CYTOKINES AND THE REGULATION OF HEAT SHOCK PROTEINS IN SYSTEMIC LUPUS ERYTHEMATOSUS
by
Barry James Marc Ripley B.Sc., PGDlp (LSHTM), M.Sc.
Thesis submitted for the degree of Ph D.
Institute of Child Health, University College London
2001 ProQuest Number: U643772
All rights reserved
INFORMATION TO ALL USERS The quality of this reproduction is dependent upon the quality of the copy submitted.
In the unlikely event that the author did not send a complete manuscript and there are missing pages, these will be noted. Also, if material had to be removed, a note will indicate the deletion. uest.
ProQuest U643772
Published by ProQuest LLC(2016). Copyright of the Dissertation is held by the Author.
All rights reserved. This work is protected against unauthorized copying under Title 17, United States Code. Microform Edition © ProQuest LLC.
ProQuest LLC 789 East Eisenhower Parkway P.O. Box 1346 Ann Arbor, Ml 48106-1346 To my family for all their love and support Abstract
Previous studies have reported elevated levels of the 90 kDa heat shock protein (Hsp 90) and autoantibodies to Hsp 90 in a subset of patients with systemic lupus erythematosus (SLE). However, the mechanisms leading to the overexpression of Hsp 90 in SLE and a link between this event and the induction of Hsp 90 autoantibodies, have remained unclear. Recent studies have shown that the cytokine IL-6, which is elevated in SLE, activates Hsp 90 gene expression in cultured cells via specific transcription factors, that include STAT-3. In view of the known role of STAT proteins in signalling pathways of the cytokine IL-10, the effect of IL-10 on Hsp 90 levels in different cells was investigated. IL-10 enhanced the expression of Hsp 90 in cultured cells. In reporter gene assays, IL=10 activated both the Hsp 90a and Hsp 90p promoters directly. Activation of the Hsp 90p promoter was mediated by a short specific region of the promoter which contains a binding site for STAT-3. Cytokines and Hsps, were quantified in individual blood samples from patients with SLE. Levels of Hsp 90 positively correlated with levels of IL-6 but not IL-10, in patients with SLE. Furthermore, this correlation was enhanced in patients who overexpressed Hsp 90 (a subset of patients associated with HLA DR3 negativity). A link between Hsp 90 overexpression and the development of Hsp 90 autoantibodies was also investigated. Antibodies to Hsps in patients with SLE were quantified. Overexpression of Hsp 90 in SLE correlated with raised levels of antibodies to Hsp 90 in the same patients. Furthermore, antibodies to Hsp 90 were associated with active general disease features of SLE. These results support a role for IL-6 in the overexpression of Hsp 90, leading to raised levels of antibodies to Hsp 90 and active disease in SLE. Table of Contents
Page Number
List of Figures ...... 12 List of Tables ...... 18 Abbreviations ...... 23 List of Publications ...... 25 Acknowledgements ...... 26
Chapter 1. Introduction
Section 1.1: The Heat Shock Proteins
Introduction ...... 28 Heat Shock Protein Families and Biological Functions ...... 29 The Hsp 90 Family...... 31 The Hsp 70 Family...... 38 The Hsp 60 Family...... 40 Small Hsps and Ubiquitin ...... 41 Conditions that Induce the Expression of Heat Shock Proteins ...... 43 Transcriptional Regulation of Heat Shock Protein Expression ...... 45 Cytokines ...... 47 Interleukin-10...... 48 Interleukin-6...... 48 IL-10 and IL-6 Activate the STAT Family of Transcription Factors...... 50 Page Number IL‘6 but not IL=10, Activates Transcription Factors of the C/EBP Family...... 53 Role of Interleukin-6 in Hsp 90 Gene Regulation...... 53 Summary...... 54
Section 1.2: The Role Of Heat Shock Proteins in the Immune System and Pathogenesis of Autoimmune Disease
Introduction ...... 56 Self/Non-Self Discrimination Within the Immune System...... 57 Lymphocytes...... 57 B Cells...... 58 T Cells...... 59 Thymic Selection of T Cells and Self-Tolerance ...... 60 Heat Shock Proteins and the Immune System ...... 61 Antigen Processing and Presentation ...... 61 Heat Shock Proteins and Infection ...... 62 Cell Surface Expression of Heat Shock Proteins...... 63 Heat Shock Proteins and Autoimmune Disease ...... 64 Antibodies to Heat Shock Proteins ...... 64 T Cell Responses to Heat Shock Proteins ...... 68 Summary...... 69
Section 1.3: Systemic Lupus Erythematosus
Introduction ...... 70 Definition and Classification ...... 70 Epidemiology ...... 73 Prevalence ...... 73 Page Number Clinical Features of SLE ...... 74 Non-Specific Features ...... 75 Dermatological Involvement ...... 75 Musculoskeletal Involvement ...... 75 Renal Disease...... 76 Involvement of the Nervous System ...... 77 Haemopoietic Involvement ...... 77 Cardio-Respiratory/-V ascular Involvement ...... 78 Management ...... 79 Prognosis and Survival ...... 79 Immunopathology in SLE ...... 80 Genetic Susceptibility ...... 80 Sex Hormones...... 81 Major Histocompatibility Complex...... 82 Complement Dejîciencies...... 83 Immune Dysregulation and Cellular Abnormalities in SLE ...... 85 B Lymphocytes...... 86 Cytokines...... 86 Apoptosis...... 89 Autoantibodies...... 90 Experimental Models of SLE...... 91 Summary ...... 93
Section 1.4: The Role of Hsps In SLE
Introduction ...... 94 The Expression of Heat Shock Proteins in SLE ...... 94 Surface Expression of Hsp 90 in SLE ...... 97 Page Number Antibodies to Hsps in SLE ...... 97 Hsp 90 Expression in MRUlpr/lpr Mice...... 98
Section 1.5: Aims of Dissertation ...... 101
Chapter 2. Materials and Methods
Section 2.1: Analysis of Hsp 90 Gene Expression
in Cultured Cells
Materials...... 103 Laboratory Chemicals ...... 103 Monoclonal Antibodies ...... 103 Cytokines ...... 104 Cells and Cell Lines ...... 104 The HepG2 Cell Line...... 104 Peripheral Blood Mononuclear Cells...... 104 Plasmids ...... 104 Enzymes ...... 105 Methods ...... 105 Isolation of PBMCs ...... 105 Cell Culture...... 105 Stimulation ...... 107 Preparation of Whole Cell Lysate ...... 107 Protein Electrophoresis and Western Blotting ...... 107 Development of Western Blots ...... 108 Bacterial Strains and Growth Conditions ...... 108 Transformation of E.coli...... I ll Page Number Small Scale Plasmid DNA Extraction from E.coli...... 112 Large Scale Plasmid DNA Extraction from E.coli...... 114 Quantitative Determination of Plasmid DNA ...... 115 Restriction Analysis of Plasmid DNA ...... 115 Agarose Gel Electrophoresis ...... 116 Transfection and Stimulation of Hsp 90 Promoter=CAT Constructs ...... 117 Chloramphenicol Acetyl Transferase (CAT) Assay ...... 118
Section 2.2: Quantitative Determination of Heat Shock Proteins in SLE
Materials...... 120 Monoclonal Antibodies ...... 120 PBMCs...... 120 Patients and Normal Individuals ...... 120 Methods ...... 123 Preparation of Whole Cell Lysate from PBMCs ...... 123 Protein Electrophoresis and Western Blotting ...... 123 Development of Western Blots ...... 123 Calculation of Heat Shock Protein Levels ...... 124 Accuracy of the Procedure for the Quantitative Determination of Hsps ...... 124
Section 2.3: Quantitative Determination of Cytokines in SLE
Quantitative Determination of Interleukin-6 ...... 129 Principle of the Assay ...... 129 Page Number Colour Development ...... 129 Experimental Measurements ...... 130 Materials...... 130 Methods ...... 131 Summary of the Assay Procedure...... 131 Calculation of Results...... 132 Precision...... 132 Calibration of Sample Values...... 132 Quantitative Determination of Interleukin-10 ...... 133 Principle of the Assay ...... 133 Materials...... 133 Methods ...... 134 Summary of the Assay Procedure...... 134 Calculation of Results...... 135 Precision...... 135
Section 2.4: ELISA for Anti-Hsp Antibodies
Principle of the Assay ...... 136 Materials...... 136 Methods ...... 137 ELISA for Anti-Hsp 90 Antibodies...... 137 ELISA for AntUHsp 70 Antibodies...... 138 Calculation of Results...... 138
Section 2.5: Disease Associations...... 139
Section 2.6: Statistical Analyses ...... 141 Page Number Chapter 3. Interleukin-10 Activates Hsp 90p Gene Expression
3.0 Introduction ...... 144 3.1 IL-10 Up-Regulates Hsp 90 in HepG2 Cells and PBMCs ...... 146 3.2 IL-10 Activates the Hsp 90a and -p Promoters ...... 149 3.3 IL-10 Activates the Hsp 90p Promoter Via a STAT-3 Binding Site ...... 149 3.4 Discussion ...... 155
Chapter 4. Cytokines and Heat Shock Protein Gene Expression in SLE
4.0 Introduction ...... 157 4.1 Quantification of Heat Shock Proteins ...... 159 4.2 Pattern of Heat Shock Protein Expression in SLE ...... 159 4.3 Cytokine Levels in SLE ...... 173 4.4 Correlation of Cytokine Levels with Heat Shock Protein Levels in SLE ...... 183 4.5 Discussion ...... 193
Chapter 5. Elevated Expression of Hsp 90 and Antibodies to Hsp 90 in SLE
5.0 Introduction ...... 196
10 Page Number 5.1 Detection of Antibodies to Hsps ...... 198 5.2 Pattern of Expression of Antibodies to Hsps in SLE ...... 198 53 Immunogenetic Analysis of Antibodies to Hsps ...... 206 5.4 Correlation of Hsp Levels with Antibodies to Hsps ...... 211 5.5 Correlation of Levels of Cytokines and Anti-Hsp Antibodies in SLE ...... 220 5.6 Discussion ...... 234
Chapter 6. Disease Associations
6.1 Introduction ...... 236 6.2 Heat Shock Proteins ...... 238 6.3 Associations of HLA with Active Clinical Subsets in SLE ...... 248 6.4 Cytokines ...... 255 6.5 Antibodies to Heat Shock Proteins ...... 263 6.6 Discussion ...... 271
Chapter 7. General Discussion ...... 275
References...... 283
Appendix A: The BILAG Index...... 335
Published Papers: (pocket attached to back cover)
11 List of Figures
Page Number Chapter 1. Introduction
Section 1.1: The Heat Shock Proteins
Figure 1.1.1 The Domain Structure of Hsp 90 and a Schematic Representation of the Native Dimeric Form ...... 32 Figure 1.1.2 The Hsp 90 Chaperone Complex Involved in Steroid Hormone Receptor Activation ...... 37 Figure 1.1.3 Inducers of the Heat Shock Response ...... 44 Figure 1.1.4 IL-10 Receptor Signalling via the JAK/STAT Pathway...... 49 Figure 1.1.5 IL-6 Receptor Signalling via the JAK/STAT Pathway...... 51
Section 1.4: Heat Shock Proteins in Systemic Lupus Erythematosus
Figure 1.4.1 A Model for Hsp Overexpression and Autoimmunity in SLE ...... 100
Chapter 2. Materials and Methods
Section 2.1: Analysis of Hsp 90 Gene Expression in Cultured Cells
Figure 2.1 Hsp 90 Promoter Sequences in PBLCAT2 ...... 106 Figure 2.2 Principle of ECL Western Blotting ...... 110
12 Page Number Section 2.2: Quantitative Determination of Heat Shock Proteins in SLE
Figure 2.3 Autoradiograph of an Hsp 90 Immunoblot ...... 125 Figure 2.4 Autoradiograph of an Hsp 70 Immunoblot ...... 126 Figure 2.5 Autoradiograph of an Actin Immunoblot ...... 127
Chapter 3. The Effect of Interleukin-10 on Hsp 90 Gene Expression
Section 3.1: IL-10 Up-Regulates Hsp 90 in HepG2 Cells and PBMCs
Figure 3.1 IL-10 Up-Regulates Hsp 90 in HepG2 Cells and PBMCs ...... 147
Section 3.2: IL-10 Activates the Hsp 90a and -p Gene Promoters
Figure 3.2 Assay of CAT activity in HepG2 Cells Transfected with Hsp 90a and -p Promoter-CAT Constructs ...... 150
Section 3.3: IL-10 Activates the Hsp 90p Promoter Via a STAT-3 Binding Site
Figure 3.3 IL-10 Activates the Hsp 90p Promoter Via a STAT-3 Binding Site ...... 152
13 Page Number Chapter 4. Cytokines and Heat Shock Protein Gene Expression in SLE
Section 4.2: Pattern of Heat Shock Protein Expression in SLE
Figure 4.2.1 Hsp 90 Levels in SLE and Control Groups ...... 163 Figure 4.2.2 Hsp 70 Levels in SLE and Control Groups ...... 164 Figure 4.2.3 Correlation Between Hsp 90 and Hsp 70 Levels in SLE ...... 167 Figure 4.2.4 Hsp 90 Mean (and 95% Cl of mean) with Respect to HLA Alleles/Haplotypes ...... 168 Figure 4.2.5 Hsp 70 Mean (and 95% Cl of mean) with Respect to HLA Alleles/Haplotypes ...... 169
Section 4.3: Cytokine Levels in SLE
Figure 4.3.1 Scattergraph of IL-6 levels in SLE ...... 177 Figure 4.3.2 Scattergraph of IL-10 levels in SLE ...... 178 Figure 4.3.3 IL-6 Mean (and 95% Cl of mean) with Respect to HLA Alleles/Haplotypes ...... 180 Figure 4.3.4 IL-10 Mean (and 95% Cl of mean) with Respect to HLA Alleles/Haplotypes ...... 181
Section 4.4: Correlation of Cytokine Levels with Heat Shock Protein Levels in SLE
Figure 4.4.1 Correlation Between Levels of IL-6 and Hsp 90 in Individual Samples from Patients with SLE ...... 184 Figure 4.4.2 Correlation Between Levels of IL-6 and Hsp 70 in Individual Samples from Patients with SLE ...... 185
14 Page Number Figure 4.4.3 Correlation Between Levels of IL-10 and Hsp 90 in Individual Samples from Patients with SLE ...... 186 Figure 4.4.4 Correlation Between Levels of IL-10 and Hsp 70 in Individual Samples from Patients with SLE ...... 187 Figure 4.4.5 Hsp 90 Mean (and 95% Cl of Mean) with Respect to IL-6 Levels in Normals and SLE ...... 188 Figure 4.4.6 IL-6 Mean (and 95% Cl of Mean) with Respect to Hsp 90 Levels in Normals and SLE ...... 189 Figure 4.4.7 Correlation of IL-6 and Hsp 90 in Patients who Overexpress Hsp 90 ...... 192
Chapter 5. Elevated Expression of Hsp 90 and Antibodies to Hsp 90 in SLE
Section 5.2: Pattern of Expression of Antibodies to Hsps in SLE
Figure 5.2.1 Levels of IgG Antibodies to Hsp 90 in SLE and Control Groups ...... 201 Figure 5.2.2 Levels of IgM Antibodies to Hsp 90 in SLE and Control Groups ...... 202 Figure 5.2.3 Levels of IgG Antibodies to Hsp 70 in SLE and Control Groups ...... 203 Figure 5.2.4 Levels of IgM Antibodies to Hsp 70 in SLE and Control Groups ...... 204
Section 5.4: Correlation of Hsp Levels with Antibodies to Hsps in SLE
Figure 5.4.1 Comparison of Levels of Hsp 90 with Levels of IgG Antibodies to Hsp 90 in SLE ...... 215
15 Page Number Figure 5.4.2 Comparison of Levels of Hsp 90 with Levels of IgM Antibodies to Hsp 90 in SLE ...... 216 Figure 5.4.3 Comparison of Levels of Hsp 70 with Levels of IgG Antibodies to Hsp 70 in SLE ...... 217 Figure 5.4.4 Comparison of Levels of Hsp 70 with Levels of IgM Antibodies to Hsp 70 in SLE ...... 218
Section 5.5: Correlation of Levels of Cytokines and Antifoodie» to Hsps in SLE
Figure 5.5.1 Correlation of IL-6 and Hsp 90 IgG Antibodies in SLE ...... 223 Figure 5.5.2 Correlation of IL-6 and Hsp 90 IgM Antibodies in SLE ...... 224 Figure 5.5.3 Correlation of IL-6 and Hsp 70 IgG Antibodies in SLE ...... 225 Figure 5.5.4 Correlation of IL-6 and Hsp 70 IgM Antibodies in SLE ...... 226 Figure 5.5.5 Correlation of IL-10 and Hsp 90 IgG Antibodies in SLE ...... 229 Figure 5.5.6 Correlation of IL-10 and Hsp 90 IgM Antibodies in SLE ...... 230 Figure 5.5.7 Correlation of IL-10 and Hsp 70 IgG Antibodies in SLE ...... 231 Figure 5.5.8 Correlation of IL-10 and Hsp 70 IgM Antibodies in SLE ...... 232
16 Page Number Chapter 6. Disease Associations
Section 6.2: Heat Shock Proteins
Figure 6.2.1 Mean Hsp 90 Levels (-2SEM to +2SEM) in Patients Graded by the BILAG Index ...... 242 Figure 6.2.2 Mean Hsp 70 Levels (-2SEM to +2SEM) in Patients Graded by the BILAG Index ...... 243 Figure 6.2.3 Mean Hsp 90 Levels (-2SEM to +2SEM) in Patients who Overexpress Hsp 90 and Graded by the BILAG Index ...... 246
Section 6.4: Cytokines
Figure 6.4.1 Mean IL-6 Levels (-2SEM to +2SEM) in Patients Graded by the BILAG Index ...... 257 Figure 6.4.2 Mean IL-10 Levels (-2SEM to +2SEM) in Patients Graded by the BILAG Index ...... 258 Figure 6.4.3 Mean IL-6 Levels (-2SEM to +2SEM) in Patients who Overexpress Hsp 90 and Graded by the BILAG Index ...... 261
Chapter 7. General Discussion
Figure 7.1 A Pathogenic Role for Hsp 90 in SLE ...... 282
17 List of Tables
Page Number Chapter 1. Introduction
Section 1.1: The Heat Shock Proteins
Table 1.1.1 Mammalian Heat Shock Proteins ...... 30 Table 1.1.2 Examples of Hsp 90 Client Proteins ...... 35
Section 1.2: The Role of Heat Shock Proteins in the Inunune System and Pathogenesis of Autoimmune Disease
Table 1.2.1 Immune Response to Hsps in Human Autoimmune Diseases ...... 65
Section 1.3: Systemic Lupus Erythematosus
Table 1.3.1 Revised Criteria of the American Rheumatism Association for the Classification of SLE ...... 72 Table 1.3.2 Chromosomal Locations of Hsp Genes in Humans and Mice ...... 84 Table 1.3.3 Features of Lupus-Prone Mouse Strains ...... 92
Chapter 2. Materials and Methods
Section 2.1: Analysis of Hsp 90 Gene Expression in Cultured Cells
18 Page Number Table 2.1 Antibody Conditions ...... 109
Section 2.2: Quantitative Determination of Heat Shock Proteins in SLE
Table 2.2 Age and Sex Distribution of Patients with SLE, RA and Normal Individuals ...... 121 Table 2.3 Ethnic Origin of Patients with SLE, RA and Normal Individuals ...... 122
Chapter 3. The Effect of Interleukin-10 on Hsp 90 Gene Expression
Table 3.1 The Effect of IL-10 on the Activity of Hsp 90 Promoter-CAT Constructs ...... 154
Chapter 4. Cytokines and Heat Shock Protein Gene Expression in SLE
Section 4.2: Pattern of Heat Shock Protein Expression in SLE
Table 4.2.1 Hsp Levels in PBMCs of Controls and Patients with SLE ...... 160 Table 4.2.2 Upper Limits of Normal Hsp Levels and Number of Patients with SLE or RA with Elevated Hsp Levels ...... 162 Table 4.2.3 Association of HLA Alleles/Haplotypes with Hsp 90 Elevation in SLE ...... 171 Table 4.2.4 Association of HLA DR3 Status with Hsp 90 Elevation in SLE ...... 172
19 Page Number Section 4.3: Cytokine levels in SLE
Table 4.3.1 Cytokine Levels in PBMCs of Normals and Patients with SLE or RA...... 174 Table 4.3.2 Upper Limits of Normal Cytokine Levels and Number of Patients with SLE or RA with Elevated Levels of Cytokines ...... 176
Chapter 5. Elevated Expression of Hsp 90 and Antibodies to Hsp 90 in SLE
Section 5.2: Pattern of Expression of Antibodies to Hsps in SLE
Table 5.2.1 Upper Limits of Normal Levels of Antibodies to Hsps and Number of Patients with SLE or RA with Raised Antibody Levels ...... 199
Section 5.3: Immunogenetic Analysis of Antibodies to Hsps
Table 5.3.1 Immunogenetic Analysis of IgG Anti-Hsp 90 Antibodies in SLE ...... 207 Table 5.3.2 Immunogenetic Analysis of IgM Anti-Hsp 90 Antibodies in SLE ...... 208 Table 5.3.3 Immunogenetic Analysis of IgG Anti-Hsp 70 Antibodies in SLE ...... 209
20 Page Number Table 5.3.4 Immunogenetic Analysis of IgM Anti-Hsp 70 Antibodies in SLE...... 210
Section 5.4: Correlation of Hsp Levels with Antibodies to Hsps in SLE
Table 5.4.1 Number of Patients with SLE with Elevated Hsps and Raised Levels of Antibodies to Hsps ...... 212 Table 5.4.2 Chi-Square Test for the Correlation of Hsp 90 Overexpression with Raised IgG Antibodies to Hsp 90 in SLE ...... 214
Section 5.5: Correlation of Levels of Cytokines and Antibodies to Hsps in SLE
Table 5.5.1 Correlation Between Levels of IL-6 and Antibodies to Hsps in SLE ...... 222 Table 5.5.2 Correlation Between Levels of IL-10 and Antibodies to Hsps in SLE ...... 228
Chapter 6. Disease Associations
Section 6.2: Heat Shock Proteins
Table 6.2.1 Number of Patients with SLE with Normal or Elevated Levels of Hsp 90 and Graded by the BILAG Index ...... 239
Section 6.3: Associations of HLA with Active Clinical Subsets in SLE
Table 6.3.1 Associations of HLA with Active Clinical Subsets in SLE ...... 249
21 Page Number Table 6.3.2 Chi-Square Tests for the Correlation of HLA Alleles/Haplotypes with Active Clinical Subsets in SLE ...... 254
Section 6.4: Cytokines
Table 6.4.1 Correlation of IL-6 and IL-10 with Global BILAG Disease Activity Scores in SLE ...... 256
Section 6.5: Antibodies to Heat Shock Proteins
Table 6.5.1 Number of Patients with SLE with Normal or Raised Levels of Antibodies to Hsp 90 and Graded by the BILAG Index ...... 264 Table 6.5.2 Chi-Square Test for the Correlation of Active BILAG General Disease Features with Raised Levels of IgG Antibodies to Hsp 90 in SLE ...... 267 Table 6.5.3 Number of Patients with SLE with Normal or Raised Levels of Antibodies to Hsp 70 and Graded by the BILAG Index ...... 268
22 Abbreviations
APS - ammonium persulphate
ATP - adenosine triphosphate a.u. - arbitrary units
CAT - chloramphenicol acetyltransferase Cl confidence interval
DNA - deoxyribonucleic acid
E.coli - Esherichia colt g - grammes hr - hours
Ig - immunoglobulin
kDa - kilodaltons
M - molar
mM - millimolar
mn - minutes
mg - milligrammes mRNA - messenger ribonucleic acid MS - multiple sclerosis n - number of subjects
ng - nannogrammes OD - optical density PAGE polyacrylamide gel electrophoresis Pg - picogrammes
P - P value
r - correlation coefficient RA - rheumatoid arthritis
RNA - ribonucleic acid
SDS - sodium dodecyl sulphate SB standard error
SLE - systemic lupus erythematosus
23 UV - ultra violet w/v - weight to volume ratio °c - degrees Celsius Jig - microgrammes jil - microlitres
24 List of Publications
Ripley, B. J., Stephanou, A., Isenberg, D. A., and Latchman, D. S. (1999). Interleukin-10 activates heat-shock protein 90beta gene expression. Immunology 97,226-31.
Ripley, B. J., Stephanou, A., Isenberg, D. A., and Latchman, D. S. (2001). Elevated levels of the 90kDa heat shock protein (hsp90) in SLE correlate with levels of IL-6 and autoantibodies to hsp90. (accepted for publication in Journal of Autoimmunity)
25 Acknowledgements
I would like to thank my supervisors Professor David Latchman and Professor David Isenberg for accepting me as a Ph.D. student. I greatly valued this opportunity every day of my time here. Thankyou for the guidance and support I have received, it has meant so much to me. Many thanks to all the members of the Medical Molecular Biology Unit who volunteered to give me blood and for your friendship. Finally, I very much hope the findings in this dissertation contribute to our understanding of lupus and one day to the rational design of a successful therapy.
26 Chapter 1 Introduction
27 Introduction
Section 1.1: The Heat Shock Proteins
Introduction
Since the discovery of heat shock proteins nearly 40 years ago, it is now clear that this group of proteins perform critical cellular roles under physiological and stressful conditions. Studies which have focussed on the importance of these proteins to cell survival, have been extended in recent years to suggest a role for Hsps in the pathogenesis of a number of human autoimmune diseases. The heat shock response was first reported in 1962 by Ritossa. In this study, exposure of the giant salivary gland chromosomes of the fruitfly Drosophila buksii to elevated temperature, resulted in areas of swelling known as puffs. These puffs, which could also be induced by sodium salicylate and dinitrophenol, were presumed to occur as a result of specific and rapid transcriptional activity (Ritossa, 1962). In 1974, the phenomenon of Drosophila chromosome puffing was further characterised by Tissières. Following heat shock or treatment with the hormone ecdysone, changes in chromosome puffing patterns were observed to correlate with temporal increases in specific polypeptide synthesis, against a background of general protein synthesis repression (Tissieres et al., 1974). The novel set of proteins described in this study are now collectively referred to as heat shock proteins (Hsps). Following observations with Drosophila, Hsps have been shown to be expressed in prokaryotes, eukaryotes and plant cells. Importantly, in all cell systems studied, levels of Hsps are increased not only by elevated temperature (heat shock), but also by a variety of other environmental stresses including viral/microbial infections, heavy metals, toxins and amino acid analogues, as well as by non-stressful conditions including cytokines and growth factors (Lindquist, 1986), (Polla, 1988). A critical role for Hsps in cell survival following exposure to various stressful conditions, appears to be an extension of their role in normal cells during non- stressful conditions. Thus, some Hsps are expressed at constitutively high levels in cells and have roles in normal cell growth and development (Bond and Schlesinger, 1987), while other Hsps are present at much lower levels in cells and their synthesis
28 Introduction is elevated in response to stressful conditions. A common link between these studies, is an association of Hsp function with the accumulation of unfolded or malfolded proteins (Gething and Sambrook, 1992).
Heat Shock Protein Families And Biological Functions
The importance of Hsps to cell survival is evidenced by a high degree of homology at the genomic level observed for Hsps between species, from bacteria to mammals (Lindquist and Craig, 1988). Thus, shared structural and cellular functions of Hsps are observed between highly diverse species and these have been conserved throughout evolution. The earliest studies of Hsp cell biology highlighted their increased synthesis in response to heat and other stresses. Studies over the last decade however, have shifted the focus of attention to understanding the role of Hsps as molecular chaperones, assisting the processes of protein synthesis, protein folding and degradation, as well as protein translocation within the cell. Many Hsps and their associated co-chaperones are constitutively expressed in cells. A number of multi=gene Hsp families are recognised and individual genes within families are observed to differ with respect to their sequence and pattern of expression, as well as their function within the cell and their sub-cellular localisation. Heat shock proteins are classified into families according to their approximate molecular weights (in kilodaltons). In mammals, the major Hsp families are designated Hsp 90 (approximately 90 kDa Hsps), Hsp 70 (approximately 70 kDa Hsps), Hsp 60 (approximately 60 kDa Hsps), small Hsps (22-32 kDa Hsps) and ubiquitin (approximately 8 kDa). In this section, the classification, structure and cellular roles of these Hsp families are outlined and a summary of the major roles of mammalian Hsps is provided in Table 1.1.1.
29 Introduction
Table 1.1.1 Mammalian Heat Shock Protein Families
Family Members Characteristics / Biological Functions
Hsp 90 Hsp 90a Hsp 90a more heat inducible than Hsp 90p. Hsp 90p In mammals it is the most abundant cytoplasmic Hsp.
Together, both isoforms constitute 1-2% of total cytoplasmic protein. Regulatory interactions with signalling proteins. Involved in cell growth and development and antigen processing/presentation. Hsp 90 Binds ATP, inherent ATPase activity.
GRP94 Glucose regulated protein (GRP). Located at plasma/nuclear membranes and within Golgi.
Hsp 70 Hsp 72 Typically stress-induced. Hsp 73 Largely constitutively expressed.
Located in cytoplasm and nucleus. Roles in protein biosynthesis (transport and folding) and cell cycle control. Bind ATP, ATPase activity.
GRP78 Present in ER and lysosomes. Expressed at high levels in secretory cells. Roles in protein import, modification and assembly. GRP75 Present in mitochondria. Similar role to GRP78.
Hsp 60 Present in mitochondria as large oligomers. Roles in protein assembly and cellular respiration. Binds ATP, weak ATPase activity.
Small Hsps Hsp 27 Oligomeric structure which re-distributes inside or around nucleus upon heat shock. Role in growth and differentiation in normal and cancerous cells associated with involvement in actin filament dynamics.
Ubiquitin Located in cytoplasm and nucleus. Role in conjugation to stress-damaged proteins, facilitating degradation by enzymes of an ubiquitin-dependent pathway. Introduction
The Hsp 90 Family
The Hsp 90 family is a group of proteins localised within the cell cytoplasm of eubacteria, yeast and multicellular organisms. Homologues have been characterised and are found in the endoplasmic reticulum (ER) of eukaryotes (known as Grp94) and in chloroplasts (known as cpHsp82). Hsp 90 is a highly abundant cytoplasmic protein and is estimated to comprise 1=2% of total soluble protein in eukaryotic cells (Welch and Feramisco, 1982) and its expression in yeast (Borkovich et al., 1989) and Drosophila melanogaster (Cutforth and Rubin, 1994), is essential. Two isoforms of Hsp 90, which are encoded by separate genes termed Hsp 90a and Hsp 90p, are found in eukaryotic cells and are differentially expressed. Both isoforms share a high degree of amino acid sequence identity to each other; approximately 85% sequence identity in humans (Hickey et al., 1989) and approximately 90% sequence identity in yeast (Borkovich et al., 1989). A further cytoplasmic Hsp 90 family member, termed TRAPl, is found in higher eukaryotes and is present at much lower levels compared to both Hsp 90a and =p (Felts et al., 2000). Studies of Hsp 90 have examined the native protein’s modular structure. Two highly conserved regions, an N-terminal domain (approximately 25 kDa) and a C- terminal region (approximately 55 kDa), are connected by a highly charged linker, which can be of variable length between species and isoforms. The complete absence of this charged region in some prokaryotic and mammalian Hsp 90 family members (see Figure 1.1.1a), would suggest it is not important for Hsp 90 function. Hsp 90 exists predominantly as a dimer in vivo and this occurs through the anti-parallel association of C-terminal regions of each monomer (shown in Figure 1.1.1b). The native forms of Hsp 90 have been studied in the mouse, using nomdenaturing gel electrophoresis (Minami et al., 1991) and in the rat liver cytosol, using isoform- specific monoclonal antibodies (Nemoto and Sato, 1998). These studies showed that the purified Hsp 90 existed predominantly as a dimer, but a considerable amount of monomer was also detected. In addition, the dimeric forms were found to consist of alpha/alpha (a/a) and beta/beta (p/p) homodimers and to a lesser extent alpha/beta (a/p) heterodimers. The monomeric form was found to consist mainly of the beta (P)
31 Figure 1.1.1 The Domain Structure of Hsp 90 and a Schematic Representation of the Native Dimeric Form, adapted from Buchner, (1999)
(a)
I ATP MEEVD Hsp 90a (human)
726
I I ATP TRAPl (human)
645
I — ATP- Hsp 90p (yeast)
704
i A'tP HtpG (Exoli)
624
(b)
Binding of ATP/ Heat Shock
(a) Domain structure of Hsp 90. The charged linker is shown in black and is much reduced in length (shown as a thin black line) in some eukaryotic and prokaryotic Hsp 90 species. Highly conserved regions are shown in light and dark grey. (b) Conformational flexibility of the native dimeric form of Hsp 90. In response to heat shock or binding of ATP (A), an open structure with N-terminal domains (N) pointing in opposite directions, changes to one which is closed and toroidal in shape. Introduction isoform. In mammals, both of the Hsp 90 isoforms have been shown to exist as oligomeric structures in the cytoplasm under both physiological and stressful cell conditions (Nemoto and Sato, 1998). Crystal structures for N-terminal domains of human (Prodromou et al., 1997a) and yeast Hsp 90 (Prodromou et al., 1997b) have been reported. These studies show the N-terminal domain to contain binding sites for nucleotide, involved in ATP binding and hydrolysis (Prodromou et al., 1997a) and the anti-tumour drug geldanamycin (which inhibits Hsp 90 function). Polypeptide binding sites have been shown to exist in both the N-terminal and C-terminal domains, with each displaying a different substrate preference (Young et al., 1997), (Scheibel et al., 1998). The overall structure of the N-terminal domain consists of an eight-stranded anti-parallel p-sheet and nine a-helices which together form an a-p sandwich. In the yeast Hsp 90 dimer, the opposing faces of the p-sheets have been suggested to form a potential peptide-binding cleft. Thus, the N-terminal domain may act as a molecular “clamp” for the binding of Hsp 90 substrates (Prodromou et al., 1997b). To date however, no structures for either a C-terminal domain or an intact Hsp 90 molecule, have been determined. Antibody and electron microscopy studies have detected conformational flexibility in the native dimeric structure. As shown in Figure 1.1.1b, heat shock or binding of ATP causes specific structural transformations, in which the relative orientation of N-terminal domains of each monomer, switches from an open conformation to one which is closed and toroidal in shape (Maruya et al., 1999). The abundance of Hsp 90 in eukaryotic cells and its elevated expression in response to various stimuli, suggest an important role for Hsp 90 during physiological and stressful conditions. Molecular chaperones are polypeptide-binding proteins which facilitate protein folding by shifting the equilibrium of the folding reaction towards a productive folded state and away from the aggregated state (Hartl, 1996), (Bukau and Horwich, 1998). Hsp chaperones, including Hsp 90 and Hsp 70, are unable to perform such a role on their own, but function within multi-component complexes involving co-chaperones and their co-factors. In comparison with other Hsp families (particularly Hsp 70), an understanding of the function of Hsp 90 in vivo
33 Introduction
remains obscure. To date, the majority of studies on Hsp 90, have revealed and focussed on Hsp 90 roles in the conformational regulation of proteins involved in signal transduction and cell-cycle control. These observations contrast with those of other Hsp families, such as Hsp 70 and Hsp 60, which in comparison with Hsp 90, display far less specificity for the type of protein with which they interact. However, the list of known Hsp 90 substrates is steadily growing (see Table 1.1.2) and these advances now suggest the Hsp 90 chaperone machinery to have a multifunctional role in protein folding in the cell and thus a more dynamic involvement than previously thought (Buchner, 1999). In particular, the recent observation (Panaretou et al., 1998) that Hsp 90 like other Hsp families, can function in an ATP-dependent manner (previously thought not to be the case), exemplifies the growing understanding of Hsp 90 function and the scope of possible roles for Hsp 90 as a molecular chaperone in the cell (discussed in Grenert et al., 1999). Hsp 90 interacts with two classes of “ classical ” Hsp 90 “client” proteins, the steroid-hormone receptors and kinases and the list of other Hsp 90 substrates is gradually increasing. As shown in Table 1.1.2, proteins whose folding is known to be affected by Hsp 90 not only include transcription factors such as heat shock factor (Zou et al., 1998) and kinases such as casein kinase 2 (Miyata and Yahara, 1995), but also proteins with functions unrelated to these “classical clients”, such as the microtubule cytoskeletal component tubulin (Gamier et al., 1998) and the enzyme nitric oxide synthase (Garcia-Cardena et al., 1998). One of the most studied roles of Hsp 90 in protein folding within the cell, involves the interaction between Hsp 90 and the steroid hormone receptors (SHRs). The SHRs comprise three separate domains which are separately involved in hormone binding, DNA binding and transcriptional activation (Evans, 1988). In this respect SHRs may be considered as direct signal transducers. Thus, in the hormone- activated state, SHRs enhance gene transcription by binding to specific DNA sequences in target genes. Studies in the 1980s established the involvement of Hsp 90 in the conformational regulation of steroid hormone receptors and is reviewed in Pratt and Toft, (1997). More recently, a greater understanding of this process has been achieved through the identification of components of the Hsp 90 chaperone complex.
34 Introduction
Table 1.1.2 Examples of Hsp 90 Client Proteins
Characteristic Client Protein
Protein Kinases PI-4 kinase v-Src Raf-1 Mek elF-2a kinase calcineurin casein kinase 2 Stell Gcn2
Receptors/Transcription FactorsGlucocorticoid receptor Progesterone receptor Eostrogen receptor Androgen receptor Mineralocorticoid receptor P53 Heat shock factor
Other Insulin Tubulin Citrate synthase Nitric oxide synthase Tumour necrosis factor P-Galactosidase Luciferase Dihydrofolate reductase Introduction
The conformational changes in SHRs which allow binding of hormone to the receptor to take place, involve the interaction of several Hsp family chaperone complexes (see Figure 1.1.2). The sequence of events which lead to a SHR in the hormone-activated state, begins with the interaction of the SHR with Hsp 70 and its co-factors. Following this, an intermediate complex forms through the interaction of an Hsp 90 chaperone complex. This appears to be facilitated via the scaffolding protein Hop, which has been demonstrated to be required for the interaction of both chaperone families (Frydman and Hohfeld, 1997). The formation of a mature complex, in which SHRs can bind hormone, requires the dissociation of Hsp 70 and its co-factors Hip and Hsp 40. The binding and hydrolysis of ATP at a site within the N-terminal domain of Hsp 90 (Prodromou et al., 1997a), allows a mammalian protein (p23) and a member of the immunophilin protein family, to bind the Hsp 90-SHR complex (Grenert et al., 1997). Finally, the SHR is released from the Hsp 90 complex, it is then able to bind hormone, dimerise and subsequently bind to DNA. As shown in Table 1.1.2, Hsp 90 also interacts with transcription factors that are not steroid receptors, including heat shock factor (HSF), the p53 tumour suppressor protein, and the dioxin receptor. The tumour suppressor protein p53, is a transcription factor whose mutant form is often found in tumour cells and Hsp 90 has been observed to form a heterocomplex with mutant p53 (Stancato et al., 1993). Members of the Src family of tyrosine kinases have been shown to be Hsp 90 clients, highlighting a role for Hsp 90 in signal transduction from receptors in the plasma membrane. For example, the oncoprotein pp60'’'^"' (v-5rc), a cytoplasmic membrane-associated tyrosine kinase, has been reported to interact with Hsp 90 and this was the first observed stable complex of a protein with Hsp 90 (Brugge, 1986). Following its synthesis in vertebrate cells and subsequent interaction with Hsp 90, v- Src is not functional as a kinase. Within the plasma membrane however, \-Src is active as a kinase and is no longer associated with Hsp 90. Genetic studies in yeast, involving mutations that lower the level of Hsp 90 (Xu and Lindquist, 1993), suggest that Hsp 90 both stabilisesw-Src and suppresses its kinase activity during transit to the plasma membrane. The Hsp 90-dependence of \-Src and other Src family members including v-Yes and \-Fps, has not been observed for their normal cellular counterparts. However, studies in yeast have shown the activation of both \-Src and
36 Figure 1.1.2 The Hsp 90 Chaperone Complex Involved in SHR Activation
HOP
Hsp 90 Hsp 90
SHR
HOP
p23 HP ATP Hsp 40 (3)
Hsp 90 Hsp 90
SHR Hormone □ = > (4)
SHR > DNA BINDING SHR
The Hsp 90 chaperone complex involved in SHR activation. The SHR interacts sequentially with several chaperone complexes. Initially this involves Hsp 70 and co factors. The scaffold protein HOP brings together the Hsp 90 chaperone machinery with Hsp 70 to form an intermediate complex (1). The Hsp 70 complex then dissociates (2) and p23 and one of the large immunophillins (IP), enter the complex (3). The SHR is released from this complex, it can then bind hormone, dimerise and bind DNA. Introduction the normal cellular homologue c-Src^ to be Hsp 90-dependent and thus these observations highlight the possibility that the activation of other normal cellular Src family members is also Hsp 90-dependent (Xu et al., 1999). Some receptors, such as the insulin and epidermal growth factor (EOF) receptor, transduce their signal from the plasma membrane through activation of members of the mitogen-activated protein (MAP) kinase family of serine/threonine kinases. Components of this system, which have been shown to form stable complexes with Hsp 90, include Src, Raf and Mek (Pratt, 1997). Not suprisingly, these observations have highlighted the suitability of Hsp 90 as a candidate target for anti-cancer drug development. In agreement with this, the antibiotic geldanamycin (which displays tumouricidal activity), has been shown to function through the inhibition of Src-Hsp 90 heterocomplex formation (Whitesell et al., 1994).
The Hsp 70 Family
The Hsp 70 family consists of a group of highly conserved proteins which are present in prokaryotic and eukaryotic cells under normal growth conditions and, as for Hsp 90, their expression is increased in response to various stressful and non- stressful conditions. Members of this family have a variety of roles which relate to their location within the cell and their function as molecular chaperones. These include the processes of protein folding and degradation, translocation of proteins across organellar membranes and auto-regulation of the heat shock response (discussed later). In mammals (see Table 1.1.1), members of this family include Hsp 72, the major heat-inducible Hsp 70 (also referred to as Hsp 70 or Hsx 70), Hsp 73 which is largely constitutively expressed (also called Hsc 70), Grp 78 which is present in the endoplasmic reticulum (also known as immunoglobulin binding protein, BiP) and Grp 75 which is present in mitochondria. The structure of Hsp 70 is highly conserved from prokaryotes to mammals. Two domains are present, an N- terminal domain (approximately 40 kDa) contains an ATP-binding site (Flaherty et al., 1990) and a C-terminal domain (approximately 25 kDa) which is less conserved between species and functions in substrate binding.
38 Introduction
A number of studies have suggested of a role for Hsp 70 in the de novo folding of proteins and are reviewed in Feldman and Frydman, (2000). The process of folding a newly translated polypeptide requires its inunediate and subsequent sequential interaction with different chaperone complexes. This is necessary to prevent vzuious processes occurring such as protein aggregation, which inhibit the formation of mature, appropriately folded and functional proteins, reviewed in Gething and Sambrook, (1992). The process of protein translation is understood to proceed within different time-scales from prokaryotes (seconds) to eukaryotes (several minutes). Thus, particularly in eukaryotic cells, it is conceivable that the folding of a polypeptide nascent chain may begin prematurely during the translation process and critically, before completion of synthesis of a complete folding domain (a structurally independent unit). A similar problem occurs during the translocation of proteins across the membranes of mitochondria, chloroplasts and endoplasmic reticulum, where an extended polypeptide conformation rather than a stably folded protein, is required for protein transport. Hsp 70 family chaperones in cooperation with co-chaperones of the Hsp 40/DnaJ family, bind small linear polypeptides with a net hydrophobic character (exposed hydrophobic regions are a characteristic feature of unfolded proteins) leading to polypeptide stabilisation in an extended conformation which prevents aggregation (Gething and Sambrook, 1992). Further studies have highlighted a role for Hsp 70 in the folding of nascent polypeptides. For example, yeast strains with a loss of Hsp 70 function, show reduced translational efficiency (Nelson et al., 1992). Furthermore, cytoplasmic Hsp 70 is observed to associate with ribosome-bound nascent chains in eukaryotic cells (Nelson et al., 1992), (Frydman et al., 1994). Hsp 40 and DnaJ (a constitutively expressed 40 kDa chaperone) family chaperone proteins are understood to be involved in the process of Hsp 70 recruitment to nascent polypeptide chains, reviewed in Bukau and Horwich, (1998), while Hsp 70s of the mitochondria and endoplasmic reticulum are observed to bind to translocating polypeptides (Manning-Krieg et al., 1991), (Brodsky and Schekman, 1993). Following translocation, mitochondrial and endoplasmic reticulum (ER) Hsp 70s are required for the subsequent folding of polypeptides. However, in many cases the Hsp 70 chaperones alone or in conjunction with the Hsp 90 chaperone complex, are
39 Introduction insufficient for the folding of polypeptides to their native state. More complex protein folding is aided by the transfer of these polypeptides to chaperonin family members such as Hsp 60.
The Hsp 60 Family
Members of this family (also known as chaperonin 60) are among the most abundant of cellular proteins found in bacteria (known as GroEL in E.coli) and are also present in the mitochondria and chloroplast of eukaryotes. A structurally related chaperonin family has been identified, the TriC family. This is found in the cytoplasm of most eukaryotes, including plants, yeast. Drosophila, archaebacterial cells, and appears to fulfil the functions of mitochondrial Hsp 60. Structurally, the Hsp 60s form large oligomeric complexes comprising 60 kDa subunits, which are arranged as two stacked heptameric rings forming a central cavity (Hendrix, 1979), (Hohn et al., 1979). Although members of the Hsp 60 family are induced under conditions which can cause protein dénaturation (Hightower, 1991), the Hsp 60s are constitutively expressed and are essential for growth. Members of the Hsp 60 family cooperate with a smaller stress-inducible protein, Hsp 10 (also known as chaperonin 10) in the process of protein folding. In E.coli, the Hsp 60 (GroEL) and the Hsp 10 (GroES), are encoded in a single operon (Tilly et al., 1981), (Chandrasekhar et al., 1986). Chaperonin proteins, which include the Hsp 60 family, function to assist de novo protein folding by sequestering proteins in the early stages of folding and providing a protective environment (the chaperonin cavity) in which to fold. The majority of client proteins for Hsp 60s tend to be of medium size (25-60 kDa). Small proteins, which generally have rapid folding kinetics, appear to require minimal assistance in preventing malfolding. Furthermore, a sub-population of medium size proteins would appear to require assistance in folding, in addition to that provided by Hsp 70 and co-chaperones. For example, the enzyme dihydrofolate reductase (DHFR), folds spontaneously in vitro, but requires several chaperone families, including mitochondrial Hsp 60 (Ostermann et al., 1989), for appropriate folding in vivo.
40 Introduction
Subsequent studies have shown Hsp 60 to be involved in the assembly of various oligomeric complexes. In yeast defective for the gene encoding mitochondrial Hsp 60, chaperonin function has been shown to be essential for the assembly of oligomeric enzyme complexes (Cheng et al., 1990) and in prokaryotes, GroEL and GroES are responsible for the folding of subunits of the enzyme ornithine transcarbamylase (OTC) (Zheng et al., 1993). A role for Hsp 60 in protein folding under conditions of heat stress conditions is supported by the observation that the exposure of mitochondria to heat stress, leads to the association of a variety of polypeptides with mitochondrial Hsp 60 (Martin et al., 1992). Under such conditions, functional Hsp 60 has been demonstrated to be required to prevent thermal inactivation of the enzyme dihydrofolate reductase (DHFR) imported into the mitochondrion. In considering the differential binding capabilities of Hsp 70 and Hsp 60, these observations would suggest that the folding of newly synthesised polypeptides follows complex pathways. Thus, it is likely that a mechanism of sequential action of Hsp 70 and Hsp 60 chaperones is involved in protein folding. Moreover, the identification of the structurally related TriC chaperone family, would suggest this mechanism could also be relevant to the eukaryotic cytosol. A major interest of Hsp 60, is its possible role in the development of autoimmunity in humans. As discussed later (see Section 1.2), a number of studies have focussed particularly on the mycobacterial Hsp 60 homologue and Hsp immune autoreactivity in various inflammatory disease states such as arthritis.
Small Hsps and Ubiquitin
The family of small Hsps (sHsps) are comparatively less conserved among species than the high molecular weight Hsp families such as Hsp 90 and Hsp 70. Members of this family have molecular weights up to 40 kDa and are characterised by a conserved region, referred to as the a-crystallin domain. This domain is located in the C-terminal region of all sHsps (Kim et al., 1998) and in Drosophila and mammals, it consists of between 80 and 100 amino acids (Ingolia and Craig, 1982). In contrast, the N-terminus of sHsps is variable in both sequence and length. Studies
41 Introduction on sHsp expression have shown that sHsp genes are highly inducible in response to heat and other stresses. Differences in the stress-induced expression of the sHsps compared to other Hsp families have been observed in Drosophila, where sHsps tend to be expressed at temperatures lower than those required for Hsp 70 induction (Yost etal., 1990). Constitutive expression of sHsps in unstressed cells has been well documented in Drosophila and appears to follow both a tissue and stage-specific pattern of expression. In mammals, Hsp 27 is constitutively expressed at low levels in various cell-lines, including human HeLa, monkey Cos and mouse Swiss-3T3 cells. Hsp 27 has also been observed to accumulate in the absence of stress within several tissues, including spinal cord neurons and Purkinje cells, during murine development (Gemold., 1993). In human and monkey cells, Hsp 27 expressed constitutively in the cytoplasm is observed, like ap-crystallin, to have a peri-nuclear localisation (Arrigo et al., 1988). In contrast, Hsp 25 of chicken, is found to be diffusely distributed in the cytoplasm (Collier and Schlesinger, 1986). Studies on the localisation of sHsps within the cell, have highlighted distribution patterns which are a consequence of either stressful or non-stressful cellular conditions. For example, heat shock or other forms of cellular stress leads to the re-distribution of Hsp 27 (Arrigo et al., 1988) and ap-crystallin (Klemenz et al., 1991), inside or around the nucleus. Subsequently, following cellular recovery from heat shock, sHsps are gradually re-distributed in the cytoplasm (Arrigo et al., 1980). All sHsps studied to date have been shown to possess oligomeric structures. During heat-shock, the size of Hsp 27 oligomers increase, resulting in a super-aggregated structure which re-distributes inside the nucleus (Arrigo et al., 1988). Furthermore, these observations do not pertain to Hsp 27 or other sHsps, in thermotolerant cells exposed to heat-shock (Arrigo and Welch, 1987). Ubiquitin is a 7-8 kDa protein of 76 amino acids, which is highly conserved between species (Lindquist, 1986). Its expression within the cell is constitutive, but also highly inducible in response to cellular stresses such as elevated temperature (Schlesinger, 1990). Ubiquitin is involved in a multienzyme system responsible for the degradation of short-lived, misfolded or damaged proteins (Peters et al., 1998). Protein degradation is mediated by the 26S proteosome, a large ATP-dependent multicatalytic protease complex. The ubiquitin-proteosome pathway represents a
42 Introduction major cellular non-lysosomal pathway of proteolysis. Proteins are usually targetted for proteosome-mediated degradation, by the covalent attachment of multiple ubiquitin molecules (a process termed polyubiquitination) via terminal glycine residues of ubiquitin to other proteins (Lindquist and Craig, 1988), (Peters et al., 1998). More recently, a role for ubiquitin in the downregulation of signal transducing receptors has been established. Binding of growth hormone to its receptor leads to receptor dimérisation, signal transduction and an increase in receptor ubiquitination. Ligand-induced endocytosis of the mammalian growth hormone receptor (GHR), occurs in a manner dependent on a functional ubiquitin conjugating system (Strous et al., 1996). Further studies have identified a ubiquitin conjugation motif within the cytoplasmic tail of the GHR. This is required for ligand-induced internalisation of the GHR, leading to degradation of the exoplasmic and cytoplasmic portions of the receptor (Govers et al., 1999). Thus, ubiquitin may be considered to have a role in the selective degradation of non-functional/malformed proteins as well as regulatory roles in a variety of receptor-associated cellular processes and is reviewed in van Kerkhof et al., (2000).
Conditions that Induce the Expression of Heat Shock Proteins
The term heat shock response clearly has a historical origin. Subsequent
studies have shown that the conditions involved in inducing the expression of Hsps
within cells are remarkably diverse, (see Figure 1.1.3). Thus, these conditions may be
broadly categorised into three different classes: (1) environmental stress, including
heat shock and exposure of cells to heavy metals, (2) cellular pathophysiology,
including viral and bacterial infection and (3) non-stress conditions, including
cytokines, cell cycle control and cellular growth and differentiation.
43 Figure 1.1.3 Inducers of the Heat Shock Response (adapted from Morimoto et aL, 1992)
Environmental Stress
TransiDon Heavy Metals Amino Add Analogues Heat Shock
Pathophysiological Stress
Fever and Inllammaljon Ischaemia 'HSF Viral and Bacterial Infection Neuronal Injury Ageing HSE
Cytokines Ce# C yde Growth Factors Non-Stressful Conditions
Figure 1.1.3 shows three broadly categorised classes of conditions which result in the elevated expression of Hsps. Also shown is the mechanism by which environmental stresses such as heat shock, induce Hsp expression via the activation (a) of heat shock factor (HSF) and its subsequent translocation into the nucleus where it binds to heat shock elements in the promoters of Hsp genes. Introduction
Regulation of Heat Shock Protein Expression
The diversity of roles observed for different Hsp families within the cell and their increased expression in response to various stimuli (Figure 1.1.3), would suggest the process of Hsp gene expression to be both highly versatile and carefully regulated. The genes encoding Hsps are highly conserved and the promoter regions of Hsp genes contain multiple sites for regulating constitutive and inducible gene expression. Transcription factors are proteins which bind to specific sequences in the promoter regions of their target genes, leading to activation or suppression of gene transcription. The induction of Hsps in response to various cellular stress conditions, is mediated by a group of transcription factors termed heat shock factors (HSFs), which bind to heat shock elements (HSEs) in the promoters of Hsps (Morimoto, 1993), (Lis and Wu, 1993). However, although HSEs are a distinguishing feature of Hsp family genes, members of the Hsp 90 and Hsp 70 families include glucose- regulated Hsp members (see Table 1.1.1) which are not heat-inducible. The cloning of HSF genes has revealed a multi-gene family and to date, three HSFs (HSFl, 2, 3)
have been cloned from different organisms, including yeast, mouse and humans and a further family member, HSF4, has been identified in humans (Nakai et al., 1997).
Subsequent studies have led to both structural and functional characterisation of these
HSF family members.
Common to all HSFs is an N-terminal-situated DNA binding domain of
approximately 100 amino acids, which contains a helix-tum-helix fold, composed of a (alpha)-helices 1, 2 and 3, where a-helix 3 is the DNA recognition motif that docks
within the major groove of the DNA helix (Harrison et al., 1994). A hydrophobic
motif which is involved in the oligomerisation of HSF prior to DNA binding, is also
N-terminal-situated. A comparison of the sequences of the vertebrate HSFs reveals
45 Introduction
that within a single species, HSF family members are approximately 40% identical at the amino acid level. This is primarily due to the conserved sequences of the DNA- binding and oligomerisation domains and is discussed in Cotto and Morimoto,
(1999). The C-terminal region is less conserved in sequence and contains a transcriptional activation domain. Only HSFl has been shown to be involved in the heat induced activation of Hsp genes and also functions in response to other stresses, including heavy metals and amino acid analogues (Sarge et al., 1993). In contrast, a role for other HSF members has been less well characterised, although these may be involved in Hsp regulation in unstressed cells in response to a variety of processes including cell differentiation and development (Morimoto et al., 1992), (Morimoto, 1993). For example, HSF2 mRNA in mouse testis is subject to both developmental and cell- stage, type regulation (Sarge et al., 1994). Under normal cellular conditions, HSFl in the cytoplasm is maintained as a latent monomeric form which is unable to bind DNA. Activation of HSFl in response to thermal stress (heat shock), leads to the formation of homotrimers which can subsequently bind to HSEs within the promoter region of Hsp genes (Wu, 1995).
Following a period of intermediate (non-extreme) heat shock, the activation of HSFl is rapid (minutes) but on returning to normal temperature, HSFl reverts to its non-
DNA binding form (Abravaya et al., 1991a). Heat-induced phosphorylation of HSFl
correlates with increased HSFl transcriptional activity, however phosphorylation is
not essential for DNA-binding (Sorger and Pelham, 1988).
Studies on the transcriptional activation of Hsp genes by HSFl in both yeast
and mammalian cells, suggest an autoregulatory role for Hsps. For example,
mutations that decrease Hsp 70 synthesis (Wemer-Washbume et al., 1987) or Hsp 90
synthesis (Duina et al., 1998), lead to increased Hsp synthesis and cellular tolerance
to heat shock. Thus, it is possible that the dissociation of HSFl from Hsp complexes
46 Introduction
during heat shock, facilitates the activation of HSFl and leads to increased Hsp synthesis in the cell. The realisation that cytokines (see Figure 1.1.3) interacting with their counterpart cell membrane receptor, can also provide a stimulus for activating Hsp gene expression, has further increased the scope of Hsp gene regulation. The involvement of cytokines in Hsp gene regulation and in particular the regulation of the Hsp 90 gene, is discussed below.
Cvtokines
Cytokines are a diverse group of small proteins (typically 15-30 kDa). They are secreted by different cell types in a number of tissues. The physiological roles of cytokines are largely concerned with the coordination of processes between different cell types or with the cellular response to environmental stresses. Consequently, their lifetime in the bloodstream or other extracellular fluids is short, suggesting limited and regulated biological activity. Cytokines influence the phenotype of their target cells by binding to a specific counterpart cell-surface receptor. However, the intracellular signalling pathways utilised by cytokines to activate their target genes, are largely to be characterised. For a comprehensive outline of the biological roles of cytokines, see Roitt., (1998).
The interleukins are a class of cytokine which are produced by, or act upon, cells of the immune system. The identification and characterisation of new interleukins is continuing, however of particular interest to this dissertation (see
Section 1.5 of this chapter) are the cytokines interleukin-10 and interleukin-6, which
are described below.
47 Introduction
Interleukin-10
Interleukin-10 (IL-10) is a multifunctional cytokine produced by B cells, T helper cells, monocyte/macrophages and kératinocytes. Human and murine IL-10 inhibit cytokine synthesis by activated T cells (Fiorentino et al., 1989; 1991), (de
Waal Malefyt et al., 1991), NK cells (Hsu et al., 1992) and monocytes/macrophages (Bogdan et al., 1991), (de Waal Malefyt et al., 1991), (Fiorentino et al., 1991). Thus,
IL-10 may be considered to have largely anti-inflammatory properties on cells of the immune system, for a review see Moore et al., (1993). These biological roles are mediated by IL-10 interacting with a counterpart cell surface receptor (see Figure 1.1.4). Characterisation of the human IL-10 receptor began in 1994 with the cloning of the human IL-10 receptor-binding chain (IL-lORl). This receptor was defined structurally as possessing a single transmembrane domain and as a member of the class II cytokine receptor family (Liu et al., 1994). Further studies have identified human CRFB4, a receptor belonging to the same cytokine family, as forming a functional constituent of the IL-10 receptor complex and it is now referred to as IL- 10R2. Members of the class II cytokine receptor family, which include the interferon
receptors, have been shown to mediate intracellular signalling via the Jak/STAT
signalling pathway (Finbloom and Winestock, 1995) and is discussed later in this section.
Interleukin-6
Interleukin-6 (IL-6) is a multifunctional cytokine with pivotal roles in
regulating the functions of lymphoid and myeloid cells as part of the normal immune
response (Kishimoto and Hirano, 1988) as well as the production of acute phase
proteins within the liver (Andus et al., 1987). The biological properties of IL-6 may
48 Figure 1.1.4 IL-10 Receptor Signalling via the JAK/STAT Pathway
IL-10
IL-1GR2
STAT-1
WÊF TAT-3
Figure 1.1.4 illustrates the structure of the IL-10 receptor following binding of IL-10 and receptor dimérisation. Phosphorylation and activation of the receptor-associated Jaks, leads to the recruitment and activation of STAT transcription factors, which subsequently enter the nucleus and bind to the promoters of their target genes. Introduction
be classified as pro-inflammatory through the promotion of differentiation and activation of T cells and macrophages. IL-6 mediates its pleiotropic effects through an interaction with a specific counterpart receptor. An 80 kDa single polypeptide chain, termed gp80, binds IL-6 within a C-terminal extracellular region of the receptor containing a type III fibronectin domain (see Figure 1.1.5). Binding of IL-6 to gp80 subsequently promotes the association of this receptor with a non-ligand binding chain termed gpl30 (Taga et al., 1989) which mediates IL-6 intracellular signalling. A soluble form of the gp80 receptor has also been identified and is generated either by proteolytic cleavage of the cognate receptor (PC-sIL-6R) or by deletion of the transmembrane domain by differential mRNA splicing (DS-sIL-6R) (Jones et al., 1998). In mice, serum soluble gp80 has been demonstrated to mediate IL-6 functions via cells expressing gpl30 (Suzuki et al., 1993). The association of gpl30 with IL-6 and other IL-6-type cytokine receptors, such as leukaemia inhibitory factor (LIF), which lack intrinsic transducing activity, therefore provides a common signalling mechanism for these receptor types. Although the pro-inflammatory effects of 11^6 contrast with the biological roles of IL-10, both cytokines share similar signalling mechanisms for activating their target genes.
IL-IO and IL-6 Activate the ST AT Family of Transcription Factors
The interaction of STATs (signal transducers and activators of transcription) with cytokine receptors which utilise the Jak/STAT signalling pathway, is mediated via SH2 (5rc homology 2) domains within the STAT protein, with specificity for a phosphotyrosine sequence motif within the intracellular domain of the cytokine receptor (Darnell, 1997). Tyrosine residues within the receptor-bound STAT
50 Figure 1.1.5 IL-6 Receptor Signalling via the JAK/STAT Pathway
gpi30 gpi30
gp80 gp80 immunoglobulin-like domain O IL-6 IL-6 0 Cytoplasm
Cytosol
767 Y814
Y905 Y915 STAT-3
Figure 1.1.5 Structure of the IL-6 receptor. Also shown is the sequence of events leading to the activation of the IL-6-activated transcription factor STAT 3, its translocation into the nucleus and binding to a promoter in a target gene. Introduction
molecule then become phosphorylated, leading to the formation of homo- or hetero- STAT dimers utilising reciprocal SH2 domains. STAT dimers are then translocated into the nucleus where binding to specific elements within the promoters of genes,
leads to activation of gene transcription (Ihle, 1996). The Janus family of tyrosine kinases (Jaks) consist of four members, Jak-1,
Jak-2, Jak-3 and TYK-2. Receptors such as the IL-10 receptor, lack intrinsic kinase activity and utilise non-covalently bound Jaks for intracellular signalling. IL-10
receptor signalling has been shown to involve the stimulation of Jakl and Tyk2 Janus tyrosine kinases (Riley et al., 1999) (see Figure 1.1.4). As shown in Figure 1.1.5, the IL-6 receptor is also associated with Jaks (Jakl and Jak2). Ligand-mediated dimérisation of receptors associated with this pathway, results in reciprocal tyrosine phosphorylation, leading to the phosphorylation and activation of the receptor associated Jaks. Activated Jaks phosphorylate tyrosine residues on the cytoplasmic tail of the receptor, leading to the recruitment of the STAT family of transcription factors. IL-10 and IL-6, have been shown to activate STAT-1 and STAT-3 leading to the formation of homo- and hetero-dimers and in some cells STAT-5 (Finbloom and
Winestock, 1995). The utilisation of specific STAT proteins by cytokines such as IL-6 and IL-
10, together with as yet unknown mechanisms of modulating the formation of homo-
and hetero-STAT dimers, may in part reflect the pleiotropic effects of these cytokines
and hence the flexibility of cytokine receptors to activate different genes in different
situations and effect a range of contrasting biological activities within target cells.
Moreover, as discussed later, signalling via the activation of STATs is interactive
with other intracellular signalling pathways. Thus, STAT dimers have been shown to
function as co-activators with several other classes of DNA-binding proteins,
exemplified by the reported interaction of STAT-5 with the glucocorticoid receptor
52 Introduction
(Stocklin., 1996). Such studies suggest that the Jak/STAT pathway may be utilised to regulate a wide range of target genes.
IL-6 but not IL-10, Activates Transcription Factors of the OEBP Family
The biological functions of IL-6 are dependent on the stimulation of two distinct signalling pathways. These leads to the activation of members of the C/EBP family of transcription factors and, as discussed above, the STAT family of transcription factors (see Figure 1.1.5). NF-IL6 (nuclear factor-IL6, also termed C/EBPp) is activated and increased synthesis of NF-IL6P (C/EBPÔ) occurs following stimulation by IL-6 and both members can form heterodimers with each other, with transcriptional activity (Kinoshita et al., 1992).
Role Of Interleukin-6 In Hsp 90 Gene Regulation
It has previously been demonstrated that IL-6 upregulates Hsp 90 expression in the HepG2 human hepatoma cell line and in peripheral blood mononuclear cells (Stephanou et al., 1997). A role for IL-6 in Hsp 90 regulation is further supported by the observation that elevation of IL^6 in transgenic mice, results in increased levels of
Hsp 90 (Stephanou et al., 1998a).
An investigation of the mechanisms of Hsp 90 promoter activation in response to IL-6, has recently been reported (Stephanou et al., 1998b). In this study, using reporter gene assays, activation of the Hsp 90p promoter occurs via the activation of the transcription factors STAT-3 and NF-IL6 and is mediated by a region of the promoter which contains binding sites for the transcription factors
STATl, STAT-3, HSFl and NF-IL6. Interestingly, STAT-3 and NF-IL6 were observed to interact differently with HSFl in mediating Hsp 90p gene expression.
53 Introduction
STAT-3 was observed to reduce the stimulatory effect of HSFl while NF-IL6 enhanced it.
Further studies of the role HSF-1 and transcription factors activated by cytokines in Hsp gene expression, have demonstrated that HSF-1 and STAT-1 interact with one another via a protein-protein interaction (Stephanou et al., 1999).
Moreover, as for NF-IL6, a strong synergistic effect of STAT-1 and HSFl in activating the Hsp 90p promoter was observed. These studies would suggest an important role for transcription factors activated by cytokines, in the regulation of
Hsp expression by non-stressful stimuli. Furthermore, the interaction of these non stressful stimuli with stress responses, may reflect the apparent flexibility of the cell to activate different Hsps under different situations. It is interesting to note that IL-6 and IL-10 utilise similar mechanisms for activating their target genes and, as discussed earlier, this is mediated via the Jak/STAT intracellular signalling pathway. Thus it is conceivable that 11^ 10, which activates the transcription factors STAT-1 and STAT-3, may be involved in the regulation of Hsp 90 gene expression in vivo.
Summary
The discovery of the heat shock response in Drosophila, has led to the identification of a multi-gene family of highly conserved proteins, which are now collectively referred to as heat shock proteins (Hsps). The regulation of Hsp gene expression by stressful stimuli has been intensively studied and further investigations into Hsp cell biology have highlighted important roles for Hsps within the cell under non-stressful conditions. However, the regulation of Hsp expression by non-stressful stimuli has been poorly understood. Recent studies have demonstrated a role for cytokines in the regulation of Hsp expression and it is likely that such non-stressful stimuli have important roles in the regulation of Hsps within the immune system. As
54 Introduction will be discussed in the following section, there is now accumulating evidence that Hsps are involved in the development of pathogenic immune responses in a number of human autoimmune diseases, including systemic lupus erythematosus.
55 Introduction
Section 1.2: The Role of Heat Shock Proteins in the Immune System and Pathogenesis of Autoimmune Disease
Introduction
A role for heat shock proteins (Hsps) as molecular chaperones involved in the assembly and disassembly of proteins within the cell, has been discussed in Section 1,1 of this chapter. Consistent with this role, the assembly of a diverse range of multimeric complexes produced by cells of the immune system, requires the assistance of chaperones. Thus, immunoglobulins, T-cell receptors and gene products of the major histocompatibility complex (MHC), which are all discussed later, are dependent on the function of Hsps. An important role for Hsps in protective immunity has also been suggested and is supported by the observation that Hsps of invading microbes serve as antigens in the host immune response to infection. Interestingly, a growing number of studies have also suggested that Hsp-specific inunune responses, may be involved in the pathogenesis of various autoimmune diseases. This section begins with a brief outline of the functions of the immune system, focussing on the mechanisms by which cells of the immune system specifically target foreign pathogens, while discriminating against a multitude of host-derived molecules. This is followed by a summary of various studies which suggest protective roles for Hsps in the immune system, particularly in the host immune response to microbial infection. Finally, various studies which suggest pathogenic roles for Hsps in the immune system are discussed in relation to their relevance to the development of autoimmune disease.
56 Introduction
Sel^on-Self Discrimination Within the Immune System
Immunology is the study of the mechanisms by which the body defends itself from invading microorganisms (including viruses and bacteria) which can cause disease, as well as the development of other disorders including cancer. Immunity is dependent on a vast array of soluble factors and cells which circulate in the blood and extracellular fluid. Three properties of this defence system are essential for virtually all immune responses: (1) a highly specific response is required for the recognition and targetting of foreign molecules; (2) the ability to discriminate foreign molecules from the multitude of host molecules is important for preventing autoreactive responses. Furthermore, (3) the acquisition of immunological memory, permits a more refined and rapid response following a repeat challenge from a specific foreign pathogen. This section focusses on the cellular and molecular basis of specificity in the immune response. However, a broad review of the immune system, can be found in Roitt, (1998). Studies into the mechanisms which give rise to immunologic specificity and memory, have focussed on a subset of cells of the immune system which are termed lymphocytes. From these studies, several different types of lymphocytes are distinguished on the basis of their functional properties, molecular genetics and organisation. As outlined below, the ability of the immune system to respond in a specific manner to a vast number of different pathogens, is facilitated by the clonal distribution of receptors, which are expressed on the cell surface of lymphocytes.
Lvmphocvtes
A normal human adult body contains approximately 10^^ lymphocytes, the majority of which appear to have uniform characteristics under light microscopy. However, several different types of lymphocytes may be distinguished on the basis of their functional properties and by their expression of specific proteins. Lymphocytes may be divided into two major lineages known as B (bone- marrow-derived) cells and T (thymus-derived) cells. These account for approximately 10% and 75% respectively, of the total number of lymphocytes in peripheral blood.
57 Introduction
The remainder of the lymphocytes belong to a separate and less well characterised lineage termed natural killer (NK) cells. B and T cells originate from stem cells in the bone marrow, which have become committed to a lymphoid (any lymphocyte lineage) line of development. A review of lymphocyte development can be found in Ikuta et al., (1992). As outlined below, the development and maturation of B and T cells from these progenitor stern cells occurs in both the bone marrow (B and T cells) and thymus (T cells only). These organs are essential for the initial production of lymphocytes (a process known as lymphopoiesis) and are collectively referred to as the primary lymphoid organs. Following dispersal into the bloodstream, these “virgin” lymphocytes migrate into various secondary (or peripheral) lymphoid organs, including the tonsils or lymph nodes. At these sites, lymphocytes frequently encounter infectious pathogens and thus it is from the secondary lymphoid organs, that immune responses are commonly initiated. A review of the role of lymphoid tissues in resistance to infection can be found in Szakal et al., (1989). Encounters of a lymphocyte with a foreign pathogen leads to a process referred to as activation, during which successive rounds of lymphocyte cell division occur. The resulting progeny may revert to a resting state (termed memory lymphocytes) or they may differentiate into effector cells, which have a short life span (a few days) and have important anti-microbial functions. A review of lymphocyte activation can be found in Janeway and Bottomly, (1994).
B Cells
Thede novo synthesis of B cells occurs exclusively in the bone marrow and is constant throughout life. The defining feature of B cells is their unique ability to synthesise an extremely diverse family of proteins called immunoglobulins. Each immunoglobulin molecule can bind specifically to a molecular determinant of foreign or self origin. Collectively these molecular determinants are referred to as antigens. B cells express immunoglobulin in two distinct functional forms. Mature non-dividing (resting) B cells secrete large quantities of immunoglobulin onto the cell surface. The interaction of surface immunoglobulin (sig) with other surface components
58 Introduction comprising disulphide-linked heterodimers of Ig-a and Ig-p, forms B cell-receptor complexes which act as specific B cell antigen receptors. Immunoglobulin expression in certain B cells (plasma cells) leads to large quantities being secreted into the blood circulation with high specificity for the antigen and these immunoglobulin molecules are referred to as antibodies.
TCelb
T cells develop from immature precursors in the bone marrow which migrate to the thymus which produces the hormones necessary for T cell maturation and proliferation. A review of T cell development can be found in Robey and Fowlkes, (1994). The surface expression of a number of molecules allows T cells to recognise foreign antigens. Unlike B cells, T cell recognition of an antigen requires presentation of the antigen to the T cell by specialised cells, termed antigen presenting cells (see antigen processing and presentation later in this section). T cells may be subdivided into major classes according to the expression of specific surface molecules. The definitive marker for T cells is the T cell receptor, which is expressed on the cell surface. In humans, the majority of T cells surface express a T cell receptor (TCR) consisting of two transmembrane polypeptides, known as the a-chain and the p-chain, as well as accessory CD4 or CDS receptors. A minor population (which is discussed later) expresses a TCR consisting of a y-chain and a 0-chain and rarely expresses either CD4 or CDS receptors, but functions like the ap-TCR-expressing T cells (Kaufmann, 1996). T cells expressing the ap-T cell receptor, usually express only one of the CD4 and CDS receptors and are commonly referred to as single-positive (CD4 or CDS) lymphocytes. CDS T cells function in the killing of host cells, particularly cells expressing foreign antigens on the cell surface. This is particularly important in the identification and disposal of viral infected cells displaying viral peptides on the cell surface. CD4 T cells function to provide help for various important immunological functions of other cells, including CDS T cells, B cells and macrophages and is reviewed in Janeway, (1999). As mentioned above, the absence of both CD4 and CDS receptors (referred to as double-negative T cells) is observed in a small
59 Introduction percentage of T cells, usually of the yô-TCR type. As discussed later, these cells appear to have important roles in the recognition of Hsp antigens of invading microbes. The individual roles of these T cell subsets and the mechanisms leading to their activation following antigen stimulation, are reviewed in Ullman et al., (1990) and Podack and Kupfer, (1991).
Thvmic Selection of T Cells and Self-Tolerance
The ability of the immune system to distinguish foreign antigens from a multitude of host molecules, involves the elimination or functional inactivation of lymphocytes which express receptors for self-determinants and is a process fundamental to maintaining self-tolerance. Since B cell responses are largely dependent on T cell help, such selective pressures occur mainly on T cells. As discussed earlier, progenitor T cells which develop from bone marrow, undergo maturation in the thymus. Positive selection of T cells expressing TCRs which can bind to antigens associated with MHC molecules, occurs in the thymic cortex. This process may involve interactions of T cells with epithelial cells which surface express self-peptides and one in which T cell activation provides a survival signal. In contrast, negative selection primarily occurs in the thymic medulla and leads to the deletion of T cells which express TCRs capable of recognising self molecules. The mechanisms of thymic selection are discussed and reviewed in Janeway, (1999).
60 Introduction
Heat Shock Proteins and the Immune System
Antigen Processing and Presentation
The capture, processing and presentation of antigen to T cells is carried out by antigen presenting cells (APCs) and is a process fundamental to the acquired immune response. Macrophages and dendritic cells of the spleen and lymph nodes, together with B cells, are the major cell types to present antigen to T cells. APCs capture antigen by a variety of methods, including phagocytosis (principally by macrophages), pinocytosis or receptor-mediated endocytosis. The processing of foreign antigens and endogenously synthesised proteins such as viral and tumour antigens, involves their proteolytic cleavage into smaller antigenic peptides, which then associate with class I or class II MHC molecules. These peptide/MHC complexes are then transported to the cell surface where they are accessible for interaction with the T cell receptor. For a review of antigen processing and presentation see Germain and Margulies, (1993). Endogenously synthesised proteins, including viral and tumour antigens, are degraded in the cytoplasm by a multimeric enzymatic complex (the proteosome) and translocated into the endoplasmic reticulum where they bind to class I MHC molecules. T cells expressing the CD8 receptor, which function in the lysis of their target cells, recognise peptides that are presented by MHC class I molecules on the surface of these cells. In contrast, CD4f T cells recognise peptides (usually of extracellular origin) that are presented by class II MHC molecules on the surface of specialised APCs, including macrophages, dendritic cells and B cells.
A number of studies have revealed roles for Hsps in the folding and assembly of MHC-peptide complexes. For example, heat shock has been demonstrated to enhance the processing and presentation of antigen entering B cells (Cristau et al..
61 Introduction
1994). Studies using a mouse fibroblast cell line, have demonstrated heat shock to dramatically increase the expression of an MHC Class I-like molecule (Qa-1), which may be involved in the presentation of Hsp antigens to yô-T cells (Imani and Soloski,
1991). Members of the Hsp 70 family have been shown to be involved in the intracellular assembly of processed-antigen/Class II MHC complexes (Vanbuskirk et al., 1989), (DeNagel and Pierce, 1992), (Schirmbeck and Reimann, 1994), (Williams and Watts, 1995) and an interaction of Hsp 70 with misfolded class II MHC molecules has been reported (Schaiff et al., 1992). Furthermore, the transport of peptides from the proteosome to the endoplasmic reticulum (ER) and their subsequent association with class I MHC molecules, is understood to involve both cytoplasmic and ER members of the Hsp 70 and Hsp 90 families (Li and Srivastava, 1993), (Srivastava, 1993), (Srivastava et al., 1994). Taken together, these studies suggest important functions for Hsps in the folding and assembly of MHC molecules in the ER. Furthermore, Hsps appear to play a key role in the transport of antigenic peptides to the ER and the formation of stable peptide/MHC complexes, for subsequent presentation to the T cell receptor.
Hsps and Infection
The induction of Hsp synthesis in parasites during infection is likely to occur
as a response to stressful conditions in the host environment and is vital for pathogen
survival (Polla, 1991). Although the role of Hsps in protective host inunune responses
is poorly understood, numerous studies have provided evidence that Hsps of invading
microbes serve as important antigenic targets in the host inunune response to such
infections (Kaufmann, 1990; 1991), (Kaufmann et al., 1990). Thus, both humoral and
cellular inunune responses to Hsps have been observed in infectious diseases
involving bacteria, (Cohen and Young, 1991), fungi (Matthews and Bumie, 1992)
and protozoa (Jendoubi and Bonnefoy, 1988).
62 Introduction
A role for Hsps as important antigenic targets in the host immune response to infection, is likely to derive from several contributory factors. The induction of Hsp synthesis in microbes in response to stressful conditions such as infection, contributes to the abundance of Hsps in infecting microbes (Polla, 1991). Moreover, as discussed in Section 1.1 of this chapter, a high degree of sequence homology is observed for
Hsp family members (particularly Hsp 90, Hsp 70 and Hsp 60), between microbial species. For example, the Mycobaterium Hsp 60 and GroEl of E.coli, share a broad range of immunological cross-reactive determinants (Shinnick et al., 1988). Thus, infection and exposure of the immune system to microbial Hsps leads to immunologic memory which following subsequent infections, is capable of recognising and responding to Hsps from a range of different organisms. These factors therefore, are likely to explain the prominence of Hsps as major antigenic determinants of infecting microbes and the importance of host anti microbial Hsp responses in promoting the efficient recognition and clearance of such infections.
Cell Surface Expression of Heat Shock Proteins
As discussed earlier in this section, Hsp determinants are expressed on the cell surface following microbial infection. Indeed, their subsequent presentation in the context of different MHC molecules and the generation of Hsp-specific immune responses, has been suggested to be important to the host immune response to microbial infection. The mechanisms responsible for the relocalisation of Hsps (which are typically cytoplasmic) to the cell surface, are at present poorly understood. Various studies have demonstrated such relocalisation in different situations. For example, Hsp 90 (La Thangue and Latchman, 1988), (Ferrarini et al., 1992), Hsp 70 (Poccia et al., 1992) and Hsp 60 (Fisch et al., 1990), (Kaur et al., 1993), have been observed to be surface expressed by lymphoma cells. Hsp surface expression has also
63 Introduction been reported for non-transformed cells. For example in patients with SLE, Hsp 90 is surface expressed by peripheral blood lymphocytes (Brkeller-Yuksel et al., 1992). As shown in Table 1.2.1 and discussed later, studies of other autoimmune diseases have also reported the surface expression of Hsps, particularly at sites of active tissue disease.
Heat Shock Proteins and Autoimmune Disease
The process of self-tolerance is therefore fundamental to the normal function of the inunune system and requires the recognition of a potentially infinite variety of foreign antigens, while discriminating against self antigens. However, as will be discussed, a loss of self-tolerance involving the activation of self-reactive lymphocytes, are defining features of autoimmunity, with possible pathological consequences. Importantly, as shown in Table 1.2.1, studies of various autoimmune diseases have reported an increased prevalence of self-Hsp-specific responses in patients, involving both antibodies and T cells. Moreover, in some cases these responses have correlated with active disease.
Antibodies to Heat Shock Proteins
Antibodies to Hsps have been found in a number of human autoimmune diseases. In SLE of both adult (Conroy et al., 1994) and childhood onset (Conroy et al., 1996), raised serum levels of antibodies to Hsp 90 have been reported. Overexpression of Hsp 90 in PBMCs has been observed in a subset of patients with SLE and in these patients this correlated with active central nervous system and cardiovascular/respiratory disease (Dhillon et al., 1994). Interestingly, increased surface expression of Hsp 90 in PBMCs and raised antibodies to Hsp 90, correlated with a subset of patients which has been associated with Hsp 90 overexpression (Conroy et al., 1994).
64 Introduction
Table 1.2.1 Immune Response to Hsps in Human Autoimmune Diseases
Autoimmune Disease Features
Systemic lupus erythematosus Overexpression of Hsp 90 in PBMCs (Dhillon et al., 1994). Raised antibody levels to Hsp 90 in patients of adult (Conroy et al., 1994) and childhood onset (Conroy et al., 1996).
Sjogren’s syndrome Gastric involvement and Helicobacter pylori previously reported. Increased prevalence of antibodies against Hsp 60 of Helicobacter pylori.
Rheumatoid arthritis Raised levels of antibodies to mycobacterial Hsp 60 (Tsoulfa et al., 1989) which react with synovial tissue (de Graeff-Meeder et al., 1990). Increased frequency of B cells specific for mycobacterial Hsp 60 in synovial fluid (Rudolphi et al., 1997).
Multiple sclerosis Increased prevalence of antibodies to alpha-crystallin correlate with active disease and disease severity (Agius et al., 1999).
Atherosclerosis Levels of antibodies against mycobacterial and human Hsp 60/65 are increased in patients with atherosclerotic lesions (Kleindienst et al., 1995), (Schett et al., 1997), (Metzler et al., 1998).
Coeliac disease Enhanced expression of epithelial cell mitochondrial Hsp 65 in patients with this disease correlates with significantly higher gammadelta+ T cell densities than those with normal Hsp 65 expression (Iltanen et al., 1999). Introduction
Behcet’s disease T cells in this disease proliferate vigorously in an antigen-dependent manner in response to a specific peptide of human Hsp 60 (Sakane, 1997), (Kaneko et al., 1997).
Psoriasis Antibodies to mycobacterial Hsp 60 in serum of patients (Rambukkana et al., 1993), (Izaki et al., 1994) Autoimmune thyroiditis Increased prevalence of IgM antibodies to human Hsp 72 and Hsp 73 in serum of patients (Appetecchia et al., 1997).
Graves’ disease Patients have increased prevelance of antibodies to Hsp 72, without correlation with any disease characteristic (Prummel et al., 1997).
Menier’s disease Increased prevalence of antibodies to Hsp 70 in serum of patients (Shin et al., 1997). Introduction
Increased levels of antibodies to Hsp 70 have also been reported in various autoimmune diseases, such as Menier’s disease (Shin et al., 1997). However, any involvement of Hsp 70 in autoimmune pathogenesis remains unclear. For example, studies of patients with autoinunune thyroiditis (Appetecchia et al., 1997) and grave’s disease (Prummel et al., 1997) showed raised antibody levels to Hsp 70, without correlation with disease activity. In multiple sclerosis (MS), increased expression of Hsps occurs in various cells of the central nervous system during active disease (Bimbaum, 1995), (Aquino et al., 1997) and Hsp 70-specific antibodies have been detected in serum and cerebrospinal fluid of patients. A role for antibodies to Hsp 70 in autoimmune pathology has been suggested and is supported by the observation that antibodies to Hsp 70 react with human myelin proteins of normal individuals and with oligodendrocytes in regions of MS demyelination (Bimbaum, 1995). Thus it is possible that immune responses to Hsps or cross-reactive Hsp epitopes, may influence disease progression, possibly involving demyelinating processes in patients with MS (Bimbaum and Kotilinek, 1999). In RA, a number of studies have reported humoral responses to Hsp 60 and in some cases, these responses have been linked to affected tissues. For example, raised levels of antibodies to mycobacterial Hsp 60 have been reported in RA (Tsoülfa et al., 1989), (McLean et al., 1990) and have been reported to be reactive with synovial tissue in patients with RA (de Graeff-Meeder et al., 1990). Furthermore, an increased frequency of B cells specific for mycobacterial Hsp 60 has been found in the synovial fluid and tissues of affected joints, in patients with RA (Rudolphi et al., 1997). Moreover, in another study, intra-articular administration of mycobacterial Hsp 65 in Wister rats sensitised to M. tuberculosis, induced the production of antibodies to Hsp 65 and joint inflammation, supporting the hypothesis that Hsp 65 is an arthritogenic stimulus (Winrow et al., 1994). In juvenile chronic arthritis (JCA), raised levels of antibodies to human Hsp 60 have been reported, particularly in those patients with polyarticular onset (de Graeff-Meeder et al., 1993). Infection with the bacterium Helicobacter pylori may be linked to gastric involvement in patients with Sjogren’s disease and raised levels of antibodies to Hsp 60 of H.pylori have been reported in this disease (Aragona et al., 1999). Increased levels of antibodies specific to mycobacterial Hsp 60 have also been reported in patients with psoriasis (Izaki et al..
67 Introduction
1994) and increased levels of these antibodies have been observed to correlate with active disease (Rambukkana et al., 1993). In atherosclerosis, (Kleindienst et al., 1995), (Schett et al., 1997), levels of antibodies to mycobacterial and human Hsp 60 are increased in patients with atherosclerotic lesions. Moreover, macrophages in atherosclerotic lesions have been shown to express high levels of Hsp 60 in these patients.
T Cell Responses to Heat Shock Proteins
T cell responses to Hsps have been reported in a number of autoimmune diseases. As for the antibody responses detailed above, it is often found that the prevalence of these responses is higher in patients than controls. Moreover, as will be discussed, overexpression of self-Hsps in regions of tissue inflanunation have been linked to these responses and shown to correlate with disease activity. In multiple sclerosis (MS), ap-T cells reactive with Hsp 70 and Hsp 60 are more prevalent in patients with this disease compared to controls (Salvetti et al., 1992), (Bimbaum et al., 1993). Moreover, the response of aP-T cells to Hsp 70, involved the cross-recognition of both microbial Hsp 70 and self Hsp 70, implicating Hsp 70 as a potential autoantigen in MS (Salvetti et al., 1996). In rheumatoid arthritis (RA), overexpression and cell-surface localisation of Hsp 60, has been reported on synovial lymphocytes (Sato et al., 1996) and ap-T cells from inflamed sites have been demonstrated to be reactive to both mycobacterial and self-Hsp 60 (Li et al., 1992). In another study of the reactivity of both synovial and peripheral blood T cells to Hsp 60, T cells isolated from patients with RA and normal controls, were found to be preferentially reactive to mycobacterial Hsp 60 and not human Hsp 60. Evidence of in vivo activation and clonal expansion of these mycobacterial Hsp 60-reactive T cells was also reported (Cells et al., 1997). In Behcet’s disease, T cells isolated from patients have been observed to proliferate vigorously to a specific peptides of both human (Sakane, 1997) and mycobacterial (Pervin et al., 1993) Hsp 60.
68 Introduction
Summary
Studies supporting a functional role for Hsps in the immune system, particularly in antigen processing and presentation, have been discussed in this section. Moreover, accumulating evidence has been summarised, supporting a role for Hsps as immunodominant antigens in protective host immune responses to microbial infection. This observation may in part be explained by high sequence homology between Hsps from different microbes. A role for Hsps in pathogenic immune responses in various human autoimmune has also been outlined. However, an understanding of the mechanisms responsible for such responses and their contribution to the development of autoimmune disease, is unclear. Interestingly, it appears likely that in certain situations, such as autoimmune disease, sequence homology between Hsps may contribute to the development of cross-reactive pathogenic T cell responses (a process known as “molecular mimicry”). As discussed earlier, this situation may occur in various autoimmune diseases and involve the elevated expression and surface localisation, of Hsps at sites of tissue inflammation. Thus, it is likely that such a role for Hsps, rather than initiating a pathogenic autoimmune response, may enhance and occur secondary to the presence of active tissue disease. As has been discussed, a number of contributory factors are involved in the development of autoimmune disease. The role of Hsps in SLE is discussed in detail in Section 1.4 after a discussion of SLE itself in Section 1.3.
69 Introduction
Section 1.3: Systemic Lupus Erythematosus
Introduction
Systemic lupus erythematosus (SLE) is a human autoimmune rheumatic disease and descriptions of its various manifestations have been described in literature dating back many centuries. Research towards an understanding of the mechanisms in SLE which leads to widespread tissue damage, involving many of the body’s organs or systems, has been aided greatly by the development of several animal models. From all these studies, it is certain that SLE develops as a result of the complex interaction of a variety of factors: genetic, hormonal and environmental. At present however, the specific aetiology of this disease remains unclear, and is the subject of much debate. In this section, the variety of clinical manifestations which characterise SLE are summarised. This is followed by an outline of the various immunological abnormalities which are observed in patients with SLE and their potential pathological consequences.
Definition And Classification
SLE may be defined as a disease with a complex multifactorial aetiology, characterised by inflammation and the involvement of most of the body’s organs and systems. Not surprisingly, the expression of this disease covers a wide clinical spectrum and may overlap with other autoimmune rheumatic diseases such as rheumatoid arthritis, dermatomyositis/polymyositis and Sjogren’s disease. Such diversity in clinical presentation with no uniquely identifying characteristic (Snaith, 1992), has necessitated a consensus of agreement on criteria for the classification of SLE. In 1971, the American Rheumatism Association (ARA), now the American College of Rheumatology (ACR), published criteria for the classification of SLE and these are discussed in Cohen and Canoso, (1972). These criteria have been revised by the Diagnostic and Therapeutic Committee of the ACR (Tan et al., 1982), (Hochberg,
70 Introduction
1997) and currently comprise 11 manifestations which may be observed in SLE. Of these (see Table 1.3.1), the ACR has proposed that a person with SLE must present with a minimum of four, serially or simultaneously, during any period of observation. These criteria have been tested for their sensitivity for SLE and specificity against other autoimmune diseases. Using an ACR database of 590 patients with SLE, scleroderma or dermatomyositis/polymyositis, the ACR criteria published in 1982 (Tan et al., 1982), were found to be 83% specific for SLE and 89% specific against the combined scleroderma and dermatomyositis/polymyositis patients (Fries, 1987). Although the ACR criteria were intended to be used as a classification tool appropriate for use in clinical trials, its use has combined with the need for a practical case definition of a patient with SLE. Problems however, with the use of the ACR criteria for diagnostic purposes have been highlighted (Levin et al., 1982). In this report, less than 50% of patients studied, fulfilled the ACR criteria at any one time. Furthermore, in some patients, fulfilling the criteria required a period of observation of between 9 and 20 years.
71 Introduction
Table 1.3.1 Criteria of the American College of Rheumatology for the Classification of SLE from Tan et al., (1982), Hochberg, (1997)
Criterion Definition
1) Malar Rash Fixed malar erythema, flat or raised. 2) Discoid Rash Erythematous-raised patches with adherent keratotic scaling and follicular plugging. Atrophic scarring may occur in older lesions. 3) Photosensitivity Skin rash occurring as an unusual reaction to sunlight.
4) Oral Ulcers Oral or nasopharyngeal ulcers. 5) Arthritis Non-erosive, characterised by tenderness, swelling or effusion. Involving two or more peripheral joints. 6) Serositis (a) Fleuritis or (b) Pericarditis. 7) Renal Disorder (a) Persistent proteinuria ( > 0.5g/24 hr or 3+) or (b) Cellular casts. 8) Neurologic Disorder (a) Seizures or (b) Psychosis (in absence of other causes). 9) Hematologic Disorder (a) Haemolytic anaemia or (b) Leukopenia (> 4.0 x 10^/ml on two or more occasions) or (c) Lymphopenia (> 1500/ml on two or more occasions) or (d) Thrombocytopenia (< 1 x 10^/ml).
10) Immunologic Disorder (a) Raised anti-double-stranded DNA antibody binding or (b) Anti-Sm antibody or (c) Positive finding of antiphospholipid antibodies based on: (1) an abnormal serum level of IgG or IgM anticardiolipin antibodies, (2) a positive test result for lupus anticoagulant using a standard method, or (3) a false-positive serologic test for syphilis known to be positive for at least 6 months and confirmed by Treponema pallidum immobilization or fluorescent treponemal antibody absorption test.
11) Anti-nuclear Antibody Raised titre of anti-nuclear antibody. Introduction
Epidemiology
The widespread use of the ACR criteria for the classification of SLE, has facilitated the epidemiological study of this disease. Thus, information obtained from patients with SLE world-wide, has allowed good estimates to be made for prevalence, morbidity and prognosis/survival in this disease. In addition, such population studies are valuable for both the development and subsequently clinical trials, of potential therapeutic drugs in SLE.
Prevalence
SLE is recognised as a disease with a world-wide distribution and primarily affects females of child-bearing years (Morrow et al., 1999). Studies to estimate the prevalence of SLE in various countries, have highlighted geographical differences in the reported figures. In Japan (Nakae et al., 1987) and Sweden (Nived et al., 1985), a prevalence of 21 and 39 cases of SLE respectively per 100,000 of total population, have been reported. In the United States, estimates of the prevalence of SLE have been reported to range between 14.6 (Siegel and Lee, 1973) and 50.8 cases (Fessel, 1974), per 100,000 persons and a later study in 1985 (Michet et al., 1985), reported an overall figure of 40 cases per 100,000 persons. More recently however, Hochberg et al., (1995) used a method of telephone-screening for physician diagnosed SLE in women aged 18 years and older and suggested previous studies to under-estimate the prevalence of SLE in the USA by at least 3-fold. In Europe, prevalence figures have ranged from 12.5 per 100,000 females in England (Hochberg, 1987) to a high of 39 per 100,000 people (both sexes combined) in Sweden (Jonsson et al., 1989). Such variability in estimates of prevalence are likely in part to reflect differences in the methodology used in each study, for case ascertainment. More significantly, an ethnic bias for the prevalence of SLE is recognised. Two studies in the United Kingdom, have highlighted significant variation in the prevalence of SLE among different ethnic groups sharing similar urban environments. In Nottingham, UK, the prevalence of SLE was reported as 3.7 and 45.4, per 100,000 males and females respectively. The authors also reported that SLE was found to be more
73 Introduction prevalent amongst Afro-Carribean groups (Hopkinson et al., 1993). A similar study in Birmingham, UK, reported significant differences in the prevalence of SLE depending on ethnic group and irrespective of place of birth. Prevalence rates of 36.2, 90.6 and 206 per 100,000 women were reported for Caucasian, Asian and Afro- Carribean ethnic groups respectively (Johnson et al., 1995). Reports however, of the relevance of ethnicity in population studies to the prevalence of SLE, can be conflicting. For example, in the United States and Carribean, approximately 1 in 500 black women suffer from SLE, however it appears that SLE is uncommon among African blacks (Fessel, 1988). Thus, all these studies ultimately suggest a complex multi-factorial aetiology in SLE and it is likely that a number of genetic (discussed later), environmental and sociological factors combine to cause this disease.
Clinical Features of SLE
A wide spectrum of clinical features are observed in patients with SLE, as varied as skin rashes and kidney failure. The treatment of any one clinical feature may be followed by the later emergence of others, with unpredictable severity. Thus, periods of mild illness, which often involve the skin and joints (Worrall et al., 1990), may be followed by major involvement of the kidney, lung or central nervous system. Although SLE may once have been thought of as fatal, the wide-spread use of tests for measuring the titres of autoantibodies associated with SLE pathogenesis (notably anti-nuclear antibodies), has facilitated the identification of mild forms of the disease. More judicious use of immunosuppressive drugs and the development of dialysis and renal and heart/lung transplant, have all combined to improve the survival figures from 50% at five years in 1950, to 95% at five years now (see Morrow et al., 1999).
74 Introduction
Non-Specific Features
A number of general and non-specific features are observed in patients with SLE and do not form part of the ACR criteria. Of these, lethargy is frequently associated with active SLE (Wysenbeek et al., 1993). Swelling of the lymph glands around the neck and in the armpits may also be observed. If profound, a tissue biopsy may be required to exclude malignancy. Patients may also experience weight loss, fever, alopecia and on occasion nausea.
Dermatological Involvement
Skin manifestations are a presenting feature in many patients with SLE and these have often been noted to occur before systemic spread of the disease (McGehee et al., 1954). Cutaneous involvement can take many forms, including malar erythema (butterfly rash), nail base erythema, mucous membrane lesions, bullous lesions, maculopapular and discoid lesions, moderate or severe livedo reticularis and splinter haemorrhages. The classic butterfly rash is an erythematous and papular facial lesion involving the cheeks and bridge of the nose and may be present in up to 68% of patients (Grigor et al., 1978). Many patients are photosensitive (commonest in white females) and exposure to UV radiation may precipitate such skin problems (Nived et al., 1993). Vasculitis affecting the fingers and toes often occurs. Patchy, diffuse scarring or non-scarring alopecia is a common feature of SLE and was listed as a criteria in the initial 1971 ARA report for the classification of SLE (Cohen and Canoso, 1972). However its non-specificity to SLE and difficulties of interpretation have caused this manifestation to be excluded subsequently.
Musculoskeletal Involvement
Arthralgia in SLE is typically polyarticular, symmetrical and flitting and occurs in approximately 90% of patients with SLE, of whom half will suffer an accompanying early morning stiffness. Joint involvement tends to be mild, however approximately 5% of these patients will suffer severe arthritis with joint deformity in
75 Introduction the hands (Jaccoud’s arthritis) (Szczepanski et al., 1992). In such cases, examination of the synovial fluid usually demonstrates a low white blood cell count(< 3000 cells/ml) and a predominance of mononuclear cells (Pekin, 1970). Immune complex deposition has been described in synovial tissue (possibly involved in observed synovial lesions) and is often found in muscles of patients with SLE, although this may not parallel significant muscle disease (Russell and Hanna, 1988).
Renal Disease
Active renal disease is associated with morbidity in SLE. Clinically this may present with weight gain, shortness of breath, ankle swelling, haematuria and/or proteinuria. Monitoring of blood pressure and various laboratory indicators such as red or white cell casts in the urine, serum urea, is important (Esdaile et al., 1989). Renal biopsy can be used to identify immune complexes (associated with nephritis pathology) in the glomerulus, which appears to be a focal point for renal disease (McCluskey, 1982). Information from a range of histological studies relating to renal biopsy, has led to the World Health Organisation (W.H.O) to describe five categories of renal disease, ranging from normal to diffuse membranous glomerulonephritis on light/electron microscopy. Subsequently, a more detailed classification of lupus nephritis has been suggested following an international study of kidney disease in children (Churg and Sobin, 1982). The major manifestations range from minimal to advanced sclerosing glomerulonephritis, on light microscopy. The significance of renal biopsy to general prognosis in patients with SLE is debatable. It has been suggested that biopsy-derived information may facilitate predictions of renal involvement, but is less relevant to the clinical stage of the disease as a whole (Schwartz et al., 1993).
76 Introduction
Involvement of the Nervous System
The involvement of the nervous system in SLE has been well documented and features range from mild migraine headaches in up to 40% of patients, epilepsy as a presenting feature in 5% of cases and major psychosis. Grand mal seizures are the most common severe form of epilepsy and eventually manifest in upto 20% of patients with SLE (Morrow et al., 1999). A range of autoantibodies, including those with specificity for small nuclear libonucleoproteins (RNPs) such as Sm RNPs (involved in RNA splicing) as well as intracytoplasmic molecules such as ribosomal- P, have been linked to neurological disease, and this is currently an area of active research (Hay and Isenberg, 1993). Nervous system involvement may be sub-divided into central/cerebral, peripheral lesions and psychological manifestations. Central/cerebral involvement may be due to a variety of mechanisms including vasculitis, cross-reacting anti-neuronal/lymphocyte antibodies and anti phospholipid antibodies. In addition, cerebral disease may occur as a secondary event due to biochemical disturbances such as uraemia. Immune complex deposition may contribute to diffuse manifestations and the presence of anti-phospholipid antibodies is also associated with various movement disorders in SLE, such as chorea. Psychiatric disturbances have been frequently reported in patients with SLE. These include depression and anxiety and may also be associated with an impairment of cognitive function (Rogers, 1983). These problems have been reported in up to 50% of patients with SLE (Hall et al., 1981).
Haemopoietic Involvement
Chronic normocytic anaemia (normochromic) is frequent in patients with SLE. Levels of ferritin in these patients are often normal, however iron deficiency may occur as a result of corticosteroid or non-steroidal anti-inflammatory drug therapy, which can lead to gastric bleeding. Haemolytic anaemia (Coombs’ positive) occurs in approximately 10% of patients, although this may not always be associated with haemolysis (Morrow etal., 1999).
77 Introduction
Leucopenia (< 4xlO^/L) and lymphopenia (< 1.5xlO^/L) are frequently observed abnormalities of white blood cell count in SLE. Such immune disturbances may be mediated by autoantibodies and lymphocytotoxic antibodies feature in approximately one third of patients. Other factors, including decreased marrow production of white blood cells and immunosuppressive consequences of corticosteroid therapy, may also contribute to the observed decrease in white blood cell count in patients with SLE. Thrombocytopenia may be chronic (< lOOxlO^/L) and is not usually associated with bleeding episodes. Acute thrombocytopenia is less frequently observed in SLE and is associated with coagulation abnormalities due to a fall in platelet count which may lead to severe or fatal bleeding episodes. These patients typically feature circulating anticardiolipin antibodies, recurrent cerebro/cardiovascular thromboses and pulmonary hypertension (Asherson et al., 1983). Such disease manifestations associated with the presence of serum anticardiolipin antibodies in patients with SLE, are collectively referred to as antiphospholipid syndrome (Cabral et al., 1992).
Cardiorespiratory/V ascular Involvement
Pulmonary disease is common in SLE and typically causes pleuritic chest pain and cough in approximately half of all patients at some time. Poor diaphragmatic movement and mild dyspnoea may be indicators of abnormal pulmonary function, particularly reduced total lung capacity and flow rates, in upto 50% of patients (Morrow et al., 1999). Diaphragmatic dysfunction in SLE may contribute to the recognised but uncommon condition “ shrinking lung syndrome” (Hoffbrand et al., 1992) involving reduced chest expansion and pathologic lesions of the alveoli. Parenchymal changes in the lung associated with SLE, have been reported in 18% of patients (Haupt et al., 1981) and include interstitial fibrosis, pulmonary vasculitis and interstitial pneumonitis. A chronic non-productive cough may be present in such patients and usually accompanies shortness of breath. Following an episode of acute lupus pneumonitis, a more acute presentation may occur (Weinrib et al., 1990). Inflammatory lesions may be present in the pleural and pericardial membrane and
78 Introduction may in a few patients correlate with immune complex deposition.
Management
Although there is no cure for SLE, the majority of the clinical manifestations observed in this disease can be successfully controlled. Thus, continuous monitoring of disease activity through clinical and laboratory parameters, is vitally important. Certain precautionary measures are recommended and these include reduced exposure to ultra-violet light and avoidance of stress where feasible. Joint pain, skin rashes and fatigue, can often be controlled by combinations of non-steroidal anti inflammatory drugs such as hydroxychloroquine. The majority of immunomodulatory treatments involve corticosteroids, such as prednisolone, particularly in the presence of renal, major central nervous system and cardiovascular/respiratory involvement. In combination with steroids, other drugs such as azathioprine and cyclophosphamide, may be used for their immunosuppressive effects. However, certain side effects have been observed as a result of the over-use of these drugs. These include infection due to the immunosuppressive effects and steroids also cause osteoporosis, high blood pressure and diabetes mellitus.
Prognosis and Survival
Although SLE may once have been considered to be both a serious and fatal disease, the wide-spread use of immunosuppressive drugs has improved both prognosis and survival in patients with this disease. Most fatalities may be attributable to active disease, particularly renal, central nervous system disease and atherosclerotic vascular disease (Ginzler et al., 1982). Renal disease, as measured by serum creatinine or urine protein assessment, is the most important predictor of poor outcome (Ginzler et al., 1982), (Reveille et al., 1990). Patients receiving high-dose steroid and cytotoxic therapy may also succumb to opportunistic infections due to the immunosuppressive effects of these drugs. A long-term study of 100 patients with
79 Introduction
SLE in the UK, showed a 5 year survival rate of 88% and a ten year survival rate of 86% (Worrall et al., 1990). During the study period there were 13 deaths (mean age 36.5 years) and half of these were attributable to SLE (including renal and central nervous system disease, thrombocytopenic purpura). An analysis of demographic factors associated with morbidity in SLE has highlighted poorer survival rates in Indian Asians in the UK, compared to their Caucasian counterparts, with an increased frequency of renal and central nervous system complications (Samanta et al., 1991). Similar observations are found for Afro-Carribeans in the United States (Ginzler et al., 1982), (Nossent, 1993) compared to Caucasians, with a greater frequency of central nervous system involvement and higher mean serum creatinine levels. In general, non-Caucasian and poor socio-economic environment are factors which have been associated with decreased survival in a number of SLE population studies (Studenski et al., 1987).
Immunopathology In SLE
In this section, the factors which may predispose an individual to developing autoimmunity are considered. Finally, the various abnormalities observed in the immune system of patients with SLE are highlighted, and their role in the disease process discussed.
Genetic Susceptibilitv
The understanding that an individual’s genetic background may influence susceptibility to acquiring SLE and be involved in the disease process derives from observations in different populations; family studies and individual genetic analysis. SLE primarily affects females of child-bearing age and as already discussed (see p. 89) displays an ethnic bias towards Afro-Carribeans and to a lesser extent Orientals. Furthermore there is a greater concordance for this disease in monozygotic twins (25%) compared to 3% in dizygotic (Deapen et al., 1992). In relatives of patients with SLE there is a 3% risk of SLE and an increased prevalence of autoantibodies in
80 Introduction the serum of healthy relatives with SLE, reviewed in Miles and Isenberg, (1993). Genome-wide linkage analyses have been used to identify regions of chromosomes or individual genes, in genetic linkage with disease. In humans, regions of chromosome 1 (lq2I-23, lq41-4) (Moser et al., 1998) and of the major histocompatibility complex (discussed later) region on chromosome 6 (6pll-21) (Gaffney et al., 1998), have been identified as potential lupus susceptibility loci. However, these studies also highlight the complexity of lupus genetics which differs among ethnic groups. Analysis of the region lq23 revealed loci in genetic-linkage with SLE in African-American pedigrees, but not in European Americans (Moser et al.,1998). Linkage analyses of murine models of SLE with different phenotypes and genotypes, have greatly aided the genetic dissection of SLE-susceptibility. The majority of these studies have been performed in New Zealand hybrid models (described later in this section) and have identified the genomic intervals Slel, Sle2 and Sle3 respectively on chromosomes 1, 4 and 7 (Kono et al., 1994; Morel et al., 1994) as conferring strong susceptibility to lupus. Comparisons of potential human susceptibility loci in SLE with studies in murine models of this disease show similarities. For example, the Slel susceptibility region on mouse chromosome 1 in New Zealand models is syntenic with the human SLE susceptibility interval lq23-31, which contains the locus for the Fc receptor gene FcyR2, shown to be in disease- linkage in both African- and European-Americans (Moser et al., 1998) and Korean populations (Song et al., 1998). Hence, genetic synteny between human and murine susceptibility regions suggests shared susceptibility genes and thus highlights the potential of murine models of SLE and congenic mouse strains to aid our understanding of lupus genetics in humans. Various candidate human SLE susceptibility genes are now discussed.
Sex Hormones
Female susceptibility to SLE over males implies the involvement of sex hormones. Experimental and human models of SLE have shown that sex hormones affect the immunological features of the disease. Androgens such as testosterone tend
81 Introduction to elicit an immunosuppressive effect, while oestrogens are immunoenhancing through B cell stimulation and regulatory suppression. (Ansar Ahmed et al., 1985). Treatment with testosterone in MRL-//?r///?r mice (a lupus-prone mouse model), is observed to reduce lupus-like symptoms without affecting lymphoproliferation (Carlsten et al., 1989).
Major Histocompatibility Complex
Research into the involvement of genes of the major histocompatibility complex (MHC) in human SLE has identified haplotypes associated with disease susceptibility and to autoantibody subsets (particularly MHC Class II genes). It is however important to note that any correlations between MHC haplotype and SLE must be understood in relation to ethnic background. In Caucasians, the composite haplotype HLA Al, B8, DR3 is present in 35% of Caucasians with SLE, compared with 6% in Caucasian normals. The presence of both Al and B8 alleles was associated with a relative risk of 8.0 and both B8 and DR3 alleles with a relative risk of 8.32 (Worrall et al., 1990). In early-onset SLE there is an association with the DRw8 haplotype, while the frequency of neuropsychiatrie involvement negatively correlates with DQ. Antibodies to DNA, nuclear and cytoplasmic antigens in patients also correlate with MHC (Arnett., 1997). HLA DR2 is associated with antibodies to Sm in Afro-Carribeans and these individuals show a reduced frequency of DR3. DR4 is associated with autoantibody production to RNP and confers a reduced risk to nephritis, while a reduced susceptibility to nephritis correlates with the DR2 haplotype. Non-HLA genes within the MHC have also been suggested to be involved in the pathogenesis of this disease. For example, polymorphisms within the tumour necrosis factor gene are associated with an increased susceptibility to SLE (Tarassi et al., 1998; Rood et al., 2000) while the genes encoding the complement components C2 and C4 (discussed below) have also been associated with SLE. Of particular interest to this dissertation is the regulation of the Hsp 90p gene in SLE (see Section 1.4 of this chapter) and the reported localisation of this gene on human Chromosome 6, in the MHC region (Durkin et al., 1993; Takahashi et al., 1994). The reported
82 Introduction differential regulation of this gene in genetic subsets of patients with SLE (Dhillon et al., 1994), implicates the existence of Hsp 90p promoter polymorphisms encoded onto particular HLA haplotypes. However, no studies have yet established the presence of such polymorphisms. Table 1.3.2 shows the chromosomal location of the genes of the major Hsp families in humans and mice.
Complement Deficiencies
Complement deficiency refers to a decrease in function of the complement system and is a feature of a number of autoimmune diseases including SLE. Activation of the complement system through the formation of immune complexes is intimately linked with autoimmune pathology. Deposition of immune complexes on tissue surfaces can result in a focus of inflammation which if perpetuated, may lead to tissue destruction. Complement deficiency in autoimmune disease can in part reflect conunitment of the complement system to immune complex clearance. The recognition of hereditary and inborn abnormalities of complement, and their correlation with clinical features of autoimmune disease, has prompted an understanding of complement genetics (for a review see Walport et al., 1997). The loci for some of the genes encoding complement components, notably C2 and C4, are found within the HLA region and are referred to as HLA class III antigens. HLA haplotype associations with complement deficiency are recognised for such components. SLE is associated with deficiencies of the early classical pathway components, most strongly with the complement components Clq, Clr, Cls, C2 and C4. Each component is inherited through two alleles and deficiency of either may therefore be partial (heterozygous deficiency), or complete (homozygous deficiency) (see Kemp et al., 1987). Congenital deficiencies of C2 and C4 are frequently in linkage disequilibrium with HLA DR2 and DR3, (see Schur, 1995). C4 comprises two distinct forms, C4A and C4B. A large number of alleles have been identified which may occupy each locus, including null alleles (Mauff et al., 1990). A C4A null allele is a common feature of complement deficiency and a single null allele increases the risk for developing SLE by 3 and two null alleles by 17. C4A deficiency is
83 Introduction
Table 1.3.2 Location of Hsp Genes in Disease Susceptibility Intervals in Human and Mouse Chromosomes
Humans
Hsp 90 lp22.1
Hsp 60 3p22.3
Hsp 90p 6p21.1- HLA
Hsp 90a 14p32.21 - not linked
Mouse
Hsp 40 C5 Sle 6
Hsp 40 C4 Mri
Hsp 105/Hsp 25 C5 Lbw3
Hsp 70-1 C17 H2
Hsp 84-1 C17 H2
Hsp 70-2/Hsp 86-1 C12 Lrdw2
Hspa4 C ll Lbw 8
Hspg2 (perlecam) C4 Lots of loci Introduction associated with impairment of immune complex clearance and solubilisation, leading to deposition in lungs and kidneys and subsequent inflammation. C4 deficiency may occur through a mutation within the C4 coding region, resulting in a deletion of C4A and part of C4B and is associated with DR3+ patients (Kemp et al., 1987). In drug- induced SLE, inhibition of C4A and C4B activity can occur due to binding of the drug to C4 active sites, thereby interfering with binding to immune complexes. The expression of complement receptors CRl and CR2 in phagocytes and T cells is significantly reduced in SLE (Wilson et al., 1986). Low CRl expression on erythrocytes (Walport et al., 1985) and peripheral blood leucocytes (Wilson et al., 1986) has been reported and is seen to correlate with disease activity. CR2 receptor expression on CD4+ and CD8+ peripheral T cells is increased in some patients with active SLE and may lead to increased T cell cytotoxicity. B cells in patients with active and inactive SLE show reduced CR2 expression (Wilson et al., 1986) and may be associated with abnormal cytokine regulation.
Immune Dvsregulation and Cellular Abnormalities in SLE
Immune disturbances involving lymphocyte and accessory cell abnormalities together with cytokine dysregulation are characteristic features of SLE immunopathology (for a review see Kirou and Crow, 1999). Circulating B and T lymphocyte numbers may be increased or decreased and may vary with disease activity. B lymphocyte hyperactivity and failure of T lymphocytes to suppress auto reactive B cell clones, leads to the formation of antibodies (40 -50 types) some of which are likely to be pathogenic, see Linker-Israeli et al., (1990). Cytokine dysregulation plays a pivotal role in the altered profile of lymphocyte populations in SLE, see Linker-Israeli, (1992). Furthermore, anomalies in the surface expression of cytokine receptors in various subsets of lymphocytes, may moderate the above cellular disturbances.
85 Introduction
B Lymphocytes
An increased number of antibody-producing B lymphocytes is observed in SLE patients compared to normals and correlates directly with disease activity (Blaese et al., 1980), (Sakane et al., 1988). Hypergammaglobulinaemia is a characteristic feature of SLE and is in part a consequence of elevated levels of activated B cells. Many factors are involved in the cause of B cell hyperactivity. Environmental and genetic factors, together with functional defects in auto-reactive B cells, have all been linked to B cell disturbances and the production of pathogenic auto-antibodies. The presence of a variety of auto-antibodies targetting in a non organ-specific manner implies the presence of polyclonal B cell activators, permitting antigen- independent stimulation of B lymphocytes. Increased sensitivity to cytokines promoting B cell proliferation may account for increased B cell numbers. A concomitant reduction in T lymphocyte subsets involved in suppressor activities, in response to dysregulated cytokine activity, may explain the ineffective regulation of hyperactive B lymphocytes in SLE. The presence of pathogenic specific auto antibodies produced by IgG secreting B lymphocytes may infer cytoplasmic overexpression of auto-antigens.
Cytokines
T cell involvement is important in the regulation of balance between humoral and cell-mediated immune responses. Understandably, alterations to lymphocyte populations will significantly alter the profile of cytokine synthesis and is a characteristic feature in SLE. Resulting abnormalities in immunoregulation form the basis for cytokine involvement in the immunopathology of SLE. SLE is characterised by an overall shift towards cells supporting humoral (antibody-producing) responses and impairment of cellular immunity. This is underlined by the recognition that IL-10 levels are raised in SLE, particularly in those patients with active disease (Park et al., 1998). The anti-inflammatory nature of this cytokine serves to promote B cell proliferation while impairing cell-mediated immunity (Moore et al., 1993).
86 Introduction
Type 1 cytokines support cell-mediated immunity while those of Type 2 support antibody production. The cytokine profile of helper T cells is used as a means of characterisation. Consequently Th 1 cells support cell-mediated immune responses, while Th 2 cells provide help for antibody production and suppress cell-mediated responses. TNF-a is produced by Thl cells, B cells, natural killer cells (NK) and macrophages. TNF-P is produced by activated lymphocytes. The genes encoding both cytokines are localised in close proximity within the major histocompatibility complex (MHC). TNF-a production in vitro by mitogen stimulated peripheral blood mononuclear cells (PBMCs) has been shown to vary according to the HLA class II make-up of the donor. A decreased production is observed in macrophages and lymphocytes from HLA-DR2 and DQwl donors, while elevated levels have been observed in HLA-DR3 and DR4 donors (Jacob et al., 1990). The DR2 and DQwl haplotype is associated with lupus nephritis. Conversely DR3 and DR4 individuals negatively correlate with lupus nephritis. It is interesting to note that TNF-a therapy has been shown to have a protective effect in B/W mice against renal pathology (Gordon et al., 1989). IFN-y serum levels are elevated in SLE (Funauchi et al., 1991), (al-Janadi et al., 1993), however inhibition of B and T cell proliferation is not observed. The number of cells spontaneously producing this cytokine are reportedly decreased. This feature appears to be most pronounced with increasing disease severity. A decrease in CD8+ T cell numbers may account for reduced IFN-y production leading to impairment of macrophage and NK cell-mediated cytotoxicity. Interleukin-1 (IL-1) production by accessory cells is decreased in most SLE patients, however monocytes spontaneously produce elevated levels of IL-1 (Sierakowski et al., 1987). Activation of T cells responding to IL-1 is diminished, even after addition of exogenous IL-1 in vitro. This suggests a possible defect at the level of the IL-1 receptor on T cells (Alcocer-Varela et al., 1984). Interleukin-2 (ID-2) is produced by CD4+ and CD8+ T cells. Production of this cytokine in mitogen-stimulated PBMCs from SLE patients may be normal or decreased (Alcocer-Varela and Alarcon-Segovia, 1982). In some cases decreased IL- 2 production has been observed to correlate with an increase in disease activity. The addition of exogenous IL-2 is not sufficient to restore T cell proliferation and
87 Introduction suppressor activity. However this may be overcome by in vitro treatment of SLE accessory cells with both IL-2 and IFN-y. Serum IL-2 levels have been reported to be increased in some SLE patients. Variations within both clinical and experimental data, require an understanding of whether significance is to be attached to levels of IL-2 or functional activity. IL-2 responsiveness in CD4+ T cells is reduced in SLE. This may in part be due to the observed aberrations in IL-2 receptor synthesis in these cells. The differential pattern of synthesis of IL-2 receptor subunits in SLE, gives rise to a heterogeneity of possible receptor affinities. Reduced responsiveness in CD4+ T cells to IL-2 may therefore reflect a sub-population expressing low-affînity receptors (Alcocer-Varela and Alarcon-Segovia, 1982). Interleukin-6 (IL-6) levels have been reported to be elevated in SLE (Linker- Israeli et al., 1991) and may contribute to B cell hyperactivity in this disease. Thus, cultured PBMCs from patients with SLE, have been demonstrated to spontaneously produce increased amounts of IL-6 (Klashman et al., 1991) and neutralising anti-IL-6 antibodies decrease immunoglobulin production (Linker-Israeli et al., 1991). Studies on the synthesis of IL-6 in patients with SLE, have identified B cells T cells and macrophages as the major cell types to synthesise this cytokine (Hashimoto, 1990). A role for IL-6 in the pathogenesis of SLE is further supported by the observation that levels of IL-6 in SLE correlated with disease activity (Linker-Israeli et al., 1991), (Spronk et al., 1992). Elevated circulating levels of interleukin-10 (IL-10) have been reported in patients with SLE, being highest in patients with active disease (Park et al., 1998). A pathogenic role for IL-10 in SLE has been supported by a number of studies. PBMCs from untreated patients with SLE spontaneously release large amounts of IL-10 in vitro (Linker-Israeli, 1992). Furthermore, spontaneous immunoglobulin production (particularly DNA antibodies) in PBMC from patients with SLE is increased in response to IL-10 and inhibited in response to a neutralising anti-IL-10 antibody (Llorente et al., 1995). Defects in the regulation of autoreactive B and T lymphocytes by a process known as programmed cell death (apoptosis) may play a pathogenic role in SLE. It has been shown that IL-10 increases Bcl-2 expression by germinal centre B lymphocytes (Mehrian et al., 1998), preventing their death and suggesting that the stimulatory effect of IL-10 on immunoglobulin production may be partly due to
88 Introduction increased B lymphocyte survival. There is also evidence that disease severity in patients with SLE correlates with an increased ratio of IL-10 secreting PBMC (Hagiwara et al., 1996). A role for IL-10 in murine SLE is supported by the observation that administration of anti-IL-10 antibodies to NZB/W lupus-prone mice, delays the onset of autoimmunity and production of autoantibodies (Ishida et al., 1994).
Apoptosis
Defects in apoptosis have been described in experimental models of SLE. Mutation of the membrane receptor Fas in MSUlpr-lpr mice (which is described later), leads to decreased apoptosis and accelerated autoimmunity (Theofilopoulos and Dixon, 1985), (Nagata and Golstein, 1995). In human SLE the study of programmed cell death is an active area of research. This has been prompted by an initial claim that 60% of SLE patients have elevated levels of a functionally abnormal soluble form of the Fas receptor, which may inhibit apoptosis in lymphocytes (Cheng et al., 1994). In this study, injection of a murine homologue to the human soluble Fas protein in normal mice led to a two to three-fold increase in splenic B and T cells as well as an increase in double negative CD4- CDS- and single positive CD4+ CD8+ T cells. Furthermore it was reported that the injection of soluble murine Fas in CDl mice led to induction of autoimmunity. However, the majority of studies now conclude that there are no Fas abnormalities of any significance in patients with SLE. Indeed, current studies suggest that there is no problem with apoptosis per se, but rather the impaired action of macrophages in the clearance of apoptopic material. Hence, such material is thought to be a source for autoantigens which induce and maintain autoimmune responses in patients with SLE (Herrmann et al,.1998) Elevated transcription of a proto-oncogene, Bcl-2 y may also contribute to autoimmunity in human SLE. Bcl-2 is found on the inner membranes of mitochondria, the endoplasmic reticulum and the nuclear membrane and is involved in the maintenance and development of adult tissue by preventing apoptosis in
89 Introduction specific cell types (Korsmeyer, 1992). Consistent with this role, Bcl-2 is found in long-lived B lymphocytes within the follicular mantle zone. Overexpression of Bcl-2 in CD4+ and CD8+ T cells from SLE patients, but not B cells, has been reported as well as in a subpopulation of CD3+ CD4- CDS- T cells. Furthermore levels of Bcl-2 in these cell populations have been shown to correlate with disease activity (Aringer etal., 1994).
Autoantibodies
SLE is characterised by the presence of a wide variety of circulating antibodies directed against self antigens. Antibodies to nuclear, cytoplasmic and plasma membrane antigens have been identified and these may provide markers for disease or play a role in tissue damage (see Section 1.2). Antibodies to nuclear antigens include those with specificity for Sm and UlRNP (Sharp et al., 1971). The former are found in 30% of black patients with SLE but only 10 % of white patients. Antibodies to Ro and La are linked to the neonatal lupus syndrome (skin rashes and heart-block in the new bom) (Lee and Weston, 1984). These antibodies are also associated with Sjogren’s syndrome secondary to a diagnosis of SLE. Autoantibodies to Clq have been reported and correlate with glomerulonephritis (Song et al., 2001) Antibodies to double-stranded (ds) DNA are mainly of the IgG isotype and a sub-population appears to be involved in tissue destruction. Deposition of IgG and complement has been identified in kidney biopsies from patients with glomerulonephritis, which have resulted from a previous focal immune response leading to inflammation. Eluates from affected kidneys show IgG molecules to have specificity and high affinity for ds DNA (Saito et al., 1994). It is not clear at present which mechanisms account for the localisation of these antibodies. Glomerular localisation may result from deposition of circulating andbody-DNA complexes, leading to activation of complement and resulting in local inflammation. Alternatively a mechanism of cross-reactivity may take place in which non-complexed antibodies with specificity for DNA, interact with components of the
90 Introduction glomerular basement membrane. In support of this hypothesis, heparan-sulphate, a major constituent of the glomerular basement membrane, is known to contribute an overall negative charge to the membrane. Antigen non-specific binding may therefore take place involving cationic anti-DNA antibodies. Furthermore heparin has been shown to inhibit DNA binding of antibodies eluted from the kidneys of humans and MKL-lpr/lpr mice. In the MKL-lpr/lpr model, administration of low, non anticoagulant doses of heparin, prevents development of nephritis in SLE (Naparstek et al., 1990). Antibody-DNA complexes have in some cases been shown to contain histones, Histones possess high affinity for heparan-sulphate (Termaat et al., 1990) and antibodies to histones feature in SLE.
Experimental Models of SLE
Ethical considerations clearly limit attempts to explore the process of pathogenesis in SLE (and other diseases). To help understand the mechanisms that give rise to clinical features, animal models have been developed, the majority of which are murine (Theofilopoulos and Dixon, 1985), dogs may also spontaneously develop SLE (Theofilopoulos and Dixon, 1985). These have aided in our understanding of the immunology of the disease. Furthermore, information gathered from animal models of SLE has contributed to the understanding of the genetic factors involved in the autoimmune response in this disease. From these studies, it is clear that multiple genes are involved in the clinical and immunological manifestations of this disease. The main features of the principle murine models of SLE are shown in Table 1.3.3. Studies of lupus-prone mice strains have shown that at least three separate genes are significant in affecting disease pathology. The MRL+^+ strain spontaneously develops late-onset lupus, especially in females. Clinical nephritis develops with increasing age and death occurs by 24 months with a 50% survival time of 18 months (Theofilopoulos and Dixon, 1981). The introduction of the Ipr (lymphoproliferation) gene, which includes an endogenous retroviral DNA sequence integrated into the Fas gene, leads to the expression of a dysfunctional Fas membrane receptor and leads to a loss of apoptosis. This mutation accelerates the disease
91 Table 1.3.3 Features of Lupus-Prone Mouse Strains
Strain H-2 Haplotype Sex In Which Disease Ciinicai Immunoioglcal
Is Most Severe Features Features
M RL +/+ H-2-C Female Lymphoproliferation Anti-nuclear antibodies, ICnephritis, rheumatoid arthritis, vasculitis, generalised lymphocyte dysfunction.
MRL-lpr/lpr H-2^ Female Same as MRL+/+ but mor severe.
NZB H-2( Male and Haemolytic anaemia. Anti-nuclear antibodies. Female glomerulonephritis, IgM hyper-antibody Erythrocyte production, generalised lymphomas lymphocyte dysfunction.
NZB/WF H-2'd/z Female Severe IC nephritis. Anti-nuclear, anti-DNA antibodies.
BXSB H-2' Male Haemolytic anaemia, Anti-DNA antibodies. Glomerulonephritis, anti-thymocyte Lymphadenopathy antibodies. Introduction process such that the 50% survival time is reduced to two to four months, with 90% death occurring by nine months. An overall amplification of antibodies from the parent MRL+/+ strain is attributed to lymphoproliferation, with an increase in CD4+ T cell help (Cohen et al., 1986). The introduction of the gld (generalised lymphoproliferative disease) gene into non-autoimmune C3H/HeJ mice results in lymphoproliferation and development of autoimmunity (Takahashi et al., 1994). In NZB/W F% mice the X-linked recessive gene, xid, is observed to protect against autoimmunity through prevention of terminal B cell differentiation and the formation of autoantibody-producing B cells.
Summary
The clinical expression of SLE is the consequence of its complex immunopathology, a combination of genetic, hormonal and environmental factors. The interaction of these factors leads to the production of pathogenic autoantibodies and the formation of immune complexes. Inappropriate regulation of cell mediated immune responses, leads to ineffective clearance of autoantibody and immune complexes and possible widespread tissue and organ damage. A growing number of studies have reported elevated levels of anti-Hsp antibodies in a number of human autoimmune diseases, including SLE. As will be discussed in the following section, anti-Hsp responses may be involved in the pathogenesis of autoimmunity.
93 Introduction
Section 1.4: Heat Shock Proteins In Systemic Lupus Erythematosus
Introduction
A role for heat shock proteins (Hsps) both in the normal immune system and in the aetiopathogenesis of autoimmunity, has been discussed earlier in this chapter (see Section 1.2). These studies have focussed particularly on a possible role for Hsp 60 and Hsp 70 family proteins of mycobacteria and other infectious organisms, in the pathogenesis of autoimmune diseases such as rheumatoid arthritis and vasculitis. More recently, a number of studies in both humans with systemic lupus erythematosus (SLE) and animal models of SLE, have suggested a role for Hsp 90 in the pathogenesis of this disease.
The Expression of Heat Shock Proteins In SLE
A role for Hsp 90 in the conformational regulation of steroid hormone receptors (see Section 1.1), has prompted a study of the effect of corticosteroid treatment on Hsp 90 levels in patients with SLE. Levels of Hsp 90 in peripheral blood mononuclear cells (PBMCs) from 60 randomly selected patients with SLE, were found not to correlate with corticosteroid treatment. However, levels of Hsp 90 were found to be highest in patients with active SLE compared to patients with inactive disease, healthy controls and patients with rheumatoid arthritis (Norton et al., 1989). In another study of Hsps in SLE, elevated levels of both Hsp 90 and Hsp 70 were observed in pooled peripheral blood mononuclear cells from 20 patients with SLE, compared with healthy controls (Deguchi et al., 1987). In agreement with these observations, a larger study of 102 patients with SLE showed significantly increased levels of Hsp 90 and Hsp 72 (the heat-inducible form of Hsp 70), compared with healthy controls. Importantly, for individual patients, this was not paralleled by
94 Introduction elevation of the heat-inducible protein Hsp 65 or the constitutively expressed protein Hsp 73 (Dhillon et al., 1993a). The specificity of Hsp 90 and Hsp 72 elevation is further underlined by the observation that transcription of the gene encoding the Hsp ubiquitin (described in Section 1.1), is also unchanged in SLE (Twomey et al., 1992). Thus, the increased expression of Hsp 90 and Hsp 72 in SLE occurs in the absence of elevation of other Hsp family members and therefore would not appear to result from a general cellular response to stress, which would involve the increased expression of a wide range of Hsp family members. It has been reported that approximately 30% of patients with SLE have elevated levels of Hsp 90 compared to normals (Dhillon et al., 1993a). In this study, only 10% of patients were found to have elevated levels of Hsp72. Furthermore, levels of Hsp 72 did not correlate with levels of Hsp 90, in individual patients. Thus, it appears that the elevation of these Hsps occurs as a result of distinct cellular processes within different sub-groups of patients with SLE. A genetic origin to Hsp 90 overexpression is suggested by the observation of significantly higher levels of Hsp 90 in PBMCs from patients with SLE who are HLA Al, B8, DR3 negative, compared to those who possess this haplotype (Dhillon et al., 1994). This haplotype is that most often found in Caucasians with lupus. Since this ethnic group comprised 60% of the study by Dhillon et al (1993a), the reported figure of 30% for patients with SLE overexpressing Hsp 90, may be understood in relation to the genetic background of the patients in this study. Overexpression of Hsp 90 in SLE, has been shown to be dependent on the enhanced transcription of the Hsp 90p gene, one of two separately coded isoforms of Hsp 90 found in mammalian cells (Twomey et al., 1993). In contrast, elevation of Hsp 72 is not dependent on enhanced transcription of the Hsp 72 gene and elevated levels of Hsp 72 mRNA were not observed. Post-transcriptional mechanisms which may include improved translatability of Hsp 72 mRNA, could account for the observed elevated protein levels of Hsp 72 in a subset of patients with SLE (Twomey et al., 1993). Any mechanism which is involved in Hsp 90 overexpression, occurs in a subset of patients with SLE and is a process distinct from that which leads to the general induction of Hsps in response to cellular stress.
95 Introduction
Genetic changes to regulatory sequences in the Hsp 90p promoter may account for the enhanced transcription of this gene, which is observed in SLE (Twomey et al., 1993). Such polymorphisms might also be found within the gene encoding a cellular transcription factor, leading to activation of the Hsp 90p promoter with increased efficiency. Alternatively, differences in transcription of the Hsp 90p gene in SLE may not be due to genetic changes of this type, but may occur as part of a normal mechanism of Hsp 90 gene regulation in cells, however one in which the regulatory mechanisms are overactive in response to cellular events specific to SLE. Since Hsp 90 is known to be induced by a number of stimuli both under physiological and stressful conditions, abnormalities of such inducers in patients with SLE is a consideration. Previous studies have highlighted cytokine abnormalities in lupus (Linker-Israeli, 1992), (Kirou and Crow, 1999). Interleukin-6 (Linker-Israeli et al., 1991) and interleukin-10 (Park et al., 1998) are observed to be elevated in SLE and particularly in those patients with active disease. Moreover, previous studies which have demonstrated that interleukin-6 (IL-6) can activate the Hsp 90p promoter in vitro (Stephanou et al., 1997), suggest a role for IL-6 in the regulation of Hsp 90 expression in the normal immune system. Thus it is conceivable that in SLE, elevated levels of circulating IL-6, may significantly contribute to the observed overexpression of Hsp 90 in this disease. A similar role for interleukin-10 (IL-10), which activates target genes using a similar signalling mechanism to IL-6, in the regulation of Hsp 90 gene expression, is also possible (see Section 1.1). Studies of Hsp expression in other autoimmune diseases, such as rheumatoid arthritis and Sjogren’s syndrome, show the pattern of Hsp 90 and Hsp 72 overexpression to be specific to SLE. Furthermore, there may be a role for Hsp 90 in distinct, clinically active patients with SLE (Dhillon et al., 1993a). Of those patients in this study overexpressing Hsp 90, patients with active neuropsychiatrie or cardiovascular disease were most likely to have elevated levels of Hsp 90. A comparison of Hsp 90 expression with the presence or absence of lupus autoantibodies has shown significant elevation of Hsp 90 in patients with lupus anti coagulant and/or anti-cardiolipin antibodies (Dhillon et al., 1994). It is conceivable that Hsp 90 overexpression and the presence of these antibodies may have a common genetic origin. Also possible is a mechanism in which the expression of these
96 Introduction autoantibodies is provoked in response to Hsp 90 epitopes which mimic those specifically recognised by these antibodies. In agreement with this hypothesis, acute thrombocytopenia, which is associated with antiphospholipid antibodies (Harris et al., 1985) correlated with Hsp 90 overexpression (Dhillon et al., 1994).
Surface Expression Of Hsp 90 in SLE
A number of studies have suggested a role for heat shock proteins as autoantigens in the pathogenesis of autoimmune disease (discussed in Section 1.2). For example, mycobacterial Hsp 65 may play a role in adjuvant arthritis in rats and increased levels of antibodies to Hsp 65 have been reported in patients with rheumatoid arthritis (Tsoulfa et al., 1989). Such a role for elevated levels of Hsp 90 in provoking an autoimmune response in SLE, is likely to require the localisation of Hsp 90 on the cell surface, providing accessibility to the immune system. Relocalisation of Hsp 90 to the cell surface is a recognised phenomenon during herpes simplex virus infection (La Thangue and Latchman, 1988), in human tumour cells (Ferrarini et al., 1992) and following mitogen activation of lymphocytes (Haire et al., 1988). Surface expression of Hsp 90 in peripheral blood mononuclear cells from patients with SLE has been reported (Erkeller-Yuksel et al., 1992). As determined by flow-cytometry, surface localisation was specific to Hsp 90 and was not observed for Hsp 70 or Hsp 60 family members. Furthermore, the degree of surface Hsp 90 expression correlated with the degree of Hsp 90 overexpression.
Antibodies to Hsps in SLE
Autoantibodies to Hsp 90 (Conroy et al., 1994) have been detected in the sera of patients with SLE and other autoimmune diseases. Previous studies (Minota et al., 1988), (Winfield and Jaijour, 1991), had produced conflicting data on the incidence of antibodies to Hsp 90 in SLE compared to normal individuals and subjects with other autoimmune diseases. However, using an ELISA system to detect antibodies to the native form of Hsp 90, rather than the reduced form used in the report by (Minota
97 Introduction et al., 1988), a study of 73 patients with SLE has shown raised IgG antibodies in 26% of subjects and raised IgM antibodies in 35% of the patients, compared to healthy normal individuals (Conroy et al., 1994). Patients with raised anti-Hsp 90 antibodies were most likely to lack the HLA A1/B8/DR3 haplotype. The absence of this haplotype was previously reported to correlate with overexpression of Hsp 90 (Dhillon et al., 1994). In contrast, elevated levels of antibodies to Hsp 70 were rarely found in individual patients with SLE.
Hsp 90 Expression inMRLUpr4pr Mice
Evidence of a primary genetic origin to Hsp 90 overexpression in SLE is supported by studies in autoimmune MSUlpr-lpr mice which develop an age- dependent lupus-like disease involving the production of high levels of double stranded DNA antibodies. Clearly, ethical considerations limit the study of lupus in humans prior to disease onset. Thus, investigations of the pathogenesis of autoimmunity in MSUlpr-lpr mice are clearly of interest. The primary genetic defect in this strain is the inactivation of the gene which encodes the Apo-1/Fas antigen which mediates apoptosis (Watanabe-Fukunaga et al., 1992). A study of Hsp 90 levels in various tissues in MKUlpr-lpr mice has been reported (Faulds et al., 1994). Levels of Hsp 90 were found to be elevated in the spleen of yWJlpr-lpr mice compared to MRL++ mice (which develops a milder version of the disease) and BALB/C mice, while an analysis of various other tissues including the kidney and liver, showed no such elevation. Elevated levels of Hsp 90 in the spleen of these mice occurred independent of age and preceded the onset of disease and antibodies to Hsp 90, suggesting a possible role for Hsp 90 in the development of autoimmunity. In contrast, elevated levels of Hsp 72 in MKUlpr-lpr mice, were found in spleen and liver tissue, compared to BALB/C mice. This elevation was age-dependent and increased with disease progression, possibly as a response to cellular stress. The observed elevation of Hsp 90 in MKUlpr-lpr mice was not paralleled by elevation of Hsp 73 or Hsp 60. The specific elevation of Hsp 90 in these mice (preceding the onset of disease) in the absence of elevation of other Hsps, parallels observations in human lupus (Dhillon et al., 1993a).
98 Introduction
Taken together, the studies outlined in this section suggest a link between Hsp 90 overexpression in SLE and the development of anti-Hsp 90 antibodies in this disease, with possible pathological consequences (see Figure 1.4.1). Any role for Hsp 90 overexpression in the pathogenesis of SLE, must involve specific environmental factors which regulate Hsp 90 gene expression. Furthermore, previous observations which highlight a genetic origin to Hsp 90 overexpression and the development of ant-Hsp 90 antibodies are consistent with a role for susceptibility genes in the development of SLE, which is discussed in Chapter 1, Section 1.3.
99 Figure 1.4.1 A Model for Hsp Overexpression and Autoimmunity In SLE
Susceptibility Genes Environmental F a c to rs eg Infection, Cytokines
Oveiiexpression And Suiface Localisation 1 r Of Hsp Immune Response To Exogenous Hsp Hsp
: : : : Human Cell
Autoimmune Reaction
Hsp
: : Human Cell
Figure 1.4.1 illustrates a hypothetical model for autoimmune pathology in SLE. An involvement of genetic and environmental factors leads to the overexpression and cell surface localisation of Hsps, with possible pathological consequences. Introduction
Section 1.5: Aims of Dissertation
It is clear from the studies presented in Sections 1.1 to 1.4, that the regulation of heat shock proteins (Hsps) within the cell is very finely tuned and occurs differentially in patients with systemic lupus erythematosus (SLE). Previous studies have associated the expression of the 90 kDa heat shock protein (Hsp 90) with active clinical subsets in SLE. Moreover, a role for Hsp 90 in the pathogenesis of SLE may involve the induction of antibodies to this protein. However, the mechanisms by which Hsp 90 is overexpressed in SLE are unclear and a link between this event and the induction of antibodies to Hsp 90 has not been established. A role for cytokines in the regulation of Hsps has been discussed in Section 1.1 and Section 1.4. Thus, interleukin-6 has already been shown to induce Hsp 90 expression and is known to be elevated in SLE. Taken together, these studies suggest that elevated levels of cytokines in SLE, may lead to the overexpression of Hsp 90 in SLE and the subsequent induction of antibodies to Hsp 90, with possible pathological consequences.
Outline of Investigations
1. The effect of the cytokine interleukin-10 (which is elevated in SLE) on Hsp 90 levels in different cells was investigated.
2. The relative effect of interleukin-10 and/or interleukin-6 on Hsp 90 levels in SLE, was studied by the quantitative determination and subsequent correlation, of the levels of Hsp 90 and cytokines in patients with this disease.
3. A role for Hsp 90 overexpression in the development of antibodies to Hsp 90 in SLE, was investigated by the quantitative determination of antibodies to Hsp 90 and subsequent correlation with levels of Hsp 90 in the same patients.
101 Introduction
Finally, associations of levels of cytokines, Hsps and antibodies to Hsps, with active disease in SLE, were studied using the BILAG index. Thus, the relative contribution of these variables to pathogenesis in SLE, was established in patients graded by active clinical subsets of this disease.
102 Chapter 2 Materials and Methods
Section 2.1: Analysis of Hsp 90 Gene Expression in Cultnred Cells
Materials
Laboratory Chemicals
General laboratory chemicals were of analytical grade and were purchased from either Sigma Chemical Company, Poole, Dorset, U.K, from Boehringer Mannheim, Lewes, East Sussex, U.K or from Merck Ltd., Poole, Dorset, U.K. Other reagents and glassware were purchased from the central supplies of the Medical Molecular Biology Unit.
Monoclonal Antibodies
Mouse monoclonal antibody AC88 was a gift from Professor D.O. Toft, Mayo Clinic Foundation, Rochester, Minnessota, USA. Antibody AC88 is raised in mice and is of class IgGp AC88 was developed following isolation of Hsp 90 from the water mould Achyla ambisexualis and has been shown to react with human Hsp 90 (Riehl et al., 1985). AC88 also reacts with a number of 25-40 kDa proteins which have been identified as heterogenous ribonuclear proteins (Minoo et al., 1989). Mouse monoclonal antibody N27F3-4 (class IgGJ was purchased from Stressgen Biotechnologies Inc., Canada and reacts with human Hsp 73 and displays faint reactivity with human Hsp 72. This was confirmed in the study of Dhillon, (1993b).
103 Methods
Cytokines
Recombinant human interleukin-10, interleukin-6 and interferon-y, were each purchased from Santa Cruz Biotechnology, Inc., USA. Each cytokine was provided as a 5 mg sample of lyophilised protein. This was subsequently resuspended in 1 ml of sterile distilled water and stored at-20°c, as single-use aliquots until required.
Cells and Cell Lines
The HepG2 Cell Line
The human hepatoma cell line HepG2, stably expressing the human IL-10 receptor (Lai et al., 1996), was a kind gift from Dr. Heinz Baumann, Roswell Park Cancer Institute, Buffalo, NY, USA.
Peripheral Blood Mononuclear Cells
Peripheral blood mononuclear cells (PBMCs) were obtained from heparinised venous blood. Approximately 20 ml of blood was drawn from each normal individual and this would yield on average 20 million cells for subsequent culture. Purification of PBMCs is detailed later.
Plasmids
The 5’ Hsp 90p promoter chloramphenicol acetyl transferase (CAT) reporter construct Hsp 90p-CAT (Rebbe et al., 1987) was a kind gift from Dr. Neil Rebbe (Washington University School of Medicine, St. Louis, MO). Construct Hsp 90p-CAT consists of a fragment -1044 to +36 relative to the transcriptional start site, coupled to a CAT vector. The 5’ Hsp 90a promoter CAT reporter construct Hsp 90a-CAT, was a kind gift from Dr. Lee Webber. Construct Hsp 90a-CAT consists of a fragment -1050 to +42 relative to the transcriptional start site, coupled to a CAT vector The Hsp 90p promoter wild-type and mutant STAT-3 vectors have been previously described (Stephanou et al., 1998b). Construct Hsp 90-S3B consists of a sequence of the Hsp 90p promoter from -643 to -623 relative to the transcriptional start site. Construct Hsp90-MS3B is identical to Hsp 90-S3B except for adeneine residues at position -641 and -631, replacing the wild type cytosine residues. Both the wild-type
104 Methods and mutant Hsp 90p promoter STAT-3 sequences (see Figure 2.1) are cloned upstream of the thymidine kinase promoter and the CAT reporter gene in the vector pBLCAT2. The wild-type STAT-3 expression vector contains the full coding region for STAT-3, cloned into the mammalian expression vector pEF-Bos and expressed under control of the elongation factor gene promoter (Minami et al., 1996).
Enzymes
Restriction enzymes and enzyme buffers were purchased from Promega, Southampton, U.K or from Life Technologies, Paisley, Renfrewshire, U.K.
Methods
Isolation of PBMCs
Fresh heparinised venous blood was diluted 1:1 with Roswell Park Memorial Institute medium (RPMI-1640) (Sigma, Dorset, U.K.). 20 ml aliquots of this mixture were layered onto 15 ml Ficoll-Pacque (Pharmacia, UK.) in Falcon tubes (Greiner, Germany). These samples were centrifuged at 800 x g for 30 mn at room temperature. The resulting buff y coat, containing PBMCs was transferred to a fresh falcon tube. PBMC obtained by Ficoll density centrifugation were pelleted at 3000 x g for 10 mn. at room temperature, washed once with PBS and pelleted in a microfuge at 13,000 x g for 1 mn at room temperature and stored at -70°c.
Cell Culture
HepG2 Cells
HepG2 Cells were maintained in monolayer cultures in Dulbecco’s Modified Eagle Medium (DMEM) supplemented with 10% (v/v) bovine foetal serum (Life Technologies, Inc.) at 37°c in a 5% CO^ incubator. For each experiment, cells were trypsinised and re-plated at a density of 10^ per well in DMEM plus 10% (v/v) foetal bovine serum in six-well plates (Nunc, Uxbridge,U.K) overnight
105 Figure 2.1 Hsp 90 Promoter Sequences in pBLCAT2
Synthetic Oligonucleotides
(a) Hsp90-S3E STAT-3 STAT-3 GCCTGGAAACTGCTGGAAAT HSE HSE HSE
(b) Hsp90-MS3E GCATGGAAACTGATGGAAAT
STAT-3 Consensus Sequence CTGGRAA
Figure 2.1 shows synthetic oligonucleotides corresponding to the wild-type Hsp 90p promoter STAT-3 binding site (a) and a mutant derivative (b), which were cloned upstream of the thymidine kinase promoter and the CAT reporter gene in the vector pBLCAT2. Methods
PBMCs
PBMCs were cultured at 10^ cells per well in RPMI-1640 medium plus 10% (v/v) bovine foetal serum (Life Technologies, Inc.) in six-well plates (Nunc, Uxbridge,U.K) at 37°c in a 5% CO^ incubator.
Stimnlation
Following plating, HepG2 cells or PBMCs were maintained in their appropriate complete medium overnight, then washed twice with PBS and subsequently cultured in complete medium containing 1% (v/v) bovine foetal serum (Life Technologies, Inc.) for a further 6 hr before addition of cytokines at the concentrations indicated in the figure legends in Chapter 3.
Analysis of Heat Shock Protein Levels in Cultured Cells
Preparation Of Whole Cell Lysate
Cells were removed from six-well plates using a rubber policeman and pelleted in a microfuge at 13,000 x g for 1 mn. The resulting pellet was resuspended in ice cold 0.1 M Tris (pH 7.5) and subjected to three cycles of freeze-thaw in liquid nitrogen and the lysate pelleted at 4°c in a microfuge for 40 mn at 13,000 x g. The resulting clear lysate was removed and stored at -70°c.
Protein Electrophoresis And Western Blotting
Following treatment with cytokines, cells were harvested as described above for Western blot analysis which was performed as described previously (Stephanou et al., 1997). Whole cell lysate samples were added in a 1:1 ratio to sample buffer containing 2.3% (w/v) sodium dodecyl sulphate (SDS), 0.0625 M Tris, 10% (v/v) glycerol, 5% (v/v) p-mercaptoethanol and bromophenol blue, at a concentration of 10^ cells to 50 pi of sample buffer. Samples were subjected to SDS-polyacrylamide gel electrophoresis (PAGE), according to the method of Laemmli., (1970). A vertical, one-dimensional electrophoresis system was used (Gibco-BRL), Paisley, Scotland). The lower resolving gel consisted of 7% (w/v) aciylamide (in 1.5 M Tris, 0.4% w/v SDS, pH 8.8) and the stacking gel consisted of 5% (w/v) acrylamide (in 0.5 M Tiis, 0.4% w/v SDS, pH 6.8).
107 Methods
Protein samples were loaded and electrophoresed in duplicate on separate gels, in running buffer (0.192 M glycine, 0.025 M Tris, 0.1% w/v SDS, pH 8.3), for 4 hr at 40 mA at room temperature. Both gels were used for Western blotting onto nitrocellulose Hybond C (Amersham, Buckinghamshire, UK). Transfer was by electrophoresis in a blotting tank (Gibco-BRL), using blotting buffer of 0.192 M glycine, 0.025 M Tris, 20% (v/v) methanol, (pH 8.0) for 12 hr at 200 mA and at 4°c.
Development Of Western Blots
Western blots were initially blocked with 4% (w/v) Marvel (Nestle, U.K.), phosphate-buffered saline (PBS) and 0.05% (v/v) Tween-20 for 1 hr at room temperature and washed 3 times in 0.3% Marvel, PBS, 0.05% Tween-20, and once in PBS, for 5 mn each wash. Blots were then incubated with a primary antibody, washed as above and subsequently a secondary antibody conjugated to horseradish peroxidase, using the conditions outlined in Table 2.1. Blots were then washed as described above and were developed (see Figure 2.2) using an enhanced chemiluminescence (ECL) kit (Amersham, U.K.), with autoradiograph exposures of 5 s to 1 min at room temperature.
Bacterial Strains and Growth Conditions
Bacterial Strains
Escherichia coli (E.coli)
DH5a: supE44 AlacU169 ( Luria Bertani (LB^ media 1% (w/v) Bacto™-tryptone l%(w/v)NaCl 0.5% (w/v) Bacto™-yeast extract LB Agar LB media (prepared as above) containing 2% (w/v) Bacto™-agar. 108 Table 2.1 Antibody Conditions Primary Conditions for Antibody Secondary Antibody Protein Primary Antibody Source Antibody Hybridisation Dilution Conditions Detected AC88 2 hr at room 1:500 1:1000 dilution of HRP-anti Hsp 90 David Toft, Mayo Clinic temperature mouse IgG (Dakopatts, (90 kDa) Foundation, Rochester, Denmark) Minnessota, USA N27F3-4 2 hr at room 1:1000 1:1000 dilution of HRP-anti Hsp 70 Stressgen Biotechnologies temperature mouse IgG (Dakopatts, (70 kDa) Inc., Canada Denmark) Actin 1 hr at room 1:2000 1:2000 dilution of HRP-anti Actin Stressgen Biotechnologies temperature goat IgG (Dakopatts, (40 kDa) Inc., Canada Denmark) Figure 2.2 Principle of ECL Western Blotting +N2 + Light NH NH o- Peroxidase NH2 (b) X-Ray Rim Secondary Aniibody-HRP Protein Peracid Oxidised Product K Primary Antibody Ught Membrane Liimlnol Enhancer Figure 2.2 shows the HRP/hydrogen peroxide catalysed reaction of luminol and the production of light (a). Following oxidation, the luminol is in an excited state which decays to a ground state via a light emitting pathway which can be detected by light- sensitive autoradiography film (b). Methods Media were autoclaved at 120°c for 30 mn at 10 lb square/inch. Ampicillin (Sigma Chemical Co. Ltd., Poole, Dorset, U.K.) was prepared as a 100 mg/ml stock, which was filter sterilised using a 0.45 p,m Sartorius filter and stored at -20 °c. Cultures of E.coli contained a final concentration of 0.1 mg/ml of ampicillin. Starter cultures of E.coli were grown in 5 ml of LB medium o/n at 37 °c in a Gallenkamp orbital shaker. Approximately 500 p,l of the overnight culture was used to inoculate a 500 ml volume of medium which was subsequently incubated at37°c o/n to the required cell turbidity. For long term storage of E.coli cultures, 500 p-1 glycerol was added to 500 p,l of an overnight culture and stored at -70°c. Altematively, cultures were streaked onto LB Agar (prepared as above), incubated overnight at37°c and stored at4°c for up to 3 months. Transformation ofE.coli, This method is adapted from Hanahan, (1985). E.coli is rendered competent to take up DNA by the treatment of cells with MgSO^ and MgCl^. Divalent cations (Mg^^) are thought to modify channels in the cell membrane thereby allowing DNA to enter cells. An overnight culture of E.coli cells (DH5a) was prepared as described earlier. A loopful of cells was used to inoculate 50 ml of SOC medium, which was subsequently cultured at 37°c to an ODgg^ of 0.4-0.55 units. Cells were then pelleted by centrifugation at 2,000 g at 4°c for 10 mn. The resulting pellet was resuspended in 15 ml ice cold RFB I and incubated on ice for 15 mn. Cells were pelleted and resuspended in 4 ml ice cold RFB II. Aliquots were snap frozen in liquid nitrogen and stored at -70 °c. For transformation of E.coli, 100 pi aliquots of cells were thawed and each aliquot was added to 10 pi of ice cold 1.74% P-mercaptoethanol solution and incubated on ice for 3 mn. DNA to be transformed (approximately 100 ng) was added to the cell mix and incubated on ice for 40 mn. Cells were heat shocked at 42°c for 3 mn, cooled briefly on ice and added to 500 pi of LB medium and incubated at 37 °c for 30 mn. Half of the cell mix was subsequently plated onto LB agar plates (prepared as described earlier) supplemented with 0.1 mg/ml ampicillin and incubated overnight at37°c. I ll Methods SOB Medium: 0.02% w/v Bacto-tryptone 0.005% w/v Bacto-yeast extract lOmM NaCl 2mMKCl SOC Medium: SOB medium supplemented with 10 mM MgSO^ and 10 mM MgCl^ E E E Ii 3 mM KAc 5 mM MnCl^. ffljO 10 mM CaCl^ 100 mM RbCl 15% Glycerol 2 mM Acetic Acid Final pH was adjusted to 5.8. The resulting solution was then filter sterilised and stored at 4°c. REBJI: 5 mM MOPS, pH 6.8 37.5 mM CaCl^ lOmMRbCl 15% glycerol Final pH was adjusted to 5.8, the resulting solution was then filter sterilised and stored at 4°c. 112 Methods Small Scale Plasmid DNA Extraction from E, coU This method is an adaption of that described by Sambrook et al., (1989) for the rapid isolation and analysis of recombinant DNA from E.coli transformants. This method exposes cells to detergent then alkali solutions, which disrupt the DNA base pairing of and hence denature, the linear chromosomal genome. Plasmid DNA is protected from dénaturation due to the supercoiled circular nature of the DNA. Denatured chromosomal DNA and proteins, can then be removed by centrifugation whereas the plasmid DNA remains in solution until precipitated with isopropanol. A 5 ml volume of LB medium supplemented with 100 |xg/ml of ampicillin, was innoculated with a single colony of E,coli containing the plasmid of interest and the culture grown overnight as described previously. The culture (1.5 ml) was then harvested at 13,000 x g in a microfuge and the cell pellet resuspended in 100 p,l ice cold Solution I and then 200 pi of Solution II added to the mix. This was then placed on ice and 150 pi ice cold solution III added to the cell suspension which was vortexed and incubated on ice for a further 5 mn. The resulting cell debris was pelleted at 13,000 x g for 5 nui in a microfuge and the supernatant transferred to a fresh tube. A few drops of phenol:chloroform (1:1 ratio) were added to the supernatant to remove contaminating proteins. This mix was vortexed and centrifuged at 13,000 x g for 2 mn. The top aqueous layer was then transferred to a fresh tube and the DNA precipitated with two volumes of ethanol and incubated at-20°c for 30 mn and then centrifuged at 13,000 x g for 10 mn. The resulting pellet was allowed to air dry for 5 mn and was subsequently resuspended in 16 pi H^O, 4 pi RNase A (10 mg/ml). Restriction analysis of DNA was subsequently performed as described later and transformation into E.coli was performed as described eariier. Solution I 50 mM Glucose 25 mM Tris (pH 8.0) 10mMEDTA(pH8.0) Stored at4°c 113 Methods Solution n 0.2 M Na OH 1% SDS Freshly Prepared Solution m 5M CH3COOK 11.5% (v/v) glacial acetic acid Large Scale Plasmid DNA Extraction from E.coli A single colony of transformed E.coli was used to inoculate 500 ml of LB medium supplemented with ampicillin (final concentration 100 pg/ml). This culture was grown overnight at 37°c, as described previously. Cells were harvested by centrifugation at 3,000 X g for 30 mn and the resulting cell pellet was resuspended in 10 ml ice cold Solution I (described earlier) and transferred to a fresh tube. This suspension was then gently mixed with 20 ml Solution II (described earlier) and then 15 ml Solution III (described earlier). This mix was then incubated on ice for 15 mn and then centrifuged at 3,000 X g for 15 nm. The resulting supernatant was removed, transferred to a fresh tube with an equal volume of isopropanol to precipitate DNA and centrifuged at 5,000 x g for 15 mn. The resulting pellet was resuspended in 3 ml TE (pH 8.0) and 3 ml ice cold 5 M LiCl. This solution was centrifuged at 10,000 x g for 15 mn to remove contaminating RNA and the supernatant transferred to a fresh tube. An equal volume of isopropanol was added to precipitate DNA and the mix centrifuged at 10,000 x g for 15 mn. The resulting pellet was resuspended in 500 ^1 0.1 M Tris (pH 8.0) and incubated at 37°c for 30 mn with 10 p,l RNase A (10 mg/ml) to remove any further contaminating RNA. DNA was precipitated with an equal volume of PEG solution and incubated at 4°c overnight. The precipitate was pelleted by centrifugation in a microfuge at 13,000 X g for 15 mn and resuspended in 400 p,l 10 mM Tris (pH 8.0), 0.5 M NaCl. Contaminating protein was removed by the addition of an equal volume of phenol, followed by vortexing and centrifugation in a microfuge at 13, 000 x g for 5 mn. The top aqueous layer was then transferred to a fresh tube and a further phenol extraction carried out. Two extractions with equal volumes of chloroform were carried out to remove any residual phenol. DNA was then precipitated with 1 ml ethanol, incubated at - 114 Methods 20°c for 30 mn and centrifuged at 13,000 x g for 15 mn in a microfuge. The resulting DNA pellet was resuspended in 100 pi ddH20 and quantified as described below. PEG Solution 20% polyethylene glycol 6000 1 M NaCl 10 mM Tris (pH 8.0) Quantitative Determination of Plasmid DNA 5 pi of a plasmid DNA preparation was diluted into 1 ml of double distilled (dd) H^O. This was subsequently quantified spectrophotometrically using the general formula: 1 A260 unit = 50 pg/ml Restriction Analysis of DNA Small scale digests were routinely used to confirm that plasmid DNA preparations contained the required plasmid. Digests were carried out in 20 pi volumes at 37 °c for 1 hr, using an appropriate enzyme with buffer as recommended by the manufacturers. Digest Mix DNA (up to 1 pg) 1 pi Buffer (10 x) 2 pi Enzyme (1 unit) 1 pi ddH^O 16 pi 115 Methods Agarose Gel Electrophoresis In general, 1% agarose gels were prepared. 1.0 g of ultra pure agarose was dissolved in 100 ml of 1 x TAE by heating in a microwave. Ethidium bromide was added to a final concentration of 0.5 mg/ml. The liquid agarose was subsequently poured into a Biometra Agagel™ gel caster and a suitable comb fitted. The agarose was allowed to solidify at room temperature, at which point the comb was removed. Samples and DNA marker, each containing 1 x loading dye, were separately loaded onto the gel. Samples were electrophoresed at 80 Volts and the gel was photographed under UV light. l& I AE 0.4 M Tris 0.2 M Sodium Acetate 20 mM EDTA pH adjusted to 8.3 with HCl 10 X Loading Buffer 0.025% bromophenol blue 0.025% xylene cyanol 30% glycerol Ethidium Bromide 10 mg/ml dissolved in ddH^O and stored in the dark at room temperature. DNA Markers 1 kilobase DNA ladder was purchased from Life Technologies, Paisley, Renfrewshire, U.K. and was diluted 1:5 with 1 x loading buffer and stored at -20°c. 116 Methods Transfection and Stimulation of Hsp 90 Promoter CAT Constructs Transfection of Hsp 90a and -p promoter CAT reporter constructs was performed as described previously (Stephanou et al., 1997) by the calcium phosphate method of Gorman, (1985). IL-10 responsive HepG2 cells were plated at a density of 10^ cells/well in six-well plates. Cells were transfected as described below with Hsp 90 promoter-CAT reporter constructs, with or without a STAT-3 expression vector at the concentrations indicated in the figure legends (see Chapter 3) and 1 pg of a control p- galactosidase expression vector. To normalise for transfection efficiency, CAT activities were corrected for p-galactosidase activities. Following a period of 4-6 hr, cells were washed five times in phosphate-buffered saline (PBS) and stimulated with cytokines as described below. 2 X HEPES Buffered Saline (HBS) 274 mM NaCl 50 mM HEPES 2mM Na^HPO^ pH was adjusted to 7.12 with 1 M NaOH, stored at 4°c and filter sterilised before use. Tube A 2 M CaClz 6 H2O 31 pi DNA (1-5 pg) 10 pi ddH^O to bring the total volume to 250 pi TubeB 2 X HBS (pH 7.12) 250 pi The contents of Tube A were added drop wise to Tube B with constant flicking and allowed to form a precipitate at room temperature for 45 mn. This mix was subsequently added drop wise to the cells and incubated for 4-6 hr as described earlier. Cells were then washed five times with complete medium to remove any precipitate and 117 Methods were maintained in complete medium for 24 hr before a 5 hr period of culturing in complete medium in which the bovine foetal serum has been previously heat-inactivated (Life Technologies, Inc.), followed by stimulation with 50 ng of 11^ 10 (or another cytokine at a concentration as indicated in the figure legends in Chapter 3). Cells were harvested after 15 hr of stimulation using a rubber policeman, pelleted at 15,000 x g. The resulting pellet was resuspended in 100 pi ice cold 0.25 M Tris (pH 7.5) and lysed by three cycles of repeated freeze-thawing in liquid nitrogen and centrifuged at 13,000 x g for 45 mn in a microfuge. The supernatant was transferred to a fresh tube and stored at - 2 0 °c. Chloramphenicol Acetyl Transferase (CAT) Assay Following transfection and transcriptional activation of Hsp 90 promoter CAT constructs, the enzyme chloramphenicol acetyl transferase is produced. This enzyme is readily assayed by a determination of its activity as measured by the acétylation of a substrate [*'*C]-chloramphenicol. The amount of the product is determined by thin-layer chromatography (TLC), as outlined below. CAT Assav Reaction Mix 0.25 M Tris (pH 7.5) 70 pi D-Threo [dichloroacetyl-l-‘'^C] chloramphenicol (54 mCi/mmol) 4 pi 4 mM acetyl Co-enzyme-A 20 pi equalised volume of cell extract plus 0.25 M Tris (pH 7.5) 55 pi Assays of CAT activity were performed on samples that had been equalised for protein content, as determined by the method of Bradford, (1976). The above reaction was incubated at 37°c for 1 hr and the chloramphenicol was extracted by addition of 1 ml of ethyl acetate. This mix was subsequently vortexed for 30 s and centrifuged at 13,000 X g for 2 mn. The top organic layer was transferred to a fresh tube and vacuum dried. Samples were resuspended in 15 pi ethyl acetate and spotted onto a silica gel TLC plate (Whatman International Ltd., Maidstone, Kent, U.K.). This plate was subjected to ascending chromatography with a 95:5 mix of chloroformimethanol, until the font had reached 1 inch from the top of the TLC plate. The plate was allowed to air 118 Methods dry, before being exposed to X-Ray film and quantified by scanning densitometry using a Bio-Rad (Hemel Hempstead, U.K.) Imaging Densitometer. 119 Methods Section 2.2 Quantitative Determination of Heat Shock Proteins in SLE Levels of Hsp 90 and Hsp 70 were determined in individual samples of PBMCs, obtained from patients with SLE or RA and normal individuals, using the method of Dhillon et al (1993a; 1993b). Materials Monoclonal Antibodies Monoclonal antibody AC 8 8 to Hsp 90 has been previously described in Section 2.1 of this chapter. The mouse monoclonal antibody N27F3-4 to Hsp 70 and a goat polyclonal antibody to actin were both purchased from StressGen (Victoria, B.C., Canada). BlQ.Qd.Samplgs PBMCs obtained from normal individuals and patients with SLE or RA, were isolated from individual heparinised venous blood samples, as described in Section 2.1 of this chapter. Patients and Normal Individuals All patients with SLE fulfilled the American Rheumatism Association (ARA) 1982 revised criteria for the classification of SLE (see Chapter 1, Section 1.1) Samples of venous blood were taken from patients attending the morning outpatient SLE Clinic or the RA Clinic of the Bloomsbury Rheumatology Unit, Department of Medicine, University College London. Normal individuals consisted of healthy volunteers from The Department of Molecular Pathology, The Windeyer Institute for Medical Sciences, University College London. The age and sex distribution of patients and normal individuals is provided in Table 2.2 and the ethnic origin of patients with SLE is provided in Table 2.3. 120 Methods Table 2.2 Age and Sex Distribution of Patients with SLE, RA and Normal Individuals n Mean Median Age Male Female Age Age Range SLE 116 36 38 18-68 113 RA 20 39 41 29-76 16 Normals 30 29 26 21-57 25 Key: n: Number of subjects SLE: Systemic lupus erythematosus RA: Rheumatoid arthritis Methods Table 2.3 Ethnic Origin of Patients with SLE, RA and Normal Individuals Ethnic Origin SLE RA Normals Caucasian 71 9 23 Afrc-Carribean 21 3 3 Caucasian/Afro-Carribean 2 1 1 Indian 1 4 2 West Indian 7 3 0 Chinese 4 0 1 Total 116 20 30 Key; SLE: Systemic lupus erythematosus RA: Rheumatoid arthritis Methods Methods Preparation of Whole Cell Lysate from PBMCs PBMCs were purified from venous blood samples as described earlier. The resulting cell pellet was resuspended in ice cold 0.1 M Tris (pH 7.5) and subjected to three cycles of freeze-thaw in liquid nitrogen and the lysate pelleted at 4°c in a microfuge for 40 nm. at 13,000 x g. The resulting clear lysate was removed and stored at -70°c. Protein Electrophoresis And Western Blotting Western blot analysis was performed as described previously (Stephanou et al., 1997), using whole cell lysate prepared as described above. Protein samples were added in a 1:1 ratio with sample buffer containing 2.3% (w/v) sodium dodecyl sulphate (SDS), 0.0625 M Tris, 10% (v/v) glycerol, 5% (v/v) p-mercaptoethanol and bromophenol blue at a concentration of 10^ cells to 50 pi of sample buffer. Protein samples were subjected to SDS-polyacrylamide gel electrophoresis (PAGE), according to the method of Laemmli., (1970). A vertical, one-dimensional electrophoresis system was used (Gibco- BRL), Paisley, Scotland). The lower resolving gel consisted of 7% (w/v) acrylamide (in 1.5 M Tris, 0.4% w/v SDS, pH 8 .8 ) and the stacking gel consisted of 5% acrylamide (in 0.5 M Tris, 0.4% w/v SDS, pH 6 .8 ). Protein samples were loaded and electrophoresed in duplicate on separate gels, in running buffer (0.192 M glycine, 0.025 M Tris, 0.1% w/v SDS, pH 83), for 4 hr at 40 mA at room temperature. Both gels were used for Western blotting onto nitrocellulose Hybond C (Amersham, Buckinghamshire, UK). Transfer was by electrophoresis in a blotting tank (Gibco-BRL), using blotting buffer of 0.192 M glycine, 0.025 M Tris, 20% methanol, (pH 8.0) for 12 hr at 200 mA at 4°c. Development Of Western Blots Western blots were initially blocked with 4% (w/v) Marvel, phosphate-buffered saline (PBS) and 0.05% Tween-20 for 1 hr at room temperature and washed 3 times in PBS containing 03% Marvel, 0.05% Tween-20, and once in PBS, for 5 mn each. Blots were then incubated with a primary antibody and washed as above and subsequently a secondary antibody conjugated to horseradish peroxidase, using the conditions outlined in Table 2.1. Blots were then washed as described above and were developed (see Figure 2.2) using an enhanced chemiluminescence (ECL) kit (Amersham, U.K.), with 123 Methods autoradiograph exposures of 5 s to 1 mn at room temperature. Representative autoradiographs of immunoblots for Hsp 90 (Figure 2.3), Hsp 70 (Figure 2.4) and actin (Figure 2.5) are illustrated in the respective figures. In Figure 2.3, the 25-40 kDa bands are heterogenous ribonuclear proteins, which are reactive with antibody AC 8S to Hsp90 as mentioned earlier (see p. 103). Calculation of Heat Shock Protein Levels Autoradiographs were scaimed within the linear range using the VD620 densitometer (Biorad, Hemel Hempstead, U.K). Levels of Hsp were calculated by the method of Dhillon et al., (1993a; 1993b). Values for Hsp levels of individual samples, obtained by densitometric scanning, were expressed as a ratio of the value obtained for the respective Hsp of the positive control standard Jurkat T cell line, which constitutively overexpresses Hsps (Dhillon, 1993b). Actin levels remain relatively constant in cells and have been demonstrated to be a good measure of total cell protein (Preston et al., 1990). Thus, Hsp values between samples were equalised for variations in protein loading, by scanning actin bands of samples and Jurkat standard. Differences in the level of an Hsp between successive generations of Jurkat standard were equalised by calibrating each new generation against the first generation. An equalisation factor (E) was therefore obtained for successive Jurkat standard generations. The formula for calculating levels of Hsp 90 or Hsp 70 in individual samples is as follows: Hsp band sample (OD x mm') -r Hsp band Jurkat (OD x mm) x B Actin band sample (OD x mm) Actin band Jurkat (OD x mm) Accuracy of the Procedure for the Quantitative Determination of Hsps Samples were sonicated for homogeneity and were loaded onto gels using a Hamilton syringe to ensure accuracy of volumes loaded. Hypeifllm-ECL (Amersham, U.K.) was used for autoradiography, which the manufacturers state produces a linear response to the light emitted by enhanced chemiluminescence. Different exposure times of autoradiographs or runs of sample over time, were assessed for intra- and inter-assay variation. The mean percentage error (mean absolute error 4- mean Hsp 90 x 100) for any Hsp 90 value was determined as 11%, which was equivalent to that reported in the study of Dhillon (1993b) which recorded a value of 13% mean percentage error. The mean percentage error for any Hsp 70 value was determined to be 9%, however no 124 Figure 2.3 Autoradiograph of an Hsp 90 Immunobiot 97kDa Hsp 90 25kDa 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 J ------SLE------► ------RA------► N N Key; J Jurkat standard sample SLE Samples from patients with systemic lupus erythematosus RA Samples from patients with rheumatoid arthritis N Sample from a normal individual Figure 2.3. Representative autoradiograph of an Hsp 90 immunobiot. Samples from the Jurkat standard (Lane 1), patients with SLE (Lanes 2-9), patients with RA (Lanes 10-14) and from normal individuals (Lanes 15-16), are indicated in the figure and were subsequently analysed by densitometric scanning. Figure 2.4 Autoradiograph of an Hsp 70 Immunobiot 97kDa Hsp 70 25kDa 1 2 3 4 5 6 7 8 9 10 11 12 13 14 SLE RA N Key: J Jurkat standard sample SLE Samples from patients with systemic lupus erythematosus RA Samples from patients with rheumatoid arthritis N Sample from a normal individual Figure 2.4. Representative autoradiograph of an Hsp 70 immunobiot. Samples from the Jurkat standard (Lane 1), patients with SLE (Lanes 2-9), patients with RA (Lanes 10-13) and from a normal individual (Lane 14), are indicated in the figure and were subsequently analysed by densitometric scanning. Figure 2.5 Autoradiograph of an Actin Imimunoblot 97kDa Actin 25kDa 12 3 4 5 6 7 9 110 11 12 13 14 SLE RA Key: J Jurkat standard sample SLE Samples from patients with systemic luppus erythematosus RA Samples from patients with rheumatoid aarthritis N Sample from a normal individual Figure 2.5. Representative autoradiograph of am actin immunobiot. Samples from the Jurkat standard (Lane 1), patients with SLE (Lames 2-9), patients with RA (Lanes 10- 13) and from a normal individual (Lane 14)i, ;are indicated in the figure and were subsequently analysed by densitometric scanniin^g. Methods corresponding value, using the exact same methods as presented in this dissertation, was recorded in the study of Dhillon, (1993b). 128 Methods Section 2.3: Quantitative Determination of Cytokines in SLE Serum samples from individual patients with SLE or RA and normal individuals, were prepared and stored using the methods outlined in Section 2.2 of this chapter. For details of patients and normal control subjects see Section 2.2, Table 2.2 and Table 2.3. The methods outlined in this section, use the principle of solid phase enzyme-linked immuno-sorbent assays (ELISAs). Alternative laboratory-calibrated methods, which utilise the same ELISA principle, had originally been employed for the measurement of IL- 6 and IL-10 in serum samples. However, these were not found to be of satisfactory sensitivity for the determination of levels of IL - 6 and IL-10 in serum from normal individuals. Hence, as outlined below, the assays detailed in this section, have been chosen for their high sensitivity in detection of 11^6 and IL-10 within serum. Quantitative Determination of Interleukin-6 Principle of the assav This assay employs the quantitative sandwich enzyme immunoassay technique. A monoclonal antibody specific for 11^6 has been pre-coated onto a microplate. Standards and samples are pipetted into the wells and any IL-6 present is bound by the immobilised antibody. After washing away any unbound substances, an enzyme-linked polyclonal antibody specific for ILr6, is added to the wells. Following a wash to remove any unbound antibody-enzyme reagent, a substrate solution is added to the wells. After an incubation period, an amplifier solution is added to the wells and colour develops in proportion to the amount of IL ^ 6 bound in the initial step. The colour development is stopped and the intensity of the colour is measured. Colour Development This assay uses an amplification system in which the alkaline phosphatase reaction provides a cofactor that activates a redox cycle leading to the formation of a coloured product. In this system, alkaline phosphatase dephosphorylates the reduced form of nicotinamide adenine dinucleotide phosphate NADPH (Substrate), to reduced nicotinamide adenine dinucleotide (NADH). The NADH subsequently serves as a 129 Methods specific cofactor that activates a redox cycle driven by the secondary enzyme system, consisting of alcohol dehydrogenase and diaphorase (Amplifier). In the reaction catalysed by diaphorase, NADH reduces a tétrazolium salt (iodonitrotetrazolium violet) to produce a coloured formazan dye and NAD^. NAD^ is in turn reduced by ethanol, in an alcohol dehydrogenase-catalysed reaction, to regenerate NADH which can then re enter the redox cycle. The rate of reduction of the tétrazolium salt and thus the amount of coloured product formed, are directly proportional to the amount of 11^6 bound in the initial step. Experimental Measurements This assay contains highly purified £.co//-expressed recombinant human 11^6 and antibodies raised against recombinant human IL- 6 . The manufacturers have stated that results obtained using natural human IL- 6 , produced linear curves which parallel the standard curves obtained when using the £.cc>//-expressed recombinant human IL - 6 standards. Thus, these results indicate that this assay can be us used to calculate relative mass values for natural IL- 6 . Moreover, since this assay was insensitive to the addition of recombinant soluble 11^6 receptor (described in Chapter 1, Section 1.1), the manufacturers also stated that experimental IL - 6 values are likely to reflect total IL -6 present (free IL -6 and 11^6 bound to soluble receptors). The calibration of experimental IL- 6 values (detailed in Appendix B) with International Units, is described later in this section. Materials Reagents were purchased from R & D Systems Europe, Abingdon, Oxon, U.K. The Quantikine™ HS (high sensitivity) Human IL - 6 Immunoassay, consisted of the following reagents. IL- 6 Microplate 96 well polystyrene microplate (12 strips of 8 wells) coated with a murine monoclonal antibody against IL^ 6 . IL- 6 Conjugate 21 ml of polyclonal antibody against IL- 6 , conjugated to alkaline phosphatase. 130 Methods Assav Diluent 6 ml of a buffered protein base with preservative. Calibrator Diluent 2 1 ml of animal serum with preservative. Wash Buffer Concentrate 1 0 0 ml of a five-fold concentrated solution of buffered surfactant. Substrate Lyophilised NADPH with stabilisers. Substrate Diluent 7ml of buffered solution with stabilisers. Amplifier Lyophilised amplifier enzymes with stabilisers. Amplifier Pilugpt 7ml of buffered solution containing ethanol and INT-violet with stabilisers. Stop solution 6 ml of 2M sulphuric acid. Methods Summary of the Assav Procedure 1. All reagents and standards were prepared according to the manufacturers instructions. 2. 50 pi of Assay Diluent was added to each well of the microtiter plate. 3. 200 pi of prepared standard or serum sample was added to each well and incubated at room temperature for 2 0 hr. 4. Wells were washed four times using the provided Wash Buffer reagent. 5. 200 pi of IL - 6 conjugate was added to each well and incubated for 60 mn at room temperature. 6 . Wells were washed six times using the provided Wash Buffer reagent. 131 Methods 1. 50 ^il of Substrate Solution was added to each well and incubated for 60 mn at room temperature. 8 . 50 pi of Amplifier Solution was added to each well and incubated for 30 mn at room temperature. 9. 50 pi of Stop Solution was added to each well and the absorbance at 490 nm was read within 30 nm and corrected with absorbance measurements at 650 nm. Calculation of Results Average readings from each standard and sample were subtracted from the average zero standard optical density. The optical density of the standards versus the concentration of the standards were plotted and a standard curve was obtained for each assay and used to determine the quantity of IL - 6 in each patient serum sample. For diluted serum samples, the concentrations read from the standard curve were multiplied by the dilution factor. Prççisioii Inter-assay precision was reported by the manufacturers as 0.09-0.81 standard deviations. Five patient serum samples (IL - 6 range: 0.2-180 pg/ml) were used in subsequent assays, and the variation of experimental 11^6 values was typically found to be in this reported range. Calibration of Sample Values The NIBSC/WHO International Reference Standard 89/548, has been evaluated by the manufacturers, in this assay. The dose response curve of this standard, parallels the curve obtained using the IL - 6 standard provided by the manufacturers. The following equation can be used to convert sample values (detailed in Appendix B) using this assay, to the equivalent International Reference Standard value: NIBSC (89/548) equivalent value (lU/ml) = 0.131 x Quantikine HS IL - 6 value (pg/ml) 132 Methods Quantitative Determination of Interleukin-10 Principkpfth? A??^y This assay is a solid phase sandwich ELISA. An antibody specific for human IL-10 has been coated onto the wells of a microtiter plate. Samples, including standards of known human IL-10 content, control and unknown serum specimens, are pipetted into these wells. During the first incubation (see Methods below) the IL-10 antigen binds to the immobilised antibody on the microtiter plate. After washing, a biotinylated antibody specific for human IL-10 is added. During the second incubation, this antibody binds to the immobilised human IL-10 “captured” during the first incubation. After removal of excess secondary antibody, Streptavidin-Peroxidase enzyme is added. This binds to the biotinylated antibody to complete the four-member sandwich. After a third incubation and washing to remove all the unbound enzyme, a substrate solution is added, which is acted upon to produce colour. The intensity of this coloured product is directly proportional to the concentration of human IL-10 present in the original specimen. Materials Reagents were purchased from Biosource International, Inc., California, U.S.A. The Biosource Cytoscreen™ Human IL^IO US (ultra sensitive) immunoassay, consisted of the following reagents. IL^IO Microplate 96 well polystyrene microplate (12 strips of 8 wells) coated with anti-human IL- 1 0 monoclonal antibodies. IL-10 Biotin Conjugate Biotin-labelled anti-human IL^IO antibody (contains 15 mM sodium azide). Assav Diluent A buffered protein base with preservative (contains 15 mM sodium azide). 133 Methods Wash Buffer Concentrate 100ml of a 25-fold concentrated solution of buffered surfactant Streptavidin-Peroxidase Streptavidin-Peroxidase (HRP), lOOx concentrate (contains 3 mM thymol) Streptavidin-Peroxidase (HRP), Diluent (contains 3 mM thymol) Stabilised Chromogen Tetramethylbenzidine (TMB) Stop solution 6 ml of 2M sulphuric acid. Methods Summary of the Assay Procedure 1 . 1 0 0 pi of standards and serum samples were added to the microtiter plate wells. 2. Plate was covered and incubated for 3 hr at 37°c. 3. Wells were thoroughly aspirated and washed six times with Wash Buffer. 4. 100 pi of biotinylated anti-IL-10 antibody was added to each well. 5. Plate was covered and incubated for 1 hr at room temperature. 6 . Wells were thoroughly aspirated and washed six times with Wash Buffer. 7. 100 pi of Streptavidin-HRP working solution was added to each well. 8 . Plate was covered and incubated for 1 hr at room temperature. 9. Wells were thoroughly aspirated and washed six times with Wash Buffer. 10. 100 pi of Stabilised Chromogen was added to each well. 11. Plate was covered and incubated for 35 mn at room temperature. 100 pi of Stop Solution was added to each well and the absorbance at 450 nm was read within 2 hr. 134 Methods Calculation of Results Average readings from each standard and sample were subtracted from the average zero standard optical density. The optical density of the standards versus the concentration of the standards were plotted and a standard curve was obtained for each assay and used to determine the quantity of IL-10 in each patient serum sample. For diluted serum samples, the concentrations read from the standard curve were multiplied by the dilution factor. Precision Inter-assay precision was reported by the manufacturers as 0.32-1.72 standard deviations. Five patient serum samples (IL-10 range: 1.8-31.2 pg/ml) were used in subsequent assays, and the inter-assay variation of these experimental IL-10 values was typically found to be in this reported range. 135 Methods Section 2.4: ELISA for Anti-Hsp Antibodies This assay utilises the protocol of Conroy et al., (1994; 1996) for the detection of antibodies to Hsp 90 or Hsp 70 within serum samples obtained from patients with SLE or RA and normal individuals. Principle of the Assay This assay is a solid phase sandwich ELISA. Purified Hsp 90 or Hsp 70 is coated onto the wells of a microititer plate. Samples, including standards (antibodies to human Hsp 90 or Hsp 70), are pipetted into these wells. During the first incubation (see Methods below) serum anti-Hsp antibodies bind to the immobilised Hsp antigen on the microtiter plate. After washing, an alkaline phosphatase-conjugated antibody specific for human IgG or IgM is added. During the second incubation, this antibody binds to the inunobilised human anti-Hsp antibody “captured” during the first incubation. After removal of excess secondary antibody, and washing to remove all the unbound moieties, a substrate solution is added, which is acted upon to produce colour. The intensity of this coloured product is directly proportional to the concentration of human anti-Hsp antibody present in the original specimen. Materials Monoclonal Antibody Standards Monoclonal antibody AC 16 was a gift from Professor D.O. Toft, Mayo Clinic Foundation, Rochester, Minnessota, USA. Antibody AC 16 is raised in rats and is of class IgGj and has been demonstrated to react with human Hsp 90 and binds to the native form with greater affinity than to denatured human Hsp 90 (private communication). Mouse monoclonal antibody N27F3-4 (class IgGJ was purchased from Stressgen Biotechnologies Inc., Canada and reacts with human Hsp 73 and with faint reactivity to human Hsp 72. This was confirmed in the study of Dhillon (1993b). 136 Methods Preparation of Purified Hsp 90 and Hsp 70 Purified Hsp 90 or Hsp 70 (Stressgen, Canada) was reconstituted in sterile phosphate buffered saline to a concentration of 1 mg/ml as single-use aliquots and stored at -70°c until required. Prior to use, frozen protein samples were subjected to SDS-polyacrylamide gel electrophoresis (PAGE), according to the method of Laemmli., (1970) and Western blotting, using the methods outlined in Section 2.1 of this chapter, to check there was only one discrete band. Methods ELISA for Anti-Hsp 90 Antibodies Nunc ELISA (Dynatech) plates were coated with 100 p,l per well of purified Hsp 90 (Stressgen, Canada) at 0.5 p-g/ml in 0.05 M bicarbonate buffer (pH 9.6), overnight at 4°c. Half of each plate was coated only with 0.05 M bicarbonate buffer. Sera were diluted 1:200 in PBS supplemented with 0.1% v/v polyoxyethane sorbitan monolaurate (Tween20), 1% (v/v) goat serum (Sigma, U.K) and 1% w/v BSA. 100 pi of diluted sera were added to the ELISA plate in duplicate wells and incubated at 37°c for 1 hr. After five washes with PBS-Tween, 100 pi of anti-human IgG or IgM F(ab ’ )2 alkahne phosphatase-conjugated antibodies, was added to wells at a dilution of 1:1 0 ,0 0 0 and 1:50,000 respectively, in PBS supplemented with 0.1% v/v polyoxy ethane sorbitan monolaurate (Tween20), 1% goat serum (Sigma, U.K) and 1% w/v BSA. This was incubated at 37°c for 1 hr, followed by five washes with PBS-Tween and then two washes with bicarbonate buffer. Colour was developed using p-nitrophenyl phosphate tablets (Sigma, U.K) as substrate. Two tablets (each 5 mg) were dissolved in 10 ml of bicarbonate buffer to which 20 pi of 1 M MgCl^ had been added. The reaction was allowed to develop at 37°c for 30 mn. Plates were read on a Dynatech 4000 ELISA reader (Dynatech, Billinghurst, UK) at 405 nm and corrected at 490 nm. 137 Methods ELISA for Anti-Hsp 70 Antibodies Nunc ELISA (Dynatech) plates were coated with 100 pi per well of purified Hsp 70 (Stressgen, Canada) at 1.0 pg/ml in 0.05 M bicarbonate buffer (pH 9.6), overnight at 4°c. Half of each plate was coated only with 0.05 M bicarbonate buffer. Sera were diluted 1:500 in PBS supplemented with 0.1% v/v polyoxyethane sorbitan monolaurate (Tween20), 1% (v/v) goat serum (Sigma, U.K) and 1% w/v BSA. 100 pi of diluted sera were added to the ELISA plate in duplicate wells and incubated atSTc for 1 hr. After five washes with PBS-Tween, 100 pi of anti-human IgG or IgM F(ab ’)2 alkaline phosphatase-conjugated antibodies were added to wells at a dilution of 1:1 0 ,0 0 0 and 1:50,000 respectively, in PBS supplemented with 0.1% v/v polyoxyethane sorbitan monolaurate (Tween20), 1% goat serum (Sigma, U.K) and 1% w/v BSA. This was incubated at 37°c for 1 hr, followed by five washes with PBS-Tween and then two washes with bicarbonate buffer. Colour was developed using p-nitrophenyl phosphate tablets (Sigma, U.K) as substrate. Two tablets (each 5 mg) were dissolved in 10 ml of bicarbonate buffer to which 20 pi of1 M MgCl^ had been added. The reaction was allowed to develop at 37°c for 2 hr. Plates were read on a Dynatech 4000 ELISA reader (Dynatech, Billinghurst, UK) at 405 nm and corrected at 490 nm. Calculation of Results Average readings from each sample were subtracted from the average OD values obtained from the wells with no antigen and was expressed as a percentage of a positive control sera (containing AC 16 to Hsp 90 or antibody N27F3-4 to Hsp 70) applied in duplicate on each plate. A known negative serum was also applied in duplicate to each plate. Patients were considered to have a raised level of anti-Hsp antibody if the corrected OD value was greater than the 97.5^^ centile of 30 healthy normal controls. 138 Methods Section 2.5: Disease Associations In the following results chapters, levels of Hsps, cytokines and antibodies to Hsps have been determined in individual blood samples from patients with SLE, using the methods outlined in this chapter. Subsequently, these data have been compared with clinical disease activity in individual patients, recorded at the time of providing a sample of blood. In a previous study (Dhillon et al., 1993a) levels of Hsps determined in individual blood samples from patients with SLE, were analysed using the UCH/Middlesex SLE disease activity scoring system (Morrow et al., 1982; 1983). Thus, using this index, disease activity in patients was classified as inactive, mildly active, moderately active or severely active. However, the yield of information using this index is relatively low. For example, disease which is scored as overall mildly active, may be masking severe disease activity in one particular organ (Dhillon, 1993b). For this reason, the study of Dhillon et al., (1993a) also used the The British Isles Lupus Assessment Group (BILAG) index for scoring clinical disease activity in patients with SLE. The BILAG index is a standardised measure of clinical disease activity in SLE, which has been validated internationally against other methods of scoring SLE activity, including the UCH/Middlesex scoring system (Symmons et al., 1988), (Isenberg, 1990), (Gladman etal., 1992). The BILAG index (Symons et al., 1988) defines disease activity within eight organs or systems: general disease, mucocutaneous disease, musculoskeletal disease, cardiovascular disease and respiratory disease, vasculitis, renal disease and haematological disease (see Chapter 1, Section 1.3 and Appendix A: The BILAG Index). An alphabetic score (A-E) is determined for each category on the basis of the physician’s intention to treat the patient. Score “A” corresponds to severely active disease requiring major immunosuppressive therapy or a substantial increase in current therapy; a score “B” corresponds to active disease without substantial alteration to current therapy; a score “ C” corresponds to mild disease or evidence of previous disease; a score “ D” corresponds to no current disease or evidence of previous disease; and a score “E” corresponds to no current/never any disease in a category. BILAG global scores in a patient with SLE are determined by the summation of BILAG scores in each organ/system, where a score “A”= 9 points, a score “B”= 3 points, a score “C” = 1 points and a score “ D” or “E” = 0 points. 139 Methods Patients attending the morning outpatient SLE Clinic at the Centre for Rheumatology, Department of Medicine, University College London, are routinely assessed for clinical disease activity, using the BILAG index. Thus, detailed records of these data have been recorded for 90% of the patients with SLE studied in this dissertation. Thus, in Chapter 6 of this dissertation, levels of Hsps, cytokines and antibodies to Hsps have been compared in patients with SLE graded by the BILAG index. For the purpose of statistical analyses (see Section 2.5 of this chapter), a patient with a score “A-C” recorded in a category, was considered as having “active disease” in that category; a score “ D” or “E ” was considered as “inactive disease” in that category. 140 Methods Section 2.6: Statistical Analyses Statistical analyses were performed using an SPSS computer software package (SPSS Science, Chicago, U.S.A) and with helpful discussions and advice from senior lecturers of the Department of Medical Statistics, Institute of Child Health, University College London. Information on the various statistical tests used in the following results chapters, is detailed in this section and has been obtained from Altman, (1999). Summary statistics, of data for levels of Hsps (Table 4.2.1) and cytokines (Table 4.3.1) in all subject groups are provided in Chapter 4 and of data for levels of antibodies to Hsps (Table 5.2.1) in Chapter 5. Following advice from the Department of Medical Statistics, levels of the cytokines IL - 6 and 11^ 10 were log^g-transformed, to reduce any influence of outlying and thus atypical values present in the raw data on subsequent statistical analyses. As described below, parametric and non-parametric methods of statistical analysis were performed on all these data. Differences between mean levels of Hsps or between mean levels of cytokines determined for two independent subject groups (see Chapter 4 and Chapter 6), were assessed for statistical significance using the student’s t-test. The use of the two sample t-test is based on the assumption that the variances of the two populations are the same. However, the t-test is known to be robust in that it is little affected by moderate failure to meet this assumption. The null hypothesis that the data from each group have equal variances is assessed by the SPSS statistics package, using the variance ratio test (F- test). The variance ratio is calculated as the square of the ratio of the sample standard deviations. The distribution of the F statistic has two values of degrees of freedom, one corresponding to each variance in the data for each group. Where variances are determined to be unequal, a modified student’s t-test is applied to the data. Analyses involving student’s t-tests are presented as bar charts with error bars representing 95% confidence intervals (95% CI:-2SEM to +2SEM) for the mean of a variable in a subject group. P values denote the significance of the difference in mean values of a variable between two independent groups. A P value of less than 0.05 (P<0.05), was considered to be significant. Analyses of the correlation between levels of Hsps and cytokines, in individual samples from patients with SLE, are presented in Chapter 4. Correlation is a method of analysis for studying the possible association between two variables. The degree of association is measured by calculating the correlation coefficient. The standard method (Pearson) leads to a quantity called “r” which measures the degree of “straight-line” association between the values of the variables and can take any value from -1 to 4-1 . 141 Methods Thus, a value of +1 or -1 is obtained, if all data points in a scattergraph of the type shown in Chapter 4, lie in a straight line. Scattergraphs in Chapter 4 showing an associated trend-line, also indicate the value for the Pearson correlation coefficient “ r” and the associated probability (?) value. Any value for “ r” which generated a P value of less than 0.05 (p<0.05), was considered to be significant. All significant correlations by this criterion presented in Chapter 4, have also attained statistical significance (p<0.05) non-parametrically, using a Spearman’s rank correlation test, which is essentially a Pearson correlation on ranked data. Correlations of BILAG global scores in individual patients with SLE with either levels of Hsps or levels of cytokines determined in individual samples from the same patients (see Chapter 6), were performed non- parametrically, using either a Spearman’s or Kendall’s rank correlation analysis. These methods of correlation were chosen, since the value of a BILAG global score in an individual patient with SUE, is obtained using a highly skewed scoring system (see Section 2.4 of this chapter). Chi-Square analyses were performed for the evaluation of possible associations between categorical data. For example, in Chapter 4, patients with SLE have been categorised as having normal or elevated levels of Hsp 90, in Chapter 5, as having normal or raised levels of antibodies to Hsp 90 and in Chapter 6, as having active or inactive disease in an organ/system. For all chi-square analyses detailed in these chapters, patients with SLE have been categorised into mutually exclusive groups in relation to two classifications. The null hypothesis of this test, is that the two classifications, for example HLA status and raised anti-Hsp 70 antibody level status, are unrelated in the relevant population. Where a chi-square analysis has been performed, a Pearson P value is provided, which if less than 0.05 (P<0.05), was considered to be significant. Included with all chi-square analyses which have generated a significant Pearson’s P value, are alternative analyses of the categorical data which each have an identical null hypothesis to the chi-square analysis. These include a Fischer’s exact test, a likelihood ratio test and a linear by linear association test (see Altman, 1999). In all cases where a significant Pearson P value has been attained and presented, these tests have also generated significant P values, in agreement with the chi-square analysis. In principal, interpretations of P values generated by any of the analyses detailed above, can lead to a rejection of a null hypothesis, when in reality it is true (a Type I error), or the acceptance of a null hypothesis, when in reality it is false (a Type II error) In all analyses detailed above, the probability of obtaining a Type I error has been pre determined as 5%. Thus, by this criterion, a P value of less than 0.05 (P<0.05) has been considered significant. In some cases however, multiple student’s t-tests and chi-square tests have been performed, involving the same variable (continuous or categorical). For 142 Methods example, in Chapter 4, six individual t-tests have been performed, to compare mean levels of Hsp 90 in patients with SLE, with respect to the presence or absence of individual/combined HLA alleles/haplotypes. Multiple analyses of this type, are likely to lead to Type I errors just by chance. There is no simple or totally satisfactory solution to this problem. Several methods have been proposed to deal with this problem, such as Bonferroni, which adjusts P values obtained from multiple student’s t-tests or chi- square analyses, such that for k paired comparisons, P is adjusted to KP. However, methods of this type may be reasonable for up to five numbers of comparisons, but for large numbers its use is highly conservative. If research aims justify large numbers of comparisons, then an alternative approach is to quote exact P values and to reduce the cut-off level for statistical significance (Altman, 1999 and advice from the Department of Medical Statistics). Thus, by this criterion, P values obtained in multiple analyses presented in the following chapters, have been considered significant if the value of P is less than 0.02 (P<0.02). Where the results of analyses of this type are in agreement with previous studies, no adjustment to the 5% cut-off level for significance has been made. 143 Results Chapter 3 The Effect Of Interleukin-10 On Hsp 90 Gene Expression 3.0 Introduction Elevated circulating levels of the cytokine interleukin-10 (IL-10) have been reported in patients with SLE and are typically highest in patients with active disease (Park et al., 1998). In view of the function of IL-10 within the normal inunune system, a pathogenic role for IL-10 in SLE has been suggested and is supported by a number of studies, which are discussed in Chapter 1, Section 1.3. Any role for IL-10 in the pathogenesis of SLE is likely to involve the induction of the expression of specific genes within its target cells, resulting in changes of cellular phenotype. The 90 kDa heat shock protein (Hsp 90) is elevated in specific subsets of patients with SLE, primarily in patients with active central nervous system and cardiovascular/respiratory disease (Dhillon et al., 1993a; 1994), see Chapter 1, Section 1.4. Further studies have shown that this elevation is dependent on enhanced transcription of the Hsp 90p gene (Twomey et al., 1993). Although the exposure of cells to elevated temperatures and other stresses results in the induction of a number of different Hsps including Hsp 90 (see Chapter 1, Section 1.1), the elevation of Hsp 90 observed in SLE is a specific effect on Hsps and is not paralleled by elevation of the constitutively expressed heat shock proteins Hsp 73 or Hsp 60. A role for Hsp 90 both in the aetiology and pathogenesis of SLE has been discussed in Chapter 1, Section 1.4, however, the precise mechanisms leading to the overexpression of Hsp 90 have remained unclear. 144 Results Recent studies (outlined in Chapter 1, Section 1.1) have shown that the cytokine interleukin-6 (IL-6), which is also present at elevated levels in patients with active SLE (Linker-Israeli et al., 1991), up-regulates Hsp 90 expression in the human hepatoma cell line HepG2 and in peripheral blood mononuclear cells (PBMCs) from normal individuals (Stephanou et al., 1997). An investigation into the mechanisms of Hsp 90 promoter activation in response to IL-6 has been recently reported (Stephanou et al., 1998b). In this study, using reporter gene assays, activation of the Hsp 90 promoter by IL-6 is mediated by a region of the promoter (from -643 to -623 relative to the transcriptional start site) which contains binding sites for the IL-6 activated transcription factors STAT-3 and NF-IL6. It is interesting to note that IL-10 and IL-6 utilise similar signalling mechanisms for modulating gene expression within their target cells. As discussed in Chapter 1, Section 1.1, IL-10 has been shown to mediate intracellular signalling via the JAK/STAT pathway, in which stimulation of receptor-associated Jaks (Jakl and Tyk2) leads to phosphorylation and activation of the transcription factors STAT-1 and STAT-3 (Finbloom and Winestock, 1995). The similarity of gene regulatory elements targetted by signalling pathways for both IL-6 and IL-10 through the activation of STATs, together with studies supporting a role for IL-10 in the pathogenesis of SLE (see Chapter 1, Section 1.3), has therefore prompted an investigation of the effect of IL-10 on Hsp 90 gene expression. 145 Results 3.1 IL-10 Upregulates Hsp 90 in HepG2 Cells and PBMCs The effect of IL-10 on Hsp 90 gene expression was investigated by treatment of both HepG2 cells and PBMCs with 11^10 and levels of Hsp 90 measured by Western blot analysis (Figure 3.1). IL-10 was clearly observed to up-regulate Hsp 90 levels within the HepG2 cell line (Figure 3.1a) which had been artificially engineered to express the human IL-10 receptor. Within the parental cell line (lacking the IL-10 receptor) no elevation of Hsp 90 was observed following treatment with IL- 10 (result not shown). The 30 kDa band in Figure 3.1 results from the cross reactivity of antibody AC88 with heterogenous ribonuclear proteins, as detailed in Chapter 2. In cells cotransfected with an expression vector for STAT-3 and then stimulated with IL-10, Hsp 90 levels were also observed to be elevated. Elevated levels of Hsp 90 were also observed in unstimulated cells transfected with STAT-3 compared to control unstimulated cells. In PBMCs, Hsp 90 levels are significantly elevated following IL-10 treatment, compared to control untreated cells (Figure 3.1b). This effect is also observed in PBMCs treated with interferon-y and lL-6, which has been reported previously (Stephanou et al., 1998b). Furthermore, an analysis of Hsp 90 levels in PBMCs following a time-course of IL-10 treatment (Figure 3.1c) shows that elevation of Hsp 90 levels is rapid. An approximate three-fold increase in the level of Hsp 90 is observed after 1 hr of IL-10 treatment compared to control untreated cells, with this increase being sustained for 48 hr of treatment. 146 Figure 3.1 IL-10 Up-Regulates Hsp 90 in HepG2 Cells and PBMCs (a) 50 100 100 0 + + MWxlO^ 3 3 97 kDa — Hsp 90 25 kDa — (b) MWxlO^ 97 kDa — Hsp 90 - Figure 3.1. (a) Western blot with antibody AC88 to Hsp 90, HepG2 ceils were treated with either 50 ng/ml (50) or 100 ng/ml (100) of IL-10 for 12 hr or untreated (0). Cells transfected with 2 pg of a STAT-3 expression vector are shown (4-3). (b) Western blot with antibody AC88 to Hsp 90. PBMCs were untreated (lanes 1 and 2), treated with interferon y (lane 3), IL-6 (lane 4), or treated for 16 hr with 50 ng/ml of either IL-6 (lane 5) or IL-10 (lane 6). Figure 3.1c Time-Course of Hsp 90 Elévation in PBMCs Treated with IL-10 Hsp 90 4 .5 . 3 .5 . 2 .5 . 0 .5 . 0 10 20 30 40 50 60 Time (hr) Figure 3.1c. Time course of Hsp 90 elevation in PBMCs treated with IL-10. PBMCs were untreated or treated with 100 ng/ml of IL-10 for the times indicated in the figure. Values for Hsp 90 levels represent fold induction relative to control untreated cells. Results 3.2 IL-10 Activates the Hsp 90a and -P Gene Promoters To investigate the mechanisms by which IL-10 induces enhanced Hsp 90 levels in HepG2 cells, the effect on the Hsp 90 promoter was studied using plasmid constructs in which the Hsp 90a and ~p promoters are coupled to a CAT reporter gene. As shown in Figure 3.2a and Table 3.1, IL-10 significantly activates both the Hsp 90a and -p promoters, as measured by CAT activity in IL-lO-treated and untreated cells. Cells cotransfected with a STAT-3 expression vector and treated with IL-10, exhibit a further slight increase in CAT activity compared to IL-10-treated cells. In view of the known effects of IL-6 on Hsp 90 gene regulation, IL-6 treatment of cells transfected with Hsp 90 promoter constructs, was used to compare the responsiveness of the reporter constructs to IL-10 and IL-6. Figure 3.2b illustrates the responsiveness of the Hsp 90p promoter construct to both IL-6 and IL-10 and indicates that the two effects are comparable. 3.3 IL-10 Activates the Hsp 90p Promoter via a STAT-3 Binding Site In view of previous studies which have shown that a region of the Hsp 90p promoter (-643 to -623 relative to the transcriptional start site) can confer responsiveness to IL-6 when linked to a heterologous promoter, the responsiveness of this region to IL-10 was investigated. HepG2 cells were transfected with construct Hsp 90-S3E, corresponding to this region linked to the heterologous thymidine kinase promoter, or with construct Hsp 90-MS3E, in which the wild-type STAT-3 site has been mutated (see Chapter 2, Figure 2.1). CAT activity of the constructs was then measured following treatment of cells with ILIO. Figure 3.3a and Table 3.1 show that the wild-type STAT-3 sequence gave high levels of CAT activity in response to IL-10 and activation was enhanced by cotransfection with a STAT-3 expression vector. Conversely the mutant construct (Figure 3.3b and Table 3.1) was not responsive to IL-10 and no significant increase in CAT activity was observed following cotransfection with STAT-3. 149 Figure 3.2 Assay of CAT Activity in HepG2 Cells Transfected with Hsp 90a and -P Promoter-CAT Constructs ( a ) 1 2 3 4 5 6 & (b ) 10 6 10 6 + + + Results Legend to Figure 3.2 ( a ) HepG2 cells were transfected with 2 pg of either Hsp 90a-CAT (lanes 1-3) or Hsp 90p-CAT (lanes 4-6). Cells were left untreated (lanes 1,4) or treated with 100 ng/ml IL-10 for 12 hr (lanes 2,3 and 5,6). Lanes 3 and 6 correspond to cells cotransfected with 1 pg of a STAT-3 expression vector. ( b ) Assay of CAT activity in HepG2 cells transfected with Hsp 90p- CAT and treated with IL-6 or IL-10. HepG2 cells were transfected with 2 pg of Hsp 90p-CAT and treated ( I ) with either 50 ng/ml IL-6 (6) or 100 ng/ml IL-10 (10) for 12 hr or untreated (-). Cells cotransfected ( S ) with 1 pg of a STAT-3 expression vector are denoted (+). 151 Figure 3.3 IL-10 Activates the Hsp 90p Promoter Via a STAT-3 Binding Site ( a ) 10 10 + (b ) 10 10 + Results Legend to Figure 3.3 Assay of CAT activity in HepG2 cells transfected with Hsp 90-S3E (a) or Hsp 90 MS3E (b). HepG2 cells were transfected with 2 pg of either CAT reporter construct and treated (I) with 100 ng/ml IL-10 (10) for 12 hr or untreated (-). Cells cotransfected (S) with 1 pg of a STAT-3 expression vector are denoted (+)• 153 Results Table 3.1 Effect of IL-10 on the Activity of Hsp 90 Promoter CAT Constructs CAT Activity (Arbritary Units) Construct — IL-10 + IL-10 +IL-10 +STAT-3 Hsp 90a-CAT 4.6 61.8 88.0 Hsp 90P-CAT 2.5 78.2 84.2 pBLCat2 vector 1.0 1.0 1.0 Hsp 90-S3E 1.3 46.0 48.1 Hsp 90-MS3E 2.9 4.0 4.7 Values represent the CAT activity of each construct either alone or cotransfected with a STAT-3 expression vector (+ STAT-3) in HepG2 cells treated (+ IL^IO) or untreated (-IL-10) with IL-10 for 12 hr before harvesting. Values were obtained by scanning densitometry of assays of the type shown in Figs 2,3,4. Results 3.4 Discussion Although the regulation of heat shock proteins by stressful stimuli has been intensively studied, much less attention has been given to their regulation by non stressful stimuli, such as cytokines. This is of particular importance, since Hsp expression has been observed to be regulated during a number of normal cellular events such as T-cell activation (Haire et al., 1988) and the differentiation of monocytes to macrophages (Twomey et al., 1993). Hence, such regulatory mechanisms may play a key role in determining the level of Hsp expression in unstressed cells, as well as being important in their overexpression in particular disease states. It has previously been demonstrated that treatment of cells with IL-6 can up- regulate Hsp 90 levels in vitro (Stephanou et al., 1997) and that such elevation of Hsp 90 levels is also observed in vivo in transgenic mice overexpressing IL-6 (Stephanou et al., 1998a). This effect is due to the IL-6-induced activation of the transcription factors NF-IL6 and STAT-3 and is mediated via a short specific region of the Hsp 90 promoter which contains binding sites for these factors (Stephanou et al., 1997; 1998b). In this chapter, these observations have been extended by showing that IL-10 can also enhance Hsp 90 levels, activating the Hsp 90p gene promoter via the same short element which is targetted by IL-6. As IL-10 activates STAT-3 but not NF-IL6 (Finbloom and Winestock, 1995), it is likely that this effect is mediated via the IL-10- induced activation of STAT-3 and in agreement with this, the ability of IL-10 to activate the Hsp 90a and -p gene promoters is enhanced by cotransfection of a STAT-3 expression vector. Hence the activation of Hsp 90 by IL-10 is likely to reflect the similarity of its signalling pathway to that of IL-6 and indicates that such activation can occur in the absence of activation of NF-IL6 which is achieved only by IL-6. This is in accordance with the observation that enhanced Hsp 90 levels are observed both in transgenic mice containing the IL-6 gene and in transgenic mice in which the gene encoding NF-IL6 has been inactivated, resulting in enhanced levels of IL-6 (Stephanou et al., 1998a). 155 Results The induction of Hsp 90 expression by both IL-6 and IL-10 infers a role for Hsp 90 expression both within the normal functioning of the immune system and as part of an immune response. It is likely that both IL-6 and IL-10 regulate the expression of Hsp 90, ensuring appropriate levels of chaperoning functions of Hsp 90 within immune cells. More importantly, these results have important implications for our understanding of the mechanisms of Hsp 90 regulation in SLE. Thus, since, IL-6 (Linker-Israeli et al., 1991) and IL-10 (Park et al., 1998), (Linker-Israeli, 1992) are elevated in SLE, whilst studies suggest a role for Hsp 90 in the pathogenesis of autoimmunity in this disease (Stephanou et al., 1998c), an investigation into the relative effects of these two factors in producing the Hsp 90 elevation observed in SLE, was carried out and is discussed in the following results chapter. 156 Results Chapter 4 Cytokines and Heat Shock Protein Gene Expression in SLE 4.0 Introduction As described in detail in Chapter 1 (Section 1.3), systemic lupus erythematosus (SLE) is an autoimmune rheumatic disease mainly confined to women during the child-bearing years, that is found in all ethnic groups but is particularly common amongst Afro-Carribeans (Johnson et al., 1995). SLE is characterised by B lymphocyte hyperactivity and decreased cellular inunune responses. Such immune disturbances may in part reflect abnormalities in the expression of cytokines which regulate the functions of cells within the normal immune system. In agreement with this, elevated levels of the cytokines interleukin-6 (Linker-Israeli et al., 1991) and interleukin-10 (Park et al., 1998) have been reported in patients with SLE and are highest in patients with active disease. Any role for cytokines in the pathogenesis of SLE, must involve the activation of specific genes in target cells. Understandably, the identification of these genes and dissection of the molecular mechanisms by which they are regulated both in vitro and in vivo, is of importance. The heat shock proteins (Hsps) are a group of proteins whose synthesis is enhanced in cells in response to a variety of stressful and non-stressful conditions (see Chapter 1, Section 1.1). Elevated levels of Hsp 90 have been reported in up to 30% of patients with SLE (Dhillon et al., 1993a). Intriguingly, this overexpression is found most often in patients who are not HLA Al, B8, DR3, the most conunon haplotype in Caucasians who have the disease (Dhillon et al., 1994). Evidence of a potential role for Hsp 90 in the aetiology and pathogenesis of this disease has been 157 Results supported by studies in autoimmune MRL-//?r//pr mice where elevated levels of Hsp 90 are detectable prior to the onset of disease (Faulds et al., 1994). The precise mechanisms leading to the observed elevation of Hsp 90 in patients with SLE, have however remained unclear. Previous studies which have suggested a role for the cytokine interleukin-6 (IL-6) in the regulation of Hsp 90 in SLE (see Chapter 1, Section 1.1), have been extended in Chapter 3 to show that in vitro, the cytokine interleukin-10 (IL-10) up-regulates Hsp 90 in a human hepatoma cell line (HepG2) and in peripheral blood mononuclear cells (PBMCs) from normal individuals (Ripley et al., 1999). Thus it is conceivable that both 11^6 and IL-10 are involved in the regulation of Hsp 90 in vivo, within cells of the normal immune system. In addition, the elevated circulating levels of IL-6 (Linker-Israeli et al., 1991) and IL-10 (Park et al., 1998) which have been reported in patients with SLE, would suggest a role for both cytokines in the overexpression of Hsp 90 in this disease. In this chapter, an investigation of the correlation between IL-6 or IL-10 levels and the levels of heat shock proteins in PBMCs, from normal individuals and patients with SLE, is outlined. In view of previous studies which suggested a genetic association with the overexpression of Hsp 90 in SLE (Dhillon et al., 1994), it was thought to be of interest to investigate in parallel, any effect of HLA status on the involvement of cytokines in the regulation of Hsps in patients with SLE. 158 Results 4.1 Quantitative Determination of Heat Shock Proteins Levels of Hsp 90 and Hsp 70 in peripheral blood mononuclear cells (PBMCs) from patients with SLE and controls (normal individuals and patients with rheumatoid arthritis) were determined by the relevant methods outlined in Chapter 2. 4.2 Pattern of Heat Shock Protein Expression in SLE Table 4.2.1 shows the mean, 95% confidence interval of the mean (-2SEM to +2SEM), median and range, of the levels of Hsp 90 and Hsp70 in normal individuals and patients with SLE or rheumatoid arthritis (RA). The mean level of Hsp 90 in patients with SLE was found to be significantly higher (p<0.01) than the mean Hsp 90 level determined for the normal control group. In contrast, the mean Hsp 90 level in patients with RA was not observed to be significantly higher than the mean Hsp 90 level of the normal control group. No significant differences in the mean Hsp 70 level were observed in patients with either SLE or RA, compared to the normal control group. Thus, the observed pattern of Hsp levels in patients with SLE or RA, compared to normal individuals, is in agreement with previous studies (Dhillon et al., 1993a) and is discussed in Chapter 1, Section 1.4. Scatter graphs representing the levels of Hsp 90 and Hsp 70 in normal individuals and patients with SLE or RA, are shown in Figure 4.2.1 and Figure 4.2.2. Table 4.2.2 shows the upper limit of the reference range for each Hsp in the normal control group. The 97.5^ centile was used as the upper limit of “normal” for both Hsp 90 and Hsp 70. Table 4.2.2 also shows the number and percentage of patients with SLE or RA, with levels of Hsps above the normal 97.5 ^ centile. Thus, by this criterion, elevated levels of Hsp 90 were observed in 30 patients with SLE (30.61%) and one patient with RA (5%). 159 Results Table 4.2.1 Hsp levels in PBMCs of Controls and Patients with SLE Hsp 90 Hsp 70 SLE n 95 107 Mean 2.46 1.81 95% CIM 1.84 - 3.08 1.67 - 1.93 Median 1.82 1.73 Range 15.6 2.32 IQR 2.68 2.14 Normals n 30 30 Mean 1.54 1.25 95% CIM 1.16 -1.92 1.08 - 1.42 Median 1.43 1.19 Range 3.02 2.42 IQR 1,79 1.58 Pvalue^ 0.01 >0.1 SE difference 0.36 0.14 Results Table 4.2.1 (continued) Hsp 90 Hsp70 Rheumatoid Arthritis n 20 20 Mean 1.83 0.67 95% CIM 1.62-2.04 0.55-0.79 Median 1.94 0.52 Range 3.68 2.13 IQR 1.83 1.12 P value^ >0.6 >0.4 SE difference 0.08 0.19 Key: n Number of subjects. CIM Confidence interval for mean. IQR: Value for the interquartile range. P value: Level of significance for the difference in mean Hsp levels between patients with SLE and normals (P value^) and between patients with RA and normals (P value^), using Student’s t-test on these data. SE Value for the standard error in the mean Hsp level Results Table 4.2.2 Upper Limits of Normal Hsp Levels and Number of Patients with SLE or RA with Elevated Hsp Levels Hsp 90 Hsp 70 Normals Mean + 2SD 3.13 1.45 97.5"^ centile 2.65 1.32 SLE Number of patients with 30 7 elevated Hsp levels (above the normal 97.5^ centile) % of patients with elevation 30.61 6.54 Rheumatoid Arthritis Number of patients with 1 1 elevated Hsp levels (above the 97.5* centile) % of patients with elevation 5.00 5.00 Figure 4.2.1 Hsp 90 k vels in SLE and Control Groups Hsp 90 18.00 16.00 P< 0.02 14.00 12.00 10.00 8.00 6.00 4.00 2.00 1 f n 95 30 20 Group SLE Normals RA Figure 4.2.2 Hsp 70 Levels in SLE and Control Groups Hsp 70 3.00 2.50 2.00 - 1.50 1.00 - 0.50 n 107 30 20 Group SLE Normals RA Results Legend to Fignres 4.2.1 and 4.2.2 Scatter graphs of Hsp levels for all groups. Dotted lines show median values and solid lines indicate the 97.5^ centile of the Hsp levels in the normal control group. (p< 0.02) - level of significance for the difference in mean values of Hsp 90 levels between SLE and normal groups, using a Student’s t-test for equality of means. No significant differences in mean Hsp 90 levels were found between RA and normal groups, or for Hsp 72 between any groups. Key to Abbreviations: SLE systemic lupus erythematosus group C normal control group RA rheumatoid arthritis group 165 Results Figore 4.2.3 is a scatter graph comparing levels of Hsp 90 with levels of Hsp 70 in individual samples from patients with SLE. No significant correlation between the levels of Hsp 90 and Hsp 70 in patients with SLE was observed. Thus, the observed elevation of Hsp 90 levels in patients with SLE, compared to normals, was specific to this Hsp, and was not paralleled by an elevation of Hsp 70 levels in patients with this disease compared to normals. In view of previous studies which have suggested a genetic origin to the elevation of Hsp 90 in SLE (Dhillon et al., 1994), mean levels of Hsp 90 were determined in patients with SLE with respect to the HLA alleles A l, B8 and DR3, which have been previously reported to be strongly associated with SLE (Worrall et al., 1990). Figure 4.2.4 shows mean Hsp 90 levels and 95% confidence limits (-2 SEM to +2 SEM), in patients with SLE with respect to the presence or absence of individual/combined HLA alleles/haplotypes. Significantly higher mean Hsp 90 levels were observed in patients who were HLA DR3-negative compared to those patients who were HLA DR3-positive (P< 0.01) and this was observed for all combined haplotypes in which DR3-negativity was present. Patients lacking the HLA B8 allele were also found to have higher mean Hsp 90 levels compared with those patients who possessed this allele (PcO.OS). In contrast. Figure 4.2.5 shows that no significant differences in mean Hsp 70 levels were associated with either the presence or absence of any of these HLA alleles. A further statistical analysis of the association of Al, B8 and DR3 HLA alleles to Hsp 90 levels in patients with SLE, was applied to the patients with elevated Hsp 90 levels. Table 4.2.3 shows chi-square analyses and levels of significance for this data and highlights HLA DR3 status as being significantly associated with a patient with SLE having elevated levels of Hsp 90. Table 4.2.4 shows the number of patients with elevated or normal levels of Hsp 90, with respect to the presence or absence of the HLA DR3 allele and also details a chi-square analysis of these data (see statistical analyses in Chapter 2). 166 Figure 4.2.3 Correlation Between Hsp 90 and Hsp 70 Levels in SLE 18.00 16.00 14.00 12.00 10.00 8.00 6.00 • • • 4.00 2.00 • • • • • • • #***«* ■# ##«#? *# # • • 20.00 40.00 60.00 80.00 100.00 120.00 140.00 Hsp 70 Scatter graph showing the correlation between levels of Hsp 90 and Hsp 70 determined for individual patients with SLE Figure 4.2.4 Hsp 90 mean (and 95% CI of mean) with respect to HLA alleles/haplotype P=0.05 P=0.01 P=0.006 P=0.004 Allele or Haplotype Present m Absent 9 84 7 87 13 81 A1B8 B8Dr3 AlDr3 Figure 4.2.5 Hsp 70 mean (and 95% CI of mean) with respect to HLA alleles/haplotypes Allele or Haplotype Present Absent g 11 87 8 91 13 86 A1B8 B8DR3 A1DR3 Results Legend to Figures 4.2.4 and 4.2.5 Bar charts of mean Hsp levels (-2SEM to +2SEM) with respect to the presence or absence of individual HLA alleles/haplotypes. P values denote the level of significance for the difference in mean values of Hsp 90 levels in patients with SLE with respect to the presence or absence of HLA alleles/haplotypes., using a Student’s t-test for equality of means. No significant differences in mean Hsp 70 levels were found in patients with SLE with respect to the presence or absence of individual HLA alleles/haplotypes. Key to Abbreviations: n number of subjects Al HLAAl B8 HLAB8 DR3 HLADR3 A1B8 HLAAl,B8 B8DR3 HLA B8, DR3 A1DR3 HLAALDR3 170 Results Table 4.2.3 Association of HLA Alleles/Haplotypes with Hsp 90 Elevation in SLE Significance Test: Chi-Square (Pearson^ Allele/Haplotvpe Value Significance Al 0.18 0.67 B8 1.95 0.16 DR3 4.27 0.04 (significant: p<0.05) A1,B8 1.96 0.38 A1DR3 2.34 0.13 B8,DR3 1.31 0.25 Results Table 4.2.4 Association of HLA DR3 Status with Hsp 90 Elevation in SLE Hsp 90 Not Elevated Hsp 90 Elevated DR3 Present n = 20 n = 4 Expected Value 15.9 Expected Value 8.1 Residual 4.1 Residual -4.1 DR3 Absent n = 39 n = 26 Expected Value 43.1 Expected Value 21.9 Residual -4.1 Residual 4.1 Significance Tests Chi-Square Value Significance Pearson 4.27 0.04 Likelihood Ratio - 3.29 0.03 Linear by Linear Association 4.64 0.04 Fisher’s Exact Test: One-Tail 0.03 Two-Tail 0.05 Results 4.3 Cytokine Levels in SLE The mean, 95% confidence limits (-2SEM to +2SEM), median and range of the levels of IL-10 and IL-6 in normals and patients with SLE or RA, are shown in Table 4.3.1, In agreement with previous studies, IL-10 was found to be significantly elevated in patients with SLE (Park et al., 1998) and IL-6 levels were found to be elevated in both patients with SLE (Linker-Israeli et al., 1991) and RA (al-Janadi et al., 1993). The upper limits of the reference ranges for each cytokine in the control normal group are shown in Table 4.3.2. The 97.5^ centile was used as the upper limit of “normal” for both IL-10 and IL-6 (see statistical analyses in Chapter 2, Section 2.6). Table 4.3.3 shows the number and percentage of patients with SLE or RA, with levels of IL-10 or IL-6 above the normal 97.5^ centile. Elevated levels of IL-10 were observed in 18 patients with SLE (17%) and two patients with RA (10%). Elevated levels of IL-6 were found in 67 patients with SLE (58%) and 12 patients with RA (60%). Scattergraphs showing the levels of IL-10 and IL-6 in patients with SLE or RA and normal individuals, are shown in Figure 4.3.1 (IL-10) and Figure 4.3.2 (IL- 6 ). The normal individual with a raised level of IL-10 (IL-10 2.25), was a 38 year old clinically well male, and is the same subject previously determined to have an elevated level of Hsp 90 (Hsp 90 2.6). There were no normal individuals with significantly elevated levels of IL-6. In view of previous studies (Dhillon et al., 1994) and the results presented here, which support a genetic origin to the overexpression of Hsp 90 in SLE and studies suggesting a role for II^IO (Ripley et al., 1999) and/or IL-6 (Stephanou et al., 1997; 1998c) in Hsp 90 regulation in SLE, levels of both IL-6 and IL-10 in serum of patients with SLE were compared with the respective patient HLA type. Mean and 95% Cl mean (-2SEM to 4-2SEM) for each cytokine relative to HLA type are shown in Figure 4.3.3 (IL-6) and Figure 4.3.4 (IL-10). No significant differences were found in the mean levels of either cytokine between HLA types in patients with SLE, as determined by a students t-test (see Chapter 2) for equality of means. 173 Results Table 4.3.1 Cytokine Levels in PBMCs of Normals and Patients with SLE or Rheumatoid Arthritis IL-6 IL-10 SLE n 116 108 Mean 1.66 1.94 95% Ci 1.53-1.79 1.83-2.05 Median 1.55 1.78 Rangé 3.90 3.14 IQR 0.89 0.79 Normals n 22 22 Mean 1.01 1.36 95% Cl 0.91-1.12 1.13-1.58 Median 0.86 1.49 Rangé 1.15 2.16 IQR 0.48 0.52 P value* <0.01 < 0.05 SE difference 0.08 0.15 Results Table 4.3.1 (continued) Rheumatoid Arthritis IL-6 IL-10 n 20 20 Mean 1.88 1.48 95% Cl 1.65-2.12 1.23-1.73 Median 1.74 1.52 Range . 1.87 2.49 IQR 1.04 0.98 P value^ <0.02 >0.3 SE difference 0.04 0.17 Key: P value: Level of significance for the difference in the mean levels of cytokines determined for the control normal group and patients with SLE (P value^) and for the control normal group and patients with rheumatoid arthritis and normals (P value^), using Student’s t-test on raw data. Note- All statistical analyses of levels of IL-6 or IL-10 in all subject groups, presented in this chapter or proceeding chapters, have been performed on log^Q transformed data (raw data), as discussed in Chapter2. Levels of IL-6 or IL-10 in the raw data were expressed in units of g'^/ml. Results Table 4.3.2 Upper Limits of Normal Cytokine Levels and Number of Patients with SLE or Rheumatoid Arthritis with Elevated Levels of Cytokines IL-6 IL-10 Normals Mean + 2SD 1.51 2.41 97.5* centile 139 2.08 SLE Number of patients 67 18 with elevation above the normal 97.5* centile % of patients with elevation 58.00 17.00 Rheumatoid Arthritis Number of patients 12 with elevation above the normal 97.5* centile % of patients with elevation 60.00 10.00 Figure 4.3.1 IL-10 Levels in SLE and Control Groups 4.50 4.00- 3.50 - 3.00 ! O) 2.50 to _j 2.00^ i 1.50 1.00 - .50 .00 n 108 22 20 Group SLE Normals RA Figure 4.3.2 IL 6 Levels in SLE and Control Groups 4.50 4.00- 3.50- 3.00- (X> _j 2.00 1.50- 1. 0 0 - .50- # .00 n 116 22 20 Group SLE Ncmnals RA Results Legend to Figures 4.3.1 and 4.3.2 Scattergraphs showing levels of cytokines in patients with SLE or RA and normal individuals 179 Figure 4.3.3 IL-6 mean (and 95% Cl of mean) with respect to HLA alleles/haplotypes in SLE 1.75 1 Allele or Haplotype Present 7 1.65 Absent % E % Ï 1.55 % 1.45 29 59 16 73 24 65 9 84 7 87 13 81 A1 B8 DR3 A1B8 B8DR3 A1DR3 Figure 4.3.4 IL-10 mean (and 95% Cl of mean) with respect to HLA alleles/haplotypes Allele or 3.5 Haplotype Present 3 Absent 2.5 m y % 7 87 13 81 B8DR3 A1DR3 Results Legend to Figures 4.3.3 and 4.3.4 Bar charts showing the mean levels of cytokines (-2SEM to +2SEM) in patients with SLE with respect to the presence or absence of individual HLA alleles/haplotypes. No significant differences in mean levels of IL-6 or IL-10 were found in patients with SLE with respect to the presence or absence of individual HLA alleles/haplotypes. Key to Abbreviations: n number of subjects A1 HLAAl B8 HLA B8 DR3 HLADR3 A1B8 HLAALB8 B8DR3 HLA B8, DR3 A1DR3 HLAALDR3 182 Results 4.4 Correlation of Cytokine Levels with Heat Shock Protein Levels in SLE In this chapter, levels of cytokines and Hsps have been determined respectively in individual serum and PBMC samples, which were isolated at the same time from normal subjects and patients with SLE or RA. An investigation of the relative effects of IL6 and IL-10 on the regulation of Hsp 90 and Hsp 70 gene expression in SLE, is outlined in this section. Levels of cytokines were compared with levels of Hsps in patients with SLE. Figure 4.4.1 is a scatter graph of IL-6 levels compared with Hsp 90 levels in patients with SLE and shows a significant correlation between the two variables (r=0.39, P< 0.05). No significant correlation was found between levels of IL-6 and Hsp 70 in patients with SLE (Figure 4.4.2). Levels of IL-10 did not correlate with levels of Hsp 90 (Figure 4.4.3) or with levels of Hsp 70 (Figure 4.4.4), in patients with this disease. Mean levels of Hsp 90 were compared in patients with elevated to those with non-elevated levels of IL-6. Figure 4.4.5 shows that patients with elevated levels of IL-6 had significantly higher mean levels of Hsp 90 in comparison with patients with non-elevated IL-6 levels (P< 0.05) and the total SLE patient cohort (P< 0.1). The reverse comparison of IL-6 levels in patients with elevated and non elevated Hsp 90 is shown in Figure 4.4.6. Significantly higher mean IL-6 levels are found in patients with elevated Hsp 90 compared to both patients with non-elevated Hsp 90 and the total studied SLE population. Interestingly, the differences in mean IL-6 levels between those patients with elevated and non-elevated Hsp 90 are significant much more than that found for the differences in mean Hsp 90 183 Figure 4.4.1 Cor relation Between Levels of IL-6 and Hsp 90 in Individual Samples from Patients with SLE Hsp 90 18.00 16.00 14.00 12.00 10.00 8.00 6.00 4.00 2.00 .00 .50 1.00 1.50 2.00 2.50 3.00 8.50 4.00 4.50 Logio IL-6 (g'^/ml) Figure 4.4.1 shows the correlation between levels of IL-6 and Hsp 90 in individual samples from patients with SLE. A trendline and the associated correlation coefficient (r=0.39) and Pearson’s p value (p< 0.01) are also shown. Figure 4.4.2 Correlation Between Levels of IL-6 and Hsp 70 in Individual Samples from Patients with SLE Hsp 70 4.50 4.00 3.50 3.00 2.50 2.00 1.50 1.00 0.50 0.50 1.00 1.50 2.00 2.50 3.00 3.50 Log ,0 IL-6 (g'^/ml) Figure 4.4.2 shows the correlation between levels of IL-6 and Hsp 70 in individual samples from patients with SLE. A trendline with the associated correlation coefficient (r=- 0.11) is also shown. Figure 4.4.3 Correlation Between Levels of IL-10 and Hsp 90 in Individual Samples from Patients with SLE Hsp 90 18.00 -I 16.00 14.00 - 12.00 10.00 8.00 6.00 r= -0.08 4.00 2.00 0 0.5 1 1.5 2 2.5 3 Figure 4.4.3 shows the correlation between levels of IL-10 and Hsp 90 in individual samples from patients with SLE. A trendline with the associated correlation coefficient (r= -0.08) is also shown. Figure 4.4.4 Correlation of Between Levels of IL-10 and Hsp 70 in Individual Sandies from Patients with SLE Hsp 70 4.50 4.00 3.50 3.00 2.50 2.00 1.50 1.00 r= -0.16 0.50 0 0.5 1 1.5 2 2.5 3 Log 10 IL-10 (g^/ml) Figure 4.4.4 is a scattergraph showing the correlation between levels of IL-10 and Hsp 70 in individual samples from patients with SLE. A trendline and associated correlation coefficient (r=-0.16) is also shown. Figure 4.4.5 Hsp 90 Mean (and 95% CI of Mean) with Respect to IL-6 Levels in Normals and SLE P<0.10 Hsp 90 P< 0.05 3.5 3 2.5 2 - 1.5 - 0.5 - 30 95 36 59 Normals All SLE IL-6 normal IL-6 elevated Figure 4.4.6 IL-6 Mean (and 95% Cl of Mean) with Respect to Hsp 90 Levels in Normals and SLE Logio IL-6 (g'^/ml) 3 - P< 0.05 P<0.02 2.5 1 2 - 1.5 0.5 30 95 67 28 Normals All SLE Hsp 90 normal Hsp 90 elevated Results Legend to Figure 4.4.5 Bar chart showing the mean levels of Hsp 90 (-2SEM to +2SEM) in the normal control group and in the total SLE patient population and patients with normal or elevated levels of IL-6. P values represent the level of significance for the difference in mean Hsp 90 levels between groups, using a students t-test for the equality of means. Legend to Figure 4.4.6 Bar chart showing the mean levels of 11^6 (-2SEM to +2SEM) in the normal control group and in the total SLE patient population and patients with normal or elevated levels of Hsp 90. P values represent the level of significance for the difference in mean IL-6 levels between groups, using a students t-test for the equality of means. 190 Results levels between patients with elevated and non-elevated IL-6 levels (0.02 191 Figure 4.4.7 Correlation of IL-6 and Hsp 90 in Patients with SLE who Overexpress Hsp 90 4.50 4.00 - - 0.49, P<0.01 3.50 - 3.00 - « 2.50 Ë 2.00 •• 1.50 1.00 .50 .00 2.00 4.00 6.00 8.00 10.00 12.00 14.00 16.00 18.00 Hsp 90 Figure 4.4.7 shows the correlation between levels of IL-6 and Hsp 90 in individual samples from patients with SLE who overexpress Hsp 90. A trendline and the associated correlation coefficient (r=0.49) and Pearson’s p value (p< 0.01) are also shown. Results 4.5 Discussion A role for Hsps in the pathogenesis of autoimmune diseases has been extensively studied (see Chapter 1, Section 1.2) and may involve the induction of Hsp-specifîc immune responses, with pathological consequences. Studies of Hsp expression in patients with SLE, have previously highlighted the specific overexpression of Hsp 90 in this disease, in the absence of elevation of other Hsp family members (Dhillon et al., 1993a). In agreement with these observations, results in this chapter have demonstrated normal levels of Hsp 70 in SLE and elevated levels of Hsp 90 in approximately 30% of patients with SLE. Furthermore, levels of Hsp 70 and Hsp 90 were not elevated in patients with rheumatoid arthritis. These results therefore support a previously held theory which suggests Hsp 90 overexpression in SLE to represent a distinct regulatory mechanism specific to the induction of Hsp 90 and not a general cellular response to stress (Latchman and Isenberg, 1994). However, until now, an understanding of the molecular events which lead to Hsp 90 overexpression has been unresolved. SLE is characterised by B cell hyperactivity, reflected in a broad spectrum of autoantibodies and decreased in vivo and in vitro cellular immune responses. Such immune responses may in part reflect abnormalities in the expression of cytokines, which regulate the function of cells within the normal immune system. Elevated levels of the cytokines IL-6 (Linker-Israeli et al., 1991) and Il^lO (Park et al., 1998) are found in patients with active SLE and a number of studies have suggested a role for these cytokines in the overexpression of Hsp 90 in SLE. Thus, IL-6 (Stephanou et al., 1997) and II^IO (Ripley et al., 1999) (see Chapter 3) have been demonstrated to activate Hsp 90 expression in vitro and Hsp 90 is specifically elevated in IL-6 transgenic mice (Stephanou et al., 1998a). The relative effects of these two factors in the overexpression of Hsp 90 in SLE has been investigated in this chapter. Levels of IL-6 were found to correlate significantly with levels of Hsp 90 but not Hsp 70 in patients with SLE. In contrast, levels of 11^ 10 did not correlate with either Hsp 90 or Hsp 70 levels in this disease. Thus, these observations support the involvement of a distinct regulatory mechanism specific for the overexpression of Hsp 90 in SLE and one which is mediated by 193 Results elevated levels of IL-6 and not elevated levels of IL-10, in patients with this disease. Of particular interest to the studies in this chapter, is the observation of an increased correlation between levels of IL-6 and Hsp 90 in patients who overexpress Hsp 90. A genetic association with Hsp 90 overexpression in SLE has been suggested by the results presented in this chapter, which show that those patients who lack the HLA B8 or DR3 alleles, are those most likely to overexpress Hsp 90. As discussed in Chapter 1, Section 3, the gene encoding Hsp 90p is located on Chromosome 6 (Durkin et al., 1993; Takahashi et al., 1994), the site of the MHC. It is conceivable that polymorphisms exist within the promoter region of the Hsp 90p gene, which in a subset of patients with SLE, contribute to the observed elevated expression of Hsp 90. Genetic upregulation of Hsp 90 protein in these patients might occur via an increase in the basal level of Hsp 90 gene expression, or indeed be mediated by an enhanced transcriptional response to elevated levels of IL-6 in SLE. However, to date, no studies have established the presence of polymorphisms within regulatory regions of the Hsp 90p gene. In criticism of the association of elevated Hsp 90 expression in SLE with patients lacking HLA B8 and/or DR3 alleles, is the absence of statistical analyses which compare Hsp 90 expression in different ethnic groups of both the total SLE population and of the control normal population. Such analyses are of importance since, as discussed in Chapter 1, Section 1.3, different HLA alleles occur with varying frequencies in different populations, according to both ethnic composition and geographic location. However, since the number of patients in some ethnic groups was low, statistical analyses of this type are unlikely to attain significance. The study of Dhillon et al., (1994), also found an association between elevated Hsp 90 levels in SLE and patients lacking HLA B8 and/or DR3 alleles. This study used a greater number of SLE patients (incorporating the study of Norton et al., 1989) and a population of similar ethnic composition (see Chapter 2) to that used in this dissertation. No differences in Hsp 90 levels were observed in White Caucasoid SLE patients compared with Non-White patients, suggesting these findings and possibly those of this dissertation, to not be the result of ethnic bias. The overexpression of Hsp 90 in SLE may therefore be mediated both by elevated levels of IL-6 in this disease and genetic factors, which for significantly 194 Results elevated levels of IL-6, sustain a linear regulatory relationship between IL-6 and Hsp 90. Taken together, these results suggest an involvement of IL-6 and not IL-10 in the regulation of Hsp 90 in SLE, which is enhanced in a genetic subset of patients with SLE. 195 Results Chapter 5 Elevated Expression of Hsp 90 and Antibodies to Hsp 90 in SLE 5.0 Introduction A role for heat shock proteins (Hsps) in the pathogenesis of autoimmune disease has been extensively reviewed. These studies have highlighted possible mechanisms by which Hsps can initiate an autoimmune response (discussed in Chapter 1, Section 1.2). For example, the mycobacterial Hsp 65 may play a role in adjuvant arthritis in rats and increased levels of antibodies to Hsp 65 have been reported in patients with rheumatoid arthritis (Tsoulfa et al., 1989). Overexpression of the 90 kDa heat shock protein (Hsp 90) is evident in a subset of patients with SLE (Dhillon et al., 1994) and antibodies to Hsp 90 have been observed in approximately one-third of patients with SLE, of both adult (Conroy et al., 1994) and childhood onset (Conroy et al., 1996). Any role for Hsp 90 in the development of Hsp 90 autoantibodies is likely to require its localisation on the surface of cells of the immune system, where it is accessible to the immune system (discussed in Chapter 1, Section 1.4). Surface expression of Hsp 90 in peripheral blood mononuclear cells (PBMCs) from patients with SLE, has been investigated by flow-cytometry (Erkeller-Yuksel et al., 1992). In this study, approximately 20% of patients with SLE were found to surface express Hsp 90, but this was rarely observed in patients with rheumatoid arthritis or Sjogren’s disease. Furthermore, surface expression of Hsp 90 in SLE, correlated with the specific subset of patients in which overexpression of Hsp 90 is a feature. A role for anti=Hsp 90 antibodies in the pathogenesis of SLE, is supported by studies in a murine model of lupus. In autoimmune MRL~lprllpr mice, the 196 Results development of antibodies to Hsp 90 precedes the onset of lupus-like disease and occurs after splenocyte elevation of Hsp 90 (Faulds et al., 1994). Moreover, antibodies to Hsp 90 have been identified in 60% of MRL-lprllpr mice and were rarely found in control BALB/C mice (Faulds et al., 1995). Studies in murine models of SLE therefore suggest a role for Hsp 90 overexpression in the development of antibodies to Hsp 90 in this disease (reviewed in Stephanou et al., 1998c). Since the presence of these antibodies precedes the onset of disease in MRL=lprllpr mice, an understanding of the mechanisms which lead to their production in human SLE, are of great interest. Interestingly, elevation of IL-6 in transgenic mice, leads to increased levels of Hsp 90 and the production of anti-Hsp 90 antibodies (Stephanou et al., 1998a). In the previous chapter, a number of patients with SLE were identified who overexpressed Hsp 90 and these patients correlated with a subset of patients which has been reported as being most likely to overexpress Hsp 90 (Dhillon et al., 1994). Using serum samples taken from normal individuals and patients with SLE, isolated at the time of Hsp quantitation, an investigation of the correlation between Hsp overexpression and the presence of raised levels of antibodies to Hsps in SLE, is now possible and is detailed in this chapter. 197 Results 5.1 Detection of Antibodies to Hsps Using serum samples from normal individuals and patients with SLE or rheumatoid arthritis (RA), levels of IgG and IgM antibodies to Hsp 90 and Hsp 70, were measured by ELISA, using the methods outlined in Chapter 2. Serum samples obtained from individual patients with SLE or RA, were considered to have a raised level of an antibody (IgG or IgM to Hsp 90 or Hsp 70) if the level was above the 97,5^ centile of the respective antibody levels in the normal control group. 5.2 Pattern of Expression of Antibodies to Hsps in SLE As shown in Table 5.2.1 and Figure 5.2.1, raised levels of IgG antibodies to Hsp 90 were detected in 46% of patients with SLE. Only 15% of patients with RA were found to have raised levels of IgG antibodies to Hsp 90. Raised IgM antibody levels to Hsp 90 (Table 5.2.1 and Figure 5.2.2) were detected in 28% of patients with SLE and 10% of patients with RA. Thus, raised levels of antibodies (IgG or IgM) to Hsp 90 were frequently detected in patients with SLE and in contrast, were rarely detected in patients with RA. Table 5.2.1 also shows the number and percentage of patients with SLE or RA with raised levels of antibodies to Hsp 70. In patients with SLE, raised levels of IgG antibodies to Hsp 70 (Figure 5.2.3) were detected in 14% of subjects. Only 5% of patients with RA were found to have raised IgG levels to Hsp 70. Levels of IgM antibodies to Hsp 70 (Figure 5.2.4) were detected in 18% of patients with SLE and were not detected in patients with RA. Taken together, the expression pattern of antibodies to Hsp 90 and Hsp 70 in patients with SLE or RA presented here, is in agreement with previous studies of patients with SLE (Conroy et al., 1994; 1996) and which used an identical ELISA system for antibody detection to that used in the studies in this chapter. 198 Results Table 5.2.1 Upper Limits of Normal Levels of Antibodies to HSPs and Number of Patients with SLE or RA with Raised Antibody Levels Antibody to Hsp Hsp 90 IgG Hsp 90 IgM Normals n 30 30 Mean +2SD 0.54 0.47 97.5^ Centile 0.61 0.56 SLE 116 102 Mean antibody level 0.71 0.42 Number of patients with raised antibody level (above the normal 97.5* centile) 53 29 % of patients with raised antibody level 46 28 Rheumatoid Arthritis 20 20 Mean antibody level 0.38 0.33 Number of patients with raised antibody level (above the normal 97.5* centile) % of patients with raised antibody level 15 10 Results Table 5.2.1 (continued) Hsp 70 IgG Hsp 70 IgM Normals n 30 30 Mean +2SD 1.1 0.61 97.5‘*‘ Centile 1.2 0.53 SLE n 98 102 Mean antibody level 1.13 0.49 Number of patients with raised antibody level (above the normal 97.5^ centile) 12 % of patients with raised antibody level 14 18 Rheumatoid Arthritis n 20 20 Mean antibody level 0.62 0.33 Number of patients with raised antibody level (above the normal 97.5^ centile) 0 % of patients with raised antibody level 0 Figure 5.2.1 Levels of IgG Antibodies to Hsp 90 in SLE and Control Groups 2.5 ! a < 0.5 n 98 30 20 Group SLE Normals RA Figure 5.2.2 Levels of IgM Antibodies to Hsp 90 in SLE and Control Groups 1 -[ 0.9 - <» <» 0.8 - i» 0.7 - 1 i u 0.6 - ii 2 0.5- « I • 1 • 1 0.4- • # < # e # • 0.3 - 1 • • • 0.2- • \ 1 1 1 0.1 - ( 1 ! 0- 1 ...... : n 102 30 20 Group SLE Normals RA Figure 5.2.3 Levels of IgG Antibodies to Hsp 70 in SLE and Control Groups E 0.8- n 98 30 20 Group SLE Normals RA Figure 5.2.4 Levels of IgM Antibodies to Hsp 70 in SLE and Control Groups 1 0.9- 0.8 - 0.7 - > 0.6-1 0.5- e 0.4 < 0.3 0.2 0.1 0 n 102 30 20 Group SLE Normals RA Results Legend to Figures 5.2.1- 5.2.4 Scatter graphs comparing the levels of anti-Hsp antibodies in patients with SLE, normal individuals and patients with rheumatoid arthritis. A solid horizontal black line indicates an antibody level corresponding to the 97.5^ centile of the antibody levels in the normal control group. Key n number of subjects SLE systemic lupus erythematosus patient group RA rheumaroid arthritis patient group 205 Results 5.3 Immunogenetic Analysis of Antibodies to Hsps Previous studies have suggested a genetic association for Hsp 90 overexpression (Dhillon et ah, 1994) and these observations are in agreement with the results presented in Chapter 4. These studies suggest that elevated levels of Hsp 90 in SLE are most likely to occur in a genetic subset of patients with SLE who have an HLA DR3 negative haplotype. An immunogenetic analysis of antibodies to Hsps in SLE has also been reported (Conroy et al., 1994) and is discussed in Chapter 1, Section 1.4. In this study, a correlation between the presence of raised levels of anti- Hsp 90 antibodies and patients who have an HLA DR3 negative haplotype, was observed. Missing from these studies however, has been a comparative immunogenetic analysis of both elevated Hsp 90 protein levels and raised anti-Hsp 90 antibody levels, in the same patients. Tables 5.3.1-4 show the number of patients with SLE who were determined to have raised (IgG or IgM) anti-Hsp 90 or anti-Hsp 70 antibodies, with respect to the presence or absence of individual/combined HLA haplotypes. As shown in Table 5.3.1, raised IgG anti-Hsp 90 antibodies were found to correlate with patients who lack the HLA DR3 haplotype (p= 0.043), as determined by a chi-square analysis (described in Chapter 2) and is in agreement with the study of (Conroy et al., 1994). Interestingly, the absence of this haplotype in patients with SLE, was observed (see Chapter 4) to correlate significantly (p= 0.042) with elevated (above the normal 97.5^ centile) Hsp 90 levels, as determined by chi-square analysis. No significant correlations were found in comparisons between either IgM anti-Hsp 90 antibodies or anti-Hsp 70 antibodies (IgG or IgM) and the presence or absence of individual/combined HLA haplotypes and this is in agreement with the study of (Conroy et al., 1994). The observation of shared immunogenetic associations, both for Hsp 90 overexpression and raised antibodies to Hsp 90, in individual patients with SLE, suggests a common link between these events. Thus, it was thought to be of interest to compare levels of Hsp 90 and levels of antibodies to Hsp 90, in individual patients with SLE. 206 Results Table 5.3.1 Immunogenetic Analysis of IgG Anti-Hsp 90 Antibodies in SLE HLA Type Hsp 90 Negative Hsp 90 Positive A1+ 15 17 Al- 37 30 NS B8+ 10 8 B8- 43 39 NS DR3+ 16 12 DR3- 24 41 p= 0.043 Al, B8+ 6 5 A1,B8- 48 47 NS A1,DR3+ 7 9 A1,DR3- 47 44 NS B8, DR3+ 5 4 B8, DR3- 49 49 NS Key: NS Not Significant Results Table 5.3.2 Immunogenetic Analysis of IgM Anti-Hsp 90 Antibodies in SLE HLA Type Hsp 90 Negative Hsp 90 Positive A1+ 12 21 Al- 39 29 NS B8+ 8 11 B8- 45 41 NS DR3+ 15 11 DR3- 36 27 NS A1,B8+ 8 6 A1,B8- 49 43 NS A1,DR3+ 5 8 A1,DR3- 44 41 NS B8, DR3+ 6 4 B8, DR3- 51 43 NS Key -NS Not Significant Results Table 5.3.3 Immunogenetic Analysis of IgG Anti-Hsp 70 Antibodies in SLE HLA Type Hsp 70 Negative Hsp 70 Positive A1+ 21 13 Al- 35 33 NS B8+ 8 8 B8- 47 36 NS DR3+ 15 10 DR3- 39 27 NS A1,B8+ 6 4 A1,B8- 47 42 NS A1,DR3+ 8 7 A1,DR3- 46 45 NS B8, DR3+ 6 3 B8, DR3- 48 42 NS Key: NS Not Significant Results Table 5.3.4 Immunogenetic Analysis of IgM Anti-Hsp 70 Antibodies in SLE HLA Type Hsp 70 Negative Hsp 70 Positive A1+ 13 15 Al- 38 34 NS B8+ 9 6 B8- 44 41 NS DR3+ 15 13 DR3- 37 28 NS A1,B8+ 5 8 A1,B8- 49 44 NS A1,DR3+ 5 11 A1,DR3- 45 46 NS B8, DR3+ 7 3 B8, DR3- AS 46 NS Key: NS Not Significant Results 5.4 Correlation of Hsp Levels with Antibodies to Hsps in SLE Levels of serum antibodies to Hsp 90 and Hsp 70 were compared with their respective Hsp protein levels in PBMCs, obtained from individual patients with SLE, using the methods described in Chapter 2. As shown in Table 5.4.1 and Table 5.4.2, raised IgG antibody levels to Hsp 90 in individual patients with SLE, correlated with elevated Hsp 90 levels in the same patients (p=0.018), as determined by a chi-square analysis (see Table 5.4.2). In contrast. Table 5.4.1 shows that levels of Hsp 90 were not found to correlate with levels of IgM antibodies to Hsp 90. Scattergraphs comparing levels of Hsp 90 with antibodies to Hsp 90 are shown in Figure 5.4.1 (IgG antibodies to Hsp 90) and Figure 5.4.2 (IgM antibodies to Hsp 90). Table 5.4.1 shows that correlations between levels of Hsp 70 and raised antibody levels to Hsp 70 (IgG or IgM) in patients with SLE, were not found to be significant, as determined by chi-square analyses. Scattergraphs comparing levels of Hsp 70 with antibodies to Hsp 70 are shown in Figure 5.4.3 (IgG antibodies to Hsp 70) and Figure 5.4.4 (IgM antibodies to Hsp 70). In previous studies, elevated levels of Hsp 90 (Dhillon et al., 1993a) and raised levels of antibodies to Hsp 90 (Conroy et al., 1994) have been observed in SLE. However, any correlation between elevated levels of cytoplasmic Hsp 90 and the presence of raised levels of anti-Hsp 90 antibody in SLE, has remained unproven A significant correlation between elevated Hsp 90 levels and raised levels of IgG antibodies to Hsp 90 in patients with SLE has been detailed above and this is the first time that such a correlation has been shown. 211 Results Table 5.4.1 Number of Patients with SLE with Elevated Hsps and Raised Hsp Antibody Levels IgG Antibodies to Hsp 90 Normal Levels Raised Levels Protein Levels Elevated Hsp 90 12 18 P= 0.018 Non-Elevated Hsp 90 39 20 IgM Antibodies to Hsp 90 Normal Levels Raised Levels Protein Levels Elevated Hsp 90 21 NS Non-Elevated Hsp 90 41 18 Results Table 5.4.1 (continued) IgG Antibodies to Hsp 70 Normal Levels Raised Levels Protein Levels Elevated Hsp 70 0 NS Non-Elevated Hsp 70 89 IgM Antibodies to Hsp 70 Normal Levels Raised Levels Protein Levels Elevated Hsp 70 NS Non-Elevated Hsp 70 81 14 Results Table 5.4.2 Chi-Square Test for the Correlation of Hsp 90 Overexpression with Raised IgG Antibodies to Hsp 90 in SLE Normal Hsp 90 Elevated Hsp 90 Total Antibody -ve 39 12 51 Antibody +ve 20 18 38 Total 59 30 Chi-Square Significance Pearson 0.018 Likelihood Ratio 0.018 Linear by Linear Association 0.019 Fisher’s Exact Test 0.016 One-tail 0.024 Two-tail Figure 5.4.1 Correlation of Hsp 90 with IgG Antibodies to Hsp 90 In SLE 2.5 1 % 0 1 0.5 - Î .V • • *• 2.00 4 .00 6.00 8.00 10.00 12.00 14.00 16.00 18.00 H sp90 Figure 5.4.2 Correlation of Hsp 90 with IgM Antibodies to Hsp 90 1 0.9 - 0.8 - 0.7 - 0.6 o 0.5 4 .Q < 0.4 - 0.3 0.2 - 0.1 - 0 2.00 4.00 6.00 8.00 10.00 12.00 14.00 16.00 18.00 HspSO Figure 5.4.3 Comparison of Levels of Hsp 70 with Levels of IgG Antibodies to Hsp 70 in SLE 1.8 - 1.6 - 1.4 - I < 0.8 4 • • 0.6 - 0.4 - 0.2 - 0 0 0.5 1 1.5 2 2.5 Hsp 70 Figure 5.4.4 Comparison of Levels of Hsp 70 with Levels of IgM Antibodies to Hsp 70 in SLE 1 0.9 - 0.8 - 0.7 - • I I 0.6 Q) —I 0.5 I c < 0.4 0.3 - 0.2 - 0.1 4 0 0 0.5 1.5 2.5 Hsp 70 Results Legend to Figures 5.4.1- 5.4.4 Scatter graphs comparing the levels of anti-Hsp antibodies and Hsps in patients with SLE. Key SLE systemic lupus erythematosus patient group RA rheumatoid arthritis patient group 219 Results 5.5 Correlation of Levels of Cytokines and Antibodies to Hsps in SLE As discussed in Chapter 1, Section 1.3, SLE is characterised by B cell hyperactivity and decreased cell mediated immunity. Such immune disturbances may be partly due to abnormalities in the expression of cytokines in this disease. For example, elevated levels of the cytokines IL-6 (Linker-Israeli et al., 1991) and IL-10 (Park et al., 1998) have been reported in SLE. Moreover, as discussed in Chapter 1, Section 1.3, the stimulatory effects of both IL-6 and IL-10 on B cell immunoglobulin synthesis together with the known inhibitory effects of IL-10 on cell-mediated inunune responses, may in part explain these inununological anomolies in patients with SLE. The studies detailed in Chapter 4, have shown a significant correlation between elevated levels of IL-6 but not IL-10, on Hsp 90 overexpression in patients with SLE. However, a general stimulatory effect of either IL-6 and/or IL-10 on B cells, may also be involved in the production of anti-Hsp antibodies, including those with specificity for Hsp 90. Levels of both cytokines have therefore been compared with levels of anti-Hsp antibodies, in samples from individual patients with SLE. Scatter graphs comparing the levels of IL-6 with levels of anti-Hsp 90 antibodies are shown in Figure 5.5.1 (IgG isotype) and Figure 5.5.2 (IgM isotype). No significant correlation was found between levels of IL-6 and anti-Hsp 90 antibodies (see Table 5.5.1). Scattergraphs showing a comparison between levels of IL-6 and levels of anti-Hsp 70 antibodies in patients with SLE are shown in Figure 5.5.3 (IgG isotype) and Figure 5.5.4 (IgM isotype). No significant correlation was found between levels of IL-6 and anti-Hsp 70 antibodies (see Table 5.5.1). A similar comparison of IL-10 levels and levels of anti-Hsp antibodies is outlined below. Scatter graphs comparing the levels of IL-10 with levels of anti-Hsp 90 antibodies are shown in Figure 5.5.5 (IgG isotype) and Figure 5.5.6 (IgM isotype). No significant correlation was found between levels of IL-10 and anti-Hsp 90 antibodies (see Table 5.5.2). Scattergraphs showing a comparison between levels of IL-10 and levels of anti-Hsp 70 antibodies in patients with SLE are shown in Figure 220 Results 5.5.7 (IgG isotype) and Figure 5.5.8 (IgM isotype). Similarly, no significant correlation was found between levels of IL-10 and anti-Hsp 70 antibodies (see Table 5.2). 221 Results Table 5.5.1 Correlation Between Levels of IL-6 and Antibodies to Hsps in SLE Antibody Correlation Coefficient Significance (Pearson) Hsp 90 IgG r=0.08 NS Hsp 90 IgM r=0.04 NS Hsp 70 IgG r=0.12 NS Hsp 70 IgM r=0.16 NS Legend to Table 5.5.1 Correlations were performed using the methods outlined in Altman (1999). Antibody level data were categorised as binary variables and coded as normal (code=0) or elevated (code=l), as defined earlier in this chapter and compared with levels of IL-6 as a continuous variable. A correlation coefficient and Pearson significance was derived. P values were considered significant if below the 5% level (p<0.05). As shown above, no significant associations were found. Key: NS not significant Figure 5.5.1 Comparison of IL-6 Levels with Levels of IgG Antibodies to Hsp 90 in SLE 2.5 I Î < 0.5 0 200 400 600 800 1000 1200 1400 1600 1800 2000 IL-6 (pg/ml) Figure 5.5.2 Comparison of IL-6 Levels with Levels of IgM Antibodies to Hsp 90 in SLE 0.8 0.7 0.6 0.5 -S' 0.4 0.3 0.2 0.1 0 200 400 600 800 1000 1200 1400 1600 1800 2000 IL-6 (pg/ml) Figure 5.5.3 Comparison of IL-6 Levels with Levels of IgG Antibodies to Hsp 70 In SLE 2 1.8 1.6 1.4 1.2 1 0.8 0.6 0.4 0.2 0 0 200 400 600 800 1000 1200 1400 1600 1800 2000 IL-6 (pg/ml) Figure 5.5.4 Comparison of IL-6 and IgM Antibodies to Hsp 70 in SLE 0.9 0.8 0.7 - 0.6 0.4 0.3 > # 0.2 0 200 400 600 800 1000 1200 1400 1600 1800 2000 IL-6 (pg/ml) Results Legend to Figures 5.5.1- 5.5.4 Scatter graphs comparing the levels of IL-6 and levels of antibodies to Hsps in patients with SLE. 227 Results Table 5.5.2 Correlation Between Levels of IL-10 and Antibodies to Hsps in SLE Antibody Correlation Coefficient Significance (Pearson) Hsp 90 IgG r=0.02 NS Hsp 90 IgM r=0.07 NS Hsp 70 IgG r=0.04 NS Hsp 70 IgM r=0.13 NS Legend to Table 5.5.2 Correlations were performed using the methods outlined in Altman (1999). Antibody level data were categorised as binary variables and coded as normal (code=0) or elevated (code=l), as defined earlier in this chapter and compared with levels of IL-6 as a continuous variable. A correlation coefficient and Pearson significance was derived. P values were considered significant if below the 5% level (p<0.05). As shown above, no significant associations were found. Key: NS not significant Figure 5.5.5 Comparison of Levels of IL-10 with Levels of IgG Antibodies to Hsp 90 in SLE 2.5 2 1.5 1 • • 0.5 %• • • yk: - 0 50.00 100.00 1 50.00 200.00 250.00 300.00 350.00 400.00 IL-10 (pg/ml) Figure 5.5.6 Comparison of Levels of IL-10 with Levels of IgM Antibodies to Hsp 90 in SLE < 0.4 50.00 100.00 150.00 200.00 250.00 300.00 350.00 400.00 IL-10 (pg/ml) Figure 5.5.7 Comparison of Levels of IL-10 with Levels of IgG Antibodies to Hsp 70 in SLE •5 0.8 0.6 0.4 0.2 50.00 100.00 150.00 200.00 250.00 300.00 350.00 400.00 IL-10 (pg/ml) Figure 5.5.8 Comparison of Levels of IL-10 with Levels of IgM Antibodies to Hsp 70 in SLE 1 0.9 - 0.8 - 0.7 - J 0.6 4.• * • 0) ^ 0.5 - < 0.4 - 0.3 - 0.2 - 0.1 - 0 50.00 100.00 150.00 200.00 250.00 300.00 350.00 400.00 IL-10 (pg/ml) Results Legend to Figures 5.5.5- 5.5.8 Scatter graphs comparing the levels of IL-10 and levels of antibodies to Hsps in patients with SLE. 233 Results 5.6 Discussion Antibodies to both Hsp 90 and Hsp 70 have been reported in patients with SLE (Conroy et al., 1994; 1996). Interestingly, an immunogenetic analysis of anti- Hsp 90 antibodies in SLE (Conroy et al., 1994) correlates with a genetic subset of patients most likely to overexpress Hsp 90 protein (Dhillon et al., 1994). A link between Hsp 90 overexpression and the development of anti-Hsp 90 antibodies has been suggested (Stephanou et al., 1998c) and may involve the altered presentation of Hsp 90 to the immune system. In agreement with this, cell surface expression of Hsp 90 has been detected in lymphocytes from patients with SLE (Erkeller-Yuksel et al., 1992). However, until now, a study comparing Hsp 90 protein levels and levels of anti-Hsp 90 antibodies from the same patients, has not been undertaken. A significant correlation between Hsp 90 overexpression in PBMCs from patients with SLE and the presence of IgG anti-Hsp 90 antibodies in the serum of these patients was found (p= 0.018). No such correlation was found between Hsp 70 and anti-Hsp 70 antibodies, in patients with SLE. Since protein levels of Hsp 90 and not Hsp 70 were found to be significantly elevated in these patients (see Chapter 4), these observations are in agreement with the aforementioned hypothesis which linked the differential expression and surface localisation of Hsps in SLE, with the development of anti-Hsp 90 antibodies (Stephanou et al., 1998c). Interestingly, observations of anti-Hsp 90 antibodies in SLE, parallels studies in autoimmune MRL/lpr mice, in which splenocyte elevation of Hsp 90 (Faulds et al., 1994) and autoantibodies to Hsp 90 (Faulds et al., 1995), preceded the onset of disease. Furthermore, studies of human breast cancer have demonstrated elevated Hsp 90 levels (Jameel et al., 1992) and autoantibodies to Hsp 90 [Conroy, 1995 #35 in this disease. A role for cytokines in the pathogenesis of SLE has been suggested and is supported by the observations that levels of IL-6 (Linker-Israeli et al., 1991) and IL- 10 (Park et al., 1998) are elevated in this disease. Such cytokine abnormalities may contribute to increased antibody and decreased cell-mediated immune responses observed in this disease (Linker-Israeli, 1992). Although a role for IL-6 and not IL-10 in the overexpression of Hsp 90 in SLE, has been detailed in Chapter 4, an analysis 234 Results of the involvement of either cytokine in the development of anti-Hsp antibody production as part of a general B cell stimulatory effect, has until now not been investigated. The results in this chapter show that levels of antibodies to Hsp 90 and Hsp 70 did not correlate with either IL-6 or IL-10 levels. Thus it is likely that any role for IL-6 in the development of anti-Hsp 90 antibodies in this disease, is mediated most strongly by the effect of IL-6 on elevated Hsp 90 levels in this disease. 235 Results Chapter 6 Disease Associations 6.1 Introduction Our understanding of the pathogenesis of systemic lupus erythematosus (SLE) has been aided by the recognition of a broad spectrum of immunological abnormalities and clinical manifestations which characterise this disease (see Chapter 1, Section 1.3. Of particular interest to the studies in this dissertation has been the role of cytokines in the expression of heat shock proteins (Hsps) in SLE and the development of antibodies to Hsps, with possible pathological consequences. As discussed in detail in Chapter 1, Section 1.4 and reviewed in Stephanou et al., (1998c), previous studies have reported the overexpression of Hsp 90 and raised levels of antibodies to Hsp 90 in patients with SLE and these observations are in agreement with the results presented in this dissertation. These studies have been extended (see Chapter 4) to show that overexpression of Hsp 90 in SLE, correlated with elevated levels of the cytokine IL-6 but not IL-10, in patients with this disease. Furthermore, in individual patients with SLE, raised levels of IgG antibodies to Hsp 90 were most strongly associated with elevated levels of Hsp 90 in the same patients (see Chapter 5). Although various studies suggest a role for Hsps (discussed in Chapter 1, Section 1.2) and cytokines (discussed in Chapter 1, Section 1.3) in SLE, the relative contribution of these factors to clinical disease has remained unclear. The British Isles Lupus Assessment Group (BILAG) index is a computerised index for measuring clinical disease activity in patients with SLE (Symmons et al., 1988). Eight organ-based systems are separately assessed (outlined in Chapter 2, Section 2.4), according to the physician’s intention to treat. This information is entered into a computer and an alphabetic score is then obtained for each organ system. Patients attending the SLE clinic at the Centre for Rheumatology, University College London (see Chapter 2), are routinely assessed using the BILAG index. 236 Results Importantly, an assessment of clinical disease activity using the BILAG index, in all the patients with SLE detailed in this dissertation, at the time of providing a sample of venous blood, has been recorded. Previous studies have detailed associations of Hsps with active clinical subsets of SLE using the BILAG index. For example, elevated levels of Hsp 90 in patients with this disease have correlated with both active central nervous system and cardio-vascularZ-respiratory disease and changes in Hsp 90 levels were observed to be sensitive to changes in haematological disease (Dhillon et al., 1993a). Levels of the cytokines IL-6 (Linker-Israeli et al., 1991) and II^IO (Park et al., 1998) have been reported to be highest in patients with active disease. However, associations of these cytokines with specific organ/system involvement, using the BILAG index, have not been reported. The objective of this chapter is to compare levels of Hsps, cytokines and antibodies to Hsps between active and inactive clinical subsets of SLE, using the BILAG index. 237 Results 6.2 Heat Shock Proteins Levels of Hsp 90 and Hsp 70 in PBMCs of patients with SLE, have been determined in Chapter 4. Levels of both Hsp 90 and Hsp 70 were assessed in patients graded by the BILAG index, by the methods outlined in Chapter 2. Table 6.2.1 shows the number of patients with SLE, graded by the BILAG index, with either normal or elevated levels of Hsp 90. No significant correlations between levels of Hsp 90 (normal or elevated, as defined in Chapter 4) and disease activity grade were found, as determined by chi-square analyses (defined in Chapter 2). As shown in Figure 6.2.1, mean levels of Hsp 90 (-2SEM to +2SEM) were determined in patients graded by the BILAG index. No significant differences in mean Hsp 90 levels between grades were found, as determined by student’s t-tests (also defined in Chapter 2) for equality of means. Previous studies (Dhillon et al., 1993a) have associated elevated levels of Hsp 90 with overall active disease as defined by the UCH/Middlesex SLE disease activity index (Morrow et al., 1982; 1983). However, clinical disease activity data for the patients outlined in this chapter, as measured by this index, was not recorded. Table 6.2.1 also shows the number of patients with SLE, graded by the BILAG index, with either normal or elevated levels of Hsp 70. Correlations between levels of Hsp 70 (normal or elevated, as defined in Chapter 4) and disease activity grade were not found to be significant, as determined by chi-square analyses. As shown in Figure 6.2.2, mean levels of Hsp 70 (-2SEM to +2SEM) were determined in patients graded by the BILAG index. Mean Hsp 70 levels were found to be significantly lower (p=0.004) in patients graded by BILAG A/B/C MUSK, compared with patients graded by BILAG D/E MUSK. No significant differences in mean Hsp 70 levels between other disease activity grades were found, as determined by student’s t-tests for equality of means. 238 Results Table 6.2.1 Number of Patients with SLE with Normal or Elevated Levels of Hsp 90 and Graded by BILAG Disease Activity Scores Hsp 90 BILAG Hsp Level Inactive Disease Active Disease Category (Score D/E) (Score A/B/C) GEN Normal 20 41 Elevated 6 24 NS SKIN Normal 41 20 Elevated 20 10 NS CNS Normal 52 Elevated 25 NS MUSK Normal 30 31 Elevated 12 18 NS CV/RP Normal 51 10 Elevated 26 4 NS VASC Normal 39 22 Elevated 19 11 NS RENAL Normal 39 22 Elevated 21 9 NS HAEM Normal 22 39 Elevated 15 15 NS Results (Table 6.2.1 continued) Hsp 70 BILAG Hsp Level Inactive Disease Active Disease Category (Score D/E) (Score A/B/C) GEN Normal 21 62 Elevated 5 2 NS SKIN Normal 71 23 Elevated 3 4 NS CNS Normal 72 24 Elevated 2 4 NS MUSK Normal 43 53 Elevated 5 2 NS CV/RP Normal 75 19 Elevated 6 1 NS VASC Normal 60 35 Elevated 3 3 NS RENAL Normal 48 48 Elevated 1 5 NS HAEM Normal 44 48 Elevated 4 3 NS Results (Table 6.2.1 continued) Key: NS not significant n number of patients GEN general non-specific disease SKIN mucocutaneous disease CNS central nervous system disease MUSK musculoskeletal disease CVRP cardiovascular and respiratory disease VASC vasculitis RENAL renal disease HAEM haematological disease Figure 6.2.1 Mean Hsp 90 Levels (-2SEM to +2SEM) in Patients Graded by the BILAG Index I I A/B/C GEN SKIN CNS MUSK CVRP VASC RENAL HAEM n 26 65 61 30 77 14 48 55 77 14 58 33 60 31 37 54 Figure 6.2.2 Mean Hsp 70 Levels (-2SEM to +2SEM) in Patients Graded by the BILAG Index p=0.004 GEN SKIN CNS MUSK CVRP VASC RENAL HAEM n 25 78 74 27 74 28 48 55 81 20 63 38 49 53 48 51 Results Legend to Figure 6.2.1 and Figure 6.2.2. Bar charts showing mean levels (-2SEM to +2SEM) of Hsps in patients with SLE and graded by the BILAG index. Blue bars represent patients with inactive disease and grey bars represent patients with active disease, in individual BILAG organ/system categories. The P value (p=0.038) denotes the value for the statistical significance of the difference in mean Hsp 70 levels, between patients graded by active and inactive musculoskeletal disease, as determined by a student’s t-test (described in Chapter 2) for equality of means. Key: n number of patients GEN general non-specific disease SKIN mucocutaneous disease CNS central nervous system disease MUSK musculoskeletal disease CVRP cardiovascular and respiratory disease VASC vasculitis RENAL renal disease HAEM haematological disease 244 Results Associations of Hsp 90 with Disease Activity in SLE in Patients who Overexpress Hsp 90 In view of the heterogeneity of factors likely to be involved in the pathogenesis of SLE (discussed in Chapter 1, Section 1.3), associations of Hsp 90 with clinical disease activity in SLE, may be most appropriately assessed in the subset of patients with elevated levels of Hsp 90. Thus, levels of Hsp 90 (in patients who were determined in Chapter 4 to overexpress Hsp 90) were compared in patients graded by the BILAG index. As shown in Figure 6.2.3, mean levels of Hsp 90 (-2SEM to +2SEM) were higher (p=0.090) in patients graded by active (BILAG A/B/C) compared with inactive (BILAG D/E) general disease features (BILAG GEN), as determined by a students t-test for equality of means. No significant differences in mean levels of Hsp 90 (-2SEM to 4-2 SEM) were observed between other disease grades. 245 Figure 6.2.3 Mean Hsp 90 Levels (-2SEM to +2SEM) in Patients who Overexpress Hsp 90 and Graded by the BILAG Index I I A/B/C I Z H d / E GEN SKIN CNS MUSK CVRP VASC RENAL HAEM n 5 22 19 8 23 4 11 16 23 4 19 8 16 11 19 Results Legend to Figure 6.2.3. Bar chart showing mean levels (-2SEM to +2SEM) of Hsp 90 in patients with SLE who overexpress Hsp 90 and graded by the BILAG index. Blue bars represent patients with inactive disease and grey bars represent patients with active disease, in individual BILAG organ/system categories. The P value (p=0.090) denotes the value for the statistical significance of the difference in mean Hsp 90 levels, between patients graded by active and inactive general non-specific disease, as determined by a student’s t-test (described in Chapter 2) for equality of means. Key: n number of patients GEN general non-specific disease SKIN mucocutaneous disease CNS central nervous system disease MUSK musculoskeletal disease CVRP cardiovascular and respiratory disease VASC vasculitis RENAL renal disease HAEM haematological disease 247 Results 6.3 Associations of HLA with Active Clinical Subsets in SLE Previous studies (Dhillon et al., 1994) and the results presented in Chapter 4, suggest a genetic association with Hsp 90 overexpression in SLE. Thus it is conceivable, that any association between levels of Hsp 90 and active clinical subsets in SLE, may be due to factors genetically linked with Hsp 90 overexpression, but with roles in this disease independent of Hsp levels. Table 6.3.1 shows the number of patients graded by the BILAG disease activity index, with respect to the presence or absence of individual HLA alleles/haplotypes. Significant associations of individual HLA alleles/haplotypes, as determined by chi-square analyses, are detailed in Table 6.3.2. The absence of HLA A l, B8, was associated with BILAG A/B/C VASC (p=0.014). The absence of HLA B8, DR3 was associated with BILAG A/B/C SKIN (p=0.018). No significant associations between HLA status and active (BILAG A/B/C) or inactive (BILAG D/E) general disease features (BILAG GEN) were observed. Reports concerning correlations of individual HLA alleles/haplotypes with active clinical subsets of patients with SLE, have been conflicting. Indeed, the associations detailed above do not correspond to previous studies. Importantly, as discussed in Chapter I, Section 1.3, such inconsistencies may derive from a complex involvement of ethnic background in patients with SLE, to any correlations involving individual HLA alleles/haplotypes with SLE. Interestingly however, in both a previous study (Dhillon et al., 1994) and in Chapter 4, the HLA DR3 negative haplotype has been associated with overexpression of Hsp 90 in patients with SLE. Moreover, Hsp 90 overexpression has been associated with active cardio-vascular/- respiratory disease, in patients with SLE (Dhillon et al., 1993a). 248 Results Table 6.3.1 Associations of HLA with Active Clinical Subsets in SLE BILAG Category Allele/Haplotype GEN D/E Am/C Al present 7 26 Al absent 22 48 NS B8 present 5 15 B8 absent 25 60 NS DR3 present 10 18 DR3 absent 19 58 NS A1/B8 present 3 8 A1/B8 absent 28 70 NS B8/DR3 present 3 6 B8/DR3 absent 28 73 NS A1/DR3 present 5 11 A1/DR3 absent 26 68 NS SKIN D/E A/B/C Al present 25 8 Al absent 43 27 NS B8 present 15 5 B8 absent 54 31 NS DR3 present 21 7 DR3 absent 48 29 NS A1/B8 present 10 1 A1/B8 absent 60 38 NS B8/DR3 present 9 0 B8/DR3 absent 61 40 p=0.0I8 A1/DR3 present 13 3 A1/DR3 absent 57 37 NS Results Table 6.3.1 (continued) BILAG Category Allele/Haplotype CNS D/E A/8/C A1 present 31 2 A1 absent 55 15 NS 88 present 17 3 88 absent 71 14 NS DR3 present 24 4 DR3 absent 64 13 NS A l/88 present 9 2 A l/88 absent 83 15 NS 88/DR3 present 8 1 88/DR3 absent 85 16 NS A1/DR3 present 15 1 A1/DR3 absent 78 16 NS MUSK D/E A/8/C A1 present 10 23 A1 absent 37 33 NS 88 present 6 14 88 absent 43 42 NS DR3 present 10 18 DR3 absent 38 39 NS A l/88 present 3 8 A l/88 absent 48 50 NS 88/DR3 present 3 6 88/DR3 absent 49 52 NS A1/DR3 present 6 10 A1/DR3 absent 46 48 NS Results Table 6.3.1 (continued) BILAG Category Allele/Haplotype CVRP D/E A/B/C Al present 24 Al absent 62 NS 38 present 14 6 38 absent 74 11 NS DR3 present 20 8 DR3 absent 68 9 NS (p=0.038) Al/38 present 8 3 A 1/38 absent 84 14 NS 38/DR3 present 7 2 38/DR3 absent 86 15 NS A1/DR3 present 12 4 A1/DR3 absent 81 13 NS VASC D/E A/3/C A1 present 24 9 A1 absent 35 35 NS 38 present 14 6 38 absent 47 38 NS DR3 present 18 10 DR3 absent 43 34 NS Al/38 present 10 1 A l/38 absent 55 43 p=0.014 38/DR3 present 8 1 38/DR3 absent 58 43 NS A1/DR3 present 12 4 A1/DR3 absent 54 40 NS Results Table 6.3.1 (continued) BILAG Category Allele/Haplotype RENAL D/E A/B/C A1 present 22 11 A1 absent 50 20 NS B8 present 14 6 B8 absent 60 25 NS DR3 present 23 5 DR3 absent 52 25 NS A1/B8 present 9 2 A1/B8 absent 68 30 NS B8/DR3 present 8 1 B8/DR3 absent 70 31 NS A1/DR3 present 14 2 A1/DR3 absent 64 30 NS HAEM D/E A/B/C A1 present 12 21 A1 absent 27 43 NS B8 present 6 14 B8 absent 32 53 NS DR3 present 7 21 DR3 absent 32 45 NS A1/B8 present 5 6 A1/B8 absent 35 63 NS B8/DR3 present 3 6 B8/DR3 absent 38 63 NS A1/DR3 present 5 11 A1/DR3 absent 36 58 NS Results Table 6.3.1 (continued) Key: NS not significant n number of patients GEN general non-specific disease SKIN mucocutaneous disease CNS central nervous system disease MUSK musculoskeletal disease CVRP cardiovascular and respiratory disease VASC vasculitis RENAL renal disease HAEM haematological disease Results Table 6.3.2 Chi-Square Tests for the Correlation of HLA AUeles/Haplotypes With Active Clinical Subsets in SLE BILAG Category: SKIN AUele/Haplotype Total D/E A/8/C 88, DR3 present 9 0 9 88, DR3 absent 61 40 101 Total 70 40 Chi-Square Significance Pearson 0.018 Likelihood Ratio 0.003 Linear by Linear Association 0.018 Fisher’s Exact Test 0.014 One-tail 0.025 Two-tail BILAG Category: VASC Allele/Haplotype Total D/E A/8/C A 1,88 present 10 1 1 Al, 88 absent 55 43 97 Total 65 44 Chi-Square Significance Pearson 0.014 Likelihood Ratio 0.005 Linear by Linear Association 0.013 Fisher’s Exact Test 0.011 One-tail 0.015 Two-tail Key: D/E inactive disease SKIN mucocutaneous disease A/8/C active disease VASC vasculitis Results 6.4 Cytokines As discussed in Chapter 1, Section 1.3, various studies have strongly supported a pathogenic role for the cytokines IL-6 and IL-10 in patients with SLE. Thus, using the BILAG index, an investigation of the associations of both cytokines with disease activity in patients with SLE, is outlined below. Mean levels of 11^6 (- 2SEM to + 2SEM) in patients graded by the BILAG index are shown in Figure 6.4.1. Higher mean IL-6 levels were found in patients graded by BILAG A/B/C HAEM, compared with those patients graded by BILAG D/E HAEM (p=0.037), as determined by a student’s t-test for equality of means. Associations of IL-6 levels in patients with SLE with other BILAG disease grades (Figure 6.4.1) or with BILAG GLOBAL disease activity scores (see Table 6.4.1), were not found to be significant. Mean levels of IL-10 (- 2SEM to + 2SEM) in patients graded by the BILAG disease activity scoring system are shown in Figure 6.4.2. Significantly higher mean IL-10 levels (p=0.020) were found in patients graded by BILAG D/E GEN, compared with those patients graded by BILAG A/B/C GEN, as determined by a student’s t-test for equality of means. Associations of IL-10 levels in patients with SLE with other BILAG disease grades, were not observed to be significant. Furthermore, as shown in Table 6.4.1, levels of IL-10 in patients with SLE were not found to correlate with BILAG GLOBAL disease activity scores. 255 Results Table 6.4.1 Correlations of IL-6 and IL-10 with Global BILAG Disease Activity Scores in SLE IL-6 n Test AU SLE 113 0.125 0.187 Spearman NS 113 0.084 0.204 Kendall NS Patients with Elevated 27 0.468 0.014 Spearman Levels of Hsp 90 27 0.370 0.009 Kendall IL-10 n Test AU SLE 105 0.120 0.223 Spearman NS 105 0.079 0.252 Kendall NS Patients with Elevated 22 -0.308 0.163 Spearman NS Levels of Hsp 90 22 -0.218 0.170 Kendall NS KEY: n number of patients r correlation coefficient p significance NS not significant Figure 6.4.1 Mean IL-6 Levels (-2SEM to +2SEM) in Patients Graded by the BILAG Index p=0.037 IIA/B/C to GEN SKIN CNS MUSK CVRP VASC RENAL HAEM n 31 79 70 40 93 17 52 58 93 17 66 44 78 32 41 69 Figure 6.4.2 Mean IL-10 Levels (-2SEM to +2SEM) in Patients Graded by the BILAG Index I I A/B/C CD D/E o 0.5 GEN SKIN CNS MUSK CVRP VASC RENAL HAEM n 28 75 64 39 87 16 48 55 87 16 62 41 73 30 38 65 Results Legend to Figure 6.4.1 and Figure 6.4.2. Bar chart showing mean levels (-2SEM to +2SEM) of cytokines in patients with SLE and graded by the BILAG index. Blue bars represent patients with inactive disease and grey bars represent patients with active disease, in individual BILAG organ/system categories. The P value (p=0.037) denotes the value for the statistical significance of the difference in mean IL-6 levels, between patients graded by active and inactive haematological disease, as determined by a student’s t-test (described in Chapter 2) for equality of means. Key: n number of patients GEN general non-specific disease SKIN mucocutaneous disease CNS central nervous system disease MUSK musculoskeletal disease CVRP cardiovascular and respiratory disease VASC vasculitis RENAL renal disease HAEM haematological disease 259 Results Associations of IL-6 with Disease Activity in SLE in Patients who Overexpress Hsp 90 Any role for IL-6 in the pathogenesis of SLE must involve the activation of specific genes in target cells. Studies in Chapter 4 have suggested that a role for IL-6 in Hsp 90 overexpression in patients with SLE, has a genetic involvement. Thus, it is conceivable that the regulation of other genes by IL-6 in SLE, may have a similar genetic involvement. In patients who were determined in Chapter 4 to overexpress Hsp 90, a comparison of IL-6 levels in these patients with BILAG GLOBAL disease activity scores (Table 6.4.1), showed significant correlations as determined by both Spearman (r=0.468, p=0.014) and Kendall (r=0.370, p=0.009) analyses. Mean levels of IL-6 (- 2SEM to + 2SEM) in patients who were determined in Chapter 4 to overexpress Hsp 90, were graded by the BILAG index as shown in Figure 6.4.3. Associations of IL-6 levels in these patients with elevated Hsp 90, who were graded by cardio-vascularZ-respiratory disease, showed significantly higher mean IL-6 levels in BILAG A/B/C patients compared with BILAG D/E patients (p=0.024). Interestingly, in a previous study of Hsp 90 associations with disease activity in SLE, overexpression of Hsp 90 in patients with this disease, was associated with active cardio-vascular/-respiratory disease (Dhillon et al., 1993a). A comparison of IL-6 levels in patients who overexpress Hsp 90 and who were graded by haematological disease activity, showed higher mean levels of IL-6 in BILAG A/B/C patients compared with BILAG D/E patients. However, an appreciably lower significance was observed in this analysis (p=0.860, see Figure 6.4.3) when compared to an analysis in patients without selection for Hsp 90 overexpression (p=0.037, see Figure 6.4.1). As shown in Figure 6.4.3, other disease associations with IL-6 levels in patients with SLE who overexpress Hsp 90, did not attain statistical significance. 260 Figure 6.43 Mean IL-6 Levels (-2SEM to +2SEM) in Patients who Overexpress Hsp 90 and Graded by the BILAG Index p = 0 .0 2 4 I I A/B/C GEN SKIN CNS MUSK CVRP VASC RENAL HAEM n 5 22 19 8 23 4 11 16 23 4 16 11 19 8 13 14 Results Legend to Figure 6.4.3. Bar chart showing mean levels (-2SEM to +2SEM) of IL-6 in patients with SLE who overexpress Hsp 90 and graded by the BILAG index. Blue bars represent patients with inactive disease and grey bars represent patients with active disease, in individual BILAG organ/system categories. The P value denotes the value for the statistical significance of the difference in mean IL-6 levels, between patients graded by active and inactive cardiovascular and respiratory disease (p=0.024) or haematological disease (p=0.860), as determined by a student’s t-test (described in Chapter 2) for equality of means. Key: n number of patients GEN general non-specific disease SKIN mucocutaneous disease CNS central nervous system disease MUSK musculoskeletal disease CVRP cardiovascular and respiratory disease VASC vasculitis RENAL renal disease HAEM haematological disease 262 Results 6.5 Antibodies to Heat Shock Proteins As discussed in Chapter 1, Section 1.2, Hsp-specific responses have been observed in a number of autoimmune diseases including SLE and may have a role in the pathogenesis of these diseases. In Chapter 5, a number of patients with SLE were determined to have raised levels of antibodies to Hsps (particularly Hsp 90). Table 6.5.1 shows the number of patients with SLE, graded by the BILAG index, with either normal or raised (defined in Chapter 5) levels of antibodies to Hsp 90. As shown in Table 6.5.2, raised levels of IgG antibodies to Hsp 90 were found to be significantly associated with BILAG A/B/C GEN patients (p=0.019), as determined by a chi-square analysis. Associations of IgG antibodies with other disease activity grades, were not found to be significant. Table 6.5.1 also shows the number of patients with SLE, graded by BILAG disease activity scores, with either normal or raised (defined in Chapter 5) levels of IgM antibodies to Hsp 90. Associations of IgM antibodies to Hsp 90 with other disease activity grades, were not found to be significant. Table 6.5.3 shows the number of patients with SLE, in each BILAG disease activity grade, with either normal or raised (defined in Chapter 5) levels of antibodies to Hsp 70. Analysis of individual organ/system involvement using the BILAG index, showed no significant disease associations with either levels of IgG or IgM antibodies to Hsp 70 in these patients, as determined by chi-square analyses. 263 Results Table 6.5.1 Number of Patients with SLE with Normal or Raised Levels of Antibodies to Hsp 90 in Inactive or Active BILAG Categories IgG Antibodies to Hsp 90 BILAG Antibody Level Inactive Disease Active Disease Category (Score A/B/C) (Score D/E) GEN Normal 23 35 Raised 7 37 P=0.01 SKIN Normal 41 17 Raised 23 21 NS CNS Normal 48 10 Raised 37 7 NS MUSK Normal 28 30 Raised 22 22 NS CV/RP Normal 51 7 Raised 35 9 NS VASC Normal 33 25 Raised 29 15 NS RENAL Normal 41 17 Raised 32 12 NS HAEM Normal 23 35 Raised 16 28 NS Results (Table 6.5.1 continued) IgM Antibodies to Hsp 90 BILAG Antibody Level Inactive Disease Active Disease Category (Score A/B/C) (Score D/E) GEN Normal 20 41 Raised 6 24 NS SKIN Normal 41 20 Raised 20 10 NS CNS Normal 52 9 Raised 25 5 NS MUSK Normal 30 31 Raised 12 18 NS CV/RP Normal 51 10 Raised 26 4 NS VASC Normal 39 22 Raised 19 11 NS RENAL Normal 39 22 Raised 21 9 NS HAEM Normal 22 39 Raised 15 15 NS Results (Table 6.5.1 continued) Key: NS not significant n number of patients GEN general non-specific disease SKIN mucocutaneous disease CNS central nervous system disease MUSK musculoskeletal disease CVRP cardiovascular and respiratory disease VASC vasculitis RENAL renal disease HAEM haematological disease Results Table 6.5.2 Chi-Square Test for the Correlation of Active BILAG General Disease with Raised IgG Antibodies to Hsp 90 in SLE Antibody Level Inactive Disease Active Disease Total Normal 23 35 58 Raised 7 37 44 Total 30 72 Chi-Square Significance Pearson 0.009 Likelihood Ratio 0.008 Linear by Linear Association 0.009 Fisher’s Exact Test 0.015 One-tail 0.008 Two-tail Results Table 6.5.3 Number of Patients with SLE with Normal or Raised Levels of Antibodies to Hsp 70 in Inactive or Active BILAG Categories IgG Antibodies to Hsp 70 BILAG Antibody Level Inactive Disease Active Disease Category (Score A/B/C) (Score D/E) GEN Normal 31 28 Raised 8 29 NS SKIN Normal 41 20 Raised 20 10 NS CNS Normal 52 Raised 25 NS MUSK Normal 30 31 Raised 12 18 NS CV/RP Normal 51 10 Raised 26 4 NS VASC Normal 39 22 Raised 19 11 NS RENAL Normal 39 22 Raised 21 9 NS HAEM Normal 22 39 Raised 15 15 NS Results (Table 6.5.3 continued) IgM Antibodies to Hsp 70 BILAG Antibody Level Inactive Disease Active Disease Category (Score A/B/C) (Score D/E) GEN Normal 20 41 Raised 6 24 NS SKIN Normal 41 20 Raised 20 10 NS CNS Normal 52 9 Raised 25 5 NS MUSK Normal 30 31 Raised 12 18 NS CV/RP Normal 51 10 Raised 26 4 NS VASC Normal 39 22 Raised 19 11 NS RENAL Normal 39 22 Raised 21 9 NS HAEM Normal 22 39 Raised 15 15 NS Results (Table 6.5.3 continued) Key: NS not significant n number of patients GEN general non-specific disease SKIN mucocutaneous disease CNS central nervous system disease MUSK musculoskeletal disease CVRP cardiovascular and respiratory disease VASC vasculitis RENAL renal disease HAEM haematological disease Results 6.6 Discussion A role for Hsp 90 in the pathogenesis of SLE has been discussed in Chapter 1, Section 1.4. These studies have suggested that overexpression of Hsp 90, which occurs in a subset of patients with SLE, may lead to the development of antibodies to Hsp 90 in these patients, with possible pathological consequences. In agreement with this hypothesis, elevated levels of Hsp 90 (Faulds et al., 1994) and antibodies to Hsp 90 (Faulds et al., 1995) have been reported in lupus-like MRL-lprllpr mice and are events which precede the onset of disease. An understanding of the regulation of Hsp 90 in SLE is clearly of interest. Previous studies (see Chapter 1, Section 1.3) have also suggested pathogenic roles for elevated levels of IL-6 and IL-10 in SLE. Of particular relevance to this dissertation, are studies which have supported a role for IL-6 (Stephanou et al., 1997), (Stephanou et al., 1998c) and IL-10 (Ripley et al., 1999), in the overexpression of Hsp 90 in SLE. Importantly, studies in Chapter 4 have shown that overexpression of Hsp 90 in SLE correlated with elevated levels of IL-6 and not IL-10, in patients with this disease. However, the relative contribution of cytokines, Hsps and antibodies to Hsps, to disease activity in SLE, has remained unclear. In this chapter, a comparison of these factors between active and inactive clinical subsets of SLE, using the BILAG index for the measurement of clinical disease activity, has been undertaken. Correlations between levels of Hsp 90 or Hsp 70 and disease activity in SLE were not found to be significant. These observations are not in agreement with the study of Dhillon et al., (1993a), which reported an association between elevated levels of Hsp 90 and active cardio-vascularZ-respiratory disease. However, in this chapter, associations of clinical subsets of SLE with the presence or absence of individual/combined HLA types, showed an association between the HLA DR3 negative haplotype and active cardio-vascularZ-respiratory disease. A common link between these two observations is that the HLA DR3 negative haplotype in SLE has been associated with Hsp 90 overexpression, both in the study of Dhillon et al., (1994) and in the results presented in Chapter 4. Thus, it is conceivable that the reported association of active cardio-vascularZ-respiratory disease with elevated levels 271 Results of Hsp 90 in the study of Dhillon et al., (1993a), may have been influenced by factors other than Hsp 90 itself. In this chapter, levels of IL-6 and IL-10 were compared both with active clinical subsets of and overall disease activity in, patients with SLE. An association between IL-6 and active haematological disease (BILAG A/B/C HAEM) was observed and was not paralleled by significant associations of IL-6 with other clinical profiles. Interestingly, in a previous study (Dhillon et al., 1993a), increasing levels of Hsp 90 in patients with SLE over time, were found to be associated with the development of haematological disease. Since the results in Chapter 4 support a role for 11^6 in the regulation of Hsp 90 expression in SLE, it is possible that the observation of Dhillon et al., (1993a) relates to an involvement of IL-6 in the development of haematological disease. Levels of IL-6 were not observed to correlate with GLOBAL BILAG disease activity scores. In a previous study, using the SLEDAI and SLAM indexes (defined in Chapterl, Section 1.3), a lack of correlation between levels of IL-6 and clinical profiles and between levels of IL-6 and overall disease activity, was reported (Grondai et al., 2000). However, in other studies, which measured levels of IL-6 both before and after disease exacerbations (Spronk et al., 1992), or at various stages of disease activity (Waszczykowska et al., 1999), levels of 11^6 correlated well with overall disease activity. It is possible therefore, that any relationship between IL-6 and disease activity in SLE may be most appropriately assessed by an assessment of changes in these variables in the same patients over time. The absence of a correlation between levels of IL-10 and active clinical subsets in SLE or overall disease activity scores (BILAG GLOBAL) is in agreement with the study of Grondai et al., (2000). However, serum IL-10 levels have been shown to correlate with overall disease activity, when serial measurements are taken in the same patients over time (Waszczykowska et al., 1999). Thus, as for IL-6, any relationship between IL-10 and disease activity, may be most appropriately assessed in patients following significant changes in either variable over time. Interestingly, studies in this chapter have shown a significant correlation between levels of IL-6 and overall disease activity scores (BILAG GLOBAL), in patients who overexpressed Hsp 90. One possible explanation for this, in agreement 272 Results with the studies in Chapter 4, is that Hsp 90 overexpression may be a marker for recent and significant elevation of 11^6 levels in these patients. Moreover, any relationship between IL-6 and disease activity, must be explainable in terms of IL-6 activation of its target genes. Thus, since a correlation between IL-6 and Hsp 90 is enhanced in patients who overexpress Hsp 90, any involvement of IL-6 in disease activity may also be enhanced in this subset of patients. A comparison of mean levels of IL-6 in patients who overexpress Hsp 90 and graded by the BILAG index, showed significantly higher levels of IL^6 in patients with active compared with inactive cardio-vascularZ-respiratory disease. This observation is of particular significance since, as discussed earlier, previous studies (Dhillon et al., 1993a) and the results presented in this dissertation, suggest an association of active cardio-vascularZ-respiratory disease with factors associated with Hsp 90 overexpression andZor with HLA DR3 negativity in this disease. Thus, the observed associations in SLE of elevated levels of Hsp 90 (Dhillon et al., 1993) and of HLA DR3 negativity, with active cardio-vascularZ-respiratory disease, may have been most strongly influenced by elevated levels of IL-6 in a subset of patients with SLE. Associations of raised levels of antibodies to Hsp 90 with active clinical subsets of patients with SLE, were not found to be significant for analyses involving any one organZsystem. Interestingly however, a significant correlation between raised IgG antibodies to Hsp 90 and active general disease features (BILAG AZBZC GEN), was found. Previous studies have correlated general disease features with severity of illness (Krupp et al., 1990), (Wysenbeek et al., 1993) in patients with SLE. Moreover, in another study, patients with active general disease features were associated with severe tissue inflammation (Amit et al., 1999). Thus, a significant correlation between the presence of raised antibodies to Hsp 90 and active general disease features (BILAG AZBZC GEN) in patients with SLE, would suggest a link between these antibodies and active disease. Furthermore, such a relationship may be most relevant to the presence of severe inflammation, rather than the prevalence of organ/system involvement. 273 Results This observation is of particular interest, since antibodies to Hsps in autoimmune diseases, have been suggested to both promote and sustain autoaggression at sites of tissue inflammation (see Chapter 1, Section 1.2). Such a role, if relevant to antibodies to Hsp 90 in the pathogenesis of SLE, would therefore involve an enhancement of and occur secondary to, the presence of active disease in one or several organ/systems. A significant correlation between raised levels of IgG antibodies to Hsp 90 and active general disease features (BILAG A/B/C GEN), in the absence of significant correlations of raised levels of these antibodies with other active clinical subsets of SLE, is consistent with a role for these antibodies in sustaining and enhancing autoagression and tissue inflammation, in a non- organ/system-specific manner. Arguably, the induction of antibodies to Hsp 90, may occur as part of, but not be involved with, active disease. However, the complexity of aetiopathogenic factors in SLE (discussed in Chapter 1, Section 1.3), would suggest this to be unlikely. Taken together, these results suggest that any involvement of Hsp 90 in the pathogenesis of SLE, is most likely to be mediated by Hsp 90 overexpression and the development of antibodies to Hsp 90 in SLE, which leads to active general disease features which suggestively are a consequence of enhanced tissue inflammation. 274 Chapter 7 General Discussion This dissertation describes an investigation of the role of cytokines in the expression of heat shock proteins (Hsps) in systemic lupus erythematosus (SLE). Studies in all the previous results chapters presented in this dissertation, are discussed together in this chapter. Previous studies have established important cellular roles for Hsps under various stressful conditions and these appear to be an extension of their role in unstressed cells. Thus, some Hsps are expressed constitutively at high levels in cells and have roles in cell growth and development (Bond and Schlesinger, 1987), while other Hsps are present at much lower levels in cells and their synthesis is enhanced in response to stressful conditions (Morimoto, 1993). Although the mechanisms underlying inducible Hsp gene expression under stressful conditions have been intensively studied, the regulation of Hsps in unstressed cells has remained less well characterised. Such studies are of importance since Hsps have been observed to be differentially expressed following viral infection (La Thangue and Latchman, 1988), in different tumour cell lines (Ferrarini et al., 1992) and in specific disease states (Jameel et al., 1992). As outlined in Chapter 1, Section 1.4, previous studies (Norton et al., 1989), (Dhillon et al., 1993a) have reported the differential expression of Hsps in SLE and are reviewed in Stephanou et al., (1998c). Thus, levels of Hsp 90 are elevated in PBMCs of up to 30% of patients with SLE, compared to normal control subjects. Moreover, this elevation occurs in the absence of elevation of other Hsps, including Hsp 73 and Hsp 60 (Dhillon et al., 1993a) and ubiquitin (Twomey et al., 1992). Thus, these studies suggest that overexpression of Hsp 90 in SLE, occurs via distinct regulatory mechanisms and not as a general response to stress, which would result in elevated levels of a wide range of Hsps. 275 Discussion Recent studies have shown that the cytokine interleukin-6 (11^6) enhances the expression of Hsp 90 in different ceils in vitro (Stephanou et al., 1997) and that such elevation of Hsp 90 levels is also observed in vivo in transgenic mice overexpressing IL-6 (Stephanou et al., 1998a). Activation of the Hsp 90 promoter is mediated by the IL-6-induced activation of the transcription factors NF-IL6 (nuclear factor for IL-6) and STAT-3 (signal transducer and activator of transcription-3) and involves a short specific region of the promoter which contains binding sites for these factors (Stephanou et al., 1997), (Stephanou et al., 1998b). Thus, in view of the known role of STAT proteins in signalling pathways mediated by the cytokine interleukin-10 (IL- 10), an investigation of the effect of IL-10 on Hsp 90 gene expression has been presented in Chapter 3. These studies have demonstrated that IL-10 can also enhance Hsp 90 levels in different cells in vitro and that this effect is mediated via the same short specific region of the Hsp 90 promoter which is targetted by IL-6 (Ripley et al., 1999). Moreover, since activation of the Hsp 90a and -p gene promoters was enhanced by cotransfection of a STAT-3 expression vector, it is likely that the observed effect of IL-10 on Hsp 90 gene expression, is mediated via the IL-lO-induced activation of STAT-3. Hence, the effect of IL-10 on Hsp 90 levels in vitro is comparable to that of IL-6 and is likely to reflect the similarity of IL-10 signalling mechanisms to that of IL-6, through the activation of STAT-3 (Lai et al., 1996). Furthermore, since murine studies have demonstrated that elevation of Hsp 90 levels by IL-6 can occur in the absence of NF-IL-6 (Stephanou et al., 1998a), which is activated by IL-6 and not IL- 10 (Finbloom and Winestock, 1995), these observations suggest a shared pathway for IL-10 and IL-6 in stimulation of Hsp 90 in vivo. Important cellular roles for Hsps in the immune system have been detailed in Chapter 1, Section 1.2 and these appear to be an extension of their known roles as molecular chaperones. Hence, any role for cytokines in the regulation of Hsp gene expression in vivo, may reflect the importance of Hsp function in such processes as antigen presentation and the production of immunoglobulins, during an immune response. Moreover, since IL-6 (Linker-Israeli et al., 1991) and IL-10 (Park et al., 1998) have been reported to be elevated in patients with SLE, these observations may 276 Discussion have important implications for the mechanisms of Hsp 90 overexpression in this disease. Studies in Chapter 4 of this dissertation, have shown a differential pattern of expression of Hsps in patients with SLE. Thus, in agreement with a previous study (Dhillon et al., 1993a), Hsp 90 was observed to be overexpressed in PBMCs in a subset of patients with SLE and not in patients with rheumatoid arthritis, compared to control normal individuals. This elevation was specific to Hsp 90 and was not paralleled by elevation of Hsp 70 in the same patients with SLE. An investigation of the relative effect of the cytokines IL-6 and IL-10, on the overexpression of Hsp 90 in SLE, has also been outlined in Chapter 4. Levels of serum IL-6 were found to correlate with levels of Hsp 90 but not Hsp 70, in PBMCs of individual patients with SLE. In contrast, levels of serum IL-10 were not found to correlate with levels of Hsp 90 nor with levels of Hsp 70. Thus, these results suggest a role for IL 6 in the overexpression of Hsp 90 in patients with SLE and is in agreement with a previous study in transgenic mice, where elevated levels of IL 6 resulted in enhanced Hsp 90 protein levels (Stephanou et al., 1998a). Levels of IL-6 were observed to be significantly higher in patients who overexpressed Hsp 90 compared to those patients with normal levels. Interestingly, this elevation of IL-6 was observed to be more significant than the observed elevation of Hsp 90 in patients with elevated levels of IL-6 compared to those with normal levels of IL-6 (see Figure 4.4.5 and Figure 4.4.6). Moreover, the observed correlation between IL 6 and Hsp 90 was stronger in patients who overexpressed Hsp 90 compared to an analysis in which all patients with SLE were studied (see Figure 4.4.1 and Figure 4.4.7). Hence, these observations suggest that an effect of IL-6 on Hsp 90 elevation in SLE, is strongest in a particular subset of patients with this disease. This may explain why elevation of Hsp 90 occurs in SLE and is not observed in other diseases such as rheumatoid arthritis (Dhillon et al., 1993a), which also features elevated levels of IL 6 (Stuart et al., 1995). Studies presented in Chapter 4 have shown that overexpression of Hsp 90 in patients with SLE, was most likely to occur in patients who had an HLA DR3- negative haplotype and this observation is in agreement with the study of (Dhillon et al., 1994). As has been discussed in Chapter 4, the gene encoding Hsp 90p has been 277 Discussion located on human chromosome 6 in the MHC region. Hence, it is conceivable that polymorphisms exist within the Hsp 90p promoter, which mediate the overexpression of Hsp 90 in response to elevated levels of IL-6, in a genetic subset of patients with SLE. However, the true significance of the association between patients lacking the HLA DR3 allele and Hsp 90 overexpression in SLE, requires the study of a greater number of patients in various ethnic groups, than those used in either this dissertation or in the study of Dhillon et al., (1994). This would permit an analysis of any effect of population ethnic bias/HLA haplotype frequencies on this reported association, which was not attainable in either study. Raised levels of antibodies to Hsp 90 have been reported in patients with SLE of both adult (Conroy et al., 1994) and childhood-onset (Conroy et al., 1996). However, the mechanisms leading to this event have remained unclear. Previous studies have reported an association of the HLA DR3-negative haplotype with the presence of antibodies to Hsp 90 in patients with SLE (Conroy et al., 1994). As discussed earlier, the HLA DR3-negative haplotype in patients with SLE, has been associated with Hsp 90 overexpression both in a previous study (Dhillon et al., 1994) and in Chapter 4. Thus, these observations have led to a hypothesis that overexpression of Hsp 90 in SLE, results in the development of antibodies to Hsp 90 in the same patients (Stephanou et al., 1998c). Studies in Chapter 5 have shown an association between overexpression of Hsp 90 and the presence of serum IgG antibodies to Hsp 90, in the same patients. Since levels of IL-6 were not observed to correlate with levels of IgG or IgM antibodies to Hsp 90, these observations suggest that the development of antibodies to Hsp 90 in SLE, occur as a result of Hsp 90 overexpression and not as a general B cell stimulatory effect mediated directly by IL^. However, a more precise time- course experiment, to explore the link between IL=6, Hsp 90 and antibodies to Hsp 90, has not been undertaken. Such a study would take into account other factors, such as the serum half-life of human IgG antibody molecules being approximately 14 days (Stites et al., 1997). As discussed in Chapter 1, Section 1.2, the development of autoantibodies within the immune system may have pathological consequences. Thus, in the normal immune system, such events are appropriately down-regulated and short-lived. Any 278 Discussion mechanism responsible for the development of antibodies to Hsp 90, is likely to require both the altered accessibility of Hsp 90 to cells of the immune system and other factors which sustain antibody production. Previous studies have reported the increased surface expression of Hsp 90 in PBMCs of patients with SLE (Erkeller- Yuksel et al., 1992). Moreover, elevated levels of IL-10, which feature in patients with SLE (Park et al., 1998), have been suggested to promote B cell hyperactivity and the increased production of antibodies in this disease (Rousset et al., 1992), (Kirou and Crow, 1999). Taken together, these observations suggest that overexpression of Hsp 90 and its surface expression (Erkeller-Yuksel et al., 1992), results in the production of antibodies to Hsp 90 in patients with SLE. Previous studies have suggested a pathogenic role for Hsp 90 in SLE. For example, elevated levels of Hsp 90 have been associated with overall active disease in patients with SLE (Dhillon et al., 1993a), using the UCH/Middlesex SLE disease activity index (Morrow et al., 1982; 1983). Furthermore, in the same study, using the BILAG disease activity index (defined in Chapter 2 and Appendix A), levels of Hsp 90 were observed to be higher in patients with active compared with inactive, cardio vascular/ -respiratory disease. Murine studies also support a role for Hsp 90 in the pathogenesis of SLE. Thus, in autoimmune y[RLJlpr-lpr mice, which develops a lupus-like disease (see Chapter 1, Section 1.3), elevation of Hsp 90 protein levels and the development of antibodies to Hsp 90, are events which precede the onset of disease (Faulds et al., 1994). A previously held hypothesis for the role of Hsp 90 in SLE, has been a process in which elevation of Hsp 90 in PBMCs results in the development of antibodies to Hsp 90, with pathological consequences for patients with this disease. In support of this hypothesis, previous studies (outlined in Chapter 1, Section 1.2) have correlated Hsp-specific antibody responses with active tissue disease. However, in patients with SLE, the relative involvement of Hsp 90 overexpression and antibodies to Hsp 90 in the pathogenesis this disease, has remained unclear. Thus, levels of these factors in the same patients, have been compared between active and inactive clinical subsets of SLE using the BILAG index. It should be noted however, that some statistical analyses of disease associations, involved small patient numbers in 279 Discussion individual clinical subsets. This is particularly evident in comparisons of mean Hsp90 (see Figure 6.2.3), IL-6 (see Figure 6.4.3) levels, in patients who overexpress Hsp 90 and graded by the BILAG index. Hence, care should be taken against over- interpretting the significance of disease associations presented in this dissertation, where statistical analyses involved small patient numbers. Studies presented in Chapter 6, have shown a correlation between levels of Hsp 90 and the presence of active general disease features of SLE. Furthermore, an enhanced correlation is observed in a comparison of raised levels of IgG antibodies to Hsp 90 and active general disease features. Associations of other clinical subsets of SLE using the BILAG index, with either levels of Hsp 90 or raised levels of antibodies to Hsp 90, did not prove to be statistically significant. Moreover, since elevated levels of IL-6 were not associated with active general disease features in patients with SLE, these observations support a role for Hsp 90 in the pathogenesis of SLE, which is mediated most strongly by the development of antibodies to Hsp 90. As discussed in Chapter 1, Section 1.2, antibodies to Hsps have been localised in regions of active tissue disease and their presence has correlated with the degree of overexpression of Hsps at these sites. These observations underline a current theory of the pathogenic involvement of Hsp-specific antibody responses in autoimmune disease. This is a process in which the development of antibodies to Hsps, in patients with active autoimmune pathology, results in binding of antibody to Hsps localised at regions of active tissue disease, leading to enhanced inflammation. Such a process is therefore non-organ-specific, but rather, specific to the presence of active tissue disease. Importantly therefore, the observation in Chapter 6 of an association between antibodies to Hsp 90 and active general disease features of SLE, in the absence of a correlation with organ-specific disease, is in agreement with this hypothesis. Moreover, this conclusion is strengthened by the reported association between active general disease features and enhanced tissue inflammation (Amit et al., 1999). A hypothetical model of the pathogenic role of Hsp 90 in SLE is provided in Figure 7.1. 280 Discussion Concluding Comments The studies presented in this dissertation have provided evidence of a role for the cytokine IL=6 in the overexpression of Hsp 90 in a subset of patients with SLE. Furthermore, a link between the overexpression of Hsp 90 and the development of antibodies to Hsp 90 in the same patients, leading to active disease, has also been established. 281 Figure 7.1 A Pathogenic Role for Hsp 90 in SLE Susceptibility Genes Overexpression and Surface Immune Response to Exogenous üocaiisalion of Hsp 90 Hsps Hsp 90 Autoantibodies Tissue Pathology and Overexpression of Hsps Enhanced Inflammalion Figure 7.1 illustrates a hypothetical model for a pathogenic role of Hsp 90 in patients with SLE. References References Abravaya, K., Phillips, B., and Morimoto, R. I. (1991). Attenuation of the heat shock response in HeLa cells is mediated by the release of bound heat shock transcription factor and is modulated by changes in growth and in heat shock temperatures. Genes Dev 5,2117-27. Abravaya, K., Phillips, B., and Morimoto, R. I. (1991). Heat shock-induced interactions of heat shock transcription factor and the human hsp70 promoter examined by in vivo footprinting. Mol Cell Biol 77, 586-92. Agius, M. A., Kirvan, C. A., Schafer, A. L., Gudipati, E., and Zhu, S. (1999). High prevalence of anti-alpha-crystallin antibodies in multiple sclerosis: correlation with severity and activity of disease. Acta Neurol Scand 100, 139-47. al-Janadi, M., al-Balla, S., al-Dalaan, A., and Raziuddin, S. (1993). Cytokine profile in systemic lupus erythematosus, rheumatoid arthritis, and other rheumatic diseases. J Clin Immunol 75, 58-67. Alcocer-Varela, J., and Alarcon-Segovia, D. (1982). Decreased production of and response to interleukin-2 by cultured lymphocytes from patients with systemic lupus erythematosus. J Clin Invest 69,1388-92. Alcocer-Varela, J., Laffon, A., and Alarcon-Segovia, D. (1984). Defective monocyte production of, and T lymphocyte response to, interleukin-1 in the peripheral blood of patients with systemic lupus erythematosus. Clin Exp Immunol 55, 125-32. Altman, D. G. (1999). Practical statistics for medical research, 9th Edition (New York: CRC Press). Amit, M., Molad, Y., Levy, O., and Wysenbeek, A. J. (1999). Headache in systemic lupus erythematosus and its relation to other disease manifestations. Clin Exp Rheumatol 77,467-70. 284 References Andus, T., Geiger, T., Hirano, T., Northoff, H., Ganter, U., Bauer, J., Kishimoto, T., and Heinrich, P. C. (1987). Recombinant human B cell stimulatory factor 2 (BSF- 2/IFN-beta 2) regulates beta-flbrinogen and albumin mRNA levels in Fao-9 cells. FEBS Lett 227, 18-22. Ansar Ahmed, S., Penhale, W. J., and Talal, N. (1985). Sex hormones, immune responses, and autoimmune diseases. Mechanisms of sex hormone action. Am J Pathol 727,531-51. Anzola, J., Luft, B. J., Gorgone, G., Dattwyler, R. J., Soderberg, C., Lahesmaa, R., and Peltz, G. (1992). Borrelia burgdorferi HSP70 homolog: characterization of an immunoreactive stress protein. Infect Immun 60,3704-13. Appetecchia, M., Castelli, M., and Delpino, A. (1997). Anti-heat shock proteins autoantibodies in autoimmune thyroiditis. Preliminary study. J Exp Clin Cancer Res 76,395-400. Aquino, D. A., Capello, E., Weisstein, J., Sanders, V., Lopez, C., Tourtellotte, W. W., Brosnan, C. F., Raine, C. S., and Norton, W. T. (1997). Multiple sclerosis: altered expression of 70- and 27-kDa heat shock proteins in lesions and myelin. J Neuropathol Exp Neurol 56,664-72. Aragona, P., Magazzu, G., Macchia, G., Bartolone, S., Di Pasquale, G., Vitali, C., and Ferreri, G. (1999). Presence of antibodies against Helicobacter pylori and its heat-shock protein 60 in the serum of patients with Sjogren's syndrome. J Rheumatol 26, 1306-11. Argaman, M., Aly, R., and Shapira, M. (1994). Expression of heat shock protein 83 in Leishmania is regulated post- transcriptionally. Mol Biochem Parasitol 64,95-110. Aringer, M., Wintersberger, W., Steiner, C. W., Kiener, H., Presterl, E., Jaeger, U., Smolen, J. S., and Graninger, W. B. (1994). High levels of bcl-2 protein in circulating 285 References T lymphocytes, but not B lymphocytes, of patients with systemic lupus erythematosus. Arthritis Rheum 37, 1423-30. Arnett, F. C. (1997). in Dubois' Lupus Erythematosus, D. Wallace and B. Hahn, eds. (Baltimore: Williams and Wilkins). Arrigo, A. P., Fakan, S., and Tissieres, A. (1980). Localization of the heat shock- induced proteins in Drosophila melanogaster tissue culture cells. Dev Biol 78, 86- 103. Arrigo, A. P., Suhan, J. P., and Welch, W. J. (1988). Dynamic changes in the structure and intracellular locale of the mammalian low-molecular-weight heat shock protein. Mol Cell Biol 8 ,5059-71. Arrigo, A. P., and Welch, W. J. (1987). Characterization and purification of the small 28,000-dalton mammalian heat shock protein. J Biol Chem 262, 15359-69. Asherson, R. A., Mackworth-Young, C. G., Boey, M. L., Hull, R. G., Saunders, A., Gharavi, A. E., and Hughes, G. R. (1983). Pulmonary hypertension in systemic lupus erythematosus. Br Med J (Clin Res Ed) 287, 1024-5. Bansal, G. S., Norton, P. M., and Latchman, D. S. (1991). The 90-kDa heat shock protein protects mammalian cells from thermal stress but not from viral infection. Exp Cell Res 195,303-6. Bamier, J. V., Bensaude, O., Morange, M., and Babinet, C. (1987). Mouse 89 kD heat shock protein. Two polypeptides with distinct developmental regulation. Exp Cell Res 170,186-94. Batchelor, J. R., Fielder, A. H., Walport, M. J., David, J., Lord, D. K., Davey, N., Dodi, I. A., Malasit, P., Wanachiwanawin, W., Bernstein, R., and et al. (1987). 286 References Family study of the major histocompatibility complex in HLA DR3 negative patients with systemic lupus erythematosus. Clin Exp Immunol 70,364-71. Beckmann, R. P., Mizzen, L. B., and Welch, W. J. (1990). Interaction of Hsp 70 with newly synthesized proteins: implications for protein folding and assembly. Science 248,850-4. Bethesda, Research, and Laboratories (1986). BRL pUC host: E.coli DH5aTM competent cells. Bethesda Research Laboratory Focus 8 ,9. Bimbaum, G. (1995). Stress proteins: their role in the normal central nervous system and in disease states, especially multiple sclerosis. Springer Semin Immunopathol 17, 107-18. Bimbaum, G., and Kotilinek, L, (1999). Immunity to heat shock proteins and neurological disorders of women. Infect Dis Obstet Gynecol 7,39-48. Bimbaum, G., Kotilinek, L., and Albrecht, L. (1993). Spinal fluid lymphocytes from a subgroup of multiple sclerosis patients respond to mycobacterial antigens [see comments]. Ann Neurol 34, 18-24. Bjorkman, P. J., and Davis, M. M. (1989). Model for the interaction of T-cell receptors with peptide/MHC complexes. Cold Spring Harb Symp Quant Biol 54,365- 73. Blaese, R. M., Grayson, J., and Steinberg, A. D. (1980). Increased immunoglobulin- secreting cells in the blood of patients with active systemic lupus erythematosus. Am J Med 69,345-50. Bogdan, C., Vodovotz, Y., and Nathan, C. (1991). Macrophage deactivation by interleukin 10. J Exp Med 174,1549-55. 287 References Bogdan, C., Vodovotz, Y., and Nathan, C. (1991). Macrophage deactivation by interleukin 10. J Exp Med 174,1549-55. Bond, U., and Schlesinger, M. J. (1987). Heat-shock proteins and development. Adv Genet 24, 1-29. Boog, C. J., de Graeff-Meeder, E. R., Lucassen, M. A., van der Zee, R., Voorhorst- Ogink, M. M., van Kooten, P. J., Geuze, H. J., and van Eden, W. (1992). Two monoclonal antibodies generated against human hsp60 show reactivity with synovial membranes of patients with juvenile chronic arthritis. J Exp Med 175, 1805-10. Borkovich, K. A., Farrelly, F. W., Finkelstein, D. B., Taulien, J., and Lindquist, S. (1989). hsp82 is an essential protein that is required in higher concentrations for growth of cells at higher temperatures. Mol Cell Biol 9,3919-30. Bradford, M. M. (1976). A rapid and sensitive method for the quantitation of microgram quantities of protein utilizing the principle of protein-dye binding. Anal Biochem 72, 248-54. Brodsky, J. L., and Schekman, R. (1993). A Sec63p-BiP complex from yeast is required for protein translocation in a reconstituted proteoliposome. J Cell Biol 123, 1355-63. Brugge, J. S. (1986). Interaction of the Rous sarcoma virus protein pp60src with the cellular proteins pp50 and pp90. Curr Top Microbiol Immunol 123,1-22. Buchner, J. (1999). Hsp90 & Co. - a holding for folding. Trends Biochem Sci 24, 136-41. Bukau, B., and Horwich, A. L. (1998). The Hsp70 and Hsp60 chaperone machines. Cell 92,351-66. 288 References Cabral, A. R., Cabiedes, J., Alarcon-Segovia, D., and Sanchez-Guerrero, J. (1992). Phospholipid specificity and requirement of beta 2-glycoprotein-I for reactivity of antibodies from patients with primary antiphospholipid syndrome. J Autoimmun 5, 787-801. Carlsten, H., Holmdahl, R., Tarkowski, A., and Nilsson, L. A. (1989). Oestradiol- and testosterone-mediated effects on the immune system in normal and autoimmune mice are genetically linked and inherited as dominant traits. Immunology 68,209-14. Castellino, F., Boucher, P. E., Eichelberg, K., Mayhew, M., Rothman, J. B., Houghton, A. N., and Germain, R. N. (2000). Receptor-mediated uptake of antigen/heat shock protein complexes results in major histocompatibility complex class I antigen presentation via two distinct processing pathways. J Exp Med 191, 1957-64. Celis, L., Vandevyver, C., Geusens, P., Dequeker, J., Raus, J., and Zhang, J. (1997). Clonal expansion of mycobacterial heat-shock protein-reactive T lymphocytes in the synovial fluid and blood of rheumatoid arthritis patients. Arthritis Rheum 40,510-9. Chandrasekhar, G. N., Tilly, K., Woolford, C., Hendrix, R., and Georgopoulos, C. (1986). Purification and properties of the groES morphogenetic protein of Escherichia coli. J Biol Chem 261, 12414-9. Cheng, J., Zhou, T., Liu, C., Shapiro, J. P., Brauer, M. J., Kiefer, M. C., Barr, P. J., and Mountz, J. D. (1994). Protection from Fas-mediated apoptosis by a soluble form of the Fas molecule. Science 263, 1759-62. Cheng, M. Y., Haiti, F. U., and Horwich, A. L. (1990). The mitochondrial chaperonin hsp60 is required for its own assembly. Nature 348,455-8. Churg, J., and Sobin J, H. (1982). Lupus nephritis: classification and atlas of glomerular diseases (Tokyo: Igaku-Shoin). 289 References Clark, J., Bourne, T., Salaman, M. R., Seifert, M. H., and Isenberg, D. A. (1996). B lymphocyte hyperactivity in families of patients with systemic lupus erythematosus. J Autoimmun 9, 59-65. Cohen, A. S., and Canoso, J. J. (1972). Criteria for the classification of systemic lupus erythematosus—status 1972. Arthritis Rheum 75,540-3. Cohen, I. R., and Young, D. B. (1991). Autoimmunity, microbial immunity and the immunological homunculus. Inmiunol Today 72, 105-10. Cohen, P. L., Rapoport, R., and Eisenberg, R. A. (1986). Characterization of functional T-cell lines derived from MRL mice. Clin Immunol Immunopathol 40, 485-96. Collier, N. C., and Schlesinger, M. J. (1986). The dynamic state of heat shock proteins in chicken embryo fibroblasts. J Cell Biol 103,1495-507. Conroy, S. E., Faulds, G. B., Williams, W., Latchman, D. S., and Isenberg, D. A. (1994). Detection of autoantibodies to the 90 kDa heat shock protein in systemic lupus erythematosus and other autoimmune diseases. Br J Rheumatol 33,923-6. Conroy, S. E., Gibson, S. L., Brunstrom, G., Isenberg, D., Luqmani, Y., and Latchman, D. S. (1995). Autoantibodies to 90 kD heat-shock protein in sera of breast cancer patients [letter]. Lancet 345, 126. Conroy, S. E., Tucker, L., Latchman, D. S., and Isenberg, D. A. (1996). Incidence of anti Hsp 90 and 70 antibodies in children with SLE, juvenile dermatomyositis and juvenile chronic arthritis. Clin Exp Rheumatol 14,99-104. 290 References Cotto, J. J., and Morimoto, R. I. (1999). Stress-induced activation of the heat-shock response: cell and molecular biology of heat-shock factors. Biochem Soc Symp 64, 105-18. Craig, E. A., and Gross, C. A. (1991). Is hsp70 the cellular thermometer? Trends Biochem Sci 16,135-40. Cristau, B., Schafer, P. H., and Pierce, S. K. (1994). Heat shock enhances antigen processing and accelerates the formation of compact class II alpha beta dimers. J Immunol 152, 1546-56. Cutforth, T., and Rubin, G. M. (1994). Mutations in Hsp83 and cdc37 impair signaling by the sevenless receptor tyrosine kinase in Drosophila. Cell 77,1027-36. Darnell, J. E., Jr. (1997). STATs and gene regulation. Science 277, 1630-5. de Benedetti, F., Massa, M., Robbioni, P., Ravelli, A., Burgio, G. R., and Martini, A. (1991). Correlation of serum interleukin-6 levels with joint involvement and thrombocytosis in systemic juvenile rheumatoid arthritis [see comments]. Arthritis Rheum 34, 1158-63. de Graeff-Meeder, E. R., Rijkers, G. T., Voorhorst-Ogink, M. M., Kuis, W., van der Zee, R., van Eden, W., and Zegers, B. J. (1993). Antibodies to human HSP60 in patients with juvenile chronic arthritis, diabetes mellitus, and cystic fibrosis. Pediatr Res 34,424-8. De Graeff-Meeder, E. R., van der Zee, R., Rijkers, G. T., Schuurman, H. J., Kuis, W., Bijlsma, J. W., Zegers, B. J., and van Eden, W. (1991). Recognition of human 60 kD heat shock protein by mononuclear cells from patients with juvenile chronic arthritis. Lancet 337,1368-72. 291 References de Graeff-Meeder, E. R., Voorhorst, M., van Eden, W., Schuurman, H. J., Huber, J., Barkley, D., Maini, R. N., Kuis, W., Rijkers, G. T., and Zegers, B. J. (1990). Antibodies to the mycobacterial 65-kd heat-shock protein are reactive with synovial tissue of adjuvant arthritic rats and patients with rheumatoid arthritis and osteoarthritis. Am J Pathol 757, 1013-7. de Waal Malefyt, R., Abrams, J., Bennett, B., Figdor, C. G., and de Vries, J. E. (1991). Interleukin lO(IL-lO) inhibits cytokine synthesis by human monocytes: an autoregulatory role of IL-10 produced by monocytes. J Exp Med 174, 1209-20. de Waal Malefyt, R., Haanen, J., Spits, H., Roncarolo, M. G., te Velde, A., Figdor, C., Johnson, K., Kastelein, R., Yssel, H., and de Vries, J. E. (1991). Interleukin 10 (IL-10) and viral IL-10 strongly reduce antigen-specific human T cell proliferation by diminishing the antigen-presenting capacity of monocytes via downregulation of class II major histocompatibility complex expression. J Exp Med 174,915-24. Deapen, D., Escalante, A., Weinrib, L., Horwitz, D., Bachman, B., Roy-Burman, P., Walker, A., and Mack, T. M. (1992). A revised estimate of twin concordance in systemic lupus erythematosus [see comments]. Arthritis Rheum 35,311-8. Deguchi, Y., Negoro, S., and Kishimoto, S. (1987). Heat-shock protein synthesis by human peripheral mononuclear cells from SLE patients. Biochem Biophys Res Commun 148,1063-8. DeNagel, D. C., and Pierce, S. K. (1992). A case for chaperones in antigen processing [see comments]. Immunol Today 13,86-9. Dhillon, V. B. (1993). Heat shock protein expression in systemic lupus erythematosus (London: University of London). 292 References Dhillon, V. B., McCallum, S., Latchman, D. S., and Isenberg, D. A. (1994). Elevation of the 90 kDa heat-shock protein in specific subsets of systemic lupus erythematosus. QJMed 87,215-22. Dhillon, V. B., McCallum, S., Norton, P., Twomey, B. M., Erkeller-Yuksel, P., Lydyard, P., Isenberg, D. A., and Latchman, D. S. (1993). Differential heat shock protein overexpression and its clinical relevance in systemic lupus erythematosus. Ann Rheum Dis 52,436-42. Donovan, J. L., and Blake, D. R. (2000). Qualitative study of interpretation of reassurance among patients attending rheumatology clinics: "just a touch of arthritis, doctor?". Bmj 320,541-4. Duina, A. A., Kalton, H. M., and Gaber, R. F. (1998). Requirement for Hsp90 and a CyP-40-type cyclophilin in negative regulation of the heat shock response. J Biol Chem 273, 18974-8. Durkin, A. S., Maglott, D. R., Vamvakopoulos, N. C., Zoghbi, H. Y., and Nierman, W. C. (1993). Assignment of an intron-containing human heat-shock protein gene (hsp90 beta, HSPCB) to chromosome 6 near TCTEl (6p21) and two intronless pseudogenes to chromosomes 4 and 15 by polymerase chain reaction amplification from a panel of hybrid cell lines. Genomics 18,452-4. Eastgate, J. A., Symons, J. A., Wood, N. C., Grinlinton, F. M., di Giovine, F. S., and Duff, G. W. (1988). Correlation of plasma interleukin 1 levels with disease activity in rheumatoid arthritis. Lancet 2,706-9. Ehrenstein, M. R., Conroy, S. E., Heath, J., Latchman, D. S., and Isenberg, D. A. (1995). The occurrence, nature and distribution of flares in a cohort of patients with systemic lupus erythematosus: a rheumatological view [see comments]. Br J Rheumatol 34, 257-60. 293 References Enk, A. H., and Katz, S. I. (1992). Identification and induction of keratinocyte- derived IL-10. J Immunol 149^ 92-5. Erkeller-Yuksel, F. M., Isenberg, D. A., Dhillon, V. B., Latchman, D. S., and Lydyard, P. M. (1992). Surface expression of heat shock protein 90 by blood mononuclear cells from patients with systemic lupus erythematosus. J Autoimmun 5, 803-14. Esdaile, J. M., Levinton, C., Federgreen, W., Hayslett, J. P., and Kashgarian, M. (1989). The clinical and renal biopsy predictors of long-term outcome in lupus nephritis: a study of 87 patients and review of the literature. Q J Med 72,779-833. Evans, R. M. (1988). The steroid and thyroid hormone receptor superfamily. Science 240,889-95. Faulds, G., Conroy, S., Madaio, M., Isenberg, D., and Latchman, D. (1995). Increased levels of antibodies to heat shock proteins with increasing age in Mrl/Mp- Ipr/lpr mice [see comments]. Br J Rheumatol 34, 610-5. Faulds, G. B., Isenberg, D. A., and Latchman, D. S. (1994). The tissue specific elevation in synthesis of the 90 kDa heat shock protein precedes the onset of disease in lupus prone MRL/lpr mice. J Rheumatol 21,234-8. Feldman, D. E., and Frydman, J. (2000). Protein folding in vivo: the importance of molecular chaperones [see comments]. Curr Opin Struct Biol 10,26-33. Felts, S. J., Owen, B. A., Nguyen, P., Trepel, J., Dormer, D. B., and Toft, D. O. (2000). The hsp90-related protein TRAPl is a mitochondrial protein with distinct functional properties. J Biol Chem 275,3305-12. Ferrarini, M., Heltai, S., Zocchi, M. R., and Rugarli, C. (1992). Unusual expression and localization of heat-shock proteins in human tumor cells. Int J Cancer 51,613-9. 294 References Fessel, W. J. (1988). Epidemiology of systemic lupus erythematosus. Rheum Dis Clin North Am 14,15-23. Fessel, W. J. (1974). Systemic lupus erythematosus in the community. Incidence, prevalence, outcome, and first symptoms; the high prevalence in black women. Arch Intern Med 134, 1027-35. Fielder, A. H., Walport, M. J., Batchelor, J. R., Rynes, R. I., Black, C. M., Dodi, I. A., and Hughes, G. R. (1983). Family study of the major histocompatibility complex in patients with systemic lupus erythematosus: importance of null alleles of C4A and C4B in determining disease susceptibility. Br Med J (Clin Res Ed) 286,425-8. Finbloom, D. S., and Winestock, K. D. (1995). IL-10 induces the tyrosine phosphorylation of tyk2 and Jakl and the differential assembly of STATl alpha and STAT3 complexes in human T cells and monocytes. J Immunol 155, 1079-90. Finck, B. K., Chan, B., and Wofsy, D. (1994). Interleukin 6 promotes murine lupus in NZB/NZW FI mice. J Clin Invest 94, 585-91. Fiorentino, D. F., Bond, M. W., and Mosmann, T. R. (1989). Two types of mouse T helper cell. IV. Th2 clones secrete a factor that inhibits cytokine production by Thl clones. J Exp Med 170, 2081-95. Fiorentino, D. F., Zlotnik, A., Mosmann, T. R., Howard, M., and O’Garra, A. (1991). IL-10 inhibits cytokine production by activated macrophages. J Immunol 147, 3815- 22 . Fiorentino, D. F., Zlotnik, A., Vieira, P., Mosmann, T. R., Howard, M., Moore, K. W., and O'Garra, A. (1991). IL-10 acts on the antigen-presenting cell to inhibit cytokine production by Thl cells. J Immunol 146,3444-51. 295 References Fisch, P., Malkovsky, M., Kovats, S., Sturm, E., Braakman, B., Klein, B. S., Voss, S. D., Morrissey, L. W., DeMars, R., Welch, W. J., and et al. (1990). Recognition by human V gamma 9/V delta 2 T cells of a GroEL homolog on Daudi Burkitt's lymphoma cells. Science 250,1269-73. Flaherty, K. M., DeLuca-Flaherty, C., and McKay, D. B. (1990). Three-dimensional structure of the ATPase fragment of a 70K heat-shock cognate protein [see comments]. Nature 346,623-8. Fries, J. F. (1987). Methodology of validation of criteria for SLE. Scand J Rheumatol Suppl 65,25-30. Frydman, J., and Hohfeld, J. (1997). Chaperones get in touch: the Hip-Hop connection. Trends Biochem Sci 22, 87-92. Frydman, J., Nimmesgem, E., Ohtsuka, K., and Hartl, F. U. (1994). Folding of nascent polypeptide chains in a high molecular mass assembly with molecular chaperones [see comments]. Nature 370, 111-7. Funauchi, M., Sugishima, H., Minoda, M., and Horiuchi, A. (1991). Serum level of interferon-gamma in autoimmune diseases. Tohoku J Exp Med 164, 259-67. Galea-Lauri, J., Latchman, D. S., and Katz, D. R. (1996). The role of the 90-kDa heat shock protein in cell cycle control and differentiation of the monoblastoid cell line U937. Exp Cell Res 226,243-54. Galea-Lauri, J., Richardson, A. J., Latchman, D. S., and Katz, D. R. (1996). Increased heat shock protein 90 (hsp90) expression leads to increased apoptosis in the monoblastoid cell line U937 following induction with TNF-alpha and cycloheximide: a possible role in immunopathology. J Immunol 157,4109-18. 296 References Garcia-Cardena, G., Fan, R., Shah, V., Sorrentino, R., Cirino, G., Papapetropoulos, A., and Sessa, W. C. (1998). Dynamic activation of endothelial nitric oxide synthase by Hsp90. Nature 392,821-4. Gamier, C., Barbier, P., Gilli, R., Lopez, C., Peyrot, V., and Briand, C. (1998). Heat- shock protein 90 (hsp90) binds in vitro to tubulin dimer and inhibits microtubule formation. Biochem Biophys Res Commun 250,414-9. Germain, R. N. (1995). The biochemistry and cell biology of antigen presentation by MHC class I and class II molecules. Implications for development of combination vaccines. Ann N Y Acad Sci 754,114-25. Germain, R. N., and Margulies, D. H. (1993). The biochemistry and cell biology of antigen processing and presentation. Annu Rev Immunol 77,403-50. Gemold, M., Knauf, U., Gaestel, M., Stahl, J., and Kloetzel, P. M. (1993). Development and tissue-specific distribution of mouse small heat shock protein hsp25. Dev Genet 74, 103-11. Gething, M. J., and Sambrook, J. (1992). Protein folding in the cell. Nature 355, 33- 45. Ginzler, E. M., Diamond, H. S., Weiner, M., Schlesinger, M., Fries, J. F., Wasner, C., Medsger, T. A., Jr., Ziegler, G., Klippel, J. H., Hadler, N. M., Albert, D. A., Hess, B. V., Spencer-Green, G., Grayzel, A., Worth, D., Hahn, B. H., and Barnett, E. V. (1982). A multicenter study of outcome in systemic lupus erythematosus. I. Entry variables as predictors of prognosis. Arthritis Rheum 25,601-11. Gladman, D. D., Goldsmith, C. H., Urowitz, M. B., Bacon, P., Bombardier, C., Isenberg, D., Kalunian, K., Liang, M. H., Maddison, P., Nived, O., and et al. (1992). Crosscultural validation and reliability of 3 disease activity indices in systemic lupus erythematosus. J Rheumatol 79,608-11. 297 References Gladman, D. D., Goldsmith, C. H., Urowitz, M. B., Bacon, P., Bombardier, C., Isenberg, D., Kalunian, K., Liang, M. H., Maddison, P., Nived, O., and et al. (1994). Sensitivity to change of 3 Systemic Lupus Erythematosus Disease Activity Indices: international validation. J Rheumatol 27,1468-71. Gordon, C., Ranges, G. E., Greenspan, J. S., and Wofsy, D. (1989). Chronic therapy with recombinant tumor necrosis factor-alpha in autoimmune NZB/N2W FI mice. Clin Immunol Immunopathol 52,421-34. Gorman, C. M. (1985). High efficiency gene transfer into mammalian cells.. Volume 2, D. M. Glover, ed. (Oxford: IRL Press). Govers, R., ten Broeke, T., van Kerkhof, P., Schwartz, A. L., and Strous, G. J. (1999). Identification of a novel ubiquitin conjugation motif, required for ligand-induced internalization of the growth hormone receptor. Embo J 78,28-36. Grenert, J. P., Johnson, B. D., and Toft, D. O. (1999). The importance of ATP binding and hydrolysis by hsp90 in formation and function of protein heterocomplexes. J Biol Chem 274, 17525-33. Grenert, J. P., Sullivan, W. P., Fadden, P., Haystead, T. A. J., Clark, J., Mimnaugh, E., Krutzsch, H., Ochel, H. J., Schulte, T. W., Sausville, E., Neckers, L. M., and Toft, D. O. (1997). The amino-terminal domain of heat shock protein 90 (hsp90) that binds geldanamycin is an ATP/ADP switch domain that regulates hsp90 conformation. J Biol Chem 272,23843-50. Grigor, R., Edmonds, J., Lewkonia, R., Bresnihan, B., and Hughes, G. R. (1978). Systemic lupus erythematosus. A prospective analysis. Ann Rheum Dis 37,121-8. 298 References Grondai, G., Gunnarsson, I., Ronnelid, J., Rogberg, S., Klareskog, L., and Lundberg, I. (2000). Cytokine production, serum levels and disease activity in systemic lupus erythematosus [In Process Citation]. Clin Exp Rheumatol 18,565-70. Grondai, G., Kristjansdottir, H., Gunnlaugsdottir, B., Amason, A., Lundberg, I., Klareskog, L., and Steinsson, K. (1999). Increased number of interleukin-10- producing cells in systemic lupus erythematosus patients and their first-degree relatives and spouses in Icelandic multicase families. Arthritis Rheum 42, 1649-54. Hagiwara, E., Gourley, M. P., Lee, S., and Klinman, D. K. (1996). Disease severity in patients with systemic lupus erythematosus correlates with an increased ratio of interleukin-10:interferon-gamma- secreting cells in the peripheral blood. Arthritis Rheum 39,379-85. Haire, R. N., Peterson, M. S., and O'Leary, J. J. (1988). Mitogen activation induces the enhanced synthesis of two heat-shock proteins in human lymphocytes. J Cell Biol 106, 883-91. Hall, R. C., Stickney, S. K., and Gardner, E. R. (1981). psychiatric symptoms in patients with systemic lupus erythematosus. Psychosomatics 22, 15-9,23-4. Hanahan, D. (1985). Techniques for transformation in E.coli. In DNA cloning: A practical approach, D. Glover, ed. (Oxford: IRL Press). Harris, E. N., Asherson, R. A., Gharavi, A. E., Morgan, S. H., Derue, G., and Hughes, G. R. (1985). Thrombocytopenia in SLE and related autoimmune disorders: association with anticardiolipin antibody. Br J Haematol 59,227-30. Harrison, C. J., Bohm, A. A., and Nelson, H. C. (1994). Crystal structure of the DNA binding domain of the heat shock transcription factor. Science 263, 224-7. 299 References Haiti, F. U. (1996). Molecular chaperones in cellular protein folding. Nature 381, 571-9. Hashimoto, C. (1990). Studies on synthesis of interleukin-6 and ganunaglobulin in peripheral blood mononuclear cells of patients with systemic lupus erythematosus. Tohoku J Exp Med 162,323-35. Haupt, H. M., Moore, G. W., and Hutchins, 0. M. (1981). The lung in systemic lupus erythematosus. Analysis of the pathologic changes in 120 patients. Am J Med 71, 791-8. Hay, E. M., Bacon, P. A., Gordon, C., Isenberg, D. A., Maddison, P., Snaith, M. L., Symmons, D. P., Viner, N., and Zoma, A. (1993). The BILAG index: a reliable and valid instrument for measuring clinical disease activity in systemic lupus erythematosus. Q J Med 86,447-58. Hay, E. M., and Isenberg, D. A. (1993). Autoantibodies in central nervous system lupus. Br J Rheumatol 32,329-32. Hendrick, J. P., and Haiti, F. U. (1993). Molecular chaperone functions of heat-shock proteins. Annu Rev Biochem 62,349-84. Hendrix, R. W. (1979). Purification and properties of groE, a host protein involved in bacteriophage assembly. J Mol Biol 129,375-92. Hermann, E., Lohse, A. W., Van der Zee, R., Van Eden, W., Mayet, W. J., Probst, P., Poralla, T., Meyer zum Buschenfelde, K. H., and Fleischer, B. (1991). Synovial fluid- derived Yersinia-reactive T cells responding to human 65- kDa heat-shock protein and heat-stressed antigen-presenting cells. Eur J Immunol 27, 2139-43. Herrmann, M., Voll, R. E., Zoller, O. M., Hagenhofer, M., Ponner, B. B., and Kalden, J. R. (1998). Impaired phagocytosis of apoptotic cell material by monocyte-derived 300 References macrophages from patients with systemic lupus erythematosus. Arthritis Rheum 41, 1241-50. Hershko, A., and Ciechanover, A. (1992). The ubiquitin system for protein degradation. Annu Rev Biochem 61,761-807. Hickey, E., Brandon, S. B., Sadis, S., Smale, G., and Weber, L. A. (1986). Molecular cloning of sequences encoding the human heat-shock proteins and their expression during hyperthermia. Gene 43, 147-54. Hickey, E., Brandon, S. E., Smale, G., Lloyd, D., and Weber, L. A. (1989). Sequence and regulation of a gene encoding a human 89-kilodalton heat shock protein. Mol Cell Biol 9 ,2615-26. Hightower, L. E. (1991). Heat shock, stress proteins, chaperones, and proteotoxicity. Cell 66, 191-7. Hochberg, M. C. (1987). Prevalence of systemic lupus erythematosus in England and Wales, 1981-2. Ann Rheum Dis 46,664-6. Hochberg, M. C. (1997). Updating the American College of Rheumatology revised criteria for the classification of systemic lupus erythematosus [letter] [see comments]. Arthritis Rheum4 0 ,1725. Hochberg, M. C., Perlmutter, D. L., Medsger, T. A., Steen, V., Weisman, M. H., White, B., and Wigley, F. M. (1995). Prevalence of self-reported physician-diagnosed systemic lupus erythematosus in the USA. Lupus 4 ,454-6. Hoffbrand, B. I., Vianna, J., Khamashta, M., Hughes, G. R., and Walters, D. V. (1992). Antiphospholipid antibodies and shrinking lungs in SLE [letter]. Lupus 1, 408. 301 References Hohn, T., Hohn, B., Engel, A., Wurtz, M., and Smith, P. R. (1979). Isolation and characterization of the host protein groB involved in bacteriophage lambda assembly. J Mol Biol 729,359-73. Hopkinson, N. D., Doherty, M., and Powell, R. J. (1993). The prevalence and incidence of systemic lupus erythematosus in Nottingham, UK, 1989-1990. Br J Rheumatol 32, 110-5. Hsu, D. H., Moore, K. W., and Spits, H. (1992). Differential effects of IL-4 and IL-10 on IL-2-induced IFN-gamma synthesis and lymphokine-activated killer activity. Int Immunol 4, 563-9. Hsu, D. H., Moore, K. W., and Spits, H. (1992). Differential effects of IL-4 and IL-10 on IL-2-induced IFN-gamma synthesis and lymphokine-activated killer activity. Int Immunol 4, 563-9. Ihle, J. N. (1996). STATs: signal transducers and activators of transcription. Cell 84, 331-4. Ikuta, K., Uchida, N., Friedman, J., and Weissman, I. L. (1992). Lymphocyte development from stem cells. Annu Rev Immunol 10,759-83. Iltanen, S., Rantala, I., Laippala, P., Holm, K., Partanen, J., and Maki, M. (1999). Expression of HSP-65 in jejunal epithelial cells in patients clinically suspected of coeliac disease. Autoimmunity 3 1 ,125-32. Imani, F., and Soloski, M. J. (1991). Heat shock proteins can regulate expression of the Tla region-encoded class Ib molecule Qa-1. Proc Natl Acad Sci U S A 88,10475- 9, 302 References Ingolia, T. D., and Craig, E. A. (1982). Four small Drosophila heat shock proteins are related to each other and to mammalian alpha-crystallin. Proc Natl Acad Sci U S A 79, 2360-4. Isenberg, D. (1990). Comparison of lupus activity indices. Clin Exp Rheumatol 8 Suppl 5,37-40. Ishida, H., Muchamuel, T., Sakaguchi, S., Andrade, S., Menon, S., and Howard, M. (1994). Continuous administration of anti-interleukin 10 antibodies delays onset of autoimmunity in N23B/W FI mice. J Exp Med 779,305-10. Izaki, S., Goto, Y., Kitamura, K., and Nomaguchi, H. (1994). Antibody to 65-kD stress protein (HSP65) of Mycobacterium leprae in various inflammatory skin diseases. A preliminary report. Nippon Rai Gakkai Zasshi 63,3-11. Jacob, C. O., Fronek, Z., Lewis, G. D., Koo, M., Hansen, J. A., and McDevitt, H. O. (1990). Heritable major histocompatibility complex class Il-associated differences in production of tumor necrosis factor alpha: relevance to genetic predisposition to systemic lupus erythematosus. Proc Natl Acad Sci U S A 87, 1233-7. Jakob, U., and Buchner, J. (1994). Assisting spontaneity: the role of Hsp90 and small Hsps as molecular chaperones. Trends Biochem Sci 79,205-11. Jameel, A., Skilton, R. A., Campbell, T. A., Chander, S. K., Coombes, R. C., and Luqmani, Y. A. (1992). Clinical and biological significance of HSP89 alpha in human breast cancer. Int J Cancer 50,409-15. Janeway, C. A., Jr. (1999). The discovery of T cell help for B cell antibody formation: a perspective from the 30th anniversary of this discovery. Immunol Cell Biol 77,177-9. 303 References Janeway, C. A., Jr. (1994). Thymic selection: two pathways to life and two to death. Immunity 7,3-6. Janeway, C. A., Jr., and Bottomly, K. (1994). Signals and signs for lymphocyte responses. Cell 75,275-85. Jendoubi, M., and Bonnefoy, S. (1988). Identification of a heat shock-like antigen in P. falciparum, related to the heat shock protein 90 family. Nucleic Acids Res 75, 10928. Jentsch, S. (1992). The ubiquitin-conjugation system. Annu Rev Genet 25, 179-207. Johnson, A. E., Gordon, C., Palmer, R. G., and Bacon, P. A. (1995). The prevalence and incidence of systemic lupus erythematosus in Birmingham, England. Relationship to ethnicity and country of birth. Arthritis Rheum 38, 551-8. Jones, S. A., Horiuchi, S., Novick, D., Yamamoto, N., and Fuller, G. M. (1998). Shedding of the soluble 11^6 receptor is triggered by Ca2+ mobilization, while basal release is predominantly the product of differential mRNA splicing in THP-1 cells. Eur J Immunol 28,3514-22. Jonsson, H., Nived, O., and Sturfelt, G. (1989). Outcome in systemic lupus erythematosus: a prospective study of patients from a defined population. Medicine (Baltimore) 6 8141-50. , Kaneko, S., Suzuki, N., Yamashita, N., Nagafuchi, H., Nakajima, T., Wakisaka, S., Yamamoto, S., and Sakane, T. (1997). Characterization of T cells specific for an epitope of human 60-kD heat shock protein (hsp) in patients with Behcet’s disease (BD) in Japan. Clin Exp Immunol 108, 204-12. 304 References Kaptein, A., Paillard, V., and Saunders, M. (1996). Dominant negative stat3 mutant inhibits interleukin-6-induced Jak-STAT signal transduction. J Biol Chem 277, 5961- 4. Kaufmann, S. H. (1996). gamma/delta and other unconventional T lymphocytes: what do they see and what do they do? Proc Natl Acad Sci U S A 93,2272-9. Kaufmann, S. H. (1994). Heat shock proteins and autoimmunity: a critical appraisal. Int Arch Allergy Immunol 703,317-22. Kaufmann, S. H. (1990). Heat shock proteins and the immune response. Immunol Today 77, 129-36. Kaufmann, S. H. (1991). Heat-shock proteins and pathogenesis of bacterial infections. Springer Semin Immunopathol 73, 25-36. Kaufmann, S. H., Schoel, B., Wand-Wurttenberger, A., Steinhoff, U., Munk, M. E., and Koga, T. (1990). T-cells, stress proteins, and pathogenesis of mycobacterial infections. CurrTop Microbiol Immunol 755, 125-41. Kaufmann, S. H. B., and Schoel, B. (1994). Heat shock proteins as antigens in immunity against infection and self. In The biology of heat shock proteins and molecular chaperones, A. T. R.I. Morimoto, C. Georgopoulos, ed. (Cold Spring Harbor, N.Y.: Cold Spring Harbor Laboratory Press), pp. 495-531. Kaur, I., Voss, S. D., Gupta, R. S., Schell, K., Fisch, P., and Sondel, P. M. (1993). Human peripheral gamma delta T cells recognize hsp60 molecules on Daudi Burkitt's lymphoma cells. J Immunol 750, 2046-55. Kemp, M. E., Atkinson, J. P., Skanes, V. M., Levine, R. P., and Chaplin, D. D. (1987). Deletion of C4A genes in patients with systemic lupus erythematosus. Arthritis Rheum 30,1015-22. 305 References Kim, K. K., Kim, R., and Kim, S. H. (1998). Crystal structure of a small heat-shock protein. Nature 394^ 595-9. Kinoshita, S., Akira, S., and Kishimoto, T. (1992). A member of the C/EBP family, NF-IL6 beta, forms a heterodimer and transcriptionally synergizes with NF-IL6. Proc Natl Acad Sci U S A 89, 1473-6. Kirou, K. A., and Crow, M. K. (1999). New pieces to the SLE cytokine puzzle [editorial]. Clin Immunol 9i, 1-5. Kishimoto, T., and Hirano, T. (1988). A new interleukin with pleiotropic activities. Bioessays 9, 11-5. Klashman, D. J., Martin, R. A., Martinez-Maza, O., and Stevens, R. H. (1991). In vitro regulation of B cell differentiation by interleukin-6 and soluble CD23 in systemic lupus erythematosus B cell subpopulations and antigen-induced normal B cells. Arthritis Rheum34, 276-86. Kleindienst, R., Schett, G., Amberger, A., Seitz, C. S., Michaelis, D., Metzler, B., Dietrich, H., Xu, Q., and Wick, G. (1995). Atherosclerosis as an autoimmune condition. Isr J Med Sci 31, 596-9. Kleindienst, R., Xu, Q., Willeit, J., Waldenberger, F. R., Weimann, S., and Wick, G. (1993). Immunology of atherosclerosis. Demonstration of heat shock protein 60 expression and T lymphocytes bearing alpha/beta or gamma/delta receptor in human atherosclerotic lesions. Am J Pathol 142,1927-37. Klemenz, R., Frohli, E., Steiger, R. H., Schafer, R., and Aoyama, A. (1991). Alpha B- crystallin is a small heat shock protein. Proc Natl Acad Sci U S A 88,3652-6. 306 References Kondo, T., Matsuda, T., Tashima, M., Umehara, H., Domae, N., Yokoyama, K., Uchiyama, T., and Okazaki, T. (2000). Suppression of heat shock protein-70 by ceramide in heat shock-induced HL-60 cell apoptosis. J Biol Chem 275,8872-9. Kono, D. H., Burlingame, R. W., Owens, D. G., Kuramochi, A., Balderas, R. S., Balomenos, D., and Theofilopoulos, A. N. (1994). Lupus susceptibility loci in New 2^aland mice. Proc Natl Acad Sci U S A 97,10168-72. Korsmeyer, S. J. (1992). Bcl-2: a repressor of lymphocyte death. Immunol Today 75, 285-8. Koyasu, S., Nishida, E., Miyata, Y., Sakai, H., and Yahara, I. (1989). HSPlOO, a 100- kDa heat shock protein, is a Ca2+-calmodulin-regulated actin-binding protein. J Biol Chem 264, 15083-7. Krupp, L. B., LaRocca, N. G., Muir, J., and Steinberg, A. D. (1990). A study of fatigue in systemic lupus erythematosus. J Rheumatol 77, 1450-2. La Thangue, N. B., and Latchman, D. S. (1988). A cellular protein related to heat- shock protein 90 accumulates during herpes simplex virus infection and is overexpressed in transformed cells. Exp Cell Res 775, 169-79. Laemmli, U. K. (1970). Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature 227, 680-5. Lai, C. P., Ripperger, J., Morelia, K. K., Jurlander, J., Hawley, T. S., Carson, W. E., Kordula, T., Caligiuri, M. A., Hawley, R. G., Fey, G. H., and Baumann, H. (1996). Receptors for interleukin (IL)-10 and IL-6-type cytokines use similar signaling mechanisms for inducing transcription through IL-6 response elements. J Biol Chem 277,13968-75. 307 References Larson, J. S., Schuetz, T. J., and Kingston, R. E. (1988). Activation in vitro of sequence-specific DNA binding by a human regulatory factor. Nature 335,372-5. Latchman, D. S., Chan, W. L., Leaver, C. E., Patel, R., Oliver, P., and La Thangue, N. B. (1987). The human Mr 90,000 heat shock protein and the Escherichia coli Lon protein share an antigenic determinant. Comp Biochem Physiol [B] 87,961-7. Latchman, D. S., and Isenberg, D. A. (1994). The role of hsp90 in SLE. Autoimmunity 19,211-8. Lee, L. A., and Weston, W. L. (1984). New findings in neonatal lupus syndrome. Am J Dis Child 138, 233-6. Lee, M. G. (1998). The 3' untranslated region of the hsp 70 genes maintains the level of steady state mRNA in Trypanosoma brucei upon heat shock. Nucleic Acids Res 26,4025-33. Levin, D. L., Roenigk, H. H., Jr., Caro, W. A., and Lyons, M. (1982). Histologic, immunofluorescent, and antinuclear antibody findings in PUVA-treated patients. J Am Acad Dermatol 6 ,328-33. Li, S. G., Quayle, A. J., Shen, Y., Kjeldsen-Kragh, J., Oftung, P., Gupta, R. S., Natvig, J. B., and Forre, O. T. (1992). Mycobacteria and human heat shock protein- specific cytotoxic T lymphocytes in rheumatoid synovial inflammation. Arthritis Rheum 35,270-81. Li, Z., and Srivastava, P. K. (1993). Tumor rejection antigen gp96/grp94 is an ATPase: implications for protein folding and antigen presentation. Embo J 72, 3143- 51. Lindquist, S. (1986). The heat-shock response. Annu Rev Biochem 55,1151-91. 308 References Lindquist, S., and Craig, E. A. (1988). The heat-shock proteins. Annu Rev Genet 22, 631-77. Linker-Israeli, M. (1992). Cytokine abnormalities in human lupus. Clin Immunol Immunopathol 63, 10-2. Linker-Israeli, M., Deans, R. J., Wallace, D. J., Prehn, J., Ozeri-Chen, T., and Klinenberg, J. R. (1991). Elevated levels of endogenous IL-6 in systemic lupus erythematosus. A putative role in pathogenesis. J Immunol 147,117-23. Linker-Israeli, M., Quismorio, F. P., Jr., and Horwitz, D. A. (1990). CD8+ lymphocytes from patients with systemic lupus erythematosus sustain, rather than suppress, spontaneous polyclonal IgG production and synergize with CD4+ cells to support autoantibody synthesis. Arthritis Rheum 33, 1216-25. Lis, J., and Wu, C. (1993). Protein traffic on the heat shock promoter: parking, stalling, and trucking along. Cell 74, 1-4. Liu, Y., Wei, S. H., Ho, A. S., de Waal Malefyt, R., and Moore, K. W. (1994). Expression cloning and characterization of a human IL-10 receptor. J Immunol 152, 1821-9. Llorente, L., Richaud-Patin, Y., Wijdenes, J., Alcocer-Varela, J., Maillot, M. C., Durand-Gasselin, I., Fourrier, B. M., Galanaud, P., and Emilie, D. (1993). Spontaneous production of interleukin-10 by B lymphocytes and monocytes in systemic lupus erythematosus. Eur Cytokine Netw 4 ,421-7. Llorente, L., Zou, W., Levy, Y., Richaud-Patin, Y., Wijdenes, J., Alcocer-Varela, J., Morel-Fourrier, B., Brouet, J. C., Alarcon-Segovia, D., Galanaud, P., and et al. (1995). Role of interleukin 10 in the B lymphocyte hyperactivity and autoantibody production of human systemic lupus erythematosus. J Exp Med 181,839-44. 309 References Mailhos, C., Howard, M. K., and Latchman, D. S. (1993). Heat shock protects neuronal cells from programmed cell death by apoptosis. Neuroscience 55,621-7. Mailhos, C., Howard, M. K., and Latchman, D. S. (1994). Heat shock proteins hsp90 and hsp70 protect neuronal cells from thermal stress but not from programmed cell death. J Neurochem 63,1787-95. Manning-Krieg, U. C., Scherer, P. E., and Schatz, G. (1991). Sequential action of mitochondrial chaperones in protein import into the matrix. Embo J 10,3273-80. Martin, J., Horwich, A. L., and Haiti, F. U. (1992). Prevention of protein dénaturation under heat stress by the chaperonin Hsp60. Science 258,995-8. Maruya, M., Sameshima, M., Nemoto, T., and Yahara, I. (1999). Monomer arrangement in HSP90 dimer as determined by decoration with N and C-terminal region specific antibodies. J Mol Biol 285,903-7. Mason, P. B., Jr., and Lis, J. T. (1997). Cooperative and competitive protein interactions at the hsp70 promoter. J Biol Chem 272,33227-33. Matthews, R., and Bumie, J. (1992). The role of hsp90 in fungal infection. Immunol Today 13,345-8. Matthews, R. C. (1992). The 14th C. L. Oakley Lecture. Candida albicans HSP 90: link between protective and auto immunity. J Med Microbiol 36,367-70. Mauff, G., Alper, C. A., Dawkins, R., Doxiadis, G., Giles, C. M., Hauptmann, G., Rittner, C., and Schneider, P. M. (1990). C4 nomenclature statement (1990). Complement Inflamm 7,261-8. Mayr, M., Metzler, B., Kiechl, S., Willeit, J., Schett, G., Xu, Q., and Wick, G. (1999). Endothelial cytotoxicity mediated by serum antibodies to heat shock proteins of 310 References Escherichia coli and Chlamydia pneumoniae: immune reactions to heat shock proteins as a possible link between infection and atherosclerosis. Circulation 99, 1560-6. McCluskey, R. T. (1982). Evidence for an inunune complex disorder in systemic lupus erythematosus (SLE). Am J Kidney Dis 2, 119-25. McGehee, H. A., Shulman, L. E., Tumulty, A. P., Lockard, C., and Schoenrich, E. H. (1954). Systemic lupus erythematosus: review of the literature and clinical analysis of 138 cases. Medicine 33,291-437. McLean, I. L., Archer, J. R., Cawley, M. I., Pegley, F. S., Kidd, B. L., and Thompson, P. W. (1990). Specific antibody response to the mycobacterial 65 kDa stress protein in ankylosing spondylitis and rheumatoid arthritis. Br J Rheumatol 29, 426-9. Mehrian, R., Quismorio, F. P., Jr., Strassmann, G., Stimmler, M. M., Horwitz, D. A., Kitridou, R. C., Gauderman, W. J., Morrison, J., Brautbar, C., and Jacob, C. O. (1998). Synergistic effect between IL-10 and bcl-2 genotypes in determining susceptibility to systemic lupus erythematosus. Arthritis Rheum 41,596-602. Metz, K., Ezemieks, J., Sebald, W., and Duschl, A. (1996). Interleukin-4 upregulates the heat shock protein Hsp90alpha and enhances transcription of a reporter gene coupled to a single heat shock element. FEBS Lett 385, 25-8. Metzler, B., Xu, Q., and Wick, G. (1998). The role of (auto-) immunity in atherogenesis [see comments]. Wien Klin Wochenschr 110,350-5. Michet, C. J., Jr., McKenna, C. H., Elveback, L. R., Kaslow, R. A., and Kurland, L. T. (1985). Epidemiology of systemic lupus erythematosus and other connective tissue diseases in Rochester, Minnesota, 1950 through 1979. Mayo Clin Proc 6 0 ,105-13. 311 References Miles, S., and Isenberg, D. (1993). A review of serological abnormalities in relatives of SLE patients. Lupus 2, 145-50. Minami, M., Inoue, M., Wei, S., Takeda, K., Matsumoto, M., Kishimoto, T., and Akira, S. (1996). STATS activation is a critical step in gplSO-mediated terminal differentiation and growth arrest of a myeloid cell line. Proc Natl Acad Sci U S A 95, 3963-6. Minami, Y., Kawasaki, H., Miyata, Y., Suzuki, K., and Yahara, I. (1991). Analysis of native forms and isoform compositions of the mouse 90-kDa heat shock protein, HSP90. J Biol Chem 266, 10099-103. Minoo, P., Sullivan, W., Solomon, L. R., Martin, T. E., Toft, D. O., and Scott, R. E. (1989). Loss of proliferative potential during terminal differentiation coincides with the decreased abundance of a subset of heterogeneous ribonuclear proteins. J Cell Biol 109, 1937-46. Minota, S., Koyasu, S., Yahara, I., and Winfield, J. (1988). Autoantibodies to the heat-shock protein hsp90 in systemic lupus erythematosus. J Clin Invest 81, 106-9. Miyata, Y., and Yahara, I. (1995). Interaction between casein kinase II and the 90- kDa stress protein, HSP90. Biochemistry 34,8123-9. Monier, J. C., Dardenne, M., Rigal, D., Costa, O., Foumel, C., and Lapras, M. (1980). Clinical and laboratory features of canine lupus syndromes. Arthritis Rheum 23, 294- 301. Moore, K. W., O'Garra, A., de Waal Malefyt, R., Vieira, P., and Mosmann, T. R. (1993). Interleukin-10. Annu Rev Immunol 11,165-90. 312 References Morel, L., Rudofsky, U. H., Longmate, J. A., Schiffenbauer, J., and Wakeland, E. K. (1994). Polygenic control of susceptibility to murine systemic lupus erythematosus. Immunity 7,219-29. Morimoto, R. I. (1993). Cells in stress: transcriptional activation of heat shock genes. Science 259,1409-10. Morimoto, R. I., Sarge, K. D., and Abravaya, K. (1992). Transcriptional regulation of heat shock genes. A paradigm for inducible genomic responses. J Biol Chem 267, 21987-90. Morrow, J., Nelson, L., Watts, R., and Isenberg, D. (1999). Autoimmune Rheumatic Diseases, 2 Edition (Oxford: Oxford University Press). Morrow, W. J., Isenberg, D. A., Todd-Pokropek, A., Parry, H. F., and Snaith, M. L. (1982). Useful laboratory measurements in the management of systemic lupus erythematosus. Q J Med 5 1125-38. , Morrow, W. J., Williams, D. J., Perec, C., Casbum-Budd, R., Isenberg, D. A., Paice, E., Snaith, M. L., Youinou, P., and Le Goff, P. (1983). The use of C3d as a means of monitoring clinical activity in systemic lupus erythematosus and rheumatoid arthritis. Ann Rheum Dis 42,668-71. Moser, K. L., Neas, B. R., Salmon, J. E., Yu, H., Gray-McGuire, C., Asundi, N., Bruner, G. R., Fox, J., Kelly, J., Henshall, S., Bacino, D., Dietz, M., Hogue, R., Koelsch, G., Nightingale, L., Shaver, T., Abdou, N. I., Albert, D. A., Carson, C., Petri, M., Treadwell, E. L., James, J. A., and Harley, J. B. (1998). Genome scan of human systemic lupus erythematosus: evidence for linkage on chromosome Iq in African-American pedigrees. Proc Natl Acad Sci U S A 95,14869-74. Nagata, S., and Golstein, P. (1995). The Fas death factor. Science 267,1449-56. 313 References Nakae, K., Funisawa, F., and Kasukawa, R. (1987). A nationwide epidemiological survey on diffuse collagen diseases: estimation of prevalence rate in Japan., R. Kasukawa and G. C. Missouri, eds. (Amsterdam: Elsevier). Nakai, A., Tanabe, M., Kawazoe, Y., Inazawa, J., Morimoto, R. I., and Nagata, K. (1997). HSF4, a new member of the human heat shock factor family which lacks properties of a transcriptional activator. Mol Cell Biol 77,469-81. Naparstek, Y., Ben-Yehuda, A., Madaio, M. P., Bar-Tana, R., Schuger, L., Pizov, G., Neeman, Z. V., and Cohen, I. R. (1990). Binding of anti-DNA antibodies and inhibition of glomerulonephritis in MRL-lpr/lpr mice by heparin [see comments]. Arthritis Rheum33, 1554-9. Nelson, R. J., Ziegelhoffer, T., Nicolet, C., Wemer-Washbume, M., and Craig, B. A. (1992). The translation machinery and 70 kd heat shock protein cooperate in protein synthesis. Cell 77,97-105. Nemoto, T., and Sato, N. (1998). Oligomeric forms of the 90-kDa heat shock protein. Biochem J 550,989-95. Nived, O., Johansen, P. B., and Sturfelt, G. (1993). Standardized ultraviolet-A exposure provokes skin reaction in systemic lupus erythematosus. Lupus 2, 247-50. Nived, O., Sturfelt, G., and Wollheim, F. (1985). Systemic lupus erythematosus in an adult population in southern Sweden: incidence, prevalence and validity of ARA revised classification criteria. Br J Rheumatol 24, 147-54. Norton, P. M., Isenberg, D. A., and Latchman, D. S. (1989). Elevated levels of the 90 kd heat shock protein in a proportion of SLE patients with active disease. J Autoimmun 2,187-95. 314 References Nossent, J. C. (1993). Course and prognostic value of Systemic Lupus Erythematosus Disease Activity Index in black Caribbean patients. Semin Arthritis Rheum 23, 16- 21. O'Farrell, A. M., Liu, Y., Moore, K. W., and Mui, A. L. (1998). IL-10 inhibits macrophage activation and proliferation by distinct signaling mechanisms: evidence for StatS-dependent and -independent pathways. Embo J 77,1006-18. Oda, A., Miyata, M., Kodama, E., Satoh, H., Sato, Y., Nishimaki, T., Nomaguchi, H., and Kasukawa, R. (1994). Antibodies to 65Kd heat-shock protein were elevated in rheumatoid arthritis. Clin Rheumatol 73,261-4. Ostermann, J., Horwich, A. L., Neupert, W., and Haiti, F. U. (1989). Protein folding in mitochondria requires complex formation with hsp60 and ATP hydrolysis. Nature 347, 125-30. Panaretou, B., Prodromou, C., Roe, S. M., O'Brien, R., Ladbury, J. E., Piper, P. W., and Pearl, L. H. (1998). ATP binding and hydrolysis are essential to the function of the Hsp90 molecular chaperone in vivo. Embo J 77,4829-36. Parag, H. A., Raboy, B., and Kulka, R. G. (1987). Effect of heat shock on protein degradation in mammalian cells: involvement of the ubiquitin system. Embo J 6, 55- 61. Parganas, E., Wang, D., Stravopodis, D., Topham, D. J., Marine, J. C., Teglund, S., Vanin, E. F., Bodner, S., Colamonici, O. R., van Deursen, J. M., Grosveld, G., and Ihle, J. N. (1998). Jak2 is essential for signaling through a variety of cytokine receptors. Cell 93,385-95. Park, Y. B., Lee, S. K., Kim, D. S., Lee, J., Lee, C. H., and Song, C. H. (1998). Elevated interleukin-10 levels correlated with disease activity in systemic lupus erythematosus. Clin Exp Rheumatol 76, 283-8. 315 References Passas, C. M., Wong, R. L., Peterson, M., Testa, M. A., and Rothfield, N. F. (1985). A comparison of the specificity of the 1971 and 1982 American Rheumatism Association criteria for the classification of systemic lupus erythematosus. Arthritis Rheum 28, 620-3. Pekin, T. J., Jr. (1970). Zvaifler NJ: Synovial fluid findings in systemic lupus erythematosus (SLE). Arthritis Rheum13, 777-85. Pelham, H., Lewis, M., and Lindquist, S. (1984). Expression of a Drosophila heat shock protein in mammalian cells: transient association with nucleoli after heat shock. Philos Trans R Soc Lond B Biol Sci 307,301-7. Pelham, H. R. (1986). Speculations on the functions of the major heat shock and glucose- regulated proteins. Cell 46,959-61. Pervin, K., Childerstone, A., Shinnick, T., Mizushima, Y., van der Zee, R., Hasan, A., Vaughan, R., and Lehner, T. (1993). T cell epitope expression of mycobacterial and homologous human 65- kilodalton heat shock protein peptides in short term cell lines from patients with Behcet's disease. J Immunol 151,2273-82. Peters, J., Harris, J. R., and Finley, D. (1998). Ubiquitin and the Biology of the Cell (New York: Plenum Press). Poccia, F., Piselli, P., Di Cesare, S., Bach, S., Colizzi, V., Mattel, M., Bolognesi, A., and Stirpe, F. (1992). Recognition and killing of tumour cells expressing heat shock protein 65 kD with immunotoxins containing saporin. Br J Cancer 66,427-32. Podack, E. R., and Kupfer, A. (1991). T-cell effector functions: mechanisms for delivery of cytotoxicity and help. Annu Rev Cell Biol 7,479-504. 316 References Polla, B. S. (1991). Heat shock proteins in host-parasite interactions. Immunol Today 72,A38-41. Polla, B. S. (1988). A role for heat shock proteins in inflammation? Immunol Today 9, 134-7. Prakken, A. B., van Eden, W., Rijkers, G. T., Kuis, W., Toebes, B. A., de Graeff- Meeder, E. R., van der Zee, R., and Zegers, B. J. (1996). Autoreactivity to human heat-shock protein 60 predicts disease remission in oligoarticular juvenile rheumatoid arthritis. Arthritis Rheum 39, 1826-32. Pratt, W. B. (1997). The role of the hsp90-based chaperone system in signal transduction by nuclear receptors and receptors signaling via MAP kinase. Annu Rev Pharmacol Toxicol 37,297-326. Pratt, W. B., and Toft, D. O. (1997). Steroid receptor interactions with heat shock protein and immunophilin chaperones. Endocr Rev 18,306-60. Preston, T. M., King, C. A., and Hyams, J. S. (1990). The molecular components of the cytoskeleton. In: The Cytoskeleton and Cell Motility (London: Blackie and Son Ltd). Prodromou, C., Roe, S. M., O'Brien, R., Ladbury, J. E., Piper, P. W., and Pearl, L. H. (1997). Identification and structural characterization of the ATP/ADP-binding site in the Hsp90 molecular chaperone. Cell 90,65-75. Prodromou, C., Roe, S. M., Piper, P. W., and Pearl, L. H. (1997). A molecular clamp in the crystal structure of the N-terminal domain of the yeast Hsp90 chaperone [see comments]. Nat Struct Biol 4 ,477-82. 317 References Prummel, M. F., Van Pareren, Y., Bakker, O., and Wiersinga, W. M. (1997). Anti heat shock protein (hsp)72 antibodies are present in patients with Graves' disease (GD) and in smoking control subjects. Clin Exp Immunol 110,292-5. Ralph, P., Nakoinz, I., Sampson-Johannes, A., Fong, S., Lowe, D., Min, H. Y., and Lin, L. (1992). IL-10, T lymphocyte inhibitor of human blood cell production of IL-1 and tumor necrosis factor. J Immunol 148,808-14. Rambukkana, A., Das, P. K., Witkamp, L., Yong, S., Meinardi, M. M., and Bos, J. D. (1993). Antibodies to mycobacterial 65-kDa heat shock protein and other immunodominant antigens in patients with psoriasis. J Invest Dermatol 100,87-92. Rasmussen, E. B., and Lis, J. T. (1993). In vivo transcriptional pausing and cap formation on three Drosophila heat shock genes. Proc Natl Acad Sci U S A 90,7923- 7. Raveche, E. S., Tjio, J. H., and Steinberg, A. D. (1980). Genetic studies in NZB mice. IV. The effect of sex hormones on the spontaneous production of anti-T cell autoantibodies. Arthritis Rheum 23,48-56. Rebbe, N. F., Ware, J., Bertina, R. M., Modrich, P., and Stafford, D. W. (1987). Nucleotide sequence of a cDNA for a member of the human 90-kDa heat- shock protein family. Gene 53,235-45. Redmond, T., Sanchez, E. R., Bresnick, E. FI., Schlesinger, M. J., Toft, D. O., Pratt, W. B., and Welsh, M. J. (1989). Immunofluorescence colocalization of the 90-kDa heat-shock protein and microtubules in interphase and mitotic mammalian cells. Eur J Cell Biol 50,66-75. Reveille, J. D., Bartolucci, A., and Alarcon, G. S. (1990). Prognosis in systemic lupus erythematosus. Negative impact of increasing age at onset, black race, and thrombocytopenia, as well as causes of death. Arthritis Rheum 33,37-48. 318 References Riehl, R. M., Sullivan, W. P., Vroman, B. T., Bauer, V. J., Pearson, G. R., and Toft, D. O. (1985). Immunological evidence that the nonhormone binding component of avian steroid receptors exists in a wide range of tissues and species. Biochemistry 24, 6586-91. Riley, J. K., Takeda, K., Akira, S., and Schreiber, R. D. (1999). Interleukin-10 receptor signaling through the JAK-STAT pathway. Requirement for two distinct receptor-derived signals for anti- inflammatory action. J Biol Chem 274, 16513-21. Ripley, B. J., Stephanou, A., Isenberg, D. A., and Latchman, D. S. (1999). Interleukin-10 activates heat-shock protein 90beta gene expression. Immunology 97, 226-31. Ritossa, R. (1962). A new puffing pattern induced by temperature shock and DNP in Drosophila. Experientia 75, 571-3. Rivas, J. M., and Ullrich, S. B. (1992). Systemic suppression of delayed-type hypersensitivity by supernatants from UV-irradiated kératinocytes. An essential role for keratinocyte- derived IL-10. J Immunol 149,3865-71. Robey, B., and Fowlkes, B. J. (1994). Selective events in T cell development. Annu Rev Immunol 72, 675-705. Rogers, M. P. (1983). Psychiatric Aspects, P. Schur, ed. (New York: Grune and Stratton). Roitt, I. (1998). Essential Immunology, 9 Edition (Oxford: Blackwell Scientific Ltd). Rood, M. J., van Krugten, M. V., Zanelli, E., van der Linden, M. W., Keijsers, V., Schreuder, G. M., Verduyn, W., Westendorp, R. G., de Vries, R. R., Breedveld, F. C., Verweij, C. L., and Huizinga, T. W. (2000). TNF-308A and HLA-DR3 alleles 319 References contribute independently to susceptibility to systemic lupus erythematosus. Arthritis Rheum43, 129-34. Rousset, F., Garcia, E., Defrance, T., Peronne, C., Vezzio, N., Hsu, D. H., Kastelein, R., Moore, K. W., and Banchereau, J. (1992). Interleukin 10 is a potent growth and differentiation factor for activated human B lymphocytes. Proc Natl Acad Sci U S A 89, 1890-3. Rudolphi, U., Rzepka, R., Batsford, S., Kaufmann, S. H., von der Mark, K., Peter, H. H., and Melchers, I. (1997). The B cell repertoire of patients with rheumatoid arthritis. II. Increased frequencies of IgO+ and IgA+ B cells specific for mycobacterial heat-shock protein 60 or human type II collagen in synovial fluid and tissue. Arthritis Rheum40, 1409-19. Russell, M. L., and Hanna, W. M. (1988). Ultrastructural pathology of skeletal muscle in various rheumatic diseases. J Rheumatol 15,445-53. Saito, M., Uechi, Y., Nakabayashi, K., and Nagasawa, T. (1994). [Measurement of antibodies to double-stranded and single-stranded DNA using bacteriophage lambda DNA as the antigen—avidity of antibodies to DNA in patients with lupus nephritis]. Nippon Jinzo Gakkai Shi 36, 1103-12. Sakane, T. (1997). New perspective on Behcet's disease. Int Rev Immunol 14, 89-96. Sakane, T., Suzuki, N., Takada, S., Ueda, Y., Murakawa, Y., Tsuchida, T., Yamauchi, Y., and Kishimoto, T. (1988). B cell hyperactivity and its relation to distinct clinical features and the degree of disease activity in patients with systemic lupus erythematosus. Arthritis Rheum31, 80-7. Salvetti, M., Buttinelli, C., Ristori, G., Carbonari, M., Cherchi, M., Fiorelli, M., Grasso, M. G., Toma, L., and Pozzilli, C. (1992). T-lymphocyte reactivity to the 320 References recombinant mycobacterial 65- and 70-kDa heat shock proteins in multiple sclerosis. J Autoimmun 5,691-702. Salvetti, M., Ristori, G., Buttinelli, C., Fiori, P., Falcone, M., Britton, W., Adams, E., Paone, G., Grasso, M. G., and Pozzilli, C. (1996). The immune response to mycobacterial 70-kDa heat shock proteins frequently involves autoreactive T cells and is quantitatively disregulated in multiple sclerosis. J Neuroimmunol 65, 143-53. Samanta, A., Feehally, J., Roy, S., Nichol, F. E., Sheldon, P. J., and Walls, J. (1991). High prevalence of systemic disease and mortality in Asian subjects with systemic lupus erythematosus. Ann Rheum Dis 50,490-2. Sambrook, J., Fritsch, E. F., and Maniatis, T. (1989). Molecular cloning, a laboratory manual, 2nd Edition (New York: Cold Spring Harbour Press). Sarge, K. D., Murphy, S. P., and Morimoto, R. I. (1993). Activation of heat shock gene transcription by heat shock factor 1 involves oligomerization, acquisition of DNA-binding activity, and nuclear localization and can occur in the absence of stress [published errata appear in Mol Cell Biol 1993 May;13(5):3122-3 and 1993 Jun;13(6):3838-9]. Mol Cell Biol 13,1392-407. Sarge, K. D., Park-Sarge, O. K., Kirby, J. D., Mayo, K. E., and Morimoto, R. I. (1994). Expression of heat shock factor 2 in mouse testis: potential role as a regulator of heat-shock protein gene expression during spermatogenesis. Biol Reprod 50, 1334- 43. Sato, H., Miyata, M., and Kasukawa, R. (1996). Expression of heat shock protein on lymphocytes in peripheral blood and synovial fluid from patients with rheumatoid arthritis. J Rheumatol23, 2027-32. 321 References Schaiff, W. T., Hruska, K. A., Jr., McCourt, D. W., Green, M., and Schwartz, B. D. (1992). HLA-DR associates with specific stress proteins and is retained in the endoplasmic reticulum in invariant chain negative cells. J Exp Med 176,657-66. Scheibel, T., Weikl, T., and Buchner, J. (1998). Two chaperone sites in Hsp90 differing in substrate specificity and ATP dependence. Proc Natl Acad Sci U S A 95, 1495-9. Schett, G., Metzler, B., Mayr, M., Amberger, A., Niederwieser, D., Gupta, R. S., Mizzen, L., Xu, Q., and Wick, G. (1997). Macrophage-lysis mediated by autoantibodies to heat shock protein 65/60. Atherosclerosis 728,27-38. Schirmbeck, R., Bohm, W., and Reimann, J. (1997). Stress protein (hsp73)-mediated, TAP-independent processing of endogenous, truncated SV40 large T antigen for Db- restricted peptide presentation. Eur J Immunol 27,2016-23. Schirmbeck, R., and Reimann, J. (1994). Peptide transporter-independent, stress protein-mediated endosomal processing of endogenous protein antigens for major histocompatibility complex class I presentation. Eur J Immunol 24, 1478-86. Schlesinger, M. J. (1990). Heat shock proteins. J Biol Chem 265,12111-4. Schmidt, S. (1999). Candidate autoantigens in multiple sclerosis. Mult Scler 5, 147- 60. Schur, P. H. (1995). Genetics of systemic lupus erythematosus. Lupus 4,425-37. Schwartz, M. M., Lan, S. P., Bernstein, J., Hill, G. S., Holley, K., and Lewis, E. J. (1993). Irreproducibility of the activity and chronicity indices limits their utility in the management of lupus nephritis. Lupus Nephritis Collaborative Study Group. Am J Kidney Dis 27,374-7. 322 References Screpanti, I., Musiani, P., Bellavia, D., Cappelletti, M., Aiello, F. B., Maroder, M., Frati, L., Modesti, A., Gulino, A., and Poli, V. (1996). Inactivation of the IL-6 gene prevents development of multicentric Castleman's disease in C/EBP beta-deficient mice. J Exp Med 184^ 1561-6. Screpanti, I., Romani, L., Musiani, P., Modesti, A., Fattori, E., Lazzaro, D., Sellitto, C., Scarpa, S., Bellavia, D., Lattanzio, G., and et al. (1995). Lymphoproliferative disorder and imbalanced T-helper response in C/EBP beta-deficient mice [published erratum appears in EMBO J 1995 Jul 17;14(14):3596]. Embo J 14, 1932-41. Seufert, W., and Jentsch, S. (1990). Ubiquitin-conjugating enzymes UBC4 and UBC5 mediate selective degradation of short-lived and abnormal proteins. Embo J 9, 543- 50. Sharp, G. C., Irvin, W. S., LaRoque, R. L., Velez, C., Daly, V., Kaiser, A. D., and Holman, H. R. (1971). Association of autoantibodies to different nuclear antigens with clinical patterns of rheumatic disease and responsiveness to therapy. J Clin Invest 50,350-9. Shibahara, S., Muller, R. M., and Taguchi, H. (1987). Transcriptional control of rat heme oxygenase by heat shock. J Biol Chem 262, 12889-92. Shin, S. O., Billings, P. B., Keithley, E. M., and Harris, J. P. (1997). Comparison of anti-heat shock protein 70 (anti-hsp70) and anti-68-kDa inner ear protein in the sera of patients with Meniere's disease. Laryngoscope 107,222-7. Shinnick, T. M. (1991). Heat shock proteins as antigens of bacterial and parasitic pathogens. CurrTop Microbiol Immunol 167,145-60. Shinnick, T. M., Vodkin, M. H., and Williams, J. C. (1988). The Mycobacterium tuberculosis 65-kilodalton antigen is a heat shock protein which corresponds to common antigen and to the Escherichia coli GroEL protein. Infect Immun 56,446-51. 323 References Shoenfeld, Y., and Isenberg, D. A. (1989). The mosaic of autoimmunity. Immunol Today 10,123-6. Siegel, M., and Lee, S. L. (1973). The epidemiology of systemic lupus erythematosus. Semin Arthritis Rheum3 ,1-54. Sierakowski, S., Kucharz, E. J., Lightfoot, R. W., Jr., and Goodwin, J. S. (1987). Interleukin-1-production by monocytes from patients with systemic lupus erythematosus. Clin Rheumatol 6 ,403-7. Smith, D. F., and Toft, D. O. (1993). Steroid receptors and their associated proteins. Mol Endocrinol 7,4-11. Snaith, M. L. (1992). Disease activity in SLE and its assessment. Lupus 1 (SI). Song, J. H., Lee, K. J., Lee, S. W., Han, J. Y., and Kim, M. J. (2001). Fatal pulmonary-renal syndrome manifested with inunune complex crescentic glomerulonephritis in a patient with mpo-anca seropositivity. Yonsei Med J 42, 425- 30. Song, Y. W., Han, C. W., Kang, S. W., Baek, H. J., Lee, E. B., Shin, C. H., Hahn, B. H., and Tsao, B. P. (1998). Abnormal distribution of Fc gamma receptor type Ila polymorphisms in Korean patients with systemic lupus erythematosus. Arthritis Rheum 47,421-6. Sorger, P. K., Lewis, M. J., and Pelham, H. R. (1987). Heat shock factor is regulated differently in yeast and HeLa cells. Nature 329,81-4. Sorger, P. K., and Pelham, H. R. (1988). Yeast heat shock factor is an essential DNA- binding protein that exhibits temperature-dependent phosphorylation. Cell 54, 855- 64. 324 References Spronk, P. E., ter Borg, E. J., Limburg, P. C., and Kallenberg, C. G. (1992). Plasma concentration of IL-6 in systemic lupus erythematosus; an indicator of disease activity? Clin Exp Inununol 90, 106-10. Srivastava, P. K. (1993). Peptide-binding heat shock proteins in the endoplasmic reticulum: role in inunune response to cancer and in antigen presentation. Adv Cancer Res 62, 153-77. Srivastava, P. K., Udono, H., Blachere, N. E., and Li, Z. (1994). Heat shock proteins transfer peptides during antigen processing and CTL priming. Immunogenetics 39, 93-8. Stancato, L. F., Chow, Y. H., Hutchison, K. A., Perdew, G. H., Jove, R., and Pratt, W. B. (1993). Raf exists in a native heterocomplex with hsp90 and p50 that can be reconstituted in a cell-free system. J Biol Chem 268, 21711-6. Stephanou, A., Amin, V., Isenberg, D. A., Akira, S., Kishimoto, T., and Latchman, D. S. (1997). Interleukin 6 activates heat-shock protein 90 beta gene expression. Biochem J 321, 103-6. Stephanou, A., Conroy, S., Isenberg, D. A., Maione, D., Poli, V., Ciliberto, G., and Latchman, D. S. (1998). Elevation of IL-6 in transgenic mice results in increased levels of the 90 kDa heat shock protein (hsp90) and the production of anti-hsp90 antibodies. J Autoimmun 11,249-53, Stephanou, A., Isenberg, D. A., Akira, S., Kishimoto, T., and Latchman, D. S. (1998). The nuclear factor interleukin-6 (NF-IL6) and signal transducer and activator of transcription-3 (STAT-3) signalling pathways co-operate to mediate the activation of the hsp90beta gene by interleukin-6 but have opposite effects on its inducibility by heat shock. Biochem J330, 189-95. 325 References Stephanou, A., Isenberg, D. A., Nakajima, K., and Latchman, D. S. (1999). Signal transducer and activator of transcription-1 and heat shock factor-1 interact and activate the transcription of the Hsp-70 and Hsp- 90beta gene promoters. J Biol Chem 274,1723-8. Stephanou, A., and Latchman, D. S. (1999). Transcriptional regulation of the heat shock protein genes by ST AT family transcription factors. Gene Expr 7,311-9. Stephanou, A., Latchman, D. S., and Isenberg, D. A. (1998). The regulation of heat shock proteins and their role in systemic lupus erythematosus. Semin Arthritis Rheum 28, 155-62. Stites, D. P., Terr, A. I., and Parslow, T. G. (1997). Medical Immunology, 9th Edition (Connecticut: Appleton & Lange). Stocklin, E., Wissler, M., Gouilleux, P., and Groner, B. (1996). Functional interactions between Stat5 and the glucocorticoid receptor. Nature 383,726-8. Strous, G. J., van Kerkhof, P., Govers, R., Ciechanover, A., and Schwartz, A. L. (1996). The ubiquitin conjugation system is required for ligand-induced endocytosis and degradation of the growth hormone receptor. Embo J 75,3806-12. Stuart, R. A., Littlewood, A. J., Maddison, P. J., and Hall, N. D. (1995). Elevated serum interleukin-6 levels associated with active disease in systemic connective tissue disorders. Clin Exp Rheumatol 13, 17-22. Studenski, S., Allen, N. B., Caldwell, D. S., Rice, J. R., and Polisson, R. P. (1987). Survival in systemic lupus erythematosus. A multivariate analysis of demographic factors. Arthritis Rheum3 0 ,1326-32. 326 References Sturfelt, G., Bengtsson, A., Klint, C., Nived, O., Sjoholm, A., and Truedsson, L. (2000). Novel roles of complement in systemic lupus erythematosus—hypothesis for a pathogenetic vicious circle. J Rheumatol 27,661-3. Suda, T., O’Garra, A., MacNeil, I., Fischer, M., Bond, M. W., and Zlotnik, A. (1990). Identification of a novel thymocyte growth-promoting factor derived from B cell lymphomas. Cell Inununol 729,228-40. Suzuki, H., Yasukawa, K., Saito, T., Narazaki, M., Hasegawa, A., Taga, T., and Kishimoto, T. (1993). Serum soluble interleukin-6 receptor in MRL/lpr mice is elevated with age and mediates the interleukin-6 signal. Eur J Immunol 23,1078-82. Symmons, D. P., Coppock, J. S., Bacon, P. A., Bresnihan, B., Isenberg, D. A., Maddison, P., McHugh, N., Snaith, M. L., and Zoma, A. S. (1988). Development and assessment of a computerized index of clinical disease activity in systemic lupus erythematosus. Members of the British Isles Lupus Assessment Group (BILAG). Q J Med 69,927-37. Szakal, A. K., Kosco, M. H., and Tew, J. G. (1989). Microanatomy of lymphoid tissue during humoral immune responses: structure function relationships. Annu Rev Inununol 7,91-109. Szczepanski, L., Targonska, B., and Piotrowski, M. (1992). Deforming arthropathy and Jaccoud's syndrome in patients with systemic lupus erythematosus. Scand J Rheumatol 27,308-9. Taga, T., Hibi, M., Hirata, Y., Yamasaki, K., Yasukawa, K., Matsuda, T., Hirano, T., and Kishimoto, T. (1989). Interleukin-6 triggers the association of its receptor with a possible signal transducer, gpl30. Cell 58, 573-81. Taga, T., and Kishimoto, T. (1992). Cytokine receptors and signal transduction. Faseb J 6,3387-96. 327 References Takahashi, I., Tanuma, R., Hirata, M., and Hashimoto, K. (1994). A cosmid clone at the D6SI82 locus on human chromosome 6pl2 contains the 90-kDa heat shock protein beta gene (HSP90 beta). Mamm Genome 5,121-2. Takahashi, T., Tanaka, M., Brannan, C. I., Jenkins, N. A., Copeland, N. G., Suda, T., and Nagata, S. (1994). Generalized lymphoproliferative disease in mice, caused by a point mutation in the Fas ligand. Cell 7(5,969-76. Tan, E. M., Cohen, A. S., Fries, J. F., Masi, A. T., McShane, D. J., Rothfield, N. F., Schaller, J. G., Talal, N., and Winchester, R. J. (1982). The 1982 revised criteria for the classification of systemic lupus erythematosus. Arthritis Rheum 25, 1271-7. Tarassi, K., Carthy, D., Papasteriades, C., Boki, K., Nikolopoulou, N., Carcassi, C., Oilier, W. B., and Hajeer, A. H. (1998). HLA-TNF haplotype heterogeneity in Greek SLE patients. Clin Exp Rheumatol 76, 66-8. Termaat, R. M., Brinkman, K., van Compel, F., van den Heuvel, L. P., Veerkamp, J. H., Smeenk, R. J., and Berden, J. H. (1990). Cross-reactivity of monoclonal anti- DNA antibodies with heparan sulfate is mediated via bound DNA/hi stone complexes. J Autoimmun 3 ,531-45. Theofilopoulos, A. N., and Dixon, F. J. (1981). Etiopathogenesis of murine SLE. Immunol Rev 55, 179-216. Theofilopoulos, A. N., and Dixon, F. J. (1985). Murine models of systemic lupus erythematosus. Adv Immunol 37,269-390. Thompson-Snipes, L., Dhar, V., Bond, M. W., Mosmann, T. R., Moore, K. W., and Rennick, D. M. (1991). Interleukin 10: a novel stimulatory factor for mast cells and their progenitors. J Exp Med 773,507-10. 328 References Tilly, K., Murialdo, H., and Georgopoulos, C. (1981). Identification of a second Escherichia coli groE gene whose product is necessary for bacteriophage morphogenesis. Proc Natl Acad Sci U S A 78, 1629-33. Tissieres, A., Mitchell, H. K., and Tracy, U. M. (1974). Protein synthesis in salivary glands of Drosophila melanogaster: relation to chromosome puffs. J Mol Biol 84, 389-98. Tsoulfa, G., Rook, G. A., Van-Embden, J. D., Young, D. B., Mehlert, A., Isenberg, D. A., Hay, F. C., and Lydyard, P. M. (1989). Raised serum IgG and IgA antibodies to mycobacterial antigens in rheumatoid arthritis. Ann Rheum Dis 48, 118-23. Twomey, B., McCallum, S., Dhillon, V., Isenberg, D., and Latchman, D. (1992). Transcription of the genes encoding the small heat shock protein ubiquitin is unchanged in patients with systemic lupus erythematosus. Autoimmunity 13, 197- 200 . Twomey, B. M., Amin, V., Isenberg, D. A., and Latchman, D. S. (1993). Elevated levels of the 70 kD heat shock protein in patients with systemic lupus erythematosus are not dependent on enhanced transcription of the hsp70 gene. Lupus 2, 297-301. Twomey, B. M., Dhillon, V. B., McCallum, S., Isenberg, D. A., and Latchman, D. S. (1993). Elevated levels of the 90 kD heat shock protein in patients with systemic lupus erythematosus are dependent upon enhanced transcription of the hsp90 beta gene. J Autoimmun 6 ,495-506. Twomey, B. M., McCallum, S., Isenberg, D. A., and Latchman, D. S. (1993). Elevation of heat shock protein synthesis and hsp gene transcription during monocyte to macrophage differentiation of U937 cells. Clin Exp Immunol 93,178-83. Ullman, K. S., Northrop, J. P., Verweij, C. L., and Crabtree, G. R. (1990). Transmission of signals from the T lymphocyte antigen receptor to the genes 329 References responsible for cell proliferation and immune function: the missing link. Annu Rev Immunol 8,421-52. Ullrich, S. J., Robinson, E. A., Law, L. W., Willingham, M., and Appella, B. (1986). A mouse tumor-specific transplantation antigen is a heat shock-related protein. Proc Natl Acad Sci U S A 83,3121-5. van Kerkhof, P., Govers, R., Alves dos Santos, C. M., and Strous, G. J. (2000). Endocytosis and degradation of the growth hormone receptor are proteasome- dependent. J Biol Chem 275, 1575-80. Vanbuskirk, A., Crump, B. L., Margoliash, E., and Pierce, S. K. (1989). A peptide binding protein having a role in antigen presentation is a member of the HSP70 heat shock family. J Exp Med 170, 1799-809. Wakeland, E. K., Wandstrat, A. E., Liu, K., and Morel, L. (1999). Genetic dissection of systemic lupus erythematosus. Curr Opin Immunol 77,701-7. Walport, M. J., Davies, K. A., Morley, B. J., and Botto, M. (1997). Complement deficiency and autoimmunity. Ann N Y Acad Sci 815,267-81. Walport, M. J., Ross, G. D., Mackworth-Young, C., Watson, J. V., Hogg, N., and Lachmann, P. J. (1985). Family studies of erythrocyte complement receptor type 1 levels: reduced levels in patients with SLE are acquired, not inherited. Clin Exp Immunol 59,547-54. Ward, M. M., Marx, A. S., and Barry, N. N. (2000). Comparison of the validity and sensitivity to change of 5 activity indices in systemic lupus erythematosus. J Rheumatol 27,664-70. 330 References Waszczykowska, E., Robak, E., Wozniacka, A., Narbutt, J., Torzecka, J. D., and Sysa-Jedrzejowska, A. (1999). Estimation of SLE activity based on the serum level of chosen cytokines and superoxide radical generation. Mediators Inflamm 5,93-100. Watanabe-Fukunaga, R., Brannan, C. I., Copeland, N. G., Jenkins, N. A., and Nagata, S. (1992). Lymphoproliferation disorder in mice explained by defects in Fas antigen that mediates apoptosis. Nature 356,314-7. Watson, A., and Latchman, D. (1995). The cyclic AMP response element in the calcitonin/calcitonin gene- related peptide gene promoter is necessary but not sufficient for its activation by nerve growth factor. J Biol Chem 270,9655-60. Weinrib, L., Sharma, O. P., and Quismorio, F. P., Jr. (1990). A long-term study of interstitial lung disease in systemic lupus erythematosus. Semin Arthritis Rheum 20, 48-56. Welch, W. J., and Feramisco, J. R. (1982). Purification of the major mammalian heat shock proteins. J Biol Chem 257, 14949-59. Wemer-Washbume, M., Stone, D. E., and Craig, E. A. (1987). Complex interactions among members of an essential subfamily of hsp70 genes in Saccharomyces cerevisiae. Mol Cell Biol 7,2568-77. Whitesell, L., Mimnaugh, E. G., De Costa, B., Myers, C. E., and Neckers, L. M. (1994). Inhibition of heat shock protein HSP90-pp60v-src heteroprotein complex formation by benzoquinone ansamycins: essential role for stress proteins in oncogenic transformation. Proc Natl Acad Sci U S A 97,8324-8. Wick, G., and Xu, Q. (1999). Atherosclerosis—an autoimmune disease. Exp Gerontol 34,559-66. 331 References Wiech, H., Buchner, J., Zimmermann, R., and Jakob, U. (1992). Hsp90 chaperones protein folding in vitro. Nature 358, 169-70. Williams, D. B., and Watts, T. H. (1995). Molecular chaperones in antigen presentation. Curr Opin Immunol 7,77-84. Wilson, J. G., Ratnoff, W. D., Schur, P. H., and Fearon, D. T. (1986). Decreased expression of the C3b/C4b receptor (CRl) and the C3d receptor (CR2) on B lymphocytes and of CRl on neutrophils of patients with systemic lupus erythematosus. Arthritis Rheum 29,739-47. Winfield, J., and Jaijour, W. (1991). Do stress proteins play a role in arthritis and autoimmunity? Immunol Rev 121, 193-220. Winfield, J. B., and Jaijour, W. N. (1991). Stress proteins, autoimmunity, and autoimmune disease. Curr Top Microbiol Immunol 167, 161-89. Winrow, V. R., Ragno, S., Morris, C. J., Colston, M. J., Mascagni, P., Leoni, P., Gromo, G., Coates, A. R., and Blake, D. R. (1994). Arthritogenic potential of the 65 kDa stress protein—an experimental model. Ann Rheum Dis 53, 197-201. Worrall, J. G., Snaith, M. L., Batchelor, J. R., and Isenberg, D. A. (1990). SLE: a rheumatological view. Analysis of the clinical features, serology and immunogenetics of 100 SLE patients during long-term follow- up. Q J Med 74,319-30. Wu, C. (1995). Heat shock transcription factors: structure and regulation. Annu Rev Cell Dev Biol 11,441-69. Wu, C., Wilson, S., Walker, B., Dawid, I., Paisley, T., Zimarino, V., and Ueda, H. (1987). Purification and properties of Drosophila heat shock activator protein. Science 238,1247-53. 332 References Wysenbeek, A. J., Leibovici, L., Weinberger, A., and Guedj, D. (1993). Fatigue in systemic lupus erythematosus. Prevalence and relation to disease expression. Br J Rheumatol 32,633-5. Xu, Q., Kiechl, S., Mayr, M., Metzler, B., Egger, G., Oberhollenzer, P., Willeit, J., and Wick, G. (1999). Association of serum antibodies to heat-shock protein 65 with carotid atherosclerosis : clinical significance determined in a follow-up study [see comments]. Circulation 100, 1169-74. Xu, Y., and Lindquist, S. (1993). Heat-shock protein hsp90 governs the activity of pp60v-src kinase. Proc Natl Acad Sci U S A 90,7074-8. Xu, Y., Singer, M. A., and Lindquist, S. (1999). Maturation of the tyrosine kinase c- src as a kinase and as a substrate depends on the molecular chaperone Hsp90. Proc Natl Acad Sci U S A 96, 109-14. Yost, H. J., and Lindquist, S. (1991). Heat shock proteins affect RNA processing during the heat shock response of Saccharomyces cerevisiae. Mol Cell Biol 11, 1062- 8. Yost, H. J., and Lindquist, S. (1986). RNA splicing is interrupted by heat shock and is rescued by heat shock protein synthesis. Cell 4 5 ,185-93. Yost, H. J., and Lindquist, S. (1988). Translation of unspliced transcripts after heat shock. Science 242,1544-8. Yost, H. J., Petersen, R. B., and Lindquist, S. (1990). RNA metabolism: strategies for regulation in the heat shock response. Trends Genet 6 ,223-7. Youde, S. J., Mower, J., Moore, D. P., and Parkes, A. B. (1998). Stress protein expression in primary and immortalized cultures of human thyroid cells: a model 333 References system for the study of stress proteins in the pathogenesis of autoimmune thyroid disease. Cell Stress Chaperones 3,89-93. Young, J. C., Schneider, C., and Haiti, F. U. (1997). In vitro evidence that hsp90 contains two independent chaperone sites. FEES Lett 418, 139-43. Zheng, X., Rosenberg, L. E., Kalousek, F., and Fenton, W. A. (1993). GroEL, GroES, and ATP-dependent folding and spontaneous assembly of ornithine transcarbamylase. J Biol Chem 268,7489-93. Zou, J., Guo, Y., Guettouche, T., Smith, D. F., and Voellmy, R. (1998). Repression of heat shock transcription factor HSFl activation by HSP90 (HSP90 complex) that forms a stress-sensitive complex with HSFl. Cell 94,471-80. Zugel, U., and Kaufmann, S. H. (1999). Immune response against heat shock proteins in infectious diseases. Immunobiology 201,22-35. Zugel, U., and Kaufmann, S. H. (1999). Role of heat shock proteins in protection from and pathogenesis of infectious diseases. Clin Microbiol Rev 12, 19-39. 334 Appendix A The BILAG Index Original Source Simmons et al., (1988) and adapted from Dhillon et al., (1993b). For further information see Hay et al., (1993). General Non-Specific Manifestations 1. Pyrexia. 2. Unintentional Weight Loss. 3. Lymphadenopathy. 4. Fatigue/malaise/weakness. 5. Anorexia/nausea/vomiting. Score A pyrexia + two other categories. Score B pyrexia or two other categories. Score C any other one category. Score D no current involvement. Score E no current/never any involvement. Mucocutaneous Disease Score A any one of the following categories: 1. Severe maculopapular, discoid or bullous eruption and/or extensive Scarring or causing disability. 2. Angio-oedema. 3. Extensive mucosal ulceration. 335 Appendix A Score B any one of the following categories: 1. Malar erythema. 2. Mild maculopapular eruption. 3. Panniculitis. 4. Localised active discoid lesions including lupus profundus. 5. Severe active alopecia. 6. Subcutaneous nodules. 7. Pemiotic skin lesions. Score C any one of the following categories: 1. Peri-ungual erythema. 2. Swollen fingers. 3. Sclerodactyly. 4. Calcinosis. 5. Telangectasia. 6. Mild alopecia. 7. Small mucosal ulceration. Score D- no current involvement. Score E- no current/never any involvement. Central Nervous System Involvement First Assessment. Score A- acute/progressive/recurring any one of the following categories: 1. Impaired consciousness. 2. Psychosis/delerium/confusional state. 336 Appendix A 3. Grand mal seizure. 4. Stroke/stroke syndrome. 5. Aseptic meningitis. 6. Mononeuritis multiplex. 7. Ascending or transverse myelitis. 8. Peripheral or cranial neuropathy. 9. Chorea. 10. Cerebellar ataxia. Score B any one of the following categories: 1. Headache (severe unremitting). 2. Organic depressive illness. 3. Chronic brain syndrome including pseudotumour cerebri. 4. Disc swelling or cytoid bodies. Score C episodic migrainous headaches. Score D- no current involvement. Score E- no current/never any involvement. Subsequent Assessments: Score A- as for the first assessment (category A1-10), scored “worse” or “new”. Score B as for the first assessment (category Bl-4) scored “worse” or “new” in the last month; or category A 1-3 scored “same” or “improving” in the last month. Score C- as for the first assessment (category Cl); or category A4-10 scored “same’ or “improving” in the last month. 337 Appendix A Score D- no current involvement. Score E- no current/never any involvement. Musculoskeletal Disease Score A- one/both of the following categories: 1. Definite myositis. 2. Severe polyarthritis with loss of function (not responsive to steroids <10 mg/day, antimalarials, non-steroidal therapy). Score B any one of the following categories: 1. Arthritis (definite synovitis). 2. Tendonitis. Score C- any one of the following categories: 1. Arthralgia. 2. Myalgia. 3. Tendon contractures and fixed deformity. 4. Aseptic necrosis. 5. Mild chronic myositis. Score D- no current involvement. Score E no current/never any involvement. 338 Appendix A Cardiovascular and Respiratory Disease Score A- cardiac failure or symptomatic effusion + any two of the following categories; or any four of the following categories: 1. Pleuropericardial pain. 2. Dysponoea. 3. Friction rub. 4. Progressive chest X-ray changes (lung fields). 5. Progressive chest X-ray changes (heart size). 6. ECG evidence of pericarditis or myocarditis. 7. Cardiac erythema including tachycardia: >100 in the absence of fever. 8. Deteriorating lung function: <20% of expected or > 20% fall. 9. Cytohistological evidence of inflammatory lung disease. Score B- any two categories from A 1-9. Score C- mild intermittent chest pain; or any category from A1-9. Score D- no current involvement. Score E- no current/never any involvement. Vasculitis Score A- any one of the following categories: 1. Major cutaneous vasculitis (including ulcers) + infarction in the past month. 2. Major abdominal crisis due to vasculitis. 3. Recurrent thrombo-embolism (excluding stroke). 339 Appendix A Score B any one of the following categories: 1. minor cutaneous vasculitis (nailfold/digital vasculitis, purpura, uticaria). 2. Superficial phlebitis. 3. Thromboembolism (excluding stroke), first episode. Score C any one of the following categories: 1. Raynaud’s phenomenon. 2. Livedo reticularis. Score D- no current involvement. Score E- no current/never any involvement. Renal Disease First Assessment. Score A- two or more of the following categories, including 1,4 or 5: 1. Proteinuria (>1 g/24 hr or S+/4+ dipstick). 2. Accelerated hypertension. 3. Creatinine clearance <50 ml/irm. 4. Active urinary sediment (on a non-centrifuged specimen): pyuria (>5wc/hpf); haematuria (>5rbc/hpf); or red cell casts in the absence of infection. 5. Histological evidence of active nephritis within the last three months. Score B any one of the following categories: 1. One category from A1 -5. 2. Urinary dipstick 2+ or more. 340 Appendix A 3. 24 hr urinary protein >0.5 g Score C- any one of the following categories: 1. Urinary dipstick 1+. 2. B.R >140/90 (5^ phase). 3. Creatinine >130 mmol/1. Score D- no current involvement. Score E no current/never any involvement. Subsequent Assessments. Score A- two or more of the following categories, including 1,4 or 5: 1. Proteinuria (urinary dipstick increased by two or more levels); or 24 hr urinary protein rising from >0.2 g to >1 g; or 24 hr urinary protein rising from > 1 g by 100% or more; or newly documented proteinuria of > Ig. 2. Accelerated hypertension. 3. Deteriorating renal function: plasma creatinine >130 mmol/1 and having risen to >130% of its previous value; or creatinine clearance having fallen to <67% of its previous value; or creatinine clearance <50 ml/mn and last time was >50 ml/nm, or was not measured. 4. Active urinary sediment (defined above). 5. Histological evidence of active nephritis (defined above). Score B any one of the following categories: 1. one category from A1 -5. 2. Moderate proteinuria: urinary dipstick 2+ or more; or 24 hr urinary protein rising from >1 g by >50% but <100%. 341 Appendix A 3. Moderate decline in renal function (plasma creatinine >130mmol/1 or having risen to 115% of previous value. Score C- any one of the following categories: 1. 24 hr urinary protein >0.25 g. 2. Urinary dipstick 1+ or more. 3. Rising B.P. (systolic rise of >30 mmHg or diastolic rise of >15 mmHg, provided recorded values are > 140/90). Score D- no current involvement. Score E no current/never any involvement. Haematological Disease Score A any one of the following categories: 1. Wcc <1000x10^/1. 2. Platelet count <25x1071. 3. Hb <8 g/dl. Score B any one of the following categories: 1. Wcc <2500x10^/1. 2. Platelet count <100x1071. 3. Hb <11 g/dl. 4. Evidence of active haemolysis (raised bilirubin +/- reticulocyte count and positive Coombs’ test). 342