DQpi alleles when considered alone (OR=0.16, Table 5.4), or in combination (OR=0.11, Table 5.15). From the data reported here, there is an observed difference in odds ratio of over 98 fold (6.87/0.07) between the high and low risk groups of alleles among the two patient populations. Yet, it is important to be cautious when interpreting this data and odds ratio since the number of individuals in each group is very small. A larger study population is needed to confirm the accuracy of these numbers. However, this information shows an increasing trend as all three loci of HLA class II markers are combined, suggesting that the combination of these genetic markers may be used for a more accurate prediction of the progression and severity of disease. Further, results of multiple logistic regression analysis show an additional predictive value for combination of HLA-DRpl, DQ pi , DPpi markers in disease severity, after controlling for other clinical prognostic factors (Chi-square=31.08; p=0.0003). This combination yields the highest predictive value in disease severity when incorporated into the logistic regression model suggesting an incremental and additive effect for each of the HLA class II markers. TCR polymorphisms were not examined in this study for three reasons. First, there is no concrete and consistent evidence of association of a particular TCR with the severity of any of the autoimmune diseases including RA. Most studies have examined only small patient populations and results are either contradictory or uninformative. Considering the plasticity of TCR changes and an autoimmune disease like RA, with unknown etiology, these results would be inevitable. Second, selective TCR gene usage may be expected to be most evident in the inflamed joint, 118 whereas this study was confined to the peripheral blood, introducing greater heterogeneity and reducing diagnostic accuracy. Third, all patients in the severe group were receiving treatment with multiple agents including Cyclosporin A. Cyclosporin A is known to suppress the growth and clonal expansion of T-cells and its long term effect on T-cell regulation and receptor gene use in RA is not known. Considering the mechanism of action of Cyclosporin A, it was a concern that the T-cell population in these individuals might not have been representative of RA-associated clonally expanded T-cells. Variable nucleotide tandem repeat polymorphisms are important measures in the study of genetic predisposition to polygenic autoimmune diseases (Epplen, C. et al 1997, Epplen, JT. et al 1995, Housman D. 1995 and Lucassen A M . et al 1993), and could be directly or indirectly involved in the progression of disease (Foissac, A. et al 1998, 1997, D'Alfonso, S. et al 1997 and Crouan-Roy, B. et al 1996). When present in the promoter, introns or in close proximity to the genes, these repeats may produce a tertiary DNA and/or RNA structure which affects the production and expression of the gene products (Li, Q. et al.1997, Montminy, M. 1997 and Klaff, P. et al 1996). Some larger polynucleotide tandem repeats contain several consensus binding sites for transcription factors which could potentially affect gene function (Furutani, Y. et al. 1986 and Ogbourne, S. et al 1998). Indirectly, VNTR polymorphisms are excellent markers which can be used to identify genes in linkage disequilibrium to these markers (Jordan, E. et al 1996) and may be involved in the progression of RA. In this study we examined polynucleotide and dinucleotide tandem repeat polymorphisms in selected pro-inflammatory cytokine genes. These cytokines were chosen because of their pivotal role in the progression of RA and other immune diseases. Certain of these polymorphisms have also been reported to be associated with higher production of the cytokine and may potentiate autoimmune injury via this mechanism. 119 TNFa is one of the major cytokines involved in the expression of injury in RA. Reducing the availability of TNFa by use of anti-TNF monoclonal antibodies can delay the progression of RA and induce remission, at least temporarily. A point mutation of Guanine to Adenine at position -308 of the TNFa promoter has been suggested to be associated with increased production of either TNFa or IL-la . Some studies have reported an increased frequency of TNF2 in RA. Analysis of this polymorphism in the data reported here did not demonstrate any significant difference in the frequency of TNF2 in patients with severe disease by comparison with patients with mild disease or controls. This marker does not, therefore, appear to be useful in the prediction of RA severity. TNF microsatellite polymorphisms occur at 5 highly polymorphic loci (designated a-e), totaling 39 different polymorphisms, including 4 recently reported dinucleotide repeats (Khani-Hanjani, A., et al. 2000). Our data demonstrated that the frequencies of TNFa2 (OR=4.27), a7 (OR=2.05), b4 (OR=3.57) and d5 (OR=2.30) were increased in the severe group in comparison to the mild group. At the same time, the frequencies of other alleles including TNFal 1 (OR=0.35; p=0.036), b9 (OR=0.27), C l (OR=0.32) and e3 (OR=0.24) were reduced in the severe group compared to the mild. However, none of these, except for TNFal 1, reached statistical significance, thus, no attempt was made in grouping these markers. Lack of a significant p value(s) could be due to the small patient populations in this study, resulting in small cell sizes with such highly polymorphic loci. TNF genes are located on the short arm of chromosome 6 telomeric to the HLA genes. Some studies have already associated some subtypes of HLADRpi with high or low production of TNF. For instance in a study of SLE HLADRpi*04 and HLA-DRpi*02 and DQwl have been 120 associated to the high and low production of TNFa respectively (Jacob, CO. et al.1990). Ideally, it is important to relate the different alleles at the TNF locus to the established HLA class II high and low risk markers in order to determine if further information is gained or if the different frequencies of TNF alleles among the two patient populations was simply due to the linkage disequilibrium between HLA class II and TNF (D'Alfonso, S. et al 1997). However, the degree of polymorphisms at TNF and HLA class II loci results in thousands of possible combinations for extended haplotypes. This made it impossible to examine the data without having some prior information on haplotypes and vertical transmission of these alleles through a family study, which was not in the scope of this study. IL-1 is mostly produced by macrophages, keratinocytes and endothelial cells and has an overlapping biological activity with TNF. IL-1 a and IL-lp are involved in the regulation and expression of the immune response of several inflammatory diseases (Duff, GW. 1989) and the pivotal importance of IL-1 in the pathogenesis of RA is well established (Van Den Loo, AAJ. et al 1995, 1995, Van Den Berg, WBL. 1994 and Van Lent, PLEM. 1995). IL-1 receptor antagonist (IL-1RN) is a competitive inhibitor of both IL-1 a and IL-lp, which acts as a powerful endogenous anti-inflammatory substance to dampen the effect of IL-1 a and IL-1P cytokines. Studies in the literature have not demonstrated an association between these IL-1 a polymorphisms and the severity of RA (Bailly, S. et al 1995). Nor were consistent and reliable results for the reported IL- la alleles achievable in this study because of failure to amplify the alleles using the primers reported. A new set of primers was designed and amplification was attempted under a number of new conditions but the amplified products were neither specific nor consistent, and the authors of the published paper could not be contacted to provide more insight. 121 IL-1P polymorphisms were clearly defined, but the frequency of individuals with different alleles did not show an important difference between severe and mild RA patients. A difference in odds ratio of 2.53 existed in homozygous individuals carrying IL-ip allele 2 that was not cut by TaqI enzyme (TaqI-/-), yet this difference did not reach a statistically significant level. Studies on the effects of IL-ip in IDDM have demonstrated that alleles which are not cut by the TaqI are associated with high secretion of IL-1 (3 and are considered as a genetic susceptibility marker for IDDM in non-DR3 and non-DR4 individuals (Picot, F. et al 1992). This group has suggested that this marker may be used as a replacement marker for non-DR3 and non-DR.4 IDDM patients since a high concentration of IL-1 P may have a pathogenic implication in IDDM. This suggestion may be in line with our data since 50% (3/6) of patients with severe RA who did not carry any of the SE-containing alleles had at least one high secretor allele for IL-lp according to Pociot, F. et al. Yet, one should appreciate that the numbers in this data are very small and cannot be used to support or refute the suggested hypothesis. In addition, the frequency of individuals homozygous for allele 1 (Taql+/+) was also decreased in the severe group; but this difference was not significant. The frequencies of the three IL-1RN alleles did not differ significantly among the two patient populations or the control. The frequency of IL-1RN allele 2 was reduced in patients with severe RA, though an increase in the frequency of IL-1RN allele 2 has been reported to be associated with nephropathy in IDDM and lichen sclerosus, and increased severity in SLE (Blakemore, A. et al 1996, 1994 and Clay, FE. et al 1994). Data presented here did not indicate any predictive value for IL-1RN polymorphisms for the severity of RA. Interleukin-2 (IL-2) is the major autocrine growth factor produced by T-cells, and stimulates the growth and differentiation of the T-lymphocyte. IL-2 is one of the essential and early cytokines 122 that are produced and released in the early stages of an autoimmune response. It results in increased synthesis and expression of IL-2 and IL-2 receptors at the cell surface. IL-2 also stimulates the production of other T-cell derived cytokines like IFNy and lymphotoxin (LT), and facilitates antibody synthesis by human B-cells. There are conflicting results regarding the role of IL-2 in patients with RA. Some studies have reported an increased IL-2 concentration in the joints of RA patients (Buchan, G. et al. 1988), while others have not confirmed this finding (Firestein, GS. et al. 1988). This lack of consistency may be indicative of a time and stage-dependent contribution of this cytokine. More specifically, this may be due to the fact that IL-2 is essential mainly at the early stages in the immune responses. This contribution of IL-2 may be diminished or replaced by other cytokines as the disease progresses. Based on this understanding the results may be different depending on the stage of the subject's disease. Thus, the presence and increased concentration of IL-2 at the early stages of the disease is expected in the joints of RA patients. However, there is no data available on the concentration and the role of IL-2 in the rheumatic joints of RA patients at the early stages. An increased concentration of IL-2 in the joints may be associated with the genetic polymorphisms of IL-2 among individuals, yet there is no data confirming this association. The data reported here indicate that there was an increase in the frequency of the A15 (114bp) allele in patients with severe RA compared to those with mild disease (OR=2.02). In contrast, the frequency of the A8 (134bp) allele was decreased in patient with severe RA compared to those with mild disease (OR=0.08). None of these values reached statistical significance, however this may be due to the small patient population studied and may require a larger patient population for full evaluation. A new IL-2 polymorphism (122bp) was discovered in addition to the ones reported in the literature. This novel allele was seen in both mild and control populations but not in the patients 123 with severe RA. However, its low frequency did not reach a statistical significance between all three groups. The physiological effects of IL-2 on the cell occur through binding to its receptor. The IL-2 receptor consists of three different peptides (a, P and y), which together bind the IL-2 molecule and transmit the signal to the nucleus of the cell. Polymorphisms at any of the three subunits could potentially affect the affinity of IL-2 binding and, consequently, its physiological effects. Further, increased affinity or expression of this receptor may be a factor responsible for breaking peripheral tolerance in the patients with RA. IL-2RP has an intermediate affinity and is responsible for the transduction of IL-2 signals to the nucleus leading to the stimulation of resting lymphocytes (Siegel, JP. et al. 1987). The data reported here showed no statistically significant difference in the frequency of the different IL-2Rp alleles among the three populations. There was a trend towards an increase in the frequency of alleles 1 (163bp) and 2 (161bp) in the severe population compared to the mild group coupled with a decrease in the frequency of alleles 5 (155bp) and 7 (15lbp). However, the magnitude of these trends did not reach statistical significance (p<0.05). Three new IL-2RP polymorphisms (147bp, 165bp and 169bp) were discovered in addition to the ones reported in the literature. These novel alleles were seen in patients with severe and mild RA and control population. However, the low frequency of these alleles did not reach a statistical significance between the three groups. Interferon gamma is produced mostly by T-cells and natural killer cells. Its physiological effects include activation of macrophages, endothelial, neutrophils and natural killer cells, which play a crucial role in the destruction of articular joints in RA (Buchan G. et al. 1988, Firestein, GS. et al. 124 1992 and Van Den Berg, W. et al. 1996). IFNy also up-regulates the expression of HLA molecules, and promotes the differentiation of T- and B-cells. The latter effect of IFNy on T- and B-cell differentiation is of importance in the progression of severe RA (Feldmann, M . 1996 and Abbas, A. et al. 1994). IFNy directly promotes differentiation of naive CD4+ T-cells to the Thl subtype and inhibits of proliferation of Th2 cells. Increased concentration of IFNy in the joints of RA patients and its role in activating macrophages and endothelial cells has been reported (Firestein, GS. et al 1990). Activation of macrophages leads to the production of TNF and IL-1 and to the destruction of articular joints (Alvaro-Garcia, JM. et al. 1993 and Van Den Berg, WB. et el. 1996). Cytokine profdes and histological data from rheumatic joints indicate that inflammation in RA is primarily a Thl driven delayed hypersensitivity type of inflammatory response in which IFNy plays a crucial role (Miltenburg, AJ. et al. 1992 and Quayle, AJ. et al. 1993). Certain IFNy polymorphisms have been associated with other autoimmune diseases like IDDM, suggesting that these polymorphisms produce specific physiological effects. Alleles identified in this study correspond closely in length to those initially reported by Ruiz-Linares (1993) which ranged from 122-134bp. Exact correlation with other reports (Awata, T. et al. 1994) is difficult because of ethnic and methodological differences between the studies. However, it appears that 6-8 alleles encompassing at least 11-16 CA (cytosine-adenine) repeats exist at this site (Ruiz-Linares, A. 1993, Awata, T. et al. 1994 and Pravica, V., et al. 1999) including the intermediate polymorphism of 128bp which was not identified in the original report by Ruiz-Linares (1993). Certain of these alleles may be uncommon in non-Caucasians. The data reported here demonstrate that the frequency of 126bp allele is significantly increased in the severe RA group compared to the mild group (OR=10.43, pO.OOOl) and to the control group 125 (OR=19.74, p<0.0001) and there is a simultaneous significant decrease in the frequency of 122bp allele in the severe group compared to the mild group (OR=0.037, pO.OOOl) and to the control group (OR=0.019, pO.OOOl). These two alleles (122bp and 126bp) seem to have a powerful capability for predicting the severity of RA, with a superior odds ratio (10.43) to that of the combined HLA class II markers (OR=6.87). Pooled analysis of RA patients according to whether they express IFNy high risk (126bp) or low risk (122bp) alleles indicates a statistically significant difference between mild and severe patients with a chi-square of 41.66. Further, results of multiple logistic regression analysis show an additional predictive value for IFNy markers in disease severity, after controlling for other clinical prognostic factors (Chi-square=34.48; pO.OOOl). IFNy markers alone have the highest single incremental predictive value in disease severity, compared to all other genetic and clinical markers, when incorporated to the logistic regression model. Even though the discriminating power of IFNy polymorphisms is much greater than that of HLA-DRpi alone or even all HLA class II markers combined, it does not negate the polygenic pathogenesis of RA. Literature reports support the effect of HLA class II genes, especially HLA-DRpi alleles, in the development of RA. The combined effects of high and low risk alleles in the IFNy and HLA-class II were therefore examined in three study groups. Results of multiple logistic regression analysis show an additional predictive value for combination of HLA-DRpl, DQ pi, DPpl to IFNy markers in disease severity, after controlling for other clinical prognostic factors (Chi-square=32.97; pO.OOOl). This combination yields the highest predictive value, with overall accuracy of over 96%, for disease severity when incorporated into the logistic regression model suggested in Table 5.31. To verify the discriminatory power of this model using the combined effects of only HLA-DRpl and IFNy, and to reduce the cost of the testing, a second, simpler model, was investigated. 126 There was a significant difference in the proportion of patients who expressed high risk alleles only at HLA-DRpi and IFNy loci (double positive) between the mild and severe RA groups (Table 5.33) with an observed odds ratio of over 53 fold. This analysis indicates a large increase in the odds ratio (from 6.23 and 23.56 for HLA-DRpi and IFNy, respectively, to 53.20) as HLA-DRpi and IFNy markers are combined. Further, as shown in Table 5.33, only one individual with double high risk alleles was seen in the mild group of RA patients while no individuals with severe RA had low risk alleles at both IFNy and HLA class II loci. Results of multiple logistic regression analysis also shows a tight correlation with disease severity when HLA-DRpi, and IFNy alleles are combined, after controlling for other clinical prognostic factors (Chi-square=28.03; pO.OOOl; OR=250.57). This combination provides an overall predictive accuracy of greater than 90% when incorporated into the reduced logistic regression model suggested in Table 5.34. A combination of HLA-DRpiand IFNy alleles therefore appears to offer a rapid, economic, simple and potentially important method for prediction of disease severity in RA, although the overall predictive accuracy is slightly less than when all three HLA class II loci are included (Table 5.31). The accuracy of this more parsimonious model was confirmed using an extended patient sample. However, the sensitivity, specificity, and predictive value of this model must now be tested in a larger clinical population. 127 CHAPTER SEVEN SUMMARY This study was designed to examine the polymorphisms within specific immune system genes which are know or implied to be associated with autoimmune disorders including RA. Unlike most previous reports, the study was designed primarily to address the question of disease severity rather than susceptibility. For this reason, it examined two groups of patients with mild or severe disease who were at opposite poles of the disease spectrum. The data reported here are consistent with the initial research hypothesis proposed, and show that two gene loci appear to be linked with the progression of severe RA. The data suggests that the combination of HLA class II and IFNy markers offers an accurate prediction of disease course in RA, and that polymorphisms at these loci may explain the differential predisposition among the two patient populations studied. These polymorphisms may therefore provide a simple, rapid and cost-effective tool for the prediction of the severity of RA. Further, because these polymorphisms do not change over time, they may be used at an early stage of the disease course in order to individualize therapy and to minimize the articular damage. Despite these encouraging findings, the study has a number of unavoidable limitations. First, the number of patients in both the mild and severe groups was relatively small in relation to the number of highly polymorphic loci examined. However, given the exploratory and hypothesis-generating nature of the investigation, practical and economic imperatives restricted the ability to examine the frequency of all these genetic markers in a substantially larger population of patients. Second, since this study did not include patients with intermediate severity, the sensitivity and 128 specificity of the genetic polymorphisms identified can not be comprehensively examined. Resolution of these issues will require the development of larger studies which include a greater number of patients representing the complete spectrum of disease severity. 129 REFERENCES 1. Abbas, AK., Lichtman, AH. and Pober, JS. Cellular and Molecular Immunology. 2nd ed Philadelphia: W. B. Saunders; 1994. 2. Albani, S. and Carson, DA. A multistep molecular mimicry hypothesis for the pathogenesis of rheumatoid arthritis. Immunology Today; 1996: 17(10):466-470 3. Albani, S. Keystone, E C , Nelson, JL., Oilier, WE., La Cava, A., Montemayor, A C , and Carson, DA. Positive selection in autoimmunity: abnormal immune responses to a bacterial dnaJ antigenic determinant in patients with early rheumatoid arthritis. Nature Medicine; 1995: l(5):448-452. 4. Adebajo, AO., and Hazelman, BL. IgG glycosylation in association with tropical infections and rheumatoid arthritis in the tropics. Clinical and Experimental Rheumatology; 1995: 13(6):737-740. 5. Albani, S., Carson, DA., and Roudier, J. Genetic and environmental factors in the immune pathogenesis of rheumatoid arthritis. Rheumatic Diseases Clinics of North America; 1992: 18(4):729-740. 6. Al-Daccak, R., Wang, FQ., Theophille, D., Lethielluex, P., Colombani, J., and Loiseau, P. Gene polymorphism of HLA-DPB1 and DPA1 loci in Caucasoid population: frequencies and DPB1-DPA1 associations. Human Immunology; 1991: 31:277-285. 7. Almagaro, JC, Vargas-Madrazo, E., Lara-Ochoa, F., and Horjales, E. Molecular modeling of a T-cell receptor bound to a major histocompatibility complex molecule: Implications for T-cell recognition. Protein Science; 1995: 4:1708-1717. 8. Alvaro-Garcia, JM., Yu, C , Zvaifler, NJ., and Firestein, GS. Mutual antagonism between interferon-gamma and tumor necrosis factor-alpha on fibroblast-like synoviocytes: Paradoxical induction of IFN-gamma and TNF-alpha receptor expression. Journal of Clinical Immunology; 1993: 13(3):212-218. 9. Anderson, L C , Beaty, JS., Nettles, JW., Seyfried, CE., Nepom, GT., and Nepom, BS. Allelic polymorphism in transcriptional regulatory regions of HLA-DQB genes. Journal of Experimental Medicine; 1991: 173:181-192. 10. Anisman, H., Baines, MG., Berczi, I., Bernstien, CN., Blennerhassett, MG., Gorzynski, RM., Greenberg, AH., Kisil, FT., Mathisn, RD., Nagy, E., Nance, DM., Perdue, MH., Pamerantz, DK., Sabbadini, ER., Stanisz, A., and Warrington, RJ. Neuroimmune mechanisms in health and disease:2. Disease. Canadian Medical Association Journal; 1996: 155(8): 10075-1082. 11. Arnett, F C , Edworthy, S., Bloch, DA., McShane, DJ., Fries, JF., Cooper, NS., Healey, LA., Kaplan, SR., Liang, MH., and Luthra, HS. The American Rheumatism Association 1987 revised criteria for the classification of rheumatoid arthritis. Arthritis and Rheumatism; 1988: 31:315-324. 130 12. Aris-Brosou, S., and Excoffiert, L. The impact of population expansion arid mutation rate heterogeneity on DNA sequence polymorphism. Molecular Biology and Evolution; 1996: 123(3):494-504. 13. Awata, T., Matsumoto, C , Urakami, T., Hagura, R., Amemiya, S., and Kanazawa, Y. Association of polymorphism in the interferon y gene with IDDM. Diabetologia; 1994: 37:1159-1162. 14. Badner, JA., Gershon, ES., and Goldin, LR. Optimal ascertainment strategies to detect linkage to common disease alleles. American Journal of Human Genetics; 1998: 63:880-888. 15. Bailly, S., Israel, N , Fay, M., Gougerot-Pocidalo, M., and Duff, GW. An intronic polymorphic repeat sequence modulates interleukin-1 alpha gene regulation. Molecular Immunology; 1996: 33 (11/12):99-1006. 16. Bailly, S., Hayem, G., Fay, M., Kahn, MF., and Gougerot-Pocidalo, M . A. Absence of correlation between IL-1 alpha intron 6 polymorphism and rheumatoid arthritis. British Journal of Rheumatology; 1995: 34(12): 1123-6. 17. Bailly, S., Di Giovine, FS., Blakemore, AIF., and Duff, GW. Genetic polymorphism of human interleukin-1 a. European Journal of Immunology; 1993: 23:1240-1245. 18. Bailly, S., Di Giovine, FS., and Duff, GW. Polymorphic tandem repeat region in interleukin-l a intron 6. Human Genetics; 1993: 91:85-86. 19. Balducci-Silano, PL., and Layrisse, ZE. HLA-DP and susceptibility to insulin-dependent diabetes mellitus in an ethnically mixed population. Associations with other HLA-alleles. Journal of Autoimmunity; 1995: 8:425-437. 20. Banchereau, J., and Steinman, RM. Dendritic cells and the control of immunity. Nature; 1998: 392(19):245-252. 21. Barron, K.S., Silverman, ED., Gonzales, J., and Reveille, JD. Clinical, serological, and immunogenetic studies in childhood-onset systemic lupus erythematosus. Arthritis and Rheumatism; 1993: 36:348-354. 22. Baxter, AG. Immunogenetics and the cause of autoimmune disease. Autoimmunity; 1997: 25:177-189. 23. Beck, S., Geraghty, D., Inoko, H., and Rowen L. Complete sequence and gene map of a human major histocompatibility complex. Nature; 1999: 401(7656):921-923 24. Bellamy, N., Duffy, D., Martin, N., and Mathews, J. Rheumatoid arthritis in twins: a study of aetiopathogenesis based on the Australian twin registry. Annals of the Rheumatic Diseases; 1992:51:588-593. 25. Bioque, G., Crusius, JBA., Koutroubakis, I., Bouma, G., Kostense, PJ, Meuwissen, SGM., and Pena, AS. Allelic polymorphism in IL-1 p and IL-1 receptor antagonist (IL-IRa) genes in inflammatory bowel disease. Clinical and Experimental Immunology; 1995: 102:379-383. 131 26. Blakemore, AL, Gonzales, A M , Maskil, JK, Hughes, M E , Wilson, R M , Ward, JD, and Duff, GW. Interleukin-1 receptor antagonist allele (IL-1RN*2) associated with nephropathy in diabetes mellitus in human. Genetics; 1996: 97(3):369-374. 27. Blakemore, A I , Watson, PF, Weetman, A P , and Duff, GW. Association of Graves' disease with an allele of the interleukin-1 receptor antagonist gene. Journal of Clinical Endocrinology and Metabolism; 1995: 80(1): 111-115. 28. Blakemore, A I , Tarlow, JK, Cork, M J , Gordon, C , Emery, P, and Duff, GW. Interleukin-1 receptor antagonist gene polymorphism as a disease severity factor in systemic lupus erythromatosus. Arthritis and Rheumatism; 1994: 3 7(9): 1380-1385. 29. Bodmer, H , Viville, S, Benoist, B , and Mathis, D. Diversity of endogenous epitopes bound to MHC class II molecules limited by invariant chain. Science; 1995: 263:1284-1289. 30. Bonnefoy, J-Y, Plater-Zyberk, P , Lecoanet-Henchoz, S, Gauchat, J-F, Aubryu, J-P, and Graver, P. A new role for CD23 in inflammation. Immunology Today; 1996: 17(9):418-420. 31. Bouma, G , Xia, B , Crusius, B A , Bioque, G , Koutroubakis, I, Von Blomberg, B M E , Meuwissen, SGM, and Pen, AS. Distribution of four polymorphisms in the tumour necrosis factor (TNF) genes in patients with inflammatory bowel disease (IBD). Clinical and Experimental Immunology; 1996: 103:391-396. 32. Bouma, G , Crusius, JB, Oudkerk Pool, M , Kolkman, JJ, von Blomberg, B M , Kostense, PJ, Giphart, M J , Schreuder, G M , Meuwissen, SG, and Pena AS. Secretion of tumor necrosis factor alpha and lymphotoxin alpha in relation to polymorphisms in the TNF genes and HLA-DR alleles. Relevance for inflammatory bowel disease. Scandinavian Journal of Immunology; 1996: 43(4):456-463. 33. Bouma, G , Crusius, JB, Oudkerk Pool, M , Kolkman, JJ, von Blomberg, B M , Kostense, PJ, Giphart, M J , Schreuder, G M , Meuwissen, SG, and Pena AS. Haplotype restricted differences in in vitro production of tumor necrosis factor alpha and lymphotoxin alpha in patients with inflammatory bowel disease and healthy controls. Gastroenterology; 1995: 108:A787. 34. Bowness, P. and Bell, J. T-cell receptors and rheumatic disease: approaches to repertoire analysis. British Journal of Rheumatology; 1992: 31:3-8. 35. Bradley, DS, Nabozny, G H , Chang, S, Zhou, P, Griffiths, M M , Lutlira, HS,. And David, CS. HLA-DQbl polymorphism determines incidence, onset, and severity of collagen-induced arthritis in transgenic mice. Implications in human rheumatoid arthritis. Journal of Clinical Investigations; 1997: 100(9):2227-2234. 36. Braverman, J M , Hudson, RR, Kaplan, N L , Langley, C H , and Stephan, W. The hitchhiking effect on the site frequency spectrum of DNA polymorphisms. Genetics; 1995: 140:783-796. 37. Breedveld, FC. New insights in the pathogenesis of rheumatoid arthritis. Journal of Rheumatology; 1998: 25(53):3-7. 38. Brennan, F M , Maini, R N , and Feldmann, M. Role of pro-inflammatory cytokines in rheumatoid arthritis. Springer Seminars in Immunopathology; 1998: 20:133-147. 132 39. Brennan, FM., Chantry, D., Jackson, M , Maini, RN., and Feldmann, M . Cytokine production in culture by cells isolated from the synovial membrane. Journal of Autoimmunology; 1989:2:177-186. 40. Brennan, FM., Chantry, D., Jackson, A., Maini, R., and Feldmann, M. Inhibitory effect of TNFa antibodies on synovial cell interleukin-1 production in rheumatoid arthritis. Lancet; 1989: 2:244-247. 41. Bresnihan, B. Pathogenesis of joint damage in rheumatoid arthritis. Journal of Rheumatology; 1999: 26:717-719. 42. Brewster, ES., Brennan, MB., and Vissing, H. Dinucleotide repeat polymorphism in the IL-2Rp gene. Nucleic Acids Research; 1991: 19:4022. 43. Brinkman, BMN., Zuijdgeest, D., Kaijzel, EL., Breedveld, F C , and Verweij, CL. Relevance of the tumor necrosis factor alpha (TNFa) -308 promoter polymorphism in TNFa gene regulation. Journal of Inflammation; 1996: 46:32-41. 44. Brinkman, BMN., Giphart, MJ., Verhoef, A., Kaijzel, EL., Naipal, AMIH., Daha, MR., Breedveld, F C , and Verweiz, CL. Tumor necrosis factor a-308 gene variants in relation to major histocompatibility complex alleles and Felty's Syndrome. Human Immunology; 1994: 41:259-266. 45. Brooks, PM., and O'Day, RO. Nonsteroidal anti-inflammatory drugs: Differences and similarities. New England Journal of Medicine; 1991: 324:1716-1725. 46. Buchan, G., Barrett, K., Fujita, T., Taniguchi, T., Maini, R., and Feldmann, M . Detection of activated T-cell products in the rheumatoid joint using cDNA probes to interleukin (IL-2) receptor and IFN-y. Clinical Experimental Immunology; 1988: 71:295-301. 47. Butler, D., Maini, RN., Feldmann, M., and Brennan, FM. Blockade of TNFa with chimeric anti TNFa monoclonal antibody, cA2 reduces (1L-6 and IL-8) release in RA NMC cultures: A comparison with IL-1RN. European Cytokine Network;\995: 6:225-230. 48. Calahan, LF., and Pincus, T. Formal education level as a significant marker of clinical status in rheumatoid arthritis. Arthritis and Rheumatism; 1988: 31(11): 1346-1357. 49. Callahan, L. The burden of rheumatoid arthritis: Facts and figures. Journal of Rheumatology; 1998: 25(53):8-12. 50. Cantagrel, A., navaux, F., Loubet-lescoulie, P., Nourhashemi, F., Enault, G., Abbal, M. , Constantin, A., Laroche, M. , and Mazieres, B. Interleukin-1 p, interleukin-1 receptor antagonist, interleukin-4, and interleukin-10 gene polymorphisms. Arthritis and Rheumatism; 1999: 42(6): 1093-1100. 51. Caramaschi, P., Zeminian, S., Carletto, A., Biasi, D., Marino, A., and Bambara, L M . Parvovirus B19 infection and rheumatic diseases. Revue Du Rhumatisme, English Edition; 1996: 63(ll):846-853. 133 52. Carthy, D , Macgregor, A , Awomoi, A , Rigby, A S , Thomson, W, Donn, R , Silman, A , and Oilier, W. HLADPB1*0201 allele is associated with particular clinical features of rheumatoid arthritis. Review of Rheumatology; 1995: 62(3): 163-168. 53. Cash, J M , and Klippel, JH. Second-line drug therapy for rheumatoid arthritis. New England Journal of Medicine; 1994: 330(19):1368-1375. 54. Cats, A , and Hazevoet, M . Significane of positive test for rheumatoid factor in the prognosis of rheumatoid arthritis. Annals of the Rheumatic Diseases; 1970: 29:254-260. 55. Cerna, M , Havelka, S, and Ivakova, E. Role of hydrophobic amino acids at position 74 of DRpT chain in rheumatoid arthritis. Archivum Immunologiae et Therapiae Expermentalis; 1995: 43(2): 139-144 56. Chan, K A , Felson, DT, Yood, R A , and Walker, AM. The lag time between onset of symptoms and diagnosis of rheumatoid arthritis. Arthritis and Rheumatism; 1994: 37(6): 814-820. 57. Chaouni, I, Radal, M , Simony-Lafontaine, J , Combe, B , Sany, J and Reme, T. Distribution of T-cell receptor-bearing lymphocytes in the synovial membrane from patients with rheumatoid arthritis. Journal of Autoimmunity; 1990: 3(6):737-745. 58. Chapman, A , Stewart, SJ, Nepom, GT, Green, WF, Crowe, D , Thomas, JW, and Miller, GG. CD1 lb+CD28-CD4+ human T-cells activation requirement and association with HLA-DR alleles. The Journal of Immunology; 1996: 157:4771-4780. 59. Chernajovsky, Y , Feldmann, M , Maini, RN. Gene therapy of rheumatoid arthritis via cytokine regulation: future perspectives. British Medical Bulletin; 1995: 51(2):503-516. 60. Choy, EHS, and Scott, DL. Drug treatment of rheumatic diseases in the 1990s. Drugs; 1997: 53(3): 337-348. 61. Clarke, B D , Collins, K L , Gandy, M S , Webb, A C , Auron, PE. Genomic sequence from human prointerleukin 1 beta: possible evolution from a reverse transcribed prointerleukin 1 alpha gene. Nucleic Acids Research; 1986: 14(20):7897-7915. 62. Clay, F E , Cork, M J , Tarlow, JK, Blakemore, AIF, Harrington, CI, Lewis, F , and Duff, GW. Interleukin 1 receptor antagonist gene polymorphism association with lichen sclerosus. Human Genetics; 1994: 94:407-410. 63. Colonna, M , Bresnahan, M , Bahram, S, Strominger, JL , and Spies, T. Allelic variants of the human putative peptide transporter involved in antigen processing. Proceedings of the National Academy of Sciences of the United States of America; 1992: 89:3932-3936. 64. Combe, B , Eliaou, J-F, Daures, J-P, Meyer, O, Clot, J,. and Sany, J. Prognositc factors in rheumatoid arthritis. Comparative study of two subsets of patients according to severity of articular damage. British Journal of Rheumatology; 1995: 34:529-534. 65. Conaghan, PG, Brooks, P. Disease-modifying antirheumatic drugs, including methotrexate, gold, sulfasalazine, antimalarials, and D-penicillamine. Current Opinion in Rheumatology; 1996: 8(3): 176-182. 134 66. Constant, SL, and Bottomly, K. Induction of Thl and Th2 CD4+ T-cell responses: The alternative approaches. Annual Review of Immunology; 1997: 15:297-322. 67. Cooper, S M , Roessner, K D , Naito-Hoopes, M , Howard, D B , Gaur, L K , and Budd, RC. Increased usage of Vb2 and Vb6 in rheumatoid synovial fluid T-cells. Arthritis and Rheumatism; 1994:37:1627-1636. 68. Cornelis, F , Faure, S, Martinez, M , Prud'homme, JF, Fritz, P , Dib, C , Alves, H , Barrera, P , De Vries, N , Balsa, A , Pascual-Salcedo, D , Maenut, K , Westhovens, R , Migliorini, P , Tran, T H , Delaye, A , Prince, N , Lefevre, C , Thomas, G , Poirier, M , Soubigou, S, Alibert, O, Lasbleiz, S, Fouix, S, Bouchier, C , Liote, F , Loste, M N , Lepage, V , Charron, D , Gyapay, G , Lopes-Vaz, A , Kuntz, D , Bardin, T, and Weissenbach, J. New susceptibility locus for rheumatoid arthritis suggested by a genome-wide linkage study. Proceedings of the National Academy of Sciences of the United States of America; 1998: 95:10746-10750. 69. Croun-Roy, B , Bouzekri, N , Carcassi, C , Clayto, J , and Contu, L. Strong association between microsatellites and an HLAB, DR haplotype (B18-DR3): Implication for microsatellite evolution. Immunogenetics; 1996: 43(5):255-260. 70. Cucca, F , Lampis, R , Fran, F , Macis, D , Angius, E , Masile, P , Chessa, M , Frongia, P , Silvetti, M , and Cao, A. The distribution of DR4 haplotypes in Sardinia suggest a primary association of type 1 diabetes with DRbl and DQbl loci. Human Immunology; 1995: 43(3):301-308. 71. Cullen, M , Noble, J , Erlich, H , Thorpe, K , Beck, S, Klitz, W, Trowsdale, J , and Carrington, M . Characterization of recombination in the HLA class II region. American Journal of Human Genetics; 1997: 60:397-407. 72. Cush, J , and Lipsky, PE. Cellular basis for rheumatoid inflammation. Clinical Orthopedics; 1991:265:9-22. 73. D'Alfonso, S, Cappello, N , Carcassi, C , Fasano, M E , and Momigliano-Richiardi, P. Microsatellites in the HLA central region are markers of HLA conserved extended haplotypes. In: Charron D , ed. Genetic Diversity of HLA: functional and medical implication, vol. II. Sevres: EDK, 1997: 125-127. 74. D'Alfonso, S, Pociot, F , Berrino, M , Momigliano Richiardi, P. Association analysis of a new intragenic TNFA polymorphism with TNFa quantitative production. In: Charron D , ed. Genetic Diversity of HLA: functional and medical implication, vol. II. Sevres: EDK, 1997: 230-232. 75. Danis, V A , Millington, M , Hyland, V , Lawford, R , Huang, Q, and Grennan, D. Increased frequency of the uncommon allele of a tumour necrosis factor alpha gene polymorphism in rheumatoid arthritis and systemic lupus erythematosus. Disease Markers; 1994: 12:127-133. 76. Davies, JM. Molecular mimicry: Can epitope mimicry induce autoimmune disease? Immunology and Cell Biology; 1997: 75:113-126. 77. De Carli, M , D'Elios, M M , Zancuoghi, G , Romagnani, S, and Del Prete, G. Human Thl and Th2 cells: Functional properties, regulation of development and role in autoimmunity. Autoimmunity; 1994: 18:301-308. 135 78. De Giovine, FS., Takhsh, E., Blakemore, AIF., and Duff, GW. Single base polymorphism at -511 in the human interleukin-ip gene (ILIP). Human Molecular Genetics; 1992: 1(6):450. 79. Deighton, CM. What is the future for the genetics of rheumatoid arthritis? British Journal of Rheumatology; 1993: 32(110):857-585. 80. Del Prete, G., De Carli, M. , Almerigogna, F., Guidizi, MG., Biagiotti, R., and Romagnani, S. Human IL-10 is produced by both type 1 helper (Thl) and type 2 helper (Th2) T-cell clones and inhibits their antigen-specific proliferation and cytokine production. Journal of Immunology; 1993: 150:353-360. 81. Del Rincon, I., and Escalante, A. HLA-DRB1 alleles associated with susceptibility or resistance to rheumatoid arthritis, articular deformities, and disability in Mexican Americans. Arthritis and Rheumatism; 1999: 42(7): 1329-1338. 82. Dinarello, CA. Biology of interleukin-1 and interleukin-1 antagonism. Blood; 1991: 77:1627-1652. 83. Drevlow, BE., Lovis, R., Haag, MA., Sinacore, JM., Jacobs, C , Blosche, C , Landay, A., Moreland, LW., and Pope, RM. Recombinant human interleukin-1 receptor type 1 in the treatment of patients with active rheumatoid arthritis. Arthritis and Rheumatism; 1996: 39:257-265. 84. Dudler, J., and Kai-Lik, A. T cells and related cytokines. Current Opinion in Rheumatology; 1998: 10:207-211. 85. Duff, GW. Peptide regulatory factors in non-malignant disease. Lancet; 1989: 1:1432-1435. 86. Eberhardt, K., Fex, E., Johnson, U., and Wolhelm, FA. Associations of HLA-DRB and DQB genes with two and five year outcome in rheumatoid arthritis. Annals of the Rheumatic Diseases; 1996: 55:34-39. 87. Elliott, MJ., Maini, RN., Feldmann, M., Kalden, JR., Antoni, C , Smolen, JS., Leeb, B., Breedweld, F C , MacFarlane, JD., and Bijl, H. Randomized double-blind comparison of chimeric monoclonal antibody to tumor necrosis factor a (cA2) versus placebo in rheumatoid arthritis. Lancet; 1994: 344(8930):1105-1110. 88. Epplen, C , Santos, EJ., Maueler, W., Van Helden, P., and Epplen, JT. On simple repetitive DNA sequences and complex diseases. Electrophoresis; 1997: 18(9): 1577-1585. 89. Epplen, JT., Buitkamp, J., Bocker, T., and Epplen, C. Indirect gene diagnosis for complex (multifactorial)-diseases. Gene; 1995: 159(l):49-55. 90. Epplen, C , Frank, G., Gomolka, M., Nurnberg, P,. and Epplen, JT. Dinuceoltide repeat polymorphism in the IL2 and IL5RA genes. Human Molecular Genetics; 1994: 3(4):679. 91. Erlich, HA., Zeidler, A., Chang, J., Shaw, S., Raffel, LJ., Klitz, W., Beshkov, Y., Costin, G., Pressman, S., and Bugavan, T. HLA class II alleles and susceptibility and resistance to insulin dependent diabetes mellitus in Mexican American families. Nature Genetics; 1993: 3(4):358-364. 136 92. Eskadale, J., Kube, D., and Gallagher, G. A second polymorphic dinucleotide repeat in the 5' flanking region of the human IL10 gene. Immunogenetics; 1996: 45:82-83. 93. Eskdale, J., and Gallagher, G. A polymorphic dinucleotide repeat in the human IL-10 promoter. Immunogenetics; 1995: 42:444-445. 94. Evans, CH., and Robbins, PD. Pathways to gene therapy in rheumatoid arthritis. Current Opinion in Rheumatology; 1996: 8:(93):230-234. 95. Feldmann, M., Brennan, FM., and Maini, RN. Role of cytokines in rheumatoid arthritis. Annual Review of Immunology; 1996: 14:397-440. 96. Feldmann, M., June, CH., McMichael, A., Maini, R., Simpson, E., and Woody, JN. T-cell targeted immunotherapy. Immunology Today; 1992: 13(3):84-85. 97. Felson, DT., Anderson, JJ., and Meenan, RF. Use of short-term efficacy/toxicity tradeoffs to select second-line drugs in rheumatoid arthritis. Arthritis and Rheumatism; 1992: 35(10): 1117-1125. 98. Felson, DT., Anderson, JJ., and Meenan, RF. The comparative efficacy and toxicity of second line drugs in rheumatoid arthritis: Results of two metaanalyses. Arthritis and Rheumatism;\990: 33(10):1449-1461. 99. Fenske, B., Lemieux, J., Hoar, D., Fitzpatrick, J., and Keown, P. A novel DRB1*03 allele identified by polymerase chain reaction-restriction fragment length polymorphism. Human Immunology; 1996: 46(l):55-57. 100. Fiorentino, DF., Bond, MW., and Mosmann, Tr. Two types of mouse T helper cell. IV. Th2 clones secrete a factor that inhibits cytokine production by Thl clones. Journal of Experimental Medicine; 1989: 170:2081-2095. 101. Firestein GS., Boyle, DL., YU, C , Paine, M M , Whisenand, TD, Zvaifler, NJ and Arend WP. Synovial interleukin-1 receptor antagonists and interleukin-1 balance in rheumatoid arthritis. Arthritis and Rheumatism; 1994: 37:644-652. 102. Firestein, GS, Berger, A E , Tracey, DE , Chosay, JG, Chapman, D L , Paine, M M , Yu, C. and Zvaifler, NJ. IL-1 receptor antagonist protein production and gene expression in rheumatoid arthritis and osteoarthritis synovium. Journal of Immunology; 1992: 149:1054-62. 103. Firestein, GS. Mechanisms of tissue destruction and cellular activation in rheumatoid arthritis. Current Opinion in Rheumatology; 1992: 4(3):348-354. 104. Firestein, GS, Alvaro-Garcia, J M , and Maki, R. Quantitative analysis of cytokine gene expression in rheumatoid arthritis. Journal of Immunology; 1990: 144:3347-3353. 105. Firestein, GS, and Zvaifler, NJ. How important are T-cells in chronic arthritis? Arthritis and Rheumatism; 1990: 33:768-773. 106. Firestein, GS, Xu, W-D, Townsend, K , Broide, D , Alvaro-Gracia,J, Glasebrook, A , and Zvaifler, NJ. Cytokines in chronic inflammatory arthritis. I. Failure to detect T-cell lymphokines (interleukin 2 and interleukin 3) and presence of macrophage colony-stimulating factor (CSF-1) 137 and a novel mast cell growth factor in rheumatoid synovitis. Journal of Experimental Medicine; 1988: 168(5):1573-1586. 107. Fitzgerald, JE, Ricalton, NS, Meyer, A-C , West, SG, Kaplan, H , Behrendt, C , and Kotzin, LB. Analysis of clonal CD8+ T-cell expansions in normal individuals and patients with rheumatoid arthritis. The Journal of Immunology; 1995: 154:3538-3547. 108. Foissac, A , and Cambon-Thomson, A. Microsatellites in the HLA region: 1998 update. Tissue Antigens; 1998: 52(4):318-352. 109. Foissac, A , Couram-Roy, B , Faure, S, Thomson, M , and Cambon-Thomson, A. Microsatellies in the HLA region: An overview. Tissue Antigens; 1997: 49(2/3): 197-214. 110. Fox, DA. Biological therapies: A novel approach to the treatment of autoimmune disease. The American Journal of Medicine; 1995: 99:82-88. 111. Friede, T, Gnau, V , Jung, G , Keilholz, W , Stevanovic, S, and Rammensee, HG. Natural ligand motifs of closely related HLA-DR4 molecules predict features of rheumatoid arthritis associated peptides. Biochimica et Biophysica Acta 1316; 1996: 85-101. 112. Fugger, L , Morling, N , Ryder, L P , Odum, N,. and Svejgaard, A. Technical aspects of typing for HLA-DP alleles using allele-specific DNA in vitro amplification and sequence-specific oligonucleotide probes. Journal of Immunological Methods; 1990: 129(2): 175-185. 113. Fujita, T, Takaoka, C , Matsui, H , and Tanuguchi, T. Structure of the human interleukin 2 gene. Proceedings of the National Academy of Sciences of the United States of America.; 1983: 80:7437-7441. 114. Fukumori, Y , Nagao, N , Ohnoki, S, Shibata, H , Okubo, Y , and Yamaguchi, H. A PCR-RFLP method with allele-specific primers for HLA-DPB1 genotyping. In Proceedings of the Eleventh International Histocompatibility Worshop and Conference held in Yokohama, Japan. 1991. Eds. Tsujik, Aizawa, M.Sasazuki, T. Oxford University Press, 1992:331-333. 115. Furutani, Y , Notake, M , Fukui, T, Ohue, M,. Nomura, H , Yamada, M , and Nakamura, S. Complete nucleotide sequence of the gene for human interleukin 1 alpha. Nucleic Acids Research; 1986: 14(8):3167-3179. 116. Ganguli, R , Brar, JS, and Robin, BS. Immune abnormalities in schizophrenia: Evidence for autoimmune hypothesis. Harvard Review of Psychiatry; 1994: 2(2):70-83. 117. Ganguli, R , and Robin, BS. Increased serum interleukin-2 receptor concentration in schizophrenic and brain-damaged subjects. Archives of General Psychiatry; 1989: 46(3):292. 118. Gao, X , Fernandez-Vina M , Olsen, NJ , Pincus, T. and Stastny, P. HLA-DPB1 0301 is a major risk factor for rheumatoid factor-negative adult rheumatoid arthritis. Arthritis and Rheumatism; 1991: 34(10): 1310-1312. 119. Gao, X , Gazit, E , Livneh, A , and Stastny, P. Rheumatoid arthritis in Israeli Jews: Shared sequences in the third hypervarible region of DRB1 alleles are associated with susceptibility. Journal of Rheumatology; 1991: 18:801-803. 138 120. Gnarra, JR, Otani, H., Wang, M G , McBride, W, Sharon, M , and Leonard, WJ. Human interleukin 2 receptor b-chain gene: Chromosomal localization and identification of 5' regulatory sequences. Proceeds of National Academy ofSciences;l990: 87:3440-3444. 121. Gorevic, PD, Kassab, HJ, Levo, N , Kohn, R , Meltzer, M , Prose, P, and Franklin, EC. Mixed cryoglobulinemia: clinical aspects and long-term follow-up of 40 patients. American Journal of Medicine; 1980: 69:287-308. 122. Gorham, JD, Guler, M L , and Murphy, K M . Genetic control of interleukin 12 responsiveness: implications for disease pathogenesis. Journal of Molecular Medicine; 1997: 75:502-511. 123. Goronzy, JJ. and Weyand, CM. T and B-cell-dependent pathways in rheumatoid arthritis. Current Opinion in Rheumatology; 1995: 7:214-221. 124. Goronzy, JJ, and Weyand, CM. Interplay of T-lymphocytes and HLA-DR molecules in rheumatoid arthritis. Current Opinion in Rheumatology; 1993: 5(2):169-177. 125. Gough, A , Faint, J , Salmon, M , Hassell, A , Wordsworth, P , Pilling, D , Birely, A , and Emery, P. Genetic typing of patients with inflammatory arthritis at presentation can be used to predict outcome. Arthritis and Rheumatism; 1995: 3 8(6): 873-874. 126. Gough, A , Faint, J , Salmon, M , Hassell, A , Wordsworth, P , Pilling, D , Birely, A , and Emery, P. Genetic typing of patients with inflammatory arthritis at presentation can be used to predict outcome. Arthritis and Rheumatism; 1995: 37(8): 1166-1170. 127. Grassi, W , De Angelis, R , Lamanna, G , and Cervini, C. The clinical features of rheumatoid arthritis. European Journal of Radiology; 1998: 27(1):S18-S24. 128. Gray, PW, and Goeddel, DV. Structure of the human immune interferon gene. Nature; 1982: 298:859-863. 129. Green, J M , Noel, PJ, Sperling, A I , Walunas, TL,. Gray, GS, Bluestone, JA, and Thompson, CB. Absence of B7-dependent responses in CD28-deficient mice. Immunity; 1994: 1:501-508. 130. Gregersen, P K , Silver, J , and Weinchester, RJ. The shared epitope hypothesis: an approach to understanding the molecular genetics of susceptibility to rheumatoid arthritis. Arthritis and Rheumatism; 1987: 30:1205-1213. 131. Guasch, JF, Bertina, R M , and Reitsma, PH. Five novel intergenic dimorphisms in the human interleukin-1 gene combine to high informativity. Cytokine; 1996: 8(8):598-602. 132. Gulwani-Akolkar, B , Shi, B , Akolkar, PN, Ito, K , Bias, WB, and Silver, S. Do HLA genes play a prominent role in determining T-cell receptor V alpha segment usage in humans? Journal of Immunology; 1995: 154(8):3843-3851. 133. Gulwani-Akolkar, B , Akolkar, PN , Gregerson, P K , Silver, J. Analysis of the peripheral blood T-cell receptor repertoire in monozygotic twins discordant for rheumatoid arthritis. Annals of the New York Academy of Sciences; 1995: 756:176-178. 134. Hang, L. and Nakamura, RM. Current concepts and advances in clinical laboratory testing for autoimmune diseases. Critical Reviews in Clinical Laboratory Sciences; 1997: 34(3):275-311. 139 135. Harris, ED Jr., Rheumatoid arthritis: Pathophysiology and implications for therapy. New England Journal of Medicine; 1990: 322(18):1277-1289. 136. Harroch, S, Gothelf, Y , Revel, M , and Chebath, J. 5' upstream sequences of MyD88, an IL-6 primary response gene in M l cells: detection of functional IRF-1 and Stat factors binding sites. Nucleic Acids Reseach; 1995: 23(17): 3539-3546. 137. Hochberg, M C , Chang, RW, Dwosh, I, Lindsey, S, Pincus, T, and Wolfe, F. The American College of Rheumatology criteria for the classification of global functional status in rheumatoid arthritis. Arthritis and Rheumatism; 1992: 35 (5): 498-502. 138. Holbrook, NJ , Lieber, M , and Crabtree, GR. DNA sequence of the 5' flanking region of the human interleukin 2 gene: homologies with adult T-cell leukemia virus. Nucleic Acids Research; 1984: 12(12): 5005-5013. 139. Hopkins, SJ, and Meager, A. Cytokines in synovial fluid II. The presence of tumour necrosis factor and interferon. Clinical and Experimental Immunology; 1988: 73:88-92. 140. Home, C , Goodfellow, PJ, McDonald, H L , Junker, A K , Keown, PA, and Hoar, DI. A new HLA-DRB1 allele formed by an intra-exonic interallelic crossover. Tissue Antigens; 1993: 42:141-143. 141. Home, C,. and Keown, PA. Rapid DNA typing for class II HLA antigens: Subtyping of DRw52-associated DRB1 alleles. Tissue Antigens; 1993: 41:243-248. 142. Horneff, G. Advances in immunotherapy in rheumatoid arthritis: clinical and immunological findings following treatment with anti-CD4 antibodies. British Journal of Rheumatology; 1993: 32(suppl 14):39-47. 143. Housman, D. Human DNA polymorphism. The New England Journal of Medicine; 1995: 332(5):318-320. 144. Hughes, A L , Yeager, M , and Carrington, M. Peptide binding function and the paradox of HLA disease associations. Immunology and Cell Biology; 1996: 74(5):444-448. 145. Imanishi, T, Akaza, T, Kimura, A , Tokunaga, K , and Gojobori, T. Estimation of allele and haplotype frequencies for HLA and complement loci. HLA 1991. In Tsuji K , Aizawa, M,. Sasazuki, T. (eds): HLA 1991, Proceedings of the Eleventh International Histocompatibility Workshop and Conference. Oxford, Oxford University Press, 1992. 146. Jacob, CO, Fronek, Z , Lewis, GD, Koo, M , Hansen, JA, and McDevitt, HO. Heritable major histocompatibility complex class Il-associated differences in production of tumor necrosis factor a: relevance to genetic predisposition to systemic lupus erythematosus. Proceeds of National Academy of Sciences USA; 1990: 87:1233-1237. 147. Jacobson, D L , Gange, SJ, Rose, NR., and Graham, MH. Epidemiology and estimated population burden of selected autoimmune diseases in the United States. Clinical Immunology and Immunopathology; 1997: 84(3): 223-243. 148. Janossy, G , Panayi, G , Duke, O, BofilfM, Poulter, L W , and Goldstein, G. Rheumatoid arthritis: a disease of T-lymphocyte/macrophage immunoregulation. Lancet; 1981:11:839-841. 140 149. Jenkins, R N , Nikaein, A , Zimmerman, A , Meek, K , and Lipsky, PE. T-cell receptor VP gene bias in rheumatoid arthritis. Journal of Clinical Investigation; 1993: 92:2688-2701. 150. Johansen, B H , Vartdall, F , Ericksen, JA, Thorsby, E , and Sollid, LM. Identification of a putative motif for binding of peptides to HLA-DQ2. International Immunology; 1996: 8(2): 177-182. 151. Jordan, E , and Collins, FS. A march of genetic maps. Nature; 1996: 380:111-112. 152. Jorgensen, C , and Gay, S. Gene therapy in osteoarticular disease: Where are we? Immunology Today; 1998: 19(9):387-391. 153. Kalsi , J , and Isenberg, D. Rheumatoid factor: primary or secondary event in the pathogenesis of RA? International Archives of Allergy and Immunology; 1993: 102(3):209-215. 154. Katsikis, P, Chu, CQ, Brennan, F M , Maini, R N , and Feldmann, M. Immunoregulatory role of interleukin 10 (IL-10) in rheumatoid arthritis. Journal of Experimental Medicine; 1994: 179:1517-1528. 155. Kavanaugh, A F , and Lipsky, PE. Gold, penicillamine, antimalarials and sulfasalazine, in Inflammation: Basic Principles and Clinical Coorelates, 2d ed, JI Gallin et al. (eds). New York, Raven Press, 1992:1083-1101. 156. Kelly, A , Powis, SH, Glynne, R , Radley, E , Beck, S, and Trowsdale, J. A proteasome-related gene between the ABC transporter loci in the class II region of the human MHC. Nature; 1990: 353:357-360. 157. Khandka, D , Tuna, M , Tal, M,Nejidat, A , and Golan-Goldhirsh, A. Variability in the pattern of random amplified polymorphic DNA. Electrophoresis; 1997: 18:2852-2856. 158. Khani-Hanjani, A , Hoar, D , Horsman, D , and Keown, PA. Identification of four novel dinucleotide repeat polymorphisms in the TNF-a and TNF-P genes. Human Immunology; 2000:61: in press. 159. Khani-Hanjani, A , Lacaille, D , Hoar, D , Home, C , Jirik, F , Chalmers, A , and Keown, PA. MHC and cytokine polymorphisms in severe rheumatoid arthritis. Abstract No. 293 in Arthritis and Rheumatism; 1998: 41(9). 160. Khani-Hanjani, A , Hoar, D , Home, C , Lacaille, D , Chalmers, A , Jirik, F , Horsman, D , Adomat, S, and Keown, PA. TNF microsatellite frequencies and disease severity in rheumatoid arthritis (RA). Abstract No. T47 in Journal of Rheumatology; 1998: 25(52): 45. 161. Khani-Hanjani, A , Home, C , Lacaille, D , Hoar, D , Jirik, F , Chalmers, A , and Keown, PA. HLA-class II gene frequencies and disease severity in rheumatoid arthritis. Abstract No. T48 in Journal of Rheumatology; 1998: 25(52): 45. 162. Kimura, A , and Sasazuki, T. Eleventh international histocompatibility workshop reference protocol for the HLA DNA-typing technique. American Society of Histocompatibility and Immunogenetics; 1996: 397-419. 141 163. Kingsley, G , and Panayi, GS. Joint destruction in rheumatoid arthritis: biological bases. Clinical and Experimental Rheumatology; 1997: 15(17): S13-S14. 164. Kirschmann, D A , Duffin, K L , Smith, C E , Welply, JK, Howard, SC, Schwartz, BD. And Woulfe, SL. Naturally processed peptides from rheumatoid arthritis associated and non-associated HLA-DR alleles. Journal of Immunology; 1995: 155: 5655-5662. 165. Kirwan, JR. The effect of glucocorticoids on joint destruction in rheumatoid arthritis. New England Journal of Medicine; 1995: 333(3): 142-146. 166. Klaff, P , Riesnre, D , and Steger, G. RNA structure and the regulation of gene expression. Molecular Biology; 1996: 32:89-106. 167. Klareskog,L, Forsum, U , Scheynius, A , Kabelitz, D , and Wigzell, H. Evidence in support of a self perpetuating HLA-DR dependent delayed type cell reaction in rheumatoid arthritis. Proceedings of the National Academy of Sciences of the United States of America; 1982: 72:3632-3636. 168. Klippel, JH. and Dieppe, JA. Practical Rheumatology. London: Mosby; 1995. 169. Klitz, W , Stephens, C , Grote, M , and Carrington, M. Discordant patterns of linkage disequilibrium of the peptide-transporter loci within the HLA class II region. American Journal of Human Genetics; 1995: 57:1436-1444. 170. Kohsaka, H , Carson, D A , Miyasaka, N . Formation of peripheral immunoreceptor repertoire for antigens. Arthritis and Rheumatism; 1998: 41(11): 1911-1918. 171. Kotzin, B L , Leung, D Y , Kappler, J , and Marrack, P. Superantigens and their potential role in human disease. Advanced Immunology; 1993: 54:99-166. 172. Kraft, S M , Panush, RS, and Longley, S. Unrecognized straphylococcal pyarthrosis with rheumatoid arthritis. Seminars in Arthritis and Rheumatism; 1985: 14(3): 196-201. 173. Krakauer, T, Oppenheim, JJ, and Jasin, HE. Human interleukin 1 mediates cartilage matrix degradation. Cell Immunology; 1985: 91:92-99. 174. Krane, S M , and Simon, LS. Rheumatoid arthritis: clinical features and pathogenetic mechanisms. Medical Clinic of North America; 1986: 70:263-284. 175. Krause, A , Kamradt, T, and Burmester, GR. Potential infectious agents in the induction of arthritides. Current Opinion in Rheumatology; 1996: 8(3):203-209. 176. Kremer, J M , Alarcon, GS, Lightfoot, RW. Jr., Willkens, RF, Furst, DE , Williams, HJ , Dent, PB , and Weinblatt, ME. Methotrexate for rheumatoid arthritis: Suggested guidelines for monitoring liver toxicity. Arthritis and Rheumatism; 1994: 37(3):316-328. 177. Kroef, MJP, Willemze, R , and Landegent, JE. Dinucleotide repeat polymorphism in the interferon regulating factor 1 (IRF1) gene. Human Molecular Genetics; 1993: 2(10): 1748-1749. 178. Kruglyak, L. The use of a genetic map of biallelic markers in linkage studies. Nature Genetics; 1997: 17:21-24. 142 179. Kwok, WW, Domeier, M E , Raymond, FC , Byers, P, and Nepom, GT. Allele-specific motifs characterize HLA-DQ interactions with a diabetes-associated peptide derived from glutamic acid decarboxylase. The Journal of Immunology; 1996: 156:2171-2177. 180. Lacaille, D , Khani-Hanjani, A , Home, C , Hoar, D , Beattie, C , Rangno, K , Chalmers, A , and Keown, PA. HLA-DRB1 frequency, TNF gene polymorphism and disease severity in patients with rheumatoid arthritis. Abstract No. 281 in Arthritis and Rheumatism; 1997: 40(9). 181. Lawrence, JS. Rheumatoid arthritis: nature or nurture? Annals of the Rheumatic Diseases; 1970: 29:357-369. 182. Lechler, R. HLA and Disease. London: Harcourt Brace; 1994. 183. Lennard, A , Gorman, P , Carrier, M , Griffiths, S, Scotney, H , Sheer, D , and Solari, R. Cloning and chromosome mapping of the human interleukin-1 receptor antagonist gene. Cytokine; 1992: 4(2):83-89. 184. Leslie, RD, and Hawa, M. Twin studies in auto-immune disease. Acta Geneticae Medicae et Gemellologiae; 1994: 43(1-2): 71-81 185. L i , Q, Wrange, O, and Eriksson, P. The role of chromatin in transcriptional regulation. The International Journal of Biochemistry and Cell Biology; 1997: 29(5):731-742. 186. L i , S, Quayle, A J , Thoen, JE, and Forre, OT. Superantigen-mediated proliferation and cytotoxicity of T-cells isolated from the inflammatory tissues and peripheral blood of arthritis patients. Clinical Immunology and Immunopathology; 1996: 79(3):278-287. 187. Lichtenstien, M J , and Pincus, T. Rheumatoid arthritis identified in population based cross sectional studies: Low prevalence of rheumatoid factor. Journal of Rheumatology; 1991: 18:989-993. 188. Lipsky, PE. Rheumatoid arthritis in Harrison's Principles of Internal Medicine, 14th ed. 1998. USA: McGraw-Hill. 189. Lipsky, PE, Davis, L S , Cush, JJ, and Oppenheimer-Marks, N . The role of cytokines in the pathogenesis of rheumatoid arthritis. Springer Seminar of Immunopathology; 1989: 11(2): 123-162. 190. Lucassen, A M , Julier, C , Beressi, JP, Boitard, C , Forguel, P, Lathrop, M , and Bell, JI. Susceptibility to insulin-dependent diabetes mellitus maps to a 4.1-kb segment of DNA spanning the insulin gene and associated VNTR. Nature Genetics; 1993: 4:305-310. 191. Ludwin, D , and Alexopoulou, I. Cyclosporin A nephropathy in patients with rheumatoid arthritis. British Journal of Rheumatology; 1993: 32(1)60-64. 192. Lynn, A H , Kwoh, C K , Venglish, C M , Aston, CE , and Chakravarti, A. Genetic epidemiology of rheumatoid arthritis. American Journal of Human Genetics; 1995: 57:150-159. 193. Maeda, M , Uryu, N , Murayama, N M , Ishii, H , Ota, M , Tsuji, K , and Inoko, H. A simple and rapid method for HLA-DP genotyping by digestion of PCR-amplified DNA with allele-specific restriction endonucleases. Human Immunology; 1990: 27:111-121. 143 194. Magzoub, M M A , Stephens, HAF, Sachs, JA, Biro, PA, Cutbush, S, Wu, Z , and Bottazzo, GF. HAL-DP polymorphism in Sudanese controls and patients with insulin-dependent diabetes mellitus. Tissue Antigens; 1992: 40:64-68. 195. Maini, R N , Elliott, M , Brennan, F M , Williams, RO, and Feldmann, M. TNF blockade in rheumatoid arthritis: Implications for therapy and pathogenesis. APMS; 1997: 105: 257-263. 196. Maini, RN. The role of cytokines in rheumatoid arthritis. Journal of the Royal College of Physicians of London; 1996: 30(4):344-351. 197. March, CJ , Mosley,B, Larsen, A , Cerretti, DP, Braedt, G , Price, V , Gillis, S, Henney, CS, Kronheim, SR., Grabstein, K , Conlon, PJ, Hopp, TP, and Cosman, D. Cloning, sequence and expression of two distinct human interleukin-1 complementary DNAs. Nature: 1985:315:641-647. 198. Marsal, S, Hall, M A , Panayi, GS, and Lanchbury, JS. Association of TAP2 polymorphism with rheumatoid arthritis is secondary to allelic association with HLA-DRB1. Arthritis and Rheumatism; 1994:37:504-513. 199. Maruyama, T, Saito,I, Miyake, S, Hashimoto,H, Sato, K , Yagita, H , Okumura, K , and Miyasaka, N . A possible role of two hydrophobic amino acids in antigen recognition by synovial T-cells in rheumatoid arthritis. European Journal of Immunology; 1993:23:2059-2065. 200. Matise, TC. Genome scanning for complex disease genes using the transmission/ disequilibrium test and haplotype-based haplotype relative risk. Genetic Epidemiology; 1995: 12:641-645. 201. McGuire, W, Hill, A V S , Allisop, C E M , Greenwood, B M , and Kwiatkowski, D. Variation in the TNF-a promoter region associated with susceptibility to cerebral malaria. Nature; 1994: 371:508-511. 202. Mclnnes, I, and Liew, FY. Interleukin 15: a proinflammatory role in rheumatoid arthritis synovitis. Immunology Today; 1998: 19(2):75-79. 203. Meunier, L , Vian, L , Lagoueyte, C , Lavabre-Bertrand, T, Duperray, C , Meynadier, J , and Cano, JP. Quantification of CD la, HLA-DR, and HLA class I expression on viable human langerhans cells and keratinocytes. Communications in Clinical Cytometry; 1996: 26:260-264. 204. Meyer, C G , May, J , and Schnittger, L. HLA-DP -part of the concert. Immunology Today; 1997: 18(2):58-64. 205. Miltenburg, A J , van Laar, J M , de Kuiper, R, Daha, MR, and Breedveld, FC. T-cells cloned from human rheumatoid synovial membrane functionally represent the Thl subset. Scandinavian Journal of Immunology; 1992: 35:603-610. 206. Mirza, N M , Relias, V , Yunis, EJ , Pachas, W N , and Dev Dasgupta, J. Defective signal transduction via T-cell recpetor-CD3 structure in T-cells from rheumatoid arthritis patients. Human Immunology; 1993: 36:91-98. 207. Mitsunaga, S, Oguchi, T, Tokunaga, K , Akaza, T, Tadokoro, K , and Juji, T. High resolution HLA-DQB1 typing by combination of group-specific amplification and RFLP. Human Immunology; 1995:42:307-314. 144 208. Mizel, SB, Dayer, J M , Krane, S M , and Mergenhagen, SE. Stimulation of rheumatoid synovial cell collagenase and prostaglandin production by partially purified lymphocyte-activating factor (interleukin-1). Proceedings of the National Academy of Sciences of the United States of America; 1981: 78:2474-2477. 209. Mody, G M , and Hammond, MG. Differences in HLA-DR association with rheumatoid arthritis among migrant Indian communities in South Africa. British Journal of Rheumatology; 1994: 33:425-427. 210. Monos, DS, Kamoun, M , Udalove, IA, Csanky, E , Cizman, B , Turetskaya, R L , Smirnova, JB, Zharkov, V G , Gasser, D , Zmijewski, C M , Spielman, RS, and Nedospasov, SA. Genetic polymorphism of the human tumor necrosis factor region in insulin-dependent diabetes mellitus linkage disequilibrium of TNFab microsatellite alleles with HLA haplotypes. Human Immunology; 1995: 44:70-79. 211. Montminy, M. Transcriptional regulation by cyclic AMP. Annual Review of Biochemistry; 1997: 66:807-822. 212. Moreland, L W , Heck, L W , Sullivan, W, Pratt, PW, and Koopman, WJ. New approaches to the therapy of autoimmune diseases: Rheumatoid arthritis as a paradigm. American Journal of Medical Sciences; 1993:305:40-51. 213. Morgan, GJ , and Chow, W-S. Clinical features, diagnosis, and prognosis in rheumatoid arthritis. Current Opinion in Rheumatology; 1993: 5:184-190. 214. Mu, H , Chen, JJ, Jiang, Y , King, M C , Thomson, G , and Criswell L. Tumor necrosis factor a microsatellite polymorphism is associated with rheumatoid arthritis severity through an interaction with the HLA-DRBl shared epitope. Arthritis and Rheumatism; 1999: 42(3):438-442. 215. Mulcahy, B , Waldron-Lynch, F , McDermott, M F , Adams, C , Amos, CI, Zhu, D K , Ward, R H , Clegg, DO, Shanahan, F , Molloy, M G , and O'Gara, F. Genetic variability in the tumor necrosis factor-lymphotoxin region influences susceptibility to rheumatoid arthritis. American Journal of Human Genetics; 1996: 59:676-683. 216. Muller-Ladner, U , Gay, RE , Gay, S. Molecular biology of cartilage and bone destruction. Current Opinion in Rheumatology; 1998: 10:212-219. 217. Murphy, T L , Cleveland, M G , Kulesza, P , Magram, J , and Murphy, K M . Regulation of interleukin 12 p40 expression through an N F - K B half-site. Molecular and Cellular Biology; 1995: 15(10):5258-5267. 218. Nabozny, G H , Baisch, J M , Cheng, S, Cosgrove, D , Griffiths, M M , Luthra, HS, and David, CS. HLA-DQ8 transgenic mice are highly susceptible to collagen-induced arthritis: A novel model for human polyarthritis. Journal of Experimental Medicine; 1996: 183:27-37. 219. Nedospasov, SA, Shakhov, A N , Turetskaya, R L , Mett, V A , Azizov, M M , Georgiev, GP, Korobko, V G , Dobrynin, V N , Filippov, SA, Bystrov, NS, Boldyreva, EF , Chuvpilo, SA, Chumakov, A M , Shingarova, L N , and Ovchinnikov, YA. Tandem arrangement of genes coding for tumor necrosis factor (TNF-a) and lymphotoxin (TNF-B) in the human genome. Proceedings of the Cold Spring Harbor symposia on quantitative biology; 1986: 51:611 -625. 145 220. Nedospasov, SA, Udalova, IA, Kuprash, D V , and Turetskaya, RL. DNA sequence polymorphism at the human tumor necrosis factor locus. Numerous TNF/Iymphotoxin alleles tagged by two closely linked microsatellites in the upstream region of the lymphotoxin (TNF-beta) gene. Journal ofImmunology; 1991: 147(3): 1053-1059. 221. Nepom, GT, Gersuk,V, and Nepom, BS. Prognostic implications of HLA genotyping in the early assessment of patients with rheumatoid arthritis. Journal of Rheumatology; 1996: 23:5-9. 222. Nepom, GT. and Nepom, BS. Prediction of susceptibility to rheumatoid arthritis by human leukocyte antigen genotyping. Rheumatic Diseases Clinics of North America; 1992: 18(4):785-792. 223. Nepom, GT, and Erlich, H. MHC class II molecules and autoimmunity. Annual Review of Immunology; 1991: 9:493-525. 224. Nepom, GT, Byers, P , Seyfried, C , Healey, L A , Wilski, K R , Stage, D , and Nepom, BS. HLA genes associated with rheumatoid arthritis: identification of susceptibility alleles using specific oligonucleotide probs. Arthritis and Rheumatisms; 1989: 32:15-21 225. Newkirk, M M , Fournier, M. and Shiroky, J. Rheumatoid factor avidity in patients with rheumatoid arthritis: identification of pathogenic RFs which correlate with disease parameters and with the gal (0) glycoform of IgG. Journal of Clinical Immunolgy; 1995: 15(5):250-257. 226. Nimgaonkar, VL. Yang, ZW, Zhang, X R , Brar, JS, Chakravarti, A , and Ganguli, R. Association study of schizophrenia and the IL-2 receptor P chain gene. American Journal of Medical Genetics (Neuropsychiatric Genetics); 1995: 60:448-451. 227. Nomura, N , Ota, M , Tsuji, K , and Inoko, H. HLA-DQB1 genotyping by modified PCR-RFLP method combined with group-specific primers. Tissue Antigens; 1991: 38:53-59. 228. Nunez, G , Moore, S, Ball, G V , Hurd, ER, and Stastny, P. Study of HLA antigens in ten multiple-case rheumatoid arthritis families. Journal of Rheumatology; 1984: 11:129-135. 229. Odeh, M. New insights into the pathogenesis and treatment of rheumatoid arthritis. Clinical Immunology and Immunopathology; 1997: 83(2): 103-116. 230. O'Dell JR. Combination DMARD therapy with hydroxychloroquine, sulfasalazine, and methotrexate. Clinical and Experimental Rheumatology; 1999: 6(18): S53-58. 231. O'Garra, A. Cytokines induce the development of functionally heterogeneous T helper cell subsets. Immunity; 1998: 8:275-283. 232. Ogbourne, S, and Antalis, TM. Transcriptional control and the role of silencers in transcriptional regulation in eukaryotes. Biochemical Journal; 1998: 331:1-14. 233. Olson, RR, De Magistris, M T , Di Tommaso, S, and Karr, RW. Mutations in the third, but not the first or second, hypervariable regions of DR (bl*0101) eliminate DR1 -restricted recognition of a pertussis toxin peptide. Journal of Immunology; 1992: 148:2703-2708. 234. Ou, D , Mitchell, L A , and Tingle, AJ. HLA-DR restrictive supertypes dominate promiscuous T-cell recognition: Association of multiple HLA-DR molecules with susceptibility to autoimmune diseases. Journal of Rheumatology; 1997: 24:253-261. 146 235. Paleolog, E M , Hunt, M , Elliott, M , Feldmann, M , Maini, R N , and Woody, JN. Deactivation of vascular endothelium by monoclonal anti-tumor necrosis factor a antibody in rheumatoid arhtirits. Arthritis and Rheumatism; 1996: 39(7):1082-1091. 236. Panayi, GS. T-cell-dependent pathways in rheumatoid arthritis. Current Opinion in Rheumatology; 1997: 9:236-240. 237. Panayi, GS, Lanchbury, JS, and Kingsley, GH. The importance of the T-cell in initiating and maintaining the chronic synovitis of rheumatoid arthritis (editorial). Arthritis and Rheumatism; 1992:35:729-735. 238. Perdriger, A , Guggenbuhl, P, Chales, G , Le Dantec, P , Yaouanq, J , Genetet, B , Pawlotsky, Y , and Semana, G. The role of HLA-DR-DR and HLA-DR-DP interactions in genetic susceptibility to rheumatoid arthritis. Human Immunology; 1996: 46:42-48. 239. Perdriger, A , Semana, G , Quillivic, F , Chales, G , Chardevel, F , Legrand, E , Meadeb, J , Fauchet, R , and Pawlotsky, Y. DPB1 polymorphism in rheumatoid arthritis: evidence of an association with allele DPB1.0401. Tissue Antigens; 1992: 39:14. 240. Petrovsky, N , and Harrison, L. HLA class II-associated polymorphism of interferon-g production; implication for HLA-disease association. Human Immunology; 1997: 53:12-16. 241. Pociot, F , Briant, L , Jongeneel, C V , Molvig, J , Worsaae, H , Abbal, M , Thomsen, M , Nerup, J , and Cambon-Thomsen, A. Association of tumor necrosis factor (TNF) and class II major histocompatibility complex alleles with the secretion of TNF-cc and TNF-P by human mononuclear cells: a possible link to insulin-dependent diabetes mellitus. European Journal of Immunology; 1993:23:224-231. 242. Pociot, F , Molvig, J , Wogensen, Worsaae, H , and Nerup, J. A TaqI polymorphism in the human interleukin-1P (IL-1P) gene correlates with IL-ip secretion in vitro. European Journal of Clinical Investigation; 1992: 22:396-402. 243. Pociot, F , Molvig, J , Wogensen, L , Worsaee, H , Dalboge, H , Baek, L , and Nerup, J. A tumor necrosis factor beta gene polymorphism in relation to monokine secretion and insulin-dependent diabetes mellitus. Scandinavian Journal of Immunology; 1991: 33:37-49. 244. Pope, RM. Rheumatoid arthritis: pathogenesis and early recognition. American Journal of Medicine; 1996: 100:2A-9S. 245. Pravica, V , Asderakis, CP, Hajeer, A , Sinnott, PJ, and Hutchinson, IV. In vitro production of IFN-y correlates with CA repeat polymorphism in the human IFN-y gene. European Journal of Immunogenetics; 1999: 26:1-3. 246. Pritchard, MH. Evidence for a hypothetical non-HLA susceptibility gene in rheumatoid arthritis. British Journal of Rheumatology; 1994: 33:475-479. 247. Puttick, A H , Briggs, DC, Welsh, K I , Vaughan, R , Williamson, E A , Boyce, M , Jacoby, R K , and Jones, VE. Genes associated with rheumatoid arthritis and mild inflammatory arthritis. I. Major histocompatibility complex class I, II, and III allotypes. Annals of Rheumatic Diseases; 1990:49:219-224. 147 248. Quayle, A J , Chomarat, P, Miossec, P , Kjeldsen-Kragh, J , Forre, O, and Natvig, JB. RJieumatoid inflammatory T-cell clones express mostly Thl but also Th2 and mixed (ThO-like) cytokine patterns. Scandinavian Journal of Immunology; 1993: 38:75-82. 249. Reveille, JD. The genetic contribution to the pathogenesis of rheumatoid arthritis. Current Opinion in Rheumatology; 1998: 10:187-200. 250. Rigby, AS. HLA haplotype sharing in rheumatoid arthritis sibships: risk estimated in siblings. Scandinavian Journal of Rheumatology; 1992: 21:68-73. 251. Ruiz-Linares, A. Dinucleotide repeat polymorphism in the interferon-gamma (IFNG) gene. Human Molecular Genetics; 1993: 2(9): 1508. 252. Salmon, M , and Gaston, JS. The role of T-lymphocytes in rheumatoid arthritis. British Medical Bulletin; 1995: 51 (2):332-345. 253. Salvarani, C , Macchioni, P , Mantovani, W, Rossi, F , Veneziani, M , Boiardi, L , Lodi, L. and Portioli, I. Extraarticular manifestations of rheumatoid arthritis and HLA antigens in northern Italy. Journal of Rheumatology; 1992: 19:242-246. 254. Sander, O, and Rau, R. Clinical trials on biologies in rheumatoid arthritis. International Journal of Clinical Pharmacology and Therapeutics; 1998: 36(11): 621-624. 255. Satter, M A , Al-Saffer, M , Guindi, RT, Sugathan, TN., and Bahbehani, K. Association between HLA-DR antigens and rheumatoid arthritis in Arabs. Annals of the Rheumatic Diseases; 1990:49:147-149. 256. Schiff, M. Emerging treatments for rheumatoid arthritis. American Journal of Medicine; 1997: 102(suppllA):llS-15S. 257. Schnitzer, TJ, and Penmetch, M. Viral arthritis. Current Opinion in Rheumatology; 1996: 8(4):341-345. 258. Sengar, DPS, and Goldstein, R. Comprehensive typing of DQB1 alleles by PCR-RFLP. Tissue Antigens; 1994: 43(4):242-248 259. Shimura, T, Hagihara, M , Takebe, K , Munkhbat, B , Ogoshi, K , Mitomi, T, Nagamachi, Y , and Tsuji, K. 10.5-kb Homozygote of tumor necrosis factor-beta gene is associated with a better prognosis in gastric cancer patients. Cancer; 1995: 75:140-143. 260. Shimura, T, Hagihara, M , Takebe, K , Munkhbat, B , Ogoshi, K , Mitomi, T, Nagamachi, Y , and Tsuji, K. The study of tumor necrosis factor beta gene polymorphism in lung cancer patients. Cancer; 1994: 73:1184-1188. 261. Shows, T, Eddy, R , Haley, L , Byers, M , Henry, M , Fujita, T, Matsui, H , and Taniguchi, T. Interleukin 2 (IL2) is assigned to human chromosome 4. Somatic Cell and Molecular Genetics; 1984: 10(3):315-318. 262. Siegel, JP, Sharon, M , Smith, PL , and Leonard, WJ. The IL-2 receptor beta chain (p70): role in mediating signals for LAK, NK and proliferative activities. Science; 1987: 238(4823): 75-78. 148 263. Siegmund, T, Usadel, K H , Donner, H , Braun, J , Walfish, PG, and Badenhoop, K. Interferon-y gene microsatellite polymorphisms in patients with Graves' disease. Thyroid; 1998: 8(11):1013-1017. 264. Silman, A J , Newman, J , and MacGregor, AJ. Cigarette smoking increases the risk of rheumatoid arthritis. Results from a nationwide study of disease-discordant twins. Arthritis and Rheumatism; 1996: 39(5):732-735. 265. Silman, AJ. Epidemiology of rheumatoid arthritis. ARMS; 1994: 102:721-728. 266. Silman, A J , MacGregor, A J , and Thomson, W. Twin concordance rates for rheumatoid arthritis: results from a nationwide study. British Journal of Rheumatology; 1993: 32:903-907. 267. Silman, A J , Hennessy, E , and Oilier, WER. Incidence of rheumatoid arthritis in a genetically predisposed population. British Journal of Rheumatology; 1992: 31:365-368. 268. Simon, LS , Weaver, A L , Graham, D Y , Kivitz, A J , Lipsky, PE, Hubbard, R C , Isakson, PC, Verburg, M K , Yu, SS, Zhao, WW, and Geis, GS. Anti-inflammatory and upper gastrointestinal effects of Celecoxib in rheumatoid arthritis. Journal of American Medical Association; 1999: 282(20): 1921-1928. 269. Simon, LS. Actions and toxicity of nonsteroidal anti-inflammatory drugs. Current Opinion in Rheumatology; 1996: 8(3): 169-175. 270. Singal, DP, D'Souza, M , Reid, B , Bensen, WG, Kassam, Y B , And Adachi, JD. HLA-DQ beta-chain polymorphism in HLA-DR4 haplotypes associated with rheumatoid arthritis. Lancet; 1987: ii: l 118-1120. 271. Smolen, JS, Kalden, JR., Scott, D L , Rozman, B , Kvien, T K , Larson, A , Loew-Friedrich, I, Oed, C , and Rosenburg, R. Efficacy and safety of leflunomide compared with placebo and sulphasalazine in active rheumatoid arthritis: a double blind, randomised, multicentre trial. Lancet; 1999: 353:259-266. 272. Stastny, P , Ball, EJ , Khan, M A , Olsen, NJ , Pincus, T, and Gao, X. HLA-DR4 and other genetic markers in rheumatoid arthritis. British Journal of Rheumatology; 1988: 27(2): 132-138. 273. Steinkasserer, A , Spurr, N K , Cox, S, Jeggo, P , and Sim, RB. The human IL-1 receptor antagonist gene (IL1RN) maps to chromosome 2ql4-q21, in the region of the IL-lcc and IL-1 P loci. Genomics; 1992: 13:654-657. 274. Stephens, HAF, Vaughan, RW, Sakkas, L I , Welsh, KI. and Panayi, GS. Southern blot analysis of HLA DP gene polymorphism in Caucasoid rheumatoid arthritis. Immunogenetics; 1989:30(3):149-155. 275. Stephens, HAF, Vaughan, RW, Sakkas, L I , Teitsson, I, Welsh, K I , and Panayi, GS. HLA-DQw7 is a disease severity marker in patient with rheumatoid arthritis. Immunogenetics; 1989: 30(2): 119-122. 276. Stem, L J , Brown, JH, Jardetzky, TS, Gorga, JC, Urban, R G , Strominger, JL , and Wiley, DC. Crystal structure of the human class II MHC protein HLA-DR1 complexed with an influenza virus peptide. Nature; 1994: 368(6468):215-221. 149 277. Strand, V , Tugwell, P , Bombardier, C , Maetzel, A , Crawford, B , Dorrier, C , Thompson, A. and Wells G. Function and health-related quality of life: Results from a randomized controlled trial of leflunomide versus methotrexate or placebo in patients with active rheumatoid arthritis. Arthritis and Rheumatism; 1999: 42(9): 1870-1878. 278. Stuber, F , Udalove, IA, Book, M , Drutskaya, L N , Kuprash, D V , Turetskaya, R L , Schade, F U , and Nedospasov, SA. -308 tumor necrosis factor (TNF) polymorphism is not associated with survival in severe sepsis and is unrelated to lipopolysaccharide inducibility of the human TNF promoter. Journal of Inflammation; 1996: 46:42-50. 279. Suarez-Almazor; M E , Tao, S, Moustarah, F , Russell A S , and Maksymowych, W. HLA-DR1, DR4, and DRB1 disease related subtypes in rheumatoid arthritis. Association with susceptibility but not severity in a city wide community based study. Journal of Rheumatology; 1995:22:2027-2033. 280. Svartz, N . The origin of rheumatoid arthritis. Rheumatology; 1975: 6:322-328. 281. Tak, PP, Taylor, PC, Breedveld, FC, Smeets, TJ, Daha, M R , Kluin, P M , Meinders, A E , and Maini, RN. Decreased cellularity and expression of adhesion molecules by anti-tumor necrosis factor alpha monoclonal antibody treatment in patients with rheumatoid arthritis. Arthritis and Rheumatism; 1997: 40(4):789-790. 282. Tak, PP, Taylor, PC, Breedveld, FC, Smeets, T, Daha, M R , Kluin, PM. Meinders, A E , and Maini, RN. Decrease in cellularity and expression of adhesion molecules by anti-tumor necrosis factor alpha monoclonal antibody treatment in patient with rheumatoid arthritis. Arthritis and Rheumatism; 1996: 39(7): 1077-1081. 283. Tak, PP, Taylor, PC, Breedveld, FC , Macfarlane, JD, Smeets, T, Daha, M R , Kluin, PM. Meinders, A E , and Maini, RN. Infdtrate analysis of rheumatoid synovial tissue before and after anti-TNFa monoclonal antibody treatment. Presented at American College of Rheumatologists Meeting, San Fransisco, October 1995. 284. Tak, PP, Van Der Lubbe,PA, Cauli, A , Daha, M R , Smeets, TJ, Kluin, P M , Meinders, A E , Yanni, G , Panayi, GS, and Breedveld, FC. Reduction of synovial inflammation after anti-CD4 monoclonal antibody treatment in early rheumatoid arthritis. Arthritis and Rheumatism; 1995: 38(10):1457-1465. 285. Tanaguchi, T, Matsui, H , Fujita, T, Takaoka, C , Kashima, N , Yoshimoto, R , and Hamuro, J. Structure and expression of a cloned cDNA for human interleukin-2. Nature; 1983: 302:395-310. 286. Taneja, V , Giphart, M J , Verduijn, W, Naipal, A , Malaviya, A N , and Mehra, NK. Polymorphism of HLA-DRB, -DQA1, and -DQB1 in rheumatoid arthritis in Asian Indians: association with DRB1*0405 and DRB1*1001. Human Immunology; 1996: 46(1): 35-41. 287. Tarlow, JK, Blakemore,AIF, Lennard, A , Solari, R , Hughes, H N , Steinkasserer, A. and Duff, GW. Polymorphism in human IL-1 receptor antagonist gene intron 2 is caused by variable numbers of an 86-bp tandem repeat. Human Genetics; 1993: 91:403-404. 150 288. Teller, K , Budhai, L , Zhang, M , Haramati, N , Keiser, HD, and Davidson, A. HLA-DRB1 and DQB typing of Hispanic American patients with rheumatoid arthritis: The \"shared epitope\" hypothesis may not apply. Journal of Rheumatology; 1996: 23:1363-1368. 289. Tett, SE. Clinical pharmacokinetics of slow-acting antirheumatic drugs. Clinical Pharmacokinetics; 1993: 25(5): 392-407. 290. The North American Rheumatoid Arthritis Disease Management Study Group. Analysis of therapeutic intervention, clinical outcome and socioeconomic impact in patients with rheumatoid arthritis. Abstract No. T46 in Journal of Rheumatology; 1998: 25(52):45. 291. Thomas, R. and Quinn, C. Functional differentiation of dendritic cells in rheumatoid arthritis. Role of CD86 in the synovium. Journal of Immunology; 1996: 156: 3074-3086. 292. Thomson, G. HLA disease associations: models for the study of complex human genetic disorders. Critical Reviews in Clinical Laboratory Sciences; 1995: 32(2): 183-219. 293. Thorsby, E , and Undlien, D. The HLA associated predisposition to type 1 diabetes and other autoimmune diseases. Journal of Pediatric Endocrinology andMetabolism;l996: 9(l):75-88. 294. Tsuchiya, N. and Williams, RC. Jr. Molecular mimicry\u00E2\u0080\u0094hypothesis or reality? Western Journal of Medicine; 1992: 157(2): 133-138. 295. Tugwell, P , Pincus, T, Yocum, D , Stein, M , Gluck, O, Cragg, G , McKendry, R , Tesser, J , Baker, P , and Wells, G. Combination therapy with cyclosporine and methotrexate in severe rheumatoid arthritis. New England Journal of Medicine; 1995: 333(3):137-141. 296. Tugwell, P , Bombarider, C , Gent, M , Bennett, K G , Benson,WG, Caretti, S, Chalmers, A , Esdaile, J M , Klinkhoff, A V , Kraag, GR. Low-dose cyclosporine versus placebo in patients with rheumatoid arthritis. Lancet; 1990: 335(8697): 1051-1055. 297. Udalova, IA , Nedospasov, SA, Webb, GS, Chaplin, DD, and Turetskaya, RL. Highly informative typing of the human TNF locus using six adjacent polymorphic markers. Genomics; 1993: 16:180-186. 298. Urbina-OJoiro, H , Cardiel, M H , and Alcocer-Varela, J. Reclassifying the pathogenesis of rheumatoid arthritis: From the susceptibility to the degenerative stages. Clinical and Experimental Rheumatology; 1998: 16: 87-91. 299. Van Den Berg, WB, and Van Lent, PLE. The role of macrophages in chronic arthritis. Immunobiology; 1996: 195:614-623. 300. Van Den Berg, WB, Joosten, L A B , Helsen, M , and Van De Loo, FAJ. Amelioration of established murine collagen-induced arthritis with anti-IL-1 treatment. Clinical Experimental Immunology; 1994: 95:237-243. 301. Vandenbroeck, K , Opdenakker, G , Goris, A , Murru, R , Billiau, A , Marrosu, MG. Interferon-y gene polymorphism-associated risk for multiple sclerosis in Sardinia. Annals of Neurology; 1998: 44(5):841-842. 151 302. Van Den Loo, A A J , Joosten, L A B , Van Lent, PLEM, Arntz, OJ, and Van Den Berg, WB. Role of interleukin-1, tumor necrosis factor a and interleukin-6 cartilage proteoglycan metabolism and destruction. Effect of in situ cytokine blocking in murine antigen- and zymosan-induced arthritis. Arthritis and Rheumatism; 1995: 38(2): 164-172. 303. Van Den Loo, A A J , Arntz, OJ, Bakker, A C , Van Lent, PLEM, Jacobs, M J M , and Van Den Berg, WB. Role of interleukin-1 in antigen-induced exacerbations of murine arthritis. American Journal of Pathology; 1995: 146(l):239-249. 304. Van Lent, PLEM, Van De Loo, FAJ, Holthuysen, A E M , Van Den Bersselaar, L A M , Vermeer, H , and Van Den Berg, WB. Major role for IL-1 but not for TNF in early cartilage damage in immune complex arthritis in mice. Journal of Rheumatology; 1995: 22(12):2250-2258. 305. Vandevyver, C , Guesen, P , Cassiman, JJ, and Raus, J. Peptide transporter genes (TAP) polymorphisms and genetic susceptibility to rheumatoid arthritis. British Journal of Rheumatology; 1995: 34:207-214. 306. Veale, D , and Maple, C. Cell adhesion molecules in rheumatoid arthritis. Drugs and Aging; 1996: 9(2):87-92. 307. Vehe, R , Nepom, K , Wilske, K. Erosive rheumatoid factor negative and positive rheumatoid arthritis are immunogenetically similar. Journal of Rheumatology; 1994: 21:194-196. 308. Veijola, R , Knip, M , Puukka, R, Reijonen, H , Cox, DW, and Ilonen, J. The immunoglobulin heavy-chain variable region in insulin-dependent diabetes mellitus: Affected-sib-pair analysis and association studies. American Journal of Human Genetics; 1996: 59:462-470. 309. Vendrell, J , Gutierrez, Pastor, R, and Richart, C. A tumor necrosis factor-P polymorphism associated with hypertriglyceridemia in non-insulin-dependent diabetes mellitus. Metabolism; 1995:44(6):691-694. 310. Versluis, L F , Rozemuller, E , Tonks, S, Marsh, SGE, Bouwens, A G M , Bodmer, JG, and Tilanus, MGJ. High-resolution HLA-DPB typing based upon computerized analysis of data obtained by fluorescent sequencing of the amplified polymorphic exon 2. Human Immunology; 1993: 38:277-283. 311. Wagner, U , Kaltenhauser, S, Sauer, H , Arnold, S, Seidel, W, Hantzschel, H , Kalden, JR., and Wassmuth, R. HLA markers and prediction of clinical course and outcome in rheumatoid arthritis. Arthritis and Rheumatism; 1997: 40(2):341-351. 312. Walser-Kuntz, DR, Weyand, C M , Weaner, A J , O'Fallon, W M , and Goronzy, JJ. Mechanisms underlying the formation of the T-cell receptor repertoire in rheumatoid arthritis. Immunity; 1995: 2:597-605. 313. Weeks, D E , and Lange, K. The affected-pedigree-member method of linkage analysis. American Journal of Human Genetics; 1988: 42:315-326. 314. Weyand, CM. and Goronzy, JJ. Inherited and noninherited risk factors in rheumatoid arthritis. Current Opinion in Rheumatology; 1995: 7:206-213. 152 315. Weyand, CM., Hunder, N N H , Hicok, K C , Hunder, G G , and Goronzy, JJ. HLA-DRbl alleles in polymyalgia rheumatica, giant cell arthritis and rheumatoid arthritis. Arthritis and Rheumatism; 1994: 37:514-520. 316. Weyand, C M , and Goronzy, JJ. Prognosis in rheumatoid arthritis: Applying new technologies to old questions. The Journal of Rheumatology; 1993: 20( 11): 1817-1820. 317. Weyand, C M , Hicok, K C , Hunder, G G , and Goronzy, JJ. The HLA-DRbl locus as a genetic component in giant cell arthritis: Mapping of disease-linked sequence motif to the antigen binding site of the HLA-DR molecule. Journal of Clinical Investigation; 1992: 90:2355-2361. 318. Whittaker, JC, and Lewis, CM. The effect of family structure on linkage test using allelic association. American Journal of Human Genetics; 1998: 63:889-897. 319. Whittemore, AS. Genome scanning for linkage: An overview. American Journal of Human Genetics; 1996: 59:704-716. 320. Williams, W V , Kieber-Emmons, T, Fang, Q, Von Feldet, J , Wang, B , Ramanujam, T, and Weiner, DB. Conserved motifs in rheumatoid arthritis synovial tissue T-cell receptor beta chain. DNA and Cell Biology; 1993:12(5):425-434. 321. Wilson, AG, De Vries, N , Van De Putte, L B A , and Duff, GW. A tumor necrosis factor a polymorphism is not associated with rheumatoid arthritis. Annals of the Rheumatic Diseases; 1995: 54:601-603. 322. Wilson, A G , Gordon, C , Di Giovine, FS, De Vries, N , Van De Putte, L B A , Emery, P , and Duff, GW. A genetic association between systemic lupus erythematosus and tumor necrosis factor a. European Journal of Immunology; 1994: 24:191-195. 323. Wilson, A G , De Vries, N , Pociot, F , Di Giovine, FS, Van Der Putte, L B A , and Duff, GW. An allelic polymorphism within the human tumor necrosis factor a promoter region is strongly associated with HLA A l , B8, and DR3 alleles. Journal of Experimental Medicine; 1993: 177:557-60. 324. Winchester, R. The molecular basis of susceptibility to rheumatoid arthritis. Advanced Immunology; 1994: 56:389-466. 325. Winchester, R , Dwyer E , and Rose, S. The genetic basis of rheumatoid arthritis. Rheumatic Diseases Clinics of North America; 1992: 18:761-783. 326. Wolfe, F. The prognosis of rheumatoid arthritis: Assessment of disease activity and disease severity in the clinic. The American Journal of Medicine; 1997: 103(6A):12S-18S. 327. Wucherpfennig, K W , Yu, B , Bhol, K , Monos, DS, Argyris, E , Karr, RW, Ahmed, A R , and Strominger, JL. Structural basis for major histocompatibility complex (MHC)-linked susceptibility to autoimmunity: charged residues of a single MHC binding pocket confer selective presentation of self-peptides in pemphigus vulgaris. Proceeds of National Academy of Sciences USA; 1995: 92 (25): 11935-11939. 153 328. Yamamoto, K , Kobayashi, H , Miura, O, Hirosawa, S, and Miyasaka, N . Assignment of IL12RB1 and IL12RB2, interleukin-12 receptor pi and P2 chains, to human chromosome 19 band pi3.1 and chromosome 1 band p31.2, respectively, by in situ hybridization..Cytogenetics and Cell Genetics; 1997: 77:257-258. 329. Yanni, G , Whelan, C , Quinlan, W, Symons, J , Duff, G , and Bresnihan, B. Contrasting levels of in vitro cytokine production by rheumatoid synovial tissues demonstrating different patterns of mononuclear cell infiltration. Clinical Experimental Immunology; 1993: 93:387-395. 330. Young, H A , and Hardy, KJ. Interferon-g: producer cells. Activation stimuli, and molecular genetic regulation. Pharmacology & Therapeutics; 1990: 45: 137-151. 331. Zanelli, E , Huizinga, TWJ, Guerne, P-A, Vischer, T L , Tiercy, J -M, Verdyun, W, Schreuder, GMT, Breedveld, FC , and De Vries, RRP. An extended HLA-DQ-DR haplotype rather than DRB1 alone contributes to RA predisposition. Immunogenetics; 1998: 48: 394-401. 332. Zanelli, E , Gonzalez,-Gay, M A , and David, CS. Could HLA-DRB1 be the protective locus in rheumatoid arthritis? Immunology Today; 1995: 16(6):274-278. 333. Zvaifler, N J , and Firestein, GS. Pannus and pannocytes: Alternative models of joint destruction in rheumatoid arthritis. Arthritis and Rheumatism; 1994: 37(6):783-789. 154 APPENDIX Figure 1 Electropherogram of ROX500 size-standard. 0 30 60 90 120 150 180 210 240 270 300 330 360 i i i i i i i 1 1 1 1 1 i i i 5 4 0 480 420, 36C 300 240 1 8 0 120 155 Figure 2a Electropherogram of samples and ROX500 size-standard. ' i \u00E2\u0080\u00A2 ' \u00E2\u0080\u00A2 ' i 1 ' > \u00E2\u0080\u00A2 i > i i > i \u00C2\u00BB < i > j < i i i j t % i i j < \u00E2\u0080\u00A2 i i | i i \u00E2\u0080\u00A2 , j i , 90 100 110 120 130 140 150 1 (JO 170 156 Figure 2b Electropherogram of multiple samples and ROX500 size-standard. 1 0 0 1 1 0 1 2 0 1 3 0 1 4 0 1 5 0 1 6 0 ' j ' I i I \u00E2\u0080\u00A2 L__ ID 3 B : 0 3 \u00C2\u00AB 0 3 F - T N F A / B / C / \u00C2\u00A3 @ 3Y:03\u00C2\u00BB03F -TNFA/8/C / Ig 3 R : 0 3 > 0 3 F - T N F A / B / C / 157 vd a _\u00C2\u00A9 u 4> \u00E2\u0080\u00A2a o E X i -a e 2 a JS 1. u I . \u00C2\u00AB JS U X i H a 158 3 co. l-H to Z H T3 to Z H to z J 3 to Z H to Z H to Z H to Q I < cu Q \u00E2\u0080\u00A2< J I i "Thesis/Dissertation"@en . "2000-05"@en . "10.14288/1.0089636"@en . "eng"@en . "Experimental Medicine"@en . "Vancouver : University of British Columbia Library"@en . "University of British Columbia"@en . "For non-commercial purposes only, such as research, private study and education. Additional conditions apply, see Terms of Use https://open.library.ubc.ca/terms_of_use."@en . "Graduate"@en . "Genetic factors associated with disease severity in rheumatoid arthritis"@en . "Text"@en . "http://hdl.handle.net/2429/10814"@en .