SEQUENCE ANALYSIS OF MONOCLONAL AUTOANTIBODIES
CELIA M LONGHURST
DEPARTMENT OF MOLECULAR PATHOLOGY
UNIVERSITY COLLEGE LONDON
A thesis submitted for the degree of Doctor of Philosophy
in the Faculty of Science
University of London, 1995
ProQuest Number: 10046176
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 10046176
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
Dedication
In loving memory of Alfred Hyde 1915-1994.
Ackno wled gements
I would like to thank Prof. David Isenberg for his kind supervision and valued
friendship and Prof David Latchman for his support throughout the course of this
project. I am grateful to the members of the Bloomsbury Rheumatology Unit, in
particular, Jo Cambridge, Ravi Raviranjan, Tindie Kalsi, Sanj Menon and Michael
Ehrenstein. I would like to give special thanks to Warren Williams for his support and
loyal friendship. I am also grateful to members of the Dept of Molecular Pathology, in
particular, Cyndi Cooper for the gallons of buffers and solutions she made for me and
for keeping my spirits up! I am especially grateful to Caroline Chapman, Southampton
University, for her invaluable technical advice and for her comments on the text of this
thesis. Last but not least, a special thanks to Mark Longhurst who was there when I
needed him most. I would like to acknowledge the Oliver Bird fund and the Gerd
Cohn bequest for funding this project.
Declaration
The work reported in this thesis was prepared by the author unless otherwise
acknowledged
Abstract
Autoimmune diseases are thought to result from the inappropriate activation of the
immune system resulting in damage to tissues in the affected individuals. The presence of circulating auto antibodies is associated with most autoimmune diseases. Systemic lupus erythematosus is characterised by the presence of high titre autoantibodies which bind to a variety of nuclear antigens including DNA. Anti-DNA antibodies have been found deposited in
the disease lesions of SLE patients, thus suggesting that these autoantibodies are involved in disease pathogenesis. Investigation of the genetic origin of autoantibodies is required to see if they are derived from different immunoglobulin genes than those utilised in the production of antibodies to exogenous antigens in healthy adults.
The immunoglobulin variable region genes utilised in a panel of monoclonal auto antibodies derived from three SLE patients with active disease were investigated and compared with immunoglobulin gene usage in autoantibodies derived from vasculitis and
polymyositisNorthern blotting analysis utilising cDNA probes specific for six human VH gene
families revealed a VH4 gene family bias had occurred in the generation of these autoantibodies. The VH4 gene family bias usage was further investigated by sequence analysis. Anti-DNA antibodies of the IgM isotype were found to be expressing immunoglobulin variable genes with close identity to known germline genes. These data support the idea that autoantibodies can be generated with little or no somatic mutation. For
example, the VH4.21 germline gene was utilised by three anti-ssDNA antibodies. The distinctive feature of these autoantibodies was the presence of arginine residues within the heavy chain variable gene CDR3 region. Positively charged amino acid residues such as arginine and lysine have been suggested as important elements in conferring DNA binding
ability in murine anti-DNA autoantibodies. These results show that these amino acids could be involved in DNA binding in human anti-DNA antibodies. The VH 4.21 gene was also found to encode an anti-myeloperoxidase auto antibody derived from a patient with vasculitis.
The anti-MPO antibody E3-MP0 did not bind DNA and did not express the arginine
residues seen in the VH4.21 encoded anti-DNA autoantibodies. Anti-DNA antibodies encoded by other variable region germline genes such as VH4.22 did not contain arginine residues suggesting that other factors apart from the presence of positively charged amino acid residues
are involved in generating DNA binding ability.The studies described here show strong evidence that human monoclonal anti-DNA
autoantibodies resemble those found in murine models of SLE.
Abbreviations
aCL Anti-cardiolipin.
ANCA Anti-neutrophil-cytoplasmic antibody.
ANA Anti-nuclear antigen.
APS Anti-phospholipid syndrome.
BSA Bovine serum albumin.
bp Base pairs.
CA Cold agglutinins.
CDR Complementarity determining regions.
CLL Chronic lymphocytic leukemia.
D Diversity gene segment.
DNA Deoxyribonucleic acid.
DIT Dithiothreitol.
EBV Epstein Barr virus.
EDTA Ethylenediaminetetra-acetic acid
FR Framework region
Id Idiotype.
IPTG Isopropyl-p-D-thiogalactopyranoside.
Jk Joining region gene segment (kappa light chain).
JX Joining region gene segment (lambda light chain).
JH Joining region gene segment (heavy chain).
kb Kilobase.
MHC Major histo compatibility complex.
MPO Myeloperoxidase.
NZB New Zealand black (mice).
NZW New Zealand white (mice)
PBMC Peripheral blood mononuclear cells
PBS Phosphate buffered saline.
PCR Polymerase chain reaction.
RA Rheumatoid arthritis.
RNA Ribonucleic acid.
RNP Ribonucleoprotein.
RT Reverse transcriptase.
SDS Sodium dodecylsulphate.
SLE Systemic lupus erythematosus.
Sm Smith antigen.
TAB Tris acetate EDTA buffer.
TBE Tris borate EDTA buffer.
Taq Pol Taq polymerase.
TEMED N,N,N,N-tetramethylethelene-diamine.
Vk Variable region gene segment (kappa light chain).
VA. Variable region gene segment (lambda light chain).
VH Variable region gene segment (heavy chain).
X-gal 5-bromo-4-chloro-3-indoly-p-galactopyranoside.
Table of Contents
TITLE PAGE.........................................................................................................1
DEDICATION............. ........................................................................................ 2
ACKNOWLEDGEMENTS...............................................................................3
ABSTRACT........................................................................................................... 4
ABBREVIATIONS...............................................................................................5
TABLE OF CONTENTS............................................................................... 7
LIST OF FIGURES....................................................................................... 14
UST OF TABLES...................................... ..........................................................16
CHAPTER 1. INTRODUCTION....................................................................... 17
CHAPTER 2. MATERIALS AND METHODS.............................................. 76
CHAPTER 3. RESULTS ..........................................................................99
CHAPTER 4. DISCUSSION.............................................................................. 169
CHAPTER 5. REFERENCES.............................................................................189
APPENDIX 1.................................................................................................... 216
TNTRODUCTTON
1.1 AUTOIMMUNE DISEASE; CLINICAL CHARACTERISTICS 18
I) SYSTEMIC LUPUS ERYTHEMATOSUS....................................... 18
II) POLYMYOSITIS.............................................................................20
in) VASCULITIS................................................................................... 2 0
IV) ANTIPHOSPHOLIPID SYNDROME.........................................21
1.2 MODELS OF SYSTEMIC LUPUS ERYTHEMATOSUS.......................... 22
1.3 IMMUNOGLOBULIN STRUCTURE..................................................24
I) HEAVY CHAIN ORGANISATION..................................................27
a) V-(D)-JH RECOMBINATION........................................................27
b)THE VH GENE LOCUS......................................................... 28
c)THEVHI FAMILY..........................................................................31
d)THE VH2 FAMILY.......................................................................... 31
e) THE VH3 FAMILY..........................................................................31
f) THE VH4 FAMILY.......................................................................... 31
g)THE VH5 FAMILY.......................................................................... 31
h)THE VH6 FAMILY.......................................................................... 32
i) THE VH7 FAMILY.......................................................................... 32
j)D REGION GENES.................................................................. 32
k)NON-CONVENTIONAL REARRANGEMENTS GENERATED
BY D SEGMENTS....................................................................33
I) JH REGION GENES...................................................................34
II) LIGHT CHAIN ORGANISATION...................................................35
a) KAPPA CHAIN........................................................................... 36
b)LAMBDA CHAIN......................................................................36
1.4 B CELL REPERTOIRE............................................................................... 36
1.5 TOLERANCE AND REGULATION OF B LYMPHOCYTES..................39
1.6 NATURAL AUTO ANTIBODIES..........................................................41
1.7 AUTOANTffiODIES IN SLE; NATURE AND SIGNIFICANCE........... 43
I) ANTI-DNA ANTIBODIES........................................................... 43
n) ANTIBODIES TO RNA...............................................................46
in) EXTRACTABLE NUCLEAR ANTIGEN.....................................46
IV) RHEUMATOID FACTOR..................................................................46
V) ANTI-HISTONE ANTIBODIES...................................................47
VI) ANTIPHOSPHOLIPID ANTIBODIES.........................................48
VH)ANTI-NEUTROPHIL CYTOPLASMIC ANTIBODIES................ 49
1.8 THE IDIOTYPE NETWORK...............................................................51
1.9 CROSS REACTIVE IDIOTYPES FOUND IN AUTOIMMUNE
DISEASE AND MALIGNANCY................................................................ 52
1.10 miOTYPES WHICH ARE ELEVATED IN SLE.................................... 54
1.11 V GENE USAGE IN AUTOIMMUNE DISEASE AND THE FETUS 60
1.12 VH GENE UTILISATION IN THE NATURAL
ANTIBODY RESPONSES................................................................ 63
1.13 THE ROLE OF SOMATIC MUTATION IN THE
GENERATION OF AUTO ANTIBODIES........................................... 64
1.14 V GENE USAGE IN AUTOANTIBODIES.........................................67
I) ANTI-HISTONE ANTIBODIES................................................... 67
II) ANTI-CARDIOLIPIN ANTIBODIES...........................................67
IE) ANTI-DNA ANTIBODIES............................................................6 8
AIM S............................................................................................................. 75
10
CHAPTER TWO
MATERIALS AND METHODS
2.1 MATERIALS............................................................................................... 77
I) BACTERIA.....................................................................................77
II) MONOCLONAL AUTOANTIBODY CELL LINES................... 77
IB) VIRUSES.........................................................................................78
IV) DNA..................................................................................................78
V) DNA PROBES............................................................................. 78
a) Human immunoglobulin constant region................................78
b) Human immunoglobulin variable region probes..................... 78
VI) POLYMERASE CHAIN REACTION PRIMERS........................79
VH) ENZYMES...................................................................................... 79
Vm) RADIOCHEMICALS.................................................................... 79
IX) BUFFERS, SOLUTIONS AND GROWTH MEDIA .............80
X) OTHER MATERIALS AND REAGENTS.................................. 80
2.2 RNA METHODS......................................................................................81
I) RNA ISOLATION............................................................................ 81
a) Large scale preparation.......................................................... 81
b) Small scale preparation (I) ....................................................81
c) Small scale preparation (2)....................................................82
II) SPECTROPHOTOMETRIC ANALYSIS OF RNA.......................82
IE) RNA ELECTROPHORESIS....................................................... 82
rV) NORTHERN BLOTTING............................................................83
V) RNA SEQUENCING................................................................... 85
2.3 DNA METHODS......................................................................................8 8
I) ISOLATION OF PLASMID DNA..............................................8 8
a) SMALL SCALE PLASMID PREPARATION..............................8 8
b) LARGE SCALE PLASMID PREPARATION............................. 89
11
n) GEL ELECTROPHORESIS OF DNA...................................... 90
a) NON DENATURING AGAROSE GELS............................... 90
b) DENATURING POLYACRYLAMIDE GELS..........................90
in) SOUTHERN BLOTTING..........................................................90
IV) RADIOLABELLING OF DOUBLE STRANDED DNA
FRAGMENTS........................................................................... 91
V) ISOLATION OF DNA FRAGMENTS......................................91
VI) cDNA SYNTHESIS................................................................... 92
VH) POLYMERASE CHAIN REACTION........................................93
a) POLYMERASE CHAIN REACTION.................................... 93
b) PURIFICATION OF PCR PRODUCTS.................................94
vm ) CLONING OF PCR PRODUCTS............................................ 94
IX) TA CLONING..............................................................................95
X) PREPARATION OF E. COLI COMPETENT CELLS AND
TRANSFORMATION WITH PLASMID DNA.......................95
XI) SELECTION OF RECOMBINANT PLASMIDS....................... 95
XH) DNA SEQUENCING..................................................................96
Xm) DIRECT SEQUENCING OF PCR PRODUCTS........................96
XTV) DOUBLE STRANDED SEQUENCING....................................97
XV) SINGLE STRAND DNA RESCUE.......................................... 98
XVI) GENE SEQUENCE ANALYSIS..................................................... 98
12
CHAPTER THREE
RESULTS
3.1 VH GENE USAGE OF MONOCLONAL AUTOANTIBODIES
DETERMINED BY NORTHERN BLOTTING......................................100
3.2 ANTI-DNA AUTOANTIBODIES ENCODED BY THE VH4.21
VARIABLE HEAVY CHAIN GERMLINE GENE............................... 107
3.3 AN ANTI-MYELOPEROXIDASE MONOCLONAL ANTIBODY
ENCODED BY THE VH4.21 GERMLINE GENE............................. 120
3.4 A MONOCLONAL AUTOANUBODY ENCODED BY THE
GERMLINE GENE VH4.22............................................................... 127
3.5 MONOCLONAL AUTOANTIBODIES ENCODED BY THE
VH4.18 GERMLINE GENE............................................................... 133
3.6 MONOCLONAL AUTOANTIBODIES ENCODED BY THE
HV71-2 GERMLINE GENE...............................................................140
3.7 MONOCLONAL ANTI-DNA ANTIBODIES ENCODED
BY THE 56P1 GERMLINE GENE................................................... 146
3.8 MONOCLONAL AUTOANTIBODIES ENCODED BY THE
VH26 GERMLINE GENE.................................................................. 151
13
CHAPTER FOITR
DISCUSSION
4.1 VH GENE FAMILY USAGE IN AUTOANTIBODIES..........................170
4.2 SEQUENCING METHODOLOGY.............................................................. 173
4.3 UTILISATION OF THE VH4.21 GENE IN AUTOANTIBODIES......... 174
4.4 AUTOANTIBODIES ENCODED BY VH4 GENES IN GENERAL.........177
4.5 AUTOANTIBODIES ENCODED BY 56P1 AND VH26 GENES............178
4.6 AMINO ACID RESIDUES ASSOCIATED WITH
BINDING TO AUTOANTIGENS......................................................... 181
4.7 THE PATHOGENIC POTENTIAL OF POLYREACTIVE
IgM AUTOANTIBODIES.......................................................................183
4 . 8 POTENTIAL DYSFUNCTION IN B CELLS PRODUCING
AUTOANTIBODIES................................................................................. 184
4.9 THERAPY..................................................................................................... 185
4.10 FURTHER WORK....................................................................................186
4.11 CONCLUSIONS.........................................................................................188
14
List of Figures
Chapter 1
Figure:-1.1 The structure of an immunoglobulin molecule..........................25
Chapter 3
Figure:- 3.01 Northern blots hybridised with the VHl and Cp. probes............. 103
Figure:- 3.02 Northern blots hybridised with the VH3 and VH5 probes 104
Figure:- 3.03 Northern blots hybridised with the VH4 probes.......................105
Figure:- 3.04 Northern blots hybridised with the VH2 and VH6 probes......... 106
Figure:-3.05 RT79 heavy chain variable..region......................................... 108
Figure:-3.06 RT79 kappa chain variable region......................................... 109
Figure:-3.07 RT84 heavy chain variable region......................................... 110
Figure:-3.08 RT84 kappa chain variable region......................................... I l l
Figure:- 3.09 IgM VH4.21 encoded anti-DNA heavy chains compared
with cold agglutinins which do not bind DNA..................... 113
Figure:-3.10 D5 heavy chain variable region........................................ 115
Figure:- 3.11 D5 IgG heavy chain variable region versus T14 an
anti-DNA VH4.21 encoded monoclonal................................117
Figure:-3.12 D5 kappa chain variable region........................................ 119
Figure:-3.13 E3 MPO heavy chain variable region....................................... 121
Figure:- 3.14 E3 MPO heavy chain variable region verses MPO
negative cold agglutinins.......................................................123
Figure:-3.15 E3 MPO light chain variable region........................................ 124
Figure:-3.16 RT72 heavy chain variable region......................................... 128
Figure:- 3.17 RT72 heavy chain variable region compared with RT72 id
positive heavy chain variable region................................... 129
Figure:-3.18 RT72 kappa chain variable region......................................... 130
Figure:- 3.19 RT72 kappa chain variable region compared with 31 id
positive kappa chain variable regions....................................132
Figure:-3.20 RT55 heavy chain variable region......................................... 134
15
Figure:- 3.21 RT55 heavy chain variable region compared with VAR24
heavy chain............................................................................. 135
Figure:-3.22 RT55 kappa chain variable region........................................ 137
Figure:- 3.23 RT55 kappa chain variable region compared with RT129
light chain.............................................................................. 138
Figure:- 3.24 YAR 24 heavy chain variable region ..................................... 139
Figure:-3.25 RT129 heavy chain variable region ....................................... 141
Figure:- 3.26 RT129 heavy chain variable region compared to BEG-2
and 2A4 heavy chains.................../......................................143
Figure:-3.27 RT129 kappa chain variable region........................................ 145
Figure:-3.28 WRI 176 heavy chain variable region ................................... 147
Figure:- 3.29 WRI 176 heavy chain variable region compared 56P1
encoded anti-DNA heavy chains...........................................148
Figure:-3.30 WRI 176 kappa chain variable region.................................... 150
Figure:-3.31 RT6 heavy chain variable region............................................152
Figure:- 3.32 RT6 heavy chain variable region compared to WRI170
VH region .............................................................................153
Figure:-3.33 RT6 light chain variable region..............................................154
Figure:- 3.34 RT6 light chain variable region compared to WRI 170
VL region................................................................................ 155
Figure:-3.35 B3 heavy chain variable region............................................ 157
Figure:-3.36 B3 light chain variable region................................................158
Figure:-3.37 B3 light chain compared to RT6 and PV Il light chains..............159
Figure:- 3.38 UK4 heavy chain variable region...........................................162
Figure:- 3.39 UK4 variable heavy chain compared with VH26 encoded
monoclonal autoantibodies.................................................... 163
Figure:- 3.40 UK4 light chain variable region.............................................165
Figure:-3.41 UK4 light chain compared to B3 and RT6 light chains............... 166
16
List of Tables
Chapter 1
Table:- 1.1
Table:- 1.2
Table:- 1.3
Table:- 1.4
Chapter 3
Table:- 3.1
Table:- 3.2
Table:- 3.3
Table:- 3.4
Autoantibodies in SLE............................................................19
Number of VH families......................................................... 30
Characterisation of human IgM monoclonal anti-DNA
antibodies ..............................................................................70
Characterisation of human IgG monoclonal anti-DNA
antibodies ..............................................................................73
Antigen specificity and Id expression of monoclonal
autoantibodies......................................................................... 1 0 1
Replacement mutations in E3 MPO V regions........................126
Replacement mutations in B3 V regions............................... 160
Replacement mutations in UK4 V regions ............................. 168
17
CHAPTER ONE
INTRODUCTION
18
INTRODUCTION
1.1 AUTOIMMUNE DISEASE; CLINICAL CHARACTERISTICS
Autoimmune diseases are thought to result from the inappropriate activation of
the immune system resulting in damage to tissues of the affected individual. The
presence of circulating autoantibodies is associated with most autoimmune diseases,
although the mechanism by which they help to induce clinical expression of the disease
is still unclear. The aim of this thesis was to investigate the molecular relationships of
the autoantibodies involved in autoimmune rheumatic diseases in order to gain greater
insight as to their origins and role in pathogenesis.
D SYSTEMIC LUPUS ERYTHEMATOSUS
Systemic lupus erythematosus (SLE) is a disease which predominantly affects
women, especially of child-bearing age, found in all racial groups though with
particularly high frequency (up to 1:250) amongst the black population in the West
Indies and the USA. Joint pain or frank arthritis, light sensitive or butterfly rashes,
severe fatigue and Raynaud’s phenomenon (a triphasic colour response to cold which
affects the hands and feet causing them to go white, blue and red) are frequently
present. Inflammation of the pericardial and pleural tissues causing effusions and
shortness of breath are also common. However, it is central nervous system and
kidney involvement which are responsible for the highest mortality. The ten year
survival rate is approximately 85%. This falls to 60% only for those with major kidney
disease.
The precise aetiology of SLE remains uncertain. However correlations with
endocrine disorders, genetic background and environmental influences have been
noted. Women are approximately ten times more likely to develop SLE compared with
males. Amongst monozygotic twins the concordance rate for SLE reaches 70% whilst
dizygotic twins have a concordance rate of only 15%. The Major Histocompatiblity
Complex (MHC) haplotypes Al, B8 , and DR3 are particularly associated with SLE in
Caucasians.
The broad diversity amongst the clinical features is matched by a seemly wide
19
Table 1.1
Autoantibodies in SLE
Antibodies to cell nucleus components Anti-nuclear antibodies Anti-dsDNA antibodies Anti-ssDNA antibodies Anti-dsRNA antibodies (including
poly A, -poly C, -poly I, -poly U) Anti-ssRNA antibodies (including
poly-rU, -poly-rA) Anti-poly(ADP-ribose) antibodies Antibodies to extracellular nuclear
antigens, anti-ribonucleoprotein antibodies, anti-Sm antibodies
Antibodies to cytoplasmic antigens Anti-SS-A (Ro) antibodies Anti-SS-B (La) antibodies
Cell-specific autoantibodies Lymphocytotoxic antibodies Anti-neurone antibodies Anti-erythrocyte antibodies Anti-platelet antibodies
Antibodies to serum components Lupus Anticoagulant Antiglobulins (rheumatoid factor)
20
variety of detectable circulating autoantibodies (Table 1.1). Recent studies using
human hybridoma derived monoclonal antibodies have suggested that some of these
autoantibodies are crossreactive (e.g. anti-DNA and anti-phospholipid antibodies), and
thus the variety of specificities recognised by autoantibodies may be greater than was
originally thought. Of the antibodies found in sera from patients with SLE, those
binding to double-stranded DNA especially in high titre are generally regarded as
disease specific and in some cases most closely reflect disease activity. Despite over
thirty years of intensive research, many questions regarding the origin, structure and
pathogenic potential of autoantibodies, especially anti-DNA autoantibodies and their
possible role in pathogenesis remain unanswered.
ID POLYMYOSmS
Polymyositis is an inflammatory disease of skeletal muscle of unknown
aetiology and pathogenesis. When the disease is accompanied by the presence of a
characteristic skin rash the term dermatomyositis is used. The clinical diagnosis of
polymyositis is made on obtaining a clinical history of proximal muscle weakness
which may occasionally be associated with pain and tenderness. Pathological
indicators include raised muscle enzymes in the plasma, myopathic features on
electromyography and perivascular cell infiltration with or without muscle fibre
degeneration on muscle biopsy (Edwards et al. 1981). Circumstantial evidence such as
elevated serum globulins, a variety of circulating autoantibodies, and a clinical
association with the other autoimmune rheumatic diseases, supports the notion that
poly and dermatomyositis are also part of the spectrum of the autoimmune disorders
The female to male preponderance of the disease is nearly 3:1 especially where
there are coexistent features of another ‘collagen-vascular disease’. Myositis most
commonly presents in an insidious fashion with weakness and wasting developing
over weeks, months or even years.
ffl) VASCULITIS.
Vasculitis is a condition in which inflammation of the blood vessels is the main
pathological feature. It includes clinical syndromes such as Wegener’s granulomatosis,
microscopic polyangiitis and polyarteritis nodosa. Initial blood vessel damage in
thought to be caused by neutrophils followed by a chronic inflammatory response to an
21
as yet unidentified target antigen in the blood vessel wall (Fauci etal. 1983).
In primary vasculitis approximately 50% of the patients have circulating
antibodies to neutrophil granule proteins (anti-neutrophil cytoplasmic antibodies -
ANCA). The main antigens recognised by ANCA are serine protease 3 and
myeloperoxidase. Although vasculitis may also occur as part of the clinical spectrum in
autoimmune rheumatic diseases such SLE, RA, and Sjogren’s syndrome, these
autoantibodies are rarely associated with vasculitis in this group of patients (Cambridge
etal. 1994a).
Since their discovery in 1982, a pathological role for ANCA has been
suggested but remains unconfirmed. The hypothesis has been supported by close
association of ANCA with vasculitis, particularly anti proteinase 3 (PR3)
autoantibodies in patients with Wegener’s granulomatosis, and the correlation of ANCA
titers with disease activity as well as ANCA isotypes with disease expression (reviewed
by Kallenberg et al 1994). In vitro, ANCA have been shown to provide an additional
activating signal to neutrophils pre-primed with cytokines; resulting in an increased
respiratory burst, degranulation and an increased ability to damage cultured endothelial
cells.
TV) ANTIPHOSPHOLIPID SYNDROME
Autoantibodies reactive with phospholipids such as anticardiolipin antibodies
(aCLs) (responsible for the false positive serological test for syphilis) and
autoantibodies reactive with phophatidylserine, phosphatidylinositol or phosphatidic
acid have received much attention over the last few years. This is not due only to their
relatively high prevalence in patients with SLE, but also because of the strong
association of anticardiolipin antibodies with recurrent episodes of venous and arterial
thrombosis, spontaneous fetal loss and thrombocytopenia (Love et al. 1990,
Sammaritano et a l 1990). A solid phase anticardiolipin assay was developed in the
early 1980’s in order to study sera from SLE patients presenting with thrombosis,
recurrent spontaneous abortions and cerebral disease (Hughes et al. 1986, Gharavi et
al. 1987) a close relationship between anticardiolipin antibody detection and the
presence of lupus anticoagulant was then demonstrated (Triplett et al 1988). The high
correlation between IgG isotype anticardiolipin antibodies and clinical thrombosis was
first documented in 1986 (Harris et a l 1986). The frequency of fetal losses and
22
thrombocytopenia was also increased in this group (Cervera et al 1990, Deleze et al
1989). By contrast only a minority of patients demonstrated high levels of IgM
anticardiolipin antibodies although thrombotic complications were also found in this
sub-group of patients. These findings led to the recognition of a condition later termed
the “antiphospholipid syndrome” (Derve etal 1985) which encompassed patients with
thrombosis, recurrent abortions and/or thrombocytopenia, elevated anticardiolipin
antibodies and lupus anticoagulant.
Antiphospholipid syndrome can occur in patients with lupus. However, this
syndrome can also occur without clinical or serological features of lupus and these
patients are defined as having primary antiphospholipid syndrome (Alarcon et a l
1989a, Asherson e ta l 1988, Asherson e ta l 1989). The prevalence of anticardiolipin
antibodies in patients with SLE varies between 20-50% (Alarcon eta l 1989b, Cervera
et a l 1990, Cronin et al 1988). Patients with persistently moderate to high titres of
anticardiolipin antibodies of the IgG isotype seem to be at higher risk of developing
clinical features of antiphospholipid syndrome (Cervera eta l 1990, Harris etal 1986).
1.2 MODELS OF SYSTEMIC LUPUS ERYTHEMATOSUS
A number of animal models of human lupus have been described which reflect
different aspects of the disease. Perhaps the best known model is that found in the
New Zealand Black mouse (NZB) first derived 30 years ago by Marianne
Bielschowsky and her colleagues at the University of Dunedin New Zealand
(Bielschowsky et al 1959). It was selected for inbreeding on the basis of a solid black
coat colour. These animals are principally a model of autoimmune haemolytic anaemia
which develops at the age of 9 months and by 12 months virtually all of the animals
have antibodies to ei*ythrocytes detectable by a Coombs test. As well as anti-red cell
antibodies, a number of other autoantibodies were detectable including anti-ss and
dsDNA. Kidney disease may also develop in these animals notably a membranous
form of glomerulonephritis. Indirect immunofluorescence studies showed deposition
of IgG on the glomerular basement membrane. These animals have a 50% survival rate
at 18 months and by this time have oflen developed splenomegaly, lymphoid
hyperplasia and invariably have circulating immune complexes.
23
A hybrid strain derived from the NZB mouse is produced by mating it with the
New Zealand White mouse. The offspring are known as (NZB/NZW) FI. This
hybrid animal is in many respects an excellent model of human lupus. As with the
human disease it is the female which is most likely to develop the disease and in a more
severe form. In these mice, the symptoms become apparent around the age of 6
months and most animals die from immune complex nephritis by 9 months of age. In
males the disease becomes apparent after 1 0 - 1 2 months with most animals dying at
around 15 months. Renal disease may be detected by the presence of proteinuria as
early as 3 months of age and this is often accompanied by the presence of anti nuclear
antibodies (ANA). Focal glomerulonephritis with immunoglobulin and complement
deposition along the mesangium and capillary basement membrane is commonly found.
The renal damage increases with age and eventually the basement membrane becomes
fibrous and necrotic with epithelial cells filling the capsular space. Serologically,
antibodies to ds and ssDNA, RNA and a number of synthetic polynucleotides and
erythrocytes have also been described. The class of antibody appears to be important
in determining the development of the disease. Up to the age of 6 months the NZB/W
female produces DNA antibodies mostly of the IgM isotype. Isotype switching to IgG
correlates with the onset of renal disease (Steward et al. 1976).
Hypergammaglobulinaemia is present in both the NZB and NZB/W strains and in the
latter infiltration of the lacrimal and parotid glands occurs indicating that they have
developed a disease analogous to Sjogren’s syndrome in humans.
MRL/1 mice were first described in 1978 by Murphy and Roths and has also
been proposed as a suitable animal model for lupus in humans. Like the NZB/W FI,
the MRL/1 develops a fatal immune complex glomerulonephritis. The life span of the
animals is shorter in females, approximately 50% having succumbed by 6 months of
age. The most notable feature of these animals is a massive lymphoproliferation
especially in the peripheral lymph nodes of the axillary and cervical regions. A small
number, perhaps 1 0 -2 0 %, develop articular lesions which bear some resemblance to
rheumatoid arthritis. Other features include vasculitic skin lesions, hair loss and
necrotizing arteritis. As with the NZB mouse a wide variety of autoantibodies binding
to nuclear antigens such as ss and dsDNA and the Sm antigen has been described.
Hypergammaglobulinaemia, hypocomplementaemia, raised circulating immune
complex levels, rheumatoid factor and cryoglobulins are also found in the serum of
24
these animals. Early thymic involution is also seen in these animals and there are major
defects within the immunoregulatory circuits of the T lymphocytes. The lymphocytes
in the swollen nodes are mostly weakly staining Ly-1 (helper cells) which appear to be
insensitive to signals from Ly-1,2,3 cells which are responsible for feedback inhibition.
As with other autoimmune strains, the MRL/1 also exhibits impaired production of
interleukin 2. A large number of macrophages bearing la antigens are found in these
animals, indicating that these cells are activated and antigen presentation may occur at
an unwarranted rate.
BXSB mice are a recombinant inbred strain resulting from the crossing of
C57BL/6J and females with SB/le males. Interestingly it is the males of the strain that
tend to develop a more typical picture of murine lupus. In this respect an analogy may
be drawn with some famihes of lupus patients as described by Lahita et al. (1983) who
clearly showed that lupus could run in families involving male rather than female
members. These animals develop haemolytic anaemia splenomegaly lymphadenopathy
and hypergammaglobulinaemia. The life span is approximately 5 months and they
usually die because of severe exudative proliferative nephritis. Serologically signs of
autoimmunity may be detected at two months, when rheumatoid factor anti-DNA and
anti-erythroctye antibodies can be found. Furthermore circulating immune complex
levels are raised and hypocomplementaemia is usually noted.
1.3 IMMUNOGLOBULIN STRUCTURE
The basic unit of secreted immunoglobulin, is composed of two identical heavy
(H) chains and two identical light (L) chains which are arranged as a pair of heavy and
light chain dimers joined by disulphide bonds (Figure 1.1). Each chain consists of
separate constant and variable domains that form the antigen binding site and one or
more constant region domains that mediate various effecter functions (p.,5, y, a and e in
the heavy chain, k and X in the light chain) (Wu and Kabat 1970, Polijak et al 1974).
Different subclasses of heavy chains have developed in different mammalian species,
which although sharing similar designations may not be functionally related. In man
subgroups of the heavy chains y and a are found:- y y2 , Y3 ..Y4 and 0 .2 . These
subgroups endow each immunoglobulin isotype with different effecter functions. For
25
Figure 1.1
STRUCTURAL ORGANISATION OF AN IMMUNOGLOBULIN MOLECULE
Antigen binding site
CH2
CH3
Constant region
VH
1. CDRl2. CDR23. CDR3
The combination of the VH and VL regions form the antigen binding site. The VH and VL domains are shown.The approximate locations of the CDR regions in relation to VH, DH, JH, VL and JL are shown.
26
example the ability to bind complement or to bind to cell surface Fc receptor.
The N terminal portion of each chain comprising approximately 110 amino
acids forms the variable region of the molecule. During B cell development, a germline
repertoire of fewer than 250 gene segments is used to generate a much larger repertoire
of variable domain structures (Tonegawa et al. 1983, Yancopoulos and Alt 1986).
Combinatorial joining of VH, DH and JH segments (for H chains) coupled with the
joining of VL and JL gene segments (for L chains) yields more than 10^ different
antibody molecules (Tonegawa et at. 1983, Yancopoulos and Alt 1986). The light
chain variable regions are encoded by two gene segments variable light (VL) and
joining light (JL). The VL and JL gene segments are divided into two groups kappa
(k ) and lambda (X) and are rearranged in a similar manner to the heavy chain genes.
Variation in the precise point of segmental joining with insertion of templated (P
junctions) and nontemplated (N regions) nucleotides in the heavy chain exponentially
amplifies the potential diversity of the preimmune repertoire (Lieber 1992). Further
somatic diversification can be achieved by a hypermutational process of insertions and
single base substitutions in rearranged antibody variable genes is associated with
exposure to antigen (Tonegawa et al. 1983). Primary amino acid sequence
comparisons have shown that heavy and light chain variable domains contain three
intervals of sequence hypervariability (termed complementarity regions or CDRs) that
are separated from each other by four intervals of relatively constant sequence (termed
frameworks or FRs) (Kabat et al. 1991). CDR’s 1 and 2 and FR 1,2 and 3 are
encoded entirely by the V region segment, FR4 by the J and CDR3 is the product of
V(D)J joining. The framework regions which consist of relatively conserved amino
acid sequences make up the framework structure of the antibody and contribute to
shape and stability. The complementarity determining regions in which the amino acid
sequences vary enormously from one immunoglobulin to another interact to form the
antigen contact area.
The genes for heavy, k and X chains are located on separate chromosomes, 12,
6 and 16 in the mouse, 14, 2 and 22 in man. Although the VH locus is located on
chromosome 14q32.3, several orphon VH genes and two orphon DH genes were
assigned to chromosome 15 and 16 (Tomlinson etal. 1994). Recently two orphons
HC16-15 and HC16-16 have appeared to be to accessible for recombination and have
27
been isolated from fetal liver as germline transcripts (Cuisinier et al 1993) Thus, the
potential exists for variable genes located on chromosomes other than the established V
gene clusters (on chromosome 14) contributing to the generation of immunoglobulin
molecules.
D HEAVY CHAIN ORGANISATION
a) V-(DVJH RECOMBINATION
There are conserved sequence motifs which flank all the variable
immunoglobulin genes known to recombine. The consensus of a recombination signal
sequence consists of a dyad-symmetric heptamer sequence directly adjacent to the
coding element and an A/T rich nonamer separated from a heptamer by a spacer region
of nonconserved nucleotides (Tonegawa 1983). There are two types of recombination
signal sequence which are defined by the length of the spacer. These are either 12(+/-)
or 23(4-/-) nucleotides long and efficient joining requires one recombination signal
sequence of both types e.g all gene segments of the same kind have the same
configuration of recombination signal sequences and joining partners have the opposite
type of signal sequence.
The first gene rearrangement during B cell development is heavy chain D to JH
rearrangement which usually occurs on both heavy chain gene alleles (Schatz et al
1992) . Precursor B cells then undergo VH to DJH rearrangement on one or both
alleles generating potentially functional VDJH heavy chain genes. During
rearrangement the intervening DNA is deleted. VH to D rearrangement were first
thought not to occur on alleles that have not first undergone D to JH rearrangement.
{PAietal 1984, Schissel e ta l 1991). However, VH-DH and DH-JH joints were
found in a human B cell line and it was proposed that a VH-DH product may
recombine with either a DH-JH or with a JH gene segment to give rise to a functional
rearrangement (Shin etal 1993).
The next rearrangement is light chain V k to Jk. The Ig k locus rearrangements
occasionally precede productive heavy chain gene rearrangement in normal human bone
marrow (Kubagawa e ta l 1989). A fraction of B cells, often ones that have deleted
one or both of their k loci then go on to rearrange some of their X light chain alleles.
28
(Seising gf a/. 1989).
The end product of this developmental lineage is a mature B cell with a unique
antigen receptor complex consisting of IgM and several accessory proteins on its cell
surface (Reth etal. 1991). The observation that in a given B cell only a single heavy
and light chain proteins are expressed on the cell surface is termed allelic exclusion
(Pemis etal. 1965).
b) THE VH GENE LOCUS
The first immunoglobulin heavy chains were sequenced at the protein level in
1969 and 1970 (Edelman et al. 1969, Cunningham et al. 1969, Winkler et al. 1969).
When these sequences were compared it became apparent that they could be arranged
into V region subgroups. The first sequenced protein (Eu) became the prototypic
member of the VHl subgroup and proteins Ou, He, Daw, and Cor were related to each
other but not to the Eu sequence and thus were designated the VHII group. Later, the
proteins Tie, Was, Jon, Zap, Tur, Nie, and Gal were sequenced and formed the VHin
subgroup.
In the early 1980s the first nucleotide sequences of human immunoglobulin
variable regions became available, and these sequences corresponded to the protein
sequences of VHI, VHII, or VHIII. On the basis of amino acid sequence analysis, it
was initially thought that human heavy chain variable regions fell into three groups of
molecules (Rechavi et al. 1983). There is no universally accepted definition of a VH
gene family, however molecular biological techniques have allowed a method of
definition, that is, VH segments that cross-hybridize by Southern filter hybrization
under standard conditions (O.IX saturated sodium citrate, 0.1% sodium dodecyl
sulphate, 65°C) are members of the same VH gene family. In practical terms this
implies that 80% nucleotide sequence identity places two genes within the same family
and less than 70% nucleotide homology classifies the genes within separate families.
The functional VH segments are located on chromosome 14q32.3 (Tonegawa 1983,
McBride et al. 1982). VH segments with an open reading frame have been located on
chromosome 15 and 16 ( Cherif etal. 1990, Matsuda etal. 1990, Tomlinson etal.
1994 ).
The VH region genes have been grouped into families on the basis of
29
nucleotide sequence. Seven human VH families have been described (Tomlinson et al
1992, Willems van Dijk et al 1993). The number of unique VH elements within each
family were initially estimated by Southern blotting digested human DNA (Kodaira et
al 1986, Lee et al 1987, Berman et al 1988). The largest of these is the VH3 family
with approximately 48 family members, 20 of these being pseudogenes (Cook et al
1994). and the smallest being VH6 with only one member (Berman et a l 1988).
Unlike the mouse, members of the different human VH families are interspersed
throughout the locus (Kodaira et a l 1986, Berman et a l 1988). It is not clear if the
extensive intermingling is a random affect of evolution or if it offers some selective
advantage.
The single VH6 member family has been identified as the most JH proximal in
the human locus. Upstream from this are members of the VH5 family which have
been found to be intermingled. The VH4 gene families seem relatively interspersed as
well. This situation is unlike that of the mouse where a more ordered VH family
grouping occurs. Walter et al (1990) have used a two dimensional mapping procedure
to assign VH containing restriction fragments to a region 1,100 kilobases upstream
from the JH segments. Overlapping yeast artificial chromosomes (YACs) and cosmids
have been utilised to generate a map of a region 800 kb upstream of the JH segments
(Shin et al 1991, Matsuda eta l 1993) Within this region there is a 50 kb insertion
polymorphism which contains an additional five VH segments (Walter eta l 1993). A
total of 50 functional VH segments can be accounted for within these maps including
an equivalent number of pseudogenes (Tomlinson e ta l 1992, Matsuda e ta l 1993).
The map of the VH locus was recently completed by Cook et a l (1994) and the
estimated number of VH gene segments on the composite map are described in Table
1.2. The map of the VH locus contains 87 VH segments of which 46 are functional
(Cook et a l 1994). The mapped segments have been compared with a database of
human immunoglobulin germline VH segments (Tomlinson eta l 1992) and nearly all
the functional segments are accounted for thus giving evidence for the completeness of
this map.
30
Table 1.2
Number of VH families
Segments with open reading frames
Seen as VDJ Not seen as VDJ Pseudogenes Not sequenced Total
VHl 8 1 4 1 14VH2 3 0 1 0 4VH3 22 4 20 2 48VH4 10 0 1 1 12VH5 1 1 0 0 2VH6 1 0 0 0 1VH7 1 1 3 1 6
Total 46 7 29 5 87
Adapted from Cook et al 1994
31
cl THE VHl FAMILY
The first VHl family member was first described in 1969 (Edelman et al.
1969). In 1980 the first nucleotide sequence of an immunoglobulin human VH gene
was shown to be homologous to the VHl Eu protein and thus the VHl family was
identified. At present there are thought to be fourteen VHl members, four of which are
pseudogenes (Cook et a l 1994). The VHl family is much smaller than was first
thought and is now the third largest human VH gene family.
d) THE VH2 FAMILY
The first VH2 family member was also first described at the protein level
(Winkler et al. 1969). The first gene reported in this family at the nucleotide level is
referred to as VCE-1 and was isolated from a rearranged gene in a B cell tumour
(Takahashi et al. 1982,1984). The VH2 family is estimated to contain four members
including one pseudogene and is now considered to be one of the smaller human VH
gene families. {Cock etal 1994)
e) THE VH3 FAMILY
The first VH3 family member was the Nie protein sequenced in the Hilshmann
laboratory (Cunningham et al 1969). At the present time the VH3 family is thought to
be the biggest VH gene family containing forty-eight members twenty of which are
pseudogenes. (Cook a/. 1994).
n THE VH4 FAMILY
The VH4 family was defined by Lee et al. (1987) from two nucleotide
sequences first described by Kodaira et a l (1986). At present VH4 is thought to
consist of twelve members including one pseudogene. (Cook etal. 1994). In terms of
functional members, VH4 is considered to be the second largest human VH gene
family.
gl THE VH5 FAMILY
The VH5 family was first described in studies of the rearranged gene of patients
with chronic lymphocytic leukemia (Shen et a l 1987). The VH5 family showed only
distant homology with the four other VH gene families and on Southern filter
32
hybrization the VHV probe detected two to four fragments. Two VH5 genes have been
described (Cook gf aZ. 1994).
hi THE VH6 FAMTI.Y
The VH6 gene was first described simultaneously by three different laboratories
(Buluwela and Rabbitts 1988, Berman etal 1988, Schroeder etal. 1987). Sanz etal.
(1989b) described several additional VH6 gene sequences. When all these sequences
were published it was apparent that they were all identical which emphasises the
conservation of this gene. Buluwela and Rabbitts (1988) observed that VH6 was the
most V proximal gene segment and several laboratories have confirmed this (Schroeder
et al 1988). This is of great interest as in the murine system it has been shown that the
proximity to D and J is critical to the expression of the immunoglobulin repertoire.
Meek et a l (1991) described the structures of the VH6 gene in a number of higher
primates and found that there was no sequence variation between these species whereas
in humans there was less than 2% nucleotide sequence variation. Therefore there is
very little variation in sequence or position between species and shows remarkable
conservation. To date only one member of the VH6 family has been isolated (Cook et
al 1994).
i) THE VH7 FAMILY
The VH7 family was first recognised in a cDNA library generated from
neonatal cord mononuclear cells (Mortari et a l 1992). Three independently derived
clones shared more than 90% homology but had only 78-82% identity with the VHl
family. These findings justified the creation of the VH7 gene family. At present there
are thought to be six members in the VH7 family, although only one of these has been
seen in a VDJH rearrangement (Cook etal 1994). Three of the members are believed
to be pseudogenes (Cook et al 1994).
OD REGION GENES
The term D segment was first proposed by Schilling et a l (1980) to highlight
the “diversity” found in the heavy chain of anti-a-1.3 dextran antibodies from position
+99 to the beginning of the JH segment spanning the third complementarity region
33
(CDR3). (Kehoe and Capra 1971). CDR3 has a crucial role in determining antigen
specificity. D gene segments contribute to heavy chain diversity in a number of ways
such as generating unique amino acid residues at the V-D and D-J junctions during the
process of rearrangement, undergoing somatic mutation and/or gene conversion, and
through nonconventional mechanisms such as D-D recombinations.
The human D gene segments are located in chromosome band 14q32. (Cox et
al. 1982, Kirsch graZ. 1982). Siebenlist er a/. (1981) identified a human D gene family
DLR whose members were encoded at regular intervals of 9 kb along a 33 kb stretch.
A unique D segment (which is the human equivalent of the mouse DQ52) was
described within the JH locus by Ravetch etal (1981). This D segment was mapped
25 nucleotides upstream of the JHl nonamer sequence. Mapping analysis located a
major D locus between VH and JH flanked by DLR4 and the DQ52 gene segments 5’
and 3’ respectively (Siebenlist eta l 1981, Buluwela et a l 1988b). There is a 20 kb
separation between the DLR4 gene segment and the closest VH gene (the single
member of the VH6 family) (Sato et a l 1988) implying that the major D locus spans
around 70kb. Several D segments have been found to be interspersed with the VH
segments (Zong etal 1988, Bulwela etal 1988b, Matsuda eta l 1988).
The D gene segments have been classified into families based on nucleotide
homology. Ichihara etal (1988a.b) sequenced a 15 kb DNA fragment corresponding
to 1-1/2 of the 9 kb repeating units previously described Siebenlist eta l (1981). Each
unit contained members of six different families; DXP, DA, DK, DN, DM, and DLR.
Sonntag et a l (1989) reported the sequence of a D gene segment from a D-JH
rearrangement that shared 60% homology with the mouse DEL 16 family. The
rearrangement containing the DEL 16 gene segment was used to probe Southern blots
of human genomic DNA. The results suggested the presence of related genes in the
germline. Including DQ52, a family with one member only, there are approximately
eight D region gene families.
k) NON CONVENTIONAL REARRANGEMENTS GENERATED BY D
SEGMENTS
In germline configuration the D gene segments are sandwiched by
recombination signal sequences that consist of a heptamer/12bp spacer/nonamer
(Sakano e ta l 1980,1981). Heptamer sequences are conserved except in DM2 and
34
DAI gene segments, whereas nonamers diverge in general from the consensus
sequence. The DIR gene segments (termed DIR due to the presence of their irregular
recombination signals) were reported to be surrounded by multiple heptamer/nonamer
like sequences as well as spacers of different lengths (Ichihara etal. 1988a,b). The
availability of 12 and 23 bp spacers flanking both ends of the coding segment
sequences suggests that the DIR segments might be a substrate for D-D recombination.
It has been postulated that up to 12% of human CDR3 sequences are best explained by
the participation of DIR sequences (Sanz 1991). A minimum of five DIR genes exist in
the human genome (Sanz et al 1994). DIR- like sequences can be found in non human
primates with remarkable conservation but do not appear to occur in other species such
as mice (Sanz et al 1994) This suggests that DIR genes either arose from exogenous
sources or sequential rearrangement of endogenous sequences.
Exonucleases and terminal deoxynucleotidyl transferase (TdT) “chew” and fill
in, respectively the coding ends of D and JH gene segments, generating unique amino
acid residues at this junction. The same process occurs at the V-D joint except the 3’
coding end of the VH gene segment is usually protected from exonuclease activity. The
mechanism of flexible recombination allows not only the deletion and or insertion of
nucleotides at the junctions but also the utilisation of the same D gene segment in
different reading frames.
It is difficult to assess the contribution somatic mutation makes to D region
diversity, although an example of this was reported by Pascual e ta l (1990) where two
clonally related antibodies used the same DLR2 gene and the pattern observed
correlated well with the rate of mutations seen in the whole heavy chain variable region
. Examples of D-D fusions have been observed in both murine and human antibodies
(van der Heijden etal 1990, Davidson etal 1990). Meek etal (1991) demonstrated
the presence of direct and inverted D-D fusions mediated by heptamer/nonamer signal
sequences or isolated heptamers in murine bone marrow. They also showed that
isolated D segments can rearrange to JH segments in either transcriptional orientation,
further increasing the potential diversity of the repertoire.
D JH REGION GENES
The human JH locus is located between the major D cluster and the
immunoglobulin heavy chain constant region locus. It spans approximately 2.6 kb of
35
DNA and contains six functional genes and three pseudogenes all in the same
transcriptional orientation. JH segments are flanked in their 3’ ends by a potential
RNA splice site and in their 5’ ends by the heptamer/spacer/nonamer signal sequences
that are involved in the rearrangement with upstream D segments. The three JH
pseudogenes have an open reading frame in their coding regions, but they contain
abnormal splice sites and their heptamer sequences differ the most from the consensus
functional heptamer. The spacer between the heptamer and nonamer varies in length
from 20-25 nucleotides in all functional JH segments and pseudogenes and only in the
JH4 segments possess the conventional 23bp spacer length. Interestingly JH4 is the
single most expressed JH segment in the human repertoire.
JH gene segments rearrange to D gene segments through their respective
flanking signal sequences. This is the first event in the assembly of the
immunoglobulin heavy chain variable region and also the first source of diversity; in
the process of joining the D and JH joining regions exonuclease nibbling back and
filling in of the 3’ and 5’ ends respectively take place with the generation of new
amino acid residues that will be part of heavy chain CDR3 one of the regions critical for
antigen contact. Six function JH gene segments have been reported (Ravetch et al.
1981). The 5’ end of certain JH segments are prone to truncation such as JH4 and JH6
(Schroeder gr fl/. 1987)
m LIGHT CHAIN ORGANISATION
al KAPPA CHAIN
The human V k locus occurs on chromosome 2 and consists of a five J k and
approximately 76 Vk genes, 32 of which are potentially functional. The V k genes have
been divided up into seven gene families (Schable & Zachau 1993). Gene families 4,
5, and 7 consist of one germline gene each, V k 6 has three members (Lautner-Rieske et
al. 1992). The V k I is the largest gene family with seventeen potentially functional
members V k 2 and V k 3 have nine and seven potentially functional members
respectively (Schable & Zachau 1993). Studies have shown that the V k 3 light chains
maybe used preferentially for IgM autoantibody synthesis (Ledford et al. 1983).
Kappa chains constitute seventy per cent of the light chains in human serum (Milstein
1965).
36
b) LAMBDA CHAIN
The VX genes are encoded on chromosome 22 in humans. Seven nonallelic
lambda constant region genes have been characterised (Hieter et ai 1981). Unlike the
K locus each JX is associated with a unique lambda constant region (Bauer & Blomberg
1991). A classification of the protein sequences corresponding to the known VX genes
led to the definition of seven VA. gene subgroups (Chuchana et al. 1994) Genes
defining two new subgroups VA.8 and VA.9 have been described by Winkler et at.
(1992) and Williams and Winter (1993) respectively. Based on the number of
hybridising bands on Southern blots, the number of VA, segments is estimated as at
least twenty and nearer seventy (Anderson etal. 1985, Lai etal. 1989) with at least
twelve VA.1 gene segments (Kabat etal. 1987), ten VX2 gene segments (Anderson et
al. 1984, Adderson gf aZ. 1992). Williams era/. (1993) estimated that there are thirty-
six VA. germline genes. However, the number of hybrising bands seen on Southern
blots indicates that there are a number of VA, segments which have not been isolated.
1.4 B CELL REPERTOIRE
Immunoglobulin (Ig) genes are assembled and their products expressed by cells
of the B lymphoid lineage. The earliest identified precursor B cells (pre B cells ) are
found in the fetal liver or adult bone marrow (Levitt et al. 1980). These cells express
cytoplasmic p heavy chains but not A. chains. The next stage in development is the
surface immunoglobulin positive B cell. Each B cell and all of its progeny expresses a
surface antigen receptor with a unique antigen binding specificity. This unique
specificity is maintained by the expression of only a single heavy and a single light
chain variable region gene by an individual cell. Thus unlike most other genes, Ig
genes are regulated by the principle of allelic exclusion. In a given B cell only one of
the two H chain alleles and only one of the several L chain alleles is expressed at the
cell surface as antibody receptor (Pemis et al. 1965). The clonality of a single B cell
and its progeny is believed to be an intrinsic requirement for the proper functioning of
an immune system based on antigen stimulated clonal expansion. Only after a B cell
expresses its surface immunoglobulin receptor is it susceptible to antigen stimulation.
When an antigen binds the surface receptor in association with appropriate
37
costimulatory signals, it triggers clonal expansion and eventual differentiation of B cells
to the terminal stage of the B cell developmental pathway, the plasma cell. Plasma cells
secrete large amounts of antibodies possessing the initial antigen binding specificity and
allelic exclusion prevents the simultaneous production of large amounts of non-antigen-
selected antibodies by individual cells involved in such a response.
The initial response to antigen results in the production of low affinity
antibodies of the IgM isotype. To produce higher affinity antibodies expressing the
effecter functions of each of the classes and subclasses, B cells stimulated by antigen
and interacting with T lymphocytes undergo two additional molecular events; they
introduce random somatic mutations into the H and L chain V regions (Berek &
Milstein 1987, Kocks & Rajewsky 1989) and they rearrange the H chain V region
downstream so that it can be expressed with each of the other C region genes (Shimuzu
& Honjo 1984).
The detailed mechanisms responsible for V region somatic mutation are unclear.
The analysis of monoclonal antibodies derived from animals immunised with a variety
of foreign antigens reveals that somatic mutation is a relatively random process, is
restricted to the V region and its immediate flanking sequences, requires T cell factors,
begins after gene rearrangements, continues after class switching but probably ends
before the formation of the fully differentiated plasma cell (Shan et al 1990, Manser
1990, Radac e ta l 1991). There is evidence that suggests V region hypermutation
occurs not during the conversion of a B lymphocyte into a clone of antibody forming
plasma cells but rather during the genesis of memory B cells derived from virgin B
cells (Siekevitz gf a/. 1987, McHeyzer-Williams graA 1991). This process is believed
to occur in the germinal centres (reviewed by Nossal 1992)
The B cells take up antigen through their immunoglobulin receptor and process
and present it to T helper cells, resulting in the preferential stimulation and proliferation
of those B cells expressing the higher affinity receptors. If a mutation occurs that
results in an increase in affinity for antigen, the B cell expressing that higher affinity
antibody will presumably preferentially bind antigen, especially if that antigen is
present in limiting amounts as must occur late in the immune response to foreign
substances. B cells that have undergone mutations that result in a decreased affinity for
antigen or a loss of the ability to bind the eliciting antigen will not activate T cells and
will not be stimulated to proliferate. These low affinity B cells are diluted out by the B
38
cells producing the higher affinity antibodies (Radac eta l 1991, Manser eta l 1987).
As somatic mutation occurs the B cell also rearranges its VDJ variable region down the
chromosome to bring it into proximity with the downstream constant region genes that
encode each of the IgG subclasses as well as Ce and Ca (Shimuzu & Honjo 1984).
The process of isotype switching occurs independently of V region mutation as IgM
antibodies can contain somatic mutations and IgG and IgA may have none (Kocks &
Rajewsky 1989). In general IgG antibodies of each subclass are likely to have
undergone significant somatic mutation.
Analysis of B cell differentiation has benefited from the availability of tumour
cell analogues representing the various stages of the developmental pathway. B cell
leukemias and lymphomas represent the surface Ig positive B cell stage, while
myelomas and plasmacytomas represent the mature Ig secreting stage. Comparison of
the configuration of Ig genes in such lymphoid tumour lines and in nonlymphoid cells
revealed that Ig variable region genes are somatically assembled. This comparison has
also elucidated some of the mechanisms involved in the rearrangement process of Ig
genes in B cells (Tonegawa 1983). However insight into the dynamics of Ig gene
rearrangement required investigation of the cell stages during which rearrangement
occurs. Direct studies of fetal liver allowed analysis of Ig gene expression in a
synchronously differentiating population of pre B cells (Levitt et a l 1980), whereas
analysis of bone marrow provided information concerning a steady state, renewable
population of pre B cells and more mature B cells (Coffman et a l 1983). Fusion of
fetal liver cells to myeloma cell lines resulted in the production of fetal liver hybiidomas
which provide a model system to study the gene rearrangements of a pre B cell within
the phenotypic background of a myeloma (Maki et a l 1980). A remarkably accurate
and dynamic representation of Ig gene rearrangements was provided by investigators
utilising the Abelson murine leukemia virus (A-MuLV) transformed cell lines. The
most valuable aspect of these lines is their ability to proceed through
immunodifferentiative events when propagated in culture.
A-MuLV is a replication defective retrovirus unique in its capacity to transform
immature B lymphoid cells in culture; transformation of fetal liver or adult bone
marrow cells yields permanent lines in which essentially all represent the pre B or
earlier stages of the B cell pathway (Alt &Baltimore et al 1982).
Analyses of such lines have provided static models of these early stages of B
39
cell development. More importantly however many of the A-MuLV transformants
undergo immunodifferentiative events when propagated in culture including the
assembly of H and/or L chain genes and H chain class switching events (Alt et al.
1981). Most apparently novel aspects of Ig gene rearrangement or expression
discovered in A-MuLV transformants have served to elucidate events later demonstrated
to occur in normal pre B cells in mice.
1.5 TOLERANCE AND REGULATION OF B LYMPHOCYTES
Several mechanisms have been proposed for the induction of self-tolerance in B
and T cells. Engagement of self-antigen may, under some circumstances, lead to cell
death, often called clonal deletion, or, for T cells in the thymus, negative selection.
Anergy refers to the state in which the autoreactive lymphocytes survive, but are
functionally inactive. Suppression is a mechanism in which the self-reactive
lymphocytes survive, and retain the ability to respond to antigen, but are held in check
by T suppressor cells or factors derived from them. Finally, many self components
may be hidden from the immune system. This can occur either because these molecules
are expressed only in so-called privileged sites such as the brain, which are not subject
to such intense immune surveillance; because they are only expressed in low levels; or
because they do not bind efficiently to major histocompatibility complex (MHC)
molecules. Binding of peptide fragments of antigens by an MHC molecule, and
subsequent presentation of this peptide-MHC complex to the T cell receptor is a
prerequisite for antigen specific T cell immune responsiveness. However
superantigens can bypass the necessity for specific interaction between T cell receptor
and MHC-peptide complex normally required for T cell activation, leading to a more
generalized non-specific T cell activation. The sequestration of self determinants is not
a mechanism of tolerance per se because the autoreactive lymphocytes for such antigens
are present and their reactivity is unhindered. The unmasking of hidden determinants,
however, maybe important for the pathogenesis of autoimmune disease.
The hypothesised mechanisms of immune tolerance were reviewed by Nossal
(1983). Immature lymphocytes in neonates, and in the bone marrow and the thymus of
the adult, are more susceptible to tolerance induction than mature cells. Although both
40
B and T lymphocytes are susceptible to self-tolerance induction, tolerance in T cells can
be induced with lower doses of antigen and is longer lasting. Tolerance induced at the
T cell level is of importance because the activation of B cells requires T cell help.
Autoreactive B cells therefore might not be too injurious in the absence of this help.
Thus, the elevated levels of autoantibodies found in autoimmune disease may result
from a dysfunction within the T cell population which have not been tolerized
effectively.
Autoimmune disease could also arise if autoreactive T and B lymphocytes are
not clonally deleted in the thymus and bone marrow. Clonal deletion occurs via a
process called apoptosis or programmed cell death. Apoptosis is characterised by the
condensation of the cytoplasm, loss of plasma membrane microvilli and fragmentation
of chromosomal DNA into lengths of around 180bp (Raff 1992). Apoptosis is thought
to be the mechanism of T cell deletion in the thymus (Smith etal. 1989) and is believed
to be involved in the maturation of antibody responses through the elimination of low
affinity antigen receptor bearing lymphocytes (Lui et al. 1989). Systemic lupus
erythematosus patients show a broad range of immunological abnormalities. These
include increased numbers of circulating activated B lymphocytes which produce large
amounts of immunoglobulin and an array of autoantibodies. The high numbers of B
cells in SLE may be due to abnormal longevity of B cells in these individuals.
Dysfunction in apoptosis the normal regulatory process governing the life span of T
and B cells in these patients could be responsible for this.
Two gene products appear to be responsible for the regulation of apoptosis.
The product of the Bcl-2 gene enhances cell survival (Graninger etal. 1987) and the
transmembrane glycoprotein product of the Fas gene promotes cell death.(Yonehara et
al. 1989, Trauth etal. 1989). The Fas antigen is not expressed on normal human
resting T cells nor on resting B cells but is present on activated T and B cells
(Miyawaka etal.1992).
Transgenic murine models have shown that the presence of an un regulated Bcl-
2 gene resulted in polyclonal expansion of B cells with a large excess of mature B cells
and plasma cells (McDonnel etal. 1989, Katsuma etal. 1991). Increased levels of Bcl-
2 mRNA have been reported in the peripheral blood lymphocytes from nineteen of
twenty-four SLE patients compared with healthy controls (Graninger 1992), however,
studies performed in the Bloomsbury Rheumatology Unit showed there were no
41
significant differences in Bcl-2 expression between SLE patients, healthy controls and
patients with other autoimmune conditions (Rose et al 1993).
The potential involvement of the Fas molecule in autoimmune disease was
highlighted by studies on the murine model of SLE, the MRIV/pr mouse (Murphy &
Roths 1979). The lymphoproliferative disorder in these mice is controlled by the
recessive autosomal gene designated Ipr and is manifested in massive
lymphoproliferation which results in an autoimmune syndrome which resembles human
SLE (Cohen & Eisenberg 1991). It has been shown that mice that carry the
lymphoproliferative {Ipr) disorder have defects in the Fas gene and therefore it has
been suggested that Ipr encodes the structural gene for the murine form of the Fas
antigen (Matsumoto eta l\99 \). There is no published data regarding Fas expression
in the peripheral blood lymphocytes of SLE patients.
There are many potential sites for dysregulation within the immuno-regulatory
mechanisms of patients with autoimmune disease. These defects could occur within the
T cell population which are responsible for controlling the B cells response. A fault in
clonal deletion/apoptosis could allow autoreactive B and T cells to escape tolerance.
Much of the work in this area has been performed on murine models and the relevance
of these findings to human autoimmune disease has yet to be confirmed.
1.6 NATURAL AUTO ANTIBODIES
It is normally assumed that the main function of the immune system is to
distinguish between self and non self and we remain healthy because this is so.
However, it has become apparent that this distinction between self and foreign elements
is not absolute. A manifestation of the blurring of this distinction is the presence of the
so called “natural autoantibodies” which have been found in the sera of normal healthy
individuals. Natural autoantibodies are defined as polyreactive, low affmity antibodies
of the IgM isotype which are found at a low concentration in normal sera. Data has
revealed a high frequency of autoreactive cells in the normal preimmune and early
ontogenic pre-B cell repertoire (Glotz et al 1988, McHeyzer-Williams et a l 1988). It
has been suggested that natural autoimmune responses originate during early B cell
development. These observations suggest that autoreactivity not only exists in normal
42
animals but also represents an integral part of early B cell ontogeny and may be
required for the development of the mature immune system.
Some of these ‘natural autoantibodies’ are directed against easily accessible self
constituents such as soluble plasma proteins, hormones, surface components of
senescent red blood cells, fetal cells and tumour cells. Others have been found to react
with more concealed self antigens including cytoskeletal and nuclear antigens or cryptic
membrane epitopes exposed after proteolytic treatment
Studies of natural autoantibodies in mice indicate that few are monospecific, the
vast majority being polyreactive, furthermore, individual murine monoclonals have
been shown to react with fairly dissimilar autoantigens including DNA, myosin,
thyroglobulin and spectrin.
These observations have led to a focussing of interest on human natural
autoantibodies and their possible involvement in autoimmune disease. Natural
autoantibodies have antigen specificities strikingly similar to autoantibodies found in
patients with autoimmune disease. However, natural autoantibodies are frequently of
the IgM isotype and bind with low affinity to autoantigens, whereas circulating
autoantibodies associated with disease are usually higher affinity IgG antibodies.
Some experiments have pointed to the pathogenic potential of these natural
autoantibodies but the role of these antibodies remains uncertain. Some natural
autoantibodies also crossreact with foreign antigens early in ontogeny, and could be
starting points for the maturation of the immune system. The situation in the normal
adult differs dramatically as, once an individual has been exposed to foreign antigen
autoreactive clones are overtaken by those reactive with foreign antigen. Natural
autoantibodies continue to be expressed at a low level in normal healthy adults. A
correlation has been found between the level of autoantibodies in the fetal state and the
phenotypes of cells present at this time. A subpopulation of B cells which express the
T cell marker CD5 have been found to be elevated in the fetal state and in some
autoimmune diseases.
43
1.7 AUTOANTIBODIES IN SLE: NATURE AND SIGNIFICANCE.
It is now over thirty years since the discovery of anti-nuclear and anti-DNA
antibodies in sera from lupus patients. Since these milestone observations numerous
diverse autoantibodies have been described including those with specificity for different
forms of DNA, RNA, histones and synthetic nucleotides. Friou and colleagues (Friou
1957, Friou 1958, Friou et al. 1958) described and partially characterised the
phenomenon by which sera from patients with SLE would bind to nuclear antigens in
fresh frozen tissue. Using fluorochrome labelled antisera it was shown that the
component binding to the nucleus was identified as antibody from the serum of the
patient. This finding suggested that an autoimmune pathology was underlying the
disease and heralded a new era of research in SLE and today. Screening for anti-
nuclear antibodies (ANA) is an important in the diagnosis of lupus.
Over 95% of SLE patients have a positive ANA test when standard
immunofluorescent methods are used. A low proportion of lupus patients while
satisfying four or more of the revised American Rheumatism Association (ARA)
criteria for the classification of the disease (Tan et at. 1982) remain consistently ANA
negative. On close examination most of these individuals invariably have antibodies to
SS-A (Ro), SS-B( La) or ssDNA and these patients are less likely to have renal
disease (Maddison 1982, Reichlin 1982).
Hybridomas can be derived from humans with SLE or lupus-prone mice
without prior immunisation, and it is generally considered that autoantibodies produced
by such hybridomas are a reflection of autoantibodies by B cells prior to
immortalisation by hybridisation. The majority of monoclonal lupus antibodies that
have been studied thus far are directed against nucleic acids.
Ï ) ANn-DNA ANTIBODIES
Coincident with Friou’s observations on ANA were reports from other
laboratories in Europe and America showing antibodies in the sera of lupus patients that
would bind to DNA (Cepellini et al. 1957, Miescher and Straessle 1957, Robbins et
al. 1957). Since this time the dsDNA binding test has remained the single most
important laboratory test for the diagnosis of SLE as approximately 75% of patients
with the disease are seropositive. Some controversy exists regarding the degree of
44
DNA binding and disease activity in lupus. Many physicians have observed that some
lupus patients with marked clinical activity have low DNA binding and vice versa.
Absence of antibodies to dsDNA has also been associated with a milder form of lupus
(Ballou & Kushner 1982). Clinically apparent nephritis in particular was found to be
less common although Raynaud’s phenomenon was found more frequently. It is also
noteworthy that anti-DNA levels at the initial evaluation of the disease seems to have no
prognostic significance (Adler etal 1975, Ludvico eta l 1980).
The role of anti-DNA antibodies in pathogenesis of SLE is supported by the
fact that some anti-DNA antibodies participate in the formation of immune complexes
that give rise to inflammatory lesions characteristic of the disease. Perfusion
experiments in vitro and in vivo with anti-dsDNA antibodies (Raz eta l 1989, Tsao et
al 1990) and the presence of such antibodies in diseased kidney of both human
(Koffler et a l 1967) and mouse (Lambert et a l 1968) have convinced most
investigators that these antibodies contribute to kidney damage. In addition, the
antibodies that appear in the circulation of patients and mice with lupus differ from
anti-DNA antibodies that can be found in non-autoimmune individuals in that they bind
dsDNAs with high affinity, are IgG, do not crossreact with a wide spectrum of
unrelated antigens and are often cationic in charge (Seigneurin eta l 1988).
Lee et al (1981) have demonstrated the influence which the base content and
sequence of a nucleotide chain can have on the binding capacity of nucleotide binding
antibodies. They produced six monoclonal autoantibodies from hybridomas derived
from NZB/W spleen cells and examined their reactivity with synthetic homo- and
heteropolymers of purines and pyrimidines. They found that the reactions of all six
antibodies were profoundly influenced by the base sequence of the antigen.
The nucleic acid binding specificities of monoclonal anti-DNA antibodies
derived from MRL/lpr mice (Andrzejewski et a l 1981) and from human patients with
SLE (Shoenfeld et al 1983a) have also been examined. These studies suggested that a
single autoantibody can bind to multiple nucleic antigens of widely different base
composition. A possible explanation for the binding pattern of these autoantigen
binding monoclonals may be the composite nature of the binding site of the antibody
which allows it to recognise (with varying degrees of affinity) different epitopes
alternatively, various ligands possess cross-reacting epitopes as determined by similar
tertiary structure. The logical candidate for anti-DNA antibodies is a structure in the
45
sugar-phosphate backbone that is common to all nucleic acids. An important feature of
the sugar-phosphate backbone is the presence of phosphate groups in the
phosphodiester linkage with carbon atoms of adjacent sugar molecules. It seems very
likely that these phosphate groups on the exterior of helical nucleic acids constitute an
important epitope for lupus antibodies. Moreover individual variations in binding
specificities among monoclonal anti-DNA antibodies may be reasonably accounted for
by structural differences in polynucleotide backbones that stem from variations in the
helical configuration and interphosphate distances.
Further analyses have shown that monoclonal lupus autoantibodies originally
selected for their ability to bind DNA can also bind to phospholipids including
cardiolipin (diphosphatidic acid). Phospholipids also contain phosphate groups in
diester linkage with carbon atoms, a feature that supports the identification of these
groups as the important epitope for the polyreactive monoclonal anti-DNA antibodies.
This view is further supported by the demonstration that absorption of such
autoantibodies with cardiolipin micelles inhibits their ability to produce the anti-nuclear
antibody reaction by indirect immunofluorescence. Moreover when injected into
rabbits and mice cardiolipin can stimulate the production of both anti-cardiolipin and
anti-DNA antibodies (Rauch et al 1984). Hybridomas prepared from the cardiolipin
immunized mice produced monoclonal antibodies that bound not only to cardiolipin but
also to ssDNA and even to dsDNA.
DNA binding autoantibodies can bind to other molecules (and in some cases
more avidly than to DNA) it is evident that DNA is not necessarily the immunogen or
even the preferred target antigen in SLE. For example, Rauch etal. (1986) confirmed
earlier suggestions that anti-DNA antibodies may react as rheumatoid factors. These
authors described eight human hybridoma derived antibodies which bound to ssDNA
and the Fc portion of human IgG
As previously alluded to, there are other possible explanations for the
crossreactivity of antibodies. Lane and Koprowski (1982) have suggested that
different antigenic structures may bind partially to a common conformational
determinant on an antibody. The determinants could consist of two or more subunits
hence the antibody appears to be binding to more antigens than may have initially been
envisaged. In addition Faaber et al. (1984) have suggested that anti-DNA antibodies
are simply binding to structures with repeating negatively charged groups hence their
46
binding to hyaluronic acid and chondroitin sulphate.
ID ANTIBODTES TO RNA
Autoantibodies to RNA can be found in SLE with a higher incidence compared
to patients with other autoimmune diseases. They appear to have little clinical
significance in lupus although it has been suggested that they may be present as part of
an immune response to a virus (Schur & Monroe 1969). Blanco eta l (1991) assessed
the frequency of anti-RNA antibodies in 138 patients with SLE. Of the sera from these
patients 9.4% had anti-RNA antibodies but no distinguishing features, clinical,
serological or immunogenetic, between those with or without these antibodies could be
identified.
m i EXTRACTABLE NUCLEAR ANTIGEN
The extractable nuclear antigens (ENA) are proteins and ribonucleic acids which
may be divided into four main types:- ribonucleoprotein RNP, the ‘Smith antigen’ Sm
both of which contain RNA and Ro and La. Using the sensitive counter-current
immunoelectrophoresis technique it has been estimated that up to 30% of lupus patients
have antibodies to Sm, although amongst Caucasians it is less than 10% (Worrall etal.
1993), while 30-50% react with RNP (Tan 1982). Antibodies to RNP alone appear to
identify a population of patients who have a low incidence of DNA binding and renal
disease. However if other autoantibodies accompany those to RNP particularly anti-
Sm, anti-Ro, and anti-DNA, nephritis may develop (Sharp et al. 1971, Reichlin &
Mattiolli 1972). The incidence of antibodies to SS-A (Ro) and SS-B (La) in lupus
patients has been estimated as 20-30% and 12% respectively (Tan 1982). Reichlin
(1982) noted that out of 66 patients who had clinical features consistent with the
diagnosis of SLE but who were ANA negative, 41 (62%) had antibodies to SS-A (Ro).
Clinically this group is associated with a low incidence of neuropsychiatrie disorders
and renal disorders and renal disease but a high incidence of photosensitivity and malar
rash.
rsn RHEUMATOID FACTOR
Rheumatoid factor are antiglobulins generally of the IgM class and with
specificity for the Fc region of IgG with which they form immune complexes. These
47
self associating IgG complexes play a significant role in the mediation of inflammatory
damage in rheumatoid arthritis. They have been found in up to 40% of sera from
patients with SLE (Estes & Christian 1971). The significance of these autoantibodies
which are present in SLE are in low titre is unknown, as they do not have an obvious
role in the pathology of the disease and have no relevance to prognosis.
VI ANTI-fflSTONE ANTIBODTES
The histones are the best understood of the structural proteins in the eukaryotic
nucleus (Butler 1983, Chambon et al. 1989, Newport et al. 1987). It has been
estimated that the mass of histone present is approximately equal to the cellular DNA
content (Chambon et al. 1989). The histones are a small group of proteins with a high
content of positively charged amino acids such as lysine and arginine. They are found
in nuclear chromatin where they are arranged together with DNA to form a structure
known as the nucleosome.
On the basis of their primary structure five types of histones have been
identified. HI is lysine rich and has the most interspecies amino acid sequence
variability. The high proportion of positively charged amino acids of these relatively
small proteins helps them bind tightly to DNA. Since these proteins rarely dissociate
from DNA they are believed to play an important role in control of gene expression and
cell division. H2A and H2B are the most lysine rich and moderately conserved
throughout evolution. They are responsible for coiling the DNA into nucleosomes.
H3 and H4 are arginine rich and are the most highly conserved of the histones. They
form a tetrameric complex of two molecules each of H3 and H4 at the interior of the
nucleosome core particle
Lupus patients can have antibodies directed against all classes of histones but
those against HI and H2B are most commonly found (Costa & Monier 1986). Studies
on the clinical features and disease activity with histone antibody titres have been
inconsistent. The studies of Muller et al. (1990) observed a correlation of disease
activity with antibodies to the core histones but not to HI. Histone antibodies have
been reported in 67-100% of sera from patients with drug induced lupus. This is
usually a mild clinical condition characterised by the presence of antibodies to ssDNA
and histones, clinically it is easy to differentiate between drug induced lupus and SLE,
48
since the former virtually never causes more than skin rashes and joint pains.
VI) ANTIPHOSPHOLIPID ANTTBODTHS
It has become evident that anticardiolipin antibodies are a heterogeneous group
of autoantibodies with different characteristics (Harris et al. 1985). They are linked
clinically to thrombosis, recurrent abortion, livido reticularis and thrombocytopenia.
Only a minority of patients with anticardiolipin antibodies develop thrombotic
complications. Possible explanations for the differences between “innocent and
pathogenic” antibodies may be attributed to isotypic diversity as well as to differences
in binding specificities, avidities, idiotypes and associated genetic and environmental
factors. Most anticardiolipin antibodies also bind to other negatively charged
phospholipids such as phosphatidic acid, phosphatidylserine, or phosphatidylinositol
and a minority even bind to neutral phospholipids such as phosphatidylethanolamine.
This crossreactivity may be due to similarity in structure charge or configuration. The
precise epitope specificity of anti cardiolipin antibodies is unknown. It was thought
that the phosphodiester groups were the sites bound by these antibodies but removal of
the glyceride portion of the molecule also resulted in loss of the antigenicity. This
suggests that the glyceride moiety is essential for antibody binding probably helping the
orientation of the phosphodiester group into a configuration best bound by
anticardiolipin antibodies. It has also been demonstrated that some antiphospholipid
antibodies require serum co-factors such as beta-2-microglobulin in order to bind
phospholipids.
Studies with monoclonal anticardiolipin antibodies have shown considerable
conformational specificity with the ability to distinguish between lamellar and
hexagonal forms (Harris et al 1990). Some investigators have reported a positive test
for anticardiolipin antibodies in patients with high levels of anti-DNA antibodies. In
these cases inhibition studies with pure cardiolipin and DNA disclosed a partial
crossreactivity between the two groups of antibodies. It has been confirmed that
monoclonal antibodies with a high affinity for anticardiolipin have no reactivity with
DNA. Some low affinity anticardiolipins may also react with other antigens
(mitochrondria, endothelial cells) and exhibit a partial crossreactivity with the type M5
anti-mitochrondrial antibodies (Meroni et a l 1987) anti-endothelial cell antibodies
(Cervera et a l 1991, Vismara et a l 1988) or IgM rheumatoid factor (Agopiai e ta l
49
1988). It has been postulated that antiphospholipid antibodies may arise in response to
foreign antigens as cardiolipin itself has been shown to be a poor immunogen when
injected into experimental animals (Lafer et al 1981). Raised levels of anticardiolipin
antibodies are frequently found in a number of infectious diseases, eg. spirochetal
infections such as Lyme disease (Mackworth-Young et a l 1988). However only on
rare occasions do these patients present with thrombotic complications.
The mechanism of action by which antiphospholipid antibodies cause
antiphospholipid syndrome is unknown, and it is still unclear whether these antibodies
are the prime cause of the clinical manifestations or are epiphenomena accompanying
more basic underlying immunological disturbances.
Since phospholipids are the main constituent of cell membranes and also play an
important role in the major haemostatic pathways, most of the efforts to elucidate the
mechanisms of thrombosis in patients with antiphospholipid syndrome have focused
on the interactive effects these antibodies have with endothelial cells and platelet
function. There may be more than one mechanism causing thrombosis in patients with
anti phospholipid syndrome and that these mechanisms vary from patient to patient.
(Khamashta gr aZ. 1989).
Vm ANTI-NEUTROPHIL CYTOPLASMIC ANTIBODIES
Until recently, the diagnosis of systemic vasculitis has relied on clinical criteria.
The presence of anti-neutrophil cytoplasmic antibodies (ANCA) in a high proportion of
sera from these patients have proved to be of major diagnostic value. In extended
Wegener’s granulomatosis (WG) for example, circulating ANCA are found in
approximately 90% of patients with active disease (Cohen et a l 1989) and in up to
60% of patients with other vasculitic syndromes (Jennette et a l 1990). They are not
simply the result of the release of neutrophil granule contents into the circulation as they
do not occur in other conditions in which sequestration and destruction of neutrophils
is widespread.
ANCA are detected by indirect immunofluorescence using ethanol fixed
neutrophils as substrate. Two staining patterns are seen: cytoplasmic (cANCA) and
perinuclear (pANCA). The majority of sera show a cANCA staining pattern, the
pattern most commonly associated with WG, and contain antibodies directed against
serine protease 3 (SP3) (Ludemann et a l 1990). Antibodies producing a pANCA
50
staining pattern are found in a wide range of diseases. Similar staining patterns are
given by antibodies reacting with various neutrophil granule enzymes including
myeloperoxidase (MPO), elastase and lactoferrin (Lesavre 1991). Antibodies reacting
with myeloperoxidase are most commonly associated with vasculitic syndromes
including idiopathic crescentic nephritis, Churg-Strauss, polyarteritis nodosa with
visceral involvement and hydralazine induced glomerulonephritis (Kallenberg 1992a).
ANCA are rarely found in SLE but up to 20% of patients with RA have antibodies to
MPO in their serum (Savige et al. 1991, Lassoued et al. 1991). Their presence has
been tentatively associated with systemic vasculitis. In a study of sera from 97 patients
with rheumatoid arthritis, 11 contained anti-MPO antibodies in whom their presence in
three patients was strongly associated with pulmonary fibrosis (Cambridge 1993).
The pathophysiology of the autoimmune response in vasculitis is unknown
although it is generally considered to reflect a T-cell mediated response to an as yet
unidentified target antigen in vessels or other organs (Kallenberg et al. 1992b).
However, the strong association of vasculitis with the presence of circulating ANCA
suggests that these autoantibodies may reflect an immune response related to the
induction of disease. There is a strong association between the development of
vasculitis and infection (DeRemee 1988, Schmitt et al. 1991). It has been suggested
that ANCA may be induced as the result of ‘molecular mimicry’ between granule
enzymes and exogenous agents with a subsequent antigen driven response resulting
from the release of granule components from activated neutrophils at sites of tissue
injury (Kallenberg etal. 1992b, Baltaro etal. 1991).
The high frequency of ANCA in vasculitic sera and the observation that ANCA
titres correlate with disease activity (Egner & Chapel 1990, Cohen Tervaert etal. 1990)
has suggested that these autoantibodies may play a pathogenetic role. In vitro, ANCA
have been shown to activate neutrophils to degranulate and produce a respiratory burst
which can damage endothelial cells (Charles et al. 1991, Ewert et al. 1992). Some
ANCA have also been shown to interfere with extracellular granule enzyme activity
(Dolman 1991).
Due to the probable diversity of antigens and epitopes recognised by anti
neutrophil antibodies in human sera (Cambridge et al. 1991), detailed analysis of
possible pathogenic antibodies has not been possible. Human monoclonal antibody
production using hybridoma technology is essential to dissect further the role of these
51
antibodies in the disease process and to analyse the fine specificity and origins of
antibodies present in patients.
1.8 THE IDIOTYPE NETWORK
Antibodies are conventionally distinguished by the antigens with which they
bind. Another way of distinguishing antibodies serologically involves an analysis of
their idiotypes. Immunoglobulin (Ig) idiotypes are essentially phenotypic markers of
the V genes used to encode Ig molecules either as soluble antibody molecules or as
lymphocyte receptors. They represent novel antigenic determinants and are recognised
in turn by anti-idiotypic antibodies.
The concept of an immune system evolving to recognise self antigens rather
than (or in addition to) foreign antigens was formulated by Jeme (1974). Experimental
evidence for this was provided by Eichmann (1978) who established in mice that
foreign antigens initiate immune responses which interconnect via idiotypes on
lymphocytes and antibodies. There is increasing evidence that a similar idiotypic
network exists in man (Geha, 1984).
The contribution of the idiotype network to autoimmune disease remains the
subject of much research (Male 1986). Idiotypes are historically defined by anti-
idiotypic antibodies raised by immunizing allogeneic or xenogeneic animals with
purified antibodies or their fragments. Depending on the manner of preparation, the
anti-idiotypic antibodies may be polyclonal, that is a number of antibodies directed
against a group of idiotopes which collectively comprise an idiotype, or a single -
monoclonal antibody directed against a single idiotope. Idiotypes are thus serologically
defined determinants present on the Fab regions of immunoglobulins. They may
represent amino acid sequences located on the light (L) or heavy (H) chains alone or in
combination (conformational determinants).
Antibodies which react only with the immunizing immunoglobulin define
restricted or private idiotypes. Sharing of crossreactive idiotypes (CRI) between
antibodies of the same or different specificities from different individuals may reflect
common amino acid sequences within the framework regions or complementarity
determining regions. Occasionally anti-idiotypic antibodies define apparent cross
52
reactive idiotypes on antibodies throughout different species. The cross-reactivity of
anti-idiotypic reagents with different antibodies may not necessarily be due to
recognition of common amino acid sequences within their V regions which define the
idiotope. This may however reflect an interaction between the V regions at the level of
tertiary structure. In these cases the tertiary structure of the anti-idiotype might
resemble that of the antigen, and such anti-idiotypes are said to carry an “internal
image” of the antigen. A further type of anti-idiotypic antibody has been described
which reacts with both the idiotypic determinant on the antigen binding site of the
antibody and also with the antigen recognised by that antibody binding site. This
peculiar anti-idiotype has been called an epibody but has only been described rarely
CBonaefaA 1982; Chen gf a/. 1985).
The use of idiotypes as phenotypic markers of populations of antibodies has been a
valuable tool in studying auto-antibodies in autoimmune diseases. The reasons for this
are threefold;
1. To provide information about the genetic background of these disorders.
2. To provide information about the mechanisms of autoantibody involvement in
autoimmune conditions.
3. Anti-idiotypic antibodies which identify idiotypes of interest can be used to analyse
cellular interactions in these diseases, to examine the tissue distribution of the
idiotypes and to explore novel therapeutic intervention.
1.9 CROSS REACTIVE IDIOTYPES FOUND IN AUTOIMMUNE
DISEASE AND MALIGNANCY
Idiotypic cross reactivity represents one other well documented property shared
by both natural autoreactive and early preimmune precursor B cells (Hartman et al.
1989, Bona et al 1988, Souroujon e ta l 1988). Many CRI that have been identified
on human antibodies have also been found on paraproteins from patients with B lineage
malignancies (Kipps et al 1988a, Kipps et a l 1988b). Sequence analyses have now
provided a structural explanation for such observations. For example, expression of
the cross reactive idiotype G6 on two recently sequenced CLL derived
immunoglobulins (AND and NEl) and by the Kim 13.1 autoantibody can be attributed
53
to their common utilisation of VHl family represented by germline gene 51P1
(Siminovitch et al. 1990, Kipps et al. 1989, Mageed et al. 1986). Expression of
another RF associated CRI 17.109 on about 20% of IgM RF CLLs appears to reflect
their frequent use of the Humku 325 germline gene (Kipps et al. 1988). This gene
encodes many RF’s and several leprosy-derived polyreactive antibodies which bind
both IgG and DNA (Dersimonian gr a/. 1989). These data highlight genetic similarities
between autoantibodies and malignant B cell paraproteins and suggests that they
originate from the same set of precursor cells in early immune repertoire. CLL is a
malignant expansion of B cell subpopulation CD5+ B cells. These cells are found
predominantly in the fetal spleen and lymphoid tissue (Antin et al. 1986, Royston et
al. 1980). CLL immunoglobulins preferentially utilise the VH251 gene. This is a
germline gene which is part of the relatively small VH5 gene family, a family which is
also associated with the early B cell repertoire (Chen 1989, Cuisinier et al. 1989). If
CLL immunoglobulins express a selected and conserved set of genes then the
malignancy might be amenable to therapy.
There is some similarity between the Kim 4.6 VXl sequence and the T2/C5 and
4G12VX gene sequences. T2/C5 is one of two almost identical VX sequences isolated
from two idiotype negative variants of a human B cell lymphoma line (Bernstein et al.
1989). In both these variant lines, loss of idiotypic expression correlated with a second
rearrangement of a previously excluded light chain allele. The 4G12VA. gene encodes a
monoclonal antibody with broad reactivity against tumour cells and is thought to
recognise a novel tumour associated antigen (Kishimoto etal. 1989, Saito etal. 1988).
As these latter sequences are identical it appears likely that they represent a germline
VA. 1 gene distinct from but closely related to the Hum IVl 17 gene (germline) encoding
the Kim 4.6 light chain. This suggests a possible relationship between autoreactivity
and malignant transformation. A human IgG antibody reactive with cytomegalovirus
(EVl-15) appears to be encoded by somatically diversified variants of the autoantibody
related VH and VL genes 51P1 and Kv 325 respectively (Newkirk & Capra 1989).
These observations suggest that autoreactive V genes serve as precursors for antibodies
against exogenous antigens and natural autoreactivity is instrumental in the
development of normal immune responses.
54
1.10 IDIOTYPES WHICH ARE ELEVATED IN SLE
A growing number of anti-DNA antibody associated idiotypes have been
described and summarised in recent reviews (Watts & Isenberg 1990). These have
been identified by monoclonal and polyclonal anti-idiotypic reagents raised against anti-
DNA antibodies of single and double strand specificity, either affinity purified from the
serum of SLE patients (Livneh et al 1987a), eluted from nephritic kidneys of SLE
mice (Hahn & Ebling 1987), or generated as human or mouse monoclonal antibodies
produced by hybridomas derived from lymphocytes of SLE patients (Shoenfeld et al
1983b, Hahn & Ebling 1984).
In order to identify those which may prove to be useful as specific markers for
disease the level and frequency of expression of anti-DNA associated idiotypes in SLE
patients has been compared to their frequency and level of expression in other disease
groups and healthy control populations.
Studies at the Bloomsbury Rheumatology Unit have focussed on the frequency
of expression of idiotypes carried by human monoclonal autoantibodies with binding
characteristics commonly found on autoantibodies in SLE patients, primarily the ability
to bind ss or dsDNA. This work has concentrated on a number of human monoclonal
antibodies generated and characterised within this department. The RT series of human
IgM monoclonal antibodies were derived from the splenocytes of an SLE patient
(Ravirajan e ta l 1992), as was the WRI176 human IgM monoclonal (Blanco e ta l
1994). The BEG2 human IgM monoclonal was derived from the human fetal liver
cells (Watts et a l 1990). The B3 and D5 IgG monoclonals were derived from
peripheral blood lymphocytes of an SLE patient with active disease (Ehrenstein et a l
1993). Analysis of the autoantigen binding profiles of the antibodies showed that these
IgM and IgG monoclonal antibodies bound predominantly to DNA, although some
reactivity with other autoantigens and polynucleotides was also noted. It was
determined that the IgM antibodies exhibited a range of binding affinities for DNA, not
just low affinity (Ravirajan et al 1992).
Polyclonal rabbit anti-idiotypic reagents raised against these monoclonal
antibodies identified determinants on the light chain for RT84 and B3, and on the heavy
chain for RT72, WRI176, BEG2, and the frequency of expression of these idiotypes in
the sera of SLE patients were as follows; RT84-19%, RT72- 12% WRI176-44%, B3-
55
26% and BEG-2-9% (Kalsi e ta l 1994, Ehrenstein e ta l 1994, Blanco e ta l 1994,
Watts era/. 1991).
Idiotypes originally identified on monoclonal antibodies with binding
specificities other than DNA have been shown to be present on anti-DNA antibodies
and found at elevated frequency in SLE patients. The 9G4 idiotope, originally
identified on cold agglutinin antibodies, has been found on anti-DNA antibodies and is
present in the sera of 45% of SLE patients (Isenberg et al 1993). This idiotope differs
from many of the previously described anti-DNA idiotypes in that its presence in
patients with autoimmune rheumatic disorders other than SLE is uncommon.
A comparative study of 19 anti-DNA associated idiotypes from 11 different
laboratories found that no single idiotype showed a strict correlation of expression and
association with disease or clinical activity, but rather individual idiotypes were present
in different affected individuals, and that individuals may possess a limited number of
idiotypes (Isenberg etal 1990). It was however found in some SLE patients that the
level of idiotype expression did mirror disease activity, in some cases more closely than
anti-DNA antibody levels themselves (16/6Id, 3IId, GN-lId, GN-2Id, Isenberg etal
1991, and NE-lId, 0-8lid Muryoi etal 1988).
Recent studies have shown that levels of B3 Id carried on IgG antibodies
(derived from an IgG anti-dsDNA autoantibody) showed a correlation with disease
activity in some SLE patients more precisely when individual clinical features were
correlated with Id expression rather than a global clinical score (Ehrenstein eta l 1994).
The levels of antibodies which carry the 9G4 Id have also been shown to correlate well
with disease activity in SLE patients (Isenberg etal 1993). It has also been shown by
affinity purification of anti-DNA antibodies from the serum of SLE patients that anti-
DNA associated idiotypes detected in these SLE patients are present predominantly on
anti-DNA antibodies (B3 Id Ehrenstein et a l 1994, 9G4 Id Williams - personal
communication).
Thus some anti-DNA associated idiotypes may be more frequently detectable in
the sera of SLE patients with increased clinical activity and may identify autoantibodies
which are involved or clearly associated with pathogenesis.
Specific amino acid (AA) sequences have also been shown to be associated
with autoreactive antibodies, for example an idiotope designated 9G4 originally
identified on cold agglutinin antibodies known to be a marker for immunoglobulins
56
which utilise the VH4.21 gene has been found on 2-20% of anti-ss and anti-dsDNA
antibodies (Stevenson et al. 1993) and have been identified in 45% of sera from
patients with SLE (Isenberg et at. 1993) The 9G4Id, as defined by a rat monoclonal
anti-idiotype, has been found to ‘map’ to the FRl region of antibodies encoded by the
VH4.21 gene. It has been shown by site directed mutagenesis studies that an amino
acid motif alanine-valine-tyrosine (AVY) in framework constitutes the 9G4Id specificity
(Potter gr aZ. 1993).
Isenberg and colleagues initially reported the presence of the 16/6 Id in SLE
patients, demonstrating that 54% of patients with active disease had raised serum levels
compared to 25% of patients with inactive disease (Isenberg etal. 1984). They also
also noted that the 16/6 Id was not disease specific being present in the serum of 25%
of rheumatoid arthritis patients compared with 4% of healthy controls.. Furthermore
the 16/6 Id+ antibodies have been found in the immune deposits in the renal and skin
lesions of the disease (Isenberg & Collins 1985b, Isenberg etal. 1985a).
N-terminal sequencing has shown that the 40 N-terminal residues of light
chains of human hybridoma derived monoclonal anti-DNA antibodies 1/17, 16/6 and
18/2 are identical and differ only by a single amino acid from the Waldenstrom
macroglobulin (WEA). Each of these bear 16/6 idiotype (Atkinson et al. 1985,
Naparstek et al. 1985). The variable regions of light and heavy chains of monoclonal
1/17 have been sequenced and the terminal 40 amino acids of the heavy chain are
identical to that predicted for VH26 a member of the VH3 family (Chen et al. 1988).
Given that the VH26 germline gene appears to encode the 16/6 Id positive anti-DNA
antibodies it is tempting to speculate that natural autoantibodies can be related to
pathogenic autoantibodies.
To determine whether these anti-DNA associated idiotypes contribute to
pathogenesis, their occurrence in immunoglobulin deposits at sites of antibody
mediated damage has been sought. One site at which antibody mediated damage occurs
is in the pathogenic skin lesions seen in SLE patients, where antibody and complex
deposition at the basement membrane of the skin initiates an inflammatory cascade with
consequent tissue damage. Anti-DNA associated Ids have been identified in skin
lesions of SLE patients in a number of studies such as 16/6 Id (Isenberg etal. 1985)
and 31 Id (Manheimer-Lory etal. Lory 1991).
However the site where antibody mediated damage leads to serious
57
complications is the kidney, where nephritis and subsequent renal failure are a major
cause of mortality in SLE patients, thus a study of anti-DNA associated idiotypes
present in inflammatory deposits and their origins is of particular importance. It has
been found that some anti-DNA associated idiotypes expressed on a large proportion of
anti-dsDNA antibodies in SLE patient serum are detectable within immunoglobulin
deposits in the kidneys of SLE patients with glomerulonephritis, and it has been
suggested that antibodies carrying certain anti-DNA associated idiotypes may
preferentially deposit in the kidneys, leading to nephritis. (16/6; Isenberg eta l 1985b,
31; Manheimer-Lory etal. 1991, 9G4; Isenberg etal. 1993, RT84, RT72; Kalsi etal.
1994, GNl, GN2; Hahn and Ebling 1987).
The charge carried by an antibody may influence its ability to form deposits in
the kidney. Studies have shown that anti-DNA antibodies which carry a cationic charge
may preferentially deposit on the negatively charged basement membrane in the
glomeruli contributing to glomerulonephritis, thus anti-idiotypic reagents recognising
idiotypes on this subset of anti-DNA antibodies may identify those which are
particularly pathogenic (Kallunian et al. 1989, Hahn et al. 1987, Gavalchin et al.
1987).
Analysis of idiotypes associated with cationic human anti-DNA antibodies have
focussed on three idiotypes designated 8.12, 31 and F4 reactive with LX, Lk, and H
chains respectively (Linveh etal. 1987b, Solomon etal. 1983, Davidson etal. 1989).
These anti-idiotypes were raised against affinity purified serum anti-dsDNA antibodies
from three SLE patients and shown to bind preferentially to cationic anti-dsDNA
antibodies. Neither 8.12 nor 31 recognise binding site related idiotypes but bind to
variable region framework determinants.
The 8.12 idiotype has been detected at elevated levels in serum from 50% of
patients with SLE and is present on antibodies in glomerular deposits. As the 8.12Id
has been identified on both IgM and IgG isotype anti-DNA antibodies and may not be
exclusively expressed by pathogenic autoantibodies (Linveh etal. 1987b).
Antibodies bearing the 31 idiotype have been detected in serum from 75% of
SLE patients and may comprise 40-90% of the anti-DNA antibody levels. In addition,
elevated levels of 31 Id positive antibodies have been shown to correlate with elevated
levels of anti-DNA antibodies (Solomon etal. 1983, Davidson etal. 1986). However
high levels of 31 idiotype positive antibodies have also been found in SLE patients in
58
remission and it was postulated that these 31 idiotype positive anti-DNA antibodies may
be present bound in immune complexes, not detectable by standard methods of
assessing DNA reactivity and were only detectable after affinity purification of 31 Id
positive antibodies using techniques which dissociate immune complexes.
However recent studies by Manheimer-Lory etal (1991) have shown that the
charge and specificity of antibodies which carry the 31 Id (a light chain framework
encoded idiotype) can be markedly influenced by the heavy chain with which it is
expressed. In this study antibody secreting Epstein-Barr \^rus (EBV) transformed B
cell lines from lymphocytes of SLE patients and a patient with multiple myeloma were
established. Those which secreted 31 id antibody were analysed for variable region
gene utilisation whilst the specificity of the secreted antibodies was studied. It was
found that single monoclonal antibodies whilst carrying the 31 id could be DNA
binding or non-DNA binding, and could carry a cationic or neutral charge. The
determining factor for charge or specificity of 31 Id positive antibodies was the heavy
chain co-expressed with the 31 Id light chain. These observations imply that the
expression of anti-DNA antibodies in SLE patients could result from the preferential
association of certain L and H chains.
The F4 anti-idiotype identifies an idiotype on heavy chains of cationic anti-
dsDNA antibodies. F4 reactive antibodies are almost exclusively IgG and found in the
serum of 60% of SLE patients and are detectable in glomerular deposits (Davidson et
al. 1986). As the F4 idiotype is almost exclusively found on IgG antibodies, it is likely
to be derived from a somatically mutated germline gene which has arisen during the
secondary immune response.
Whilst the 31 Id and F4Id can be detected on separate antibodies, they can also
be found expressed together on the same antibody where a 31 expressing light chain
and an F4 expressing heavy chain form a single antibody. In such cases
immunoglobulins expressing both 31 and F4 are cationic whether or not they are
isolated from an SLE patient. However, only those 3I/F4 Id antibodies derived from
SLE patients display dsDNA binding activity.
Northern blot analysis of a 31 reactive cell line has shown 22/26 clonal lines use
the Vkl gene family to encode 31 light chain sequence and that only two VkI genes are
used. The high frequency of autoreactive IgM antibodies from EBV lines of SLE
patients might reflect Ig gene polymorphisms that lead to germline encoded DNA
59
binding. If such polymorphisms were specifically associated with SLE, they would
constitute a genetic basis for the familial predisposition to SLE (Brodeur et al 1984).
F4 reactive B cell lines producing IgG, often use VH3 and VH4 to encode their heavy
chains.
The 2A4/EBV transformed B cell line expresses both 31 and F4 idiotypes and
possesses high affinity anti-dsDNA binding activity (Davidson eta l 1989). The kappa
chain of the 2A4 antibody is encoded by a member of the VkI gene family. The most
homologous VkI germline sequence differs significantly from the 2A4 light chain
sequence. The F4 reactive heavy chain of 2A4 is encoded by a VH4 gene whose
sequence has already been reported (Davidson et a l 1989, Lee et a l 1987). The
closest reported VH4 germline sequence differs by only 24 base pairs. (Sanz et a l
1989c, Lee e ta l 1987, Berman e ta l 1988). Some somatic mutation has occurred in
2A4 in the CDR2 region. Of the 24 presumed somatic mutations in the heavy chain
variable region gene, 14 code for amino acid substitutions and 10 are silent (Davidson
et a l 1989). Nine of the presumed changes are in the CDR regions and all nine are
replacement mutations. This suggests antigen driven somatic mutation may have taken
place as these replacement mutations occur in CDR regions which contribute to form
the antigen binding site. Two of these mutations which occur in CDRs introduce
positive charges which are believed to be important in binding to negatively charged
DNA. Therefore it appears that somatic mutation plays a role in creating high affinity
anti-DNA activity. This is likely to make the antibody more pathogenic (Sharon et al
1989, Kocks et a l 1988). The light chain has homology with WEA, GAl, HAU,
HKlOl and DEE (Lampman et a l 1989). Sequencing of N terminal amino acid
sequencing of four H chains from myeloma proteins highly reactive with F4 revealed
that three were encoded by VH3 and one VH2. The F4 idiotype activity is associated
with more than one gene family.
It is not yet clear if antigen stimulates the autoantibody response in SLE, but it
is possible that environmental antigens and even autoantigens other than DNA could
elicit the production of anti-DNA antibodies. In susceptible individuals this might
occur in one of three ways:- An autoantibody might arise with a specificity such that it
cross reacts with both self and foreign antigens (Male et a l 1988, Monestier et a l
1987, Carroll eta l 1985). Alternatively, regulation within the idiotypic network might
induce expression of ‘homologous members’ of an idiotypic family that bind to self
60
rather than foreign antigens (Male et al. 1986, Rajewsky et al. 1983, Williams etal.
1988) or it could be that protective antibodies became autoreactive via somatic
mutation.
1.11 GENE USAGE IN AUTOIMMUNE DISEASE AND THE FETUS
One of the hallmarks of the murine B cell repertoire during ontogeny is its
limited heterogeneity matched by the production of low affinity poly reactive
autoantibodies of the IgM isotype (Perhnutter et al. 1985, Yancopoulos et al. 1984). B
cells in the developing human fetus express a restricted antibody repertoire
( Schroeder a/. 1987, Hillson etal. 1989).
JH proximal VH genes are major constituents of immunoglobulin heavy region
messenger RNA produced by B cell lineages from murine fetal liver. The frequency of
VH usage does not appear to be based upon the complexity of each VH gene family.
However, in the normal adult, utilisation of VH genes becomes randomised. It is
possible that the restricted usage of the VH segments could be due to the position of a
recombinase recognition sequence enhancing the ability of a particular VH sequence(s)
to interact with recombinase. It is difficult to distinguish the relative importance of
chromosomal position versus sequence specific elements or whether the sequences are
particularly chosen in the fetal state to play a vital role in the development of the
immune system eg. in moulding the idiotypic interactions.
An apparently biased usage of the small VH5 family, in up to 30%, of patients
with chronic lymphocytic leukemia has been reported. Furthermore these are mainly
found in CD5+ B cells. It is possible that CD5+ cells, high levels of which are found in
the fetus, have a biased repertoire. The germline gene preferentially expressed in CLL
is VH251, a member of the VH5 family.
VH genes encoding autoantibodies have been described which are identical or
very similar to VH sequences found to be preferentially expressed in fetal repertoires,
for example, over-expression of a VH3 encoded anti-Sm autoantibody has been
demonstrated in two fetal liver samples. It is possible that potentially autoimmune
polyreactive VH genes are preferentially expressed in immature B cell repertoires to
establish a protective idiotypic network. However, in the autoimmune state a
perturbation of the network occurs as the result of defects in the normal developmental
61
process of the repertoire and this results in the over-production of autoantibodies.
As mentioned previously, it appears that there is a preferential rearrangement of
3’ JH proximal VH genes in the murine and human fetal pre B cells. This phenomenon
is interesting as studies of autoimmune murine systems have shown a fetal like pattern
of VH gene usage (Bona 1988, Trepicchio eta l 1987a, Wysocki etal. 1985). It is for
this reason that there has been much interest in human natural fetal autoimmune
responses. This in turn implies that autoreactivity plays an important physiological role
in the developing immune system.
Monoclonal antibody BEG 2 is a dsDNA binding IgM derived from a 12 week
old human fetus. Two binding site idiotypes were defined as BEG2 Ida light chain
associated and BEG2 Idp located on the heavy chain. This latter idiotype was located
on an EBV derived Mabs from human fetal liver or spleen (5%) human cord blood
(2.7%) and adult peripheral blood (1%) (Watts et al. 1991). The idiotype is also
present on 8.5% of adult derived hybridomas antibodies and 6% of RA synovium
derived antibodies. The BEG2 heavy chain was sequenced and was found to be
encoded by a member of the VH4 family joined by a JH5 variant and a very short
diversity region. Of the BEG2 Idp positive monoclonals, five expressed VH4 and one
expressed VH6. Therefore BEG2 Idp identifies a set of polyreactive antibodies that are
common in fetal life, persist into adult life and are encoded by a VH6 and a subset of
VH4 genes.
Sequence analysis of Ig heavy chains from human 130 day fetal liver showed
expression of a restricted set of VH genes. These were two VHl (51P1, 20P3), five
VH3 (56P1 30P1 38P1 60P2 and 20P1), one VH4 (58P2) and one VH6 (15P1) genes.
30P1 was found to be identical to VH26 (Barrett etal. 1987, Mathyssen and Babbitts
1980, Chen et al. 1988). In addition 20P1 is identical to the heavy chains of the
human-human hybridoma derived 4B4 anti-Sm antibody and the heavy chain of 51P1
is identical to those of two polyreactive autoantibodies L16 and MLl (Sanz et al.
1989b, Dersimonian et al. 1987). Schroeder & Wang (1990b) constructed a cDNA
library from a 104 day old fetal liver and apart from the presence of two members of
the VH2 family and one member of the VH5 family found a similar situation to that
found in the 130 day old fetus. Cuisinier etal. (1989) analysed Ig transcripts from a 7
week old fetal liver by dot blot hybridization using probes specific for the six human
VH gene families. Their results suggest that the VH5 and VH6 families are first
62
expressed at the time when Ig rearrangement begins in the fetal liver (Gathings et al.
1977). Berman & Alt (1990) studied by Northern blotting analysis, the hybridization
ratio of different VH probes to mu RNA from fetal liver 16-24 weeks gestation and
compared it to adult lymphoid tissue. They concluded that the most D-proximal VH
family VH6 is over-represented in fetal liver as compared to adult tissue whereas the
expression of VH3 and VH4 is similar at both fetal and adult stages of development.
Logtenberg et al. (1989) analysed by Northern blotting analysis a total of 187
monoclonal IgM secreting EBV lines derived from human adult and fetal tissues. They
found a correlation between the frequency of VH family use with the complexity of
each family suggesting that the population of transformable B cells was randomised in
both repertoires. From a study of fetal cDNA 5 VH3 genes were found to contribute to
60% of the repertoire thus there may be a restriction within a limited set of the VH3
gene segments.
To explain the above, Coutinho et al. (1988) proposed a network model
whereby in a sterile fetal environment stimulation by autoantigens and/or idiotype -anti
idiotype interactions, selectively expands those B cells expressing the autoreactive V
genes to form an initial functional network. However, with maturity and exposure to
exogenous antigen the self reactive B cell pool diminishes in size to about 20%,
interacting with the remaining 80% of non-autoimmune resting B cells that are not
connected with the network. With repeated exposure to non-self antigens, the non-
autoimmune resting B cells respond and, via somatic mutation and gene conversion,
produce classic high affinity antibodies.
Antigen selection cannot account for the programmed expression of certain V
genes. In humans the most proximal VH6 gene is frequently expressed and it may be
that the physical distance between the V and DJ loci can influence their expression
during early development. There appears to be evidence for the conservation of fetally
associated V genes with the outbred population. This suggests that there is
evolutionary pressure for preservation of these V genes, implying that autoreactivity
plays an important role physiological role in the developing immune system.
63
1.12 VH GENE UTILISATION TN THE NATURAL ANTIBODY
RESPONSES
The occurrence of autoantibodies in apparently healthy individuals is well
established. Their physiological significance and relationship to disease associated
autoantibodies is not clear. It is possible that natural immune responses originate
during early B cell development. Natural autoantibodies can have immunochemical
specificities with a striking resemblance to those found in patients with autoimmune
disease (Souroujon etal. 1988).
The characterisation of human natural antibodies has allowed a number of
observations to be made. Most seem to be of the IgM isotype and bind antigen in a
polyreactive manner and each VH gene family is represented. Sanz et al. (1989b)
sequenced seven natural antibodies derived from EBV transformed CD5+ B cells from
normals, they found that members of the VH4 family were expressed in three of them
implying some kind of restriction in VH gene usage. However no restriction appears
to occur in natural antibodies with anti-DNA specificity since members of five VH
families have been found to be expressed in this particular system; five express VH3,
five express VH4, two express VHl and one partial sequence expresses a VH2
segment. (Sanz etal. 1989b, Dersimonian etal. 1989, Cairns etal. 1989).
Finally there are several examples of natural autoantibodies using VH gene
segments that are related to the fetal repertoire and have 100% homology to their
germline gene of origin.
A human tonsillar derived natural autoantibody Kim 13.1 has been studied. The
heavy chain variable region was found to be identical (except for one nucleotide ) to a
VHl gene 51 PI. This gene is over-represented in the fetal repertoire (Schroeder etal.
1987). The Kim 13.1 V k gene shows extensive similarity with a germline V k3
sequence Vg which is identical to 38K a cDNA recently isolated from a 130 day fetal
liver (Siminovitch et al. 1990, Pech et al. 1984). It appears that both the heavy and the
light chain V genes expressed by Kim 13.1 represent germline encoded genes belonging
to the selected set of V genes expressed early in ontogeny. A second tonsillar derived
natural autoantibody Kim4.6 differs by only one nucleotide from a fetally derived VH
gene FL2-2VH (Nickerson et al. 1989) and closely resembles a second fetal associated
VH3 gene 56P1 (Schroeder et al. 1987). It has also been reported that not only
64
polyreactive antibodies but also those binding Sm, the Fc of IgG, dsDNA and
thyroglobulin are encoded by a restricted fetal VH gene repertoire (Sanz et al. 1989a,
Sanz era/. 1989b, Newkirk & Capra 1989) .These results suggest a fetal like
restriction of autoimmune responses and reinforce the notion that natural autoantibodies
reflect early B cell repertoire. This is also supported by the report from Logtenberg et
al. (1989) which shows a correlation between expression of auto reactivity and
utilisation of highly conserved JH proximal VH6 gene. Also unexpected was the
conservation of these genes among the human population, an example being the
relationship between two rearranged genes Humha 356 and fetal derived VH1.9III and
humhv 3005 which have been isolated from unrelated individuals and are 99% or more
related (Chen 1989). The VHl genes 51P1 halLR AND and NEI derived from
different individuals are also identical. This evidence suggests that there is some
evolutionary pressure at work for the conservation of sequences which encode natural
autoantibodies.
1.13 THE ROLE OF SOMATIC MUTATION IN THE GENERATION OF
AUTOANTIBODIES
In detailed studies of the immune response to foreign antigen, most of the initial
IgM antibodies have not undergone somatic mutation and have relatively low affinities.
Somatic mutation of germline sequences is required to generate high affinity binding to
the eliciting antigen. (French gfaZ. 1989, Radac a/. 1991). The somatic mutations
that occur during the course of the anti-DNA response in autoimmune mice have been
analysed by comparing the ratio of replacement (R) to silent (S) mutations within the
CDRs that form the antigen binding sites with the R/S ratio in the framework regions.
This study has also been done on the response of antibodies to foreign antigen. If
mutations occur randomly then there would be 2.9 times as many replacement as silent
mutations (Shlomchik etal. 1987). A complilation of R/S ratios in the framework
regions reveals a ratio 1.5 which is evidence of negative selection for replacements
within frameworks which are responsible for the overall folding and assembly of the
antibody. R/S ratios of 3.0 or above are presumed to be evidence for positive selection
for mutation in the areas of the antibody that they are seen. Thus high R/S ratios within
65
the CDR regions could suggest an antigen driven response. Distinctions between
germline encoded natural autoantibodies and disease related IgG antibodies have been
described for both RF and anti-DNA antibodies (Carson et al. 1987, Eilat et al. 1986,
Smeenk^ra/. 1988). However, Shefner gtaZ. (1991) showed that non-autoimmune
strains of mice contain B cells that can produce germline encoded high affinity IgG
antibodies.
Some experiments point to the pathogenic potential of germline encoded antibodies
whereas others show that pathogenic autoantibodies in SLE arise from somatic
mutation with DNA as a selecting antigen (Shlomchik etal. 1987).
A lupus derived anti-DNA antibody 21/28 is encoded by a germline gene from
the VHl gene family. VH21/28 was derived from a woman of Iranian decent and is
identical in nucleic acid sequence to 8E10, an anti-DNA antibody from a Thai patient
with lepromatous leprosy. This observation argues that these genes represent one
unmutated germline VH gene and shows evidence of conservation of this gene in the
population. The closest known sequence to 21/28 and 8E10 is VHl germline gene
1.92 which is 96% homologous.
Monoclonal IgM RFs have been examined in detail and the human autoantibody
associated kappa light chain V gene Hum Kv 325 (Chen 1990) was isolated. This
sequence has been found to be identical in four RFs and one antibody directed against
intermediate filaments. Analysis of further light chain V sequences revealed an
additional eight RFs and one anti-LDL antibody with little or no somatic mutation.
These clones were all isolated from different individuals. This provides further
evidence that autoantibodies can be encoded by inherited germline genes without
somatic mutation, their genes being conserved in the population and this V k gene can
encode autoantibodies of different specificities. This gene has also been associated
with an antibody against cytomeglovirus and other foreign antigens.
A VH gene sequence VH51P1, a member of the VHl family, has been found to
partly encode a rheumatoid factor-associated cross reactive idiotype designated
G6,which was also present on two antibodies derived from patients with CLL. These
autoantibodies may therefore be considered identical or close to germline sequence
encoded (Newkirk etal. 1987, Mageed etal. 1986, Kipps etal. 1989).
The human IgM anti ss/dsDNA antibody NE-13 was derived from a patient
with SLE and carried the NE-1 Id which has been detected on glomeruli-deposited
66
anti-DNA antibodies ( Hirabayashi er a/. 1993). The VH and Vk gene segments of
NE-13 were identical with the germline genes VH4.21 and Vb respectively. This
suggests that some IgM anti-DNA antibodies which express the NE-1 Id associated
with lupus nephritis use germline genes without mutation.
It has been proposed that some cells expressing the gene HumKv 325 may be
stimulated by autoantigens to proliferate constantly and that if these cells are continually
cycling they could be more susceptible to abnormal clonal expansion and malignant
transformation (Chen 1990). This theory would also apply to other autoreactive V
genes such as VH51P1 and VH26 which encode the 18/2 anti-DNA antibody, Ab255
an antithyroglobulin antibody and polyreactive autoantibodies AblS and Ab21 which
are also expressed by 10% B cell CLLs (Sanz etal. 1989b, Fong etal 1985).
A substantial proportion of the VH genes expressed in autoantibodies thus far
studied show considerable homology or complete identity with germ line sequences.
Thus somatic mutation of VH genes is not obligatory for the generation of
autoantibodies, including specificities other than for DNA (Dersimonian etal. 1990).
There is evidence that IgG anti-DNA antibodies are associated with clinically
active lupus whereas IgM antibodies against denatured DNA can occur without any
disease. Class switch from IgM to IgG is associated with the diversification of
responding population of antibodies through somatic mutation of V genes. Results
obtained by sequencing a large number of mRNAs of MRL Ipr/lpr hybridomas provide
support for the role of somatic mutation in the development of IgG anti-DNA
antibodies (Shlomchik et al. 1990a). In these analyses clonally related hybridomas
from a single mouse defined a series of mutations towards the inclusion of arginine and
asparagine in the CDRs in association with increasing affinity for denatured DNA.
Winkler et al (1992) produced six IgG monoclonal anti-DNA antibodies from three
patients with active SLE. The distribution of presumed somatic mutations in these
antibodies were analysed. A high replacement to silent mutation ratio was found in the
CDR regions of all six heavy and light chains. This non-random pattern of somatic
mutations is characteristic for antigen driven immune responses with affinity maturation
(Berek gr a/. 1987).
In conclusion, it is clear that IgM autoantibodies can generated by V regions
that have close identity with germline genes. However, IgG anti-DNA antibodies, the
isotype thought to be most closely associated with pathogenesis in SLE, are somatically
67
mutated and show evidence of an antigen driven mechanism.
1.14 V GENE USAGE IN ATJTOANTIBODIES
D ANn-HLSTONE ANTTBODTES
Anti-histone antibodies can be found in all the major immunoglobulin classes,
but there is little agreement as to the predominant isotype. Earlier studies have
indicated that some anti-DNA antibodies, a common serological feature of SLE, can
arise from unmutated germline genes, whereas some contain multiple somatic
mutations. Two reports of the nucleotide sequences of murine anti-histone antibodies
demonstrated that these antibodies were not clonally related and that diverse V, D and J
genes were represented (Monestier eta l 1993). Seven out of eight had VH segments
encoded by genes from the J558 family. This is of interest as the J558 family is
commonly used amongst autoantibodies of various specificities (Koffler et al. 1987,
Komisar et al 1989). The second CDR of the heavy chain of MRA12 (the most acidic
and the strongest histone reactive antibody) included only two positively charged but
five negatively charged amino acid residues. This feature is unusual since the CDRs of
most of the VHJ558 genes are not comprised predominantly of acidic residues.
Two human IgM anti-Hl antibodies WRI 170 and BEN27 have been analysed
(Tuaillon e ta l 1994). The variable regions were found to use VH 1 -DN1 -JH4/VA.3-
JX2 and VH3-DIR2D 2 1 /9-JH 1/VX2-JX2 gene segment combinations respectively. The
CDRs of these monoclonals contained several negatively charged amino acids and
striking similarities were found between the CDR2 regions of BEN27 and a murine
anti-Hl antibody MRA12. These data points to the potential importance of acidic
amino acid residues in binding to basic histone proteins.
ID ANTI-CARDIOLIPIN ANTIBODIES
The anti-cardiolipin antibodies from patients with SLE constitute a
heterogeneous population that may cross-react with other phospholipids and
polynucleotides (Rauch et al 1984). Importantly, DNA/cardiolipin cross reactivity is
characteristic for autoantibodies eluted from the kidneys of patients with active lupus
nephritis and from lupus prone mice (Sabbaga eta l 1990, Pankewycz etal. 1987).
68
An IgG anti-cardiolipin /ssDNA antibody R149, was isolated from a patient
with active SLE (van Es et al 1992b). The nucleotide sequences of the V regions
encoding R149 had 94.6% homology with the VHl member 51P1 and 99.4%
homology with the Vk2 member A3. Comparisons between the V regions of R149 and
their germline counterparts revealed that all replacement mutations resided in CDRl
and CDR2 in both heavy and light chains. A high frequency of arginine residues were
found in the CDR regions of the variable region, particularly in the CDR3 region of the
heavy chain where five out of twenty residues were arginines. These data points to a
role for basic residues in binding to not only DNA but also cardiolipin.
ANTI-DNA ANTIBODIES
Natural antibodies produced by auto-reactive clones found in the repertoire of
normal humans are known to contain a DNA binding population. These natural
antibodies however are of the IgM isotype, are often polyreactive, and may bind to
several auto and/or foreign antigens. Although autoantibodies of the IgM isotype are
deposited in the kidneys of lupus patients, those thought to mediate pathogenesis are of
IgG isotype. IgG anti-DNA antibodies have high affinity and are found more
specifically in lupus patients, thus the key to a better understanding of the
characteristics of ‘pathogenic’ anti-DNA antibodies could lie within the amino acid
sequence and structure of IgG anti-DNA antibodies. Until recently it has been
extremely difficult to immortalise human B cells (including those from SLE patients)
secreting antibody of IgG isotype which are stable and secrete monoclonal antibody in
amounts sufficient for detailed analysis. Thus only a limited number of IgG anti-DNA
antibody sequences are available with which to perform a valid sequence/structure
analysis. Tables 1.3 and 1.4 summarise the anti-DNA antibodies sequenced to date.
Amino acids that may have the greatest effect on protein binding to DNA are
arginine, asparagine, glutamine, tyrosine, lysine and histidine (Pabo & Sauer 1984).
Because of their charge and ability to form hydrogen bonds with either deoxyribose
phosphate or base pair structures of DNA, arginines are thought to be important for
protein recognition and binding to DNA (Seeman era/. 1976). Arginine can potentially
form hydrogen bonds with an individual G-C pair through the major groove of a DNA
double helix. Sequencing studies of a large number of mRNAs of MRL Ipr/lpr
hybridomas revealed that clonally related hybridomas from a single mouse defined a
69
series of mutations towards the inclusion of arginine and asparagine in the CDRs in
association with increasing affinity for denatured DNA (Shlomchik et al. 1990).
Crystallographic analysis and computer modelling of murine anti-DNA antibodies have
revealed that amino acids 31-33 in VL CDRl, 92-94 in VL CDR3, 52 in VH CDR2
and 100-100a in VH CDR3 are sites for DNA interaction and are often occupied by
polar amino acids (Tillman et a l 1992, Herron e ta l 1991, Eilat e ta l 1988). In an
analysis of a large panel of monoclonal anti-DNA antibodies derived from NZB/NZW
mice Tillman eta l (1992) found the presence of arginine at positions lOO-lOOa as well
as more than one arginine in VH CDR3 were associated with high avidity to ds DNA.
Thus it is likely that the appearance of arginine residues within the CDR regions of
human anti-DNA antibodies involved in DNA binding. Site directed mutagenesis
studies with murine anti-DNA antibodies have supported the role of arginine residues
in DNA binding (Radie et al 1993).
A summary of the IgM monoclonal anti-DNA antibodies sequenced are shown
in Table 1.3. Dersimonian et al (1987) isolated four IgM anti-DNA antibodies, three
were derived from an SLE patient, one from a patient with leprosy. Two of the SLE
monoclonals 18/2 and 1/17 carried the 16/6 idiotype levels of which have been found to
be elevated in the sera of patients with active lupus (Isenberg et a l 1984). The
nucleotide sequences of these antibodies were found to be identical and had 99%
homology with the germline gene segment VH26. 21/28 the third SLE monoclonal
was found to have 93% homology to the VHl germline gene segment HA2 and was
idiotypically related to 8E10, a monoclonal antibody derived from a patient with
Leprosy. 21/28 and 8E10 shared 100% homology at the nucleotide level thus showing
strong evidence of the conservation of this germline gene.
Interestingly 21/28 and 8E10 shared homology with the VH26 encoded anti-
DNA antibodies in the first CDR. 21/28 also shared homology with the Id 16/6+
antibodies within the the third CDR region.
C6B2, (Hoch & Schwaber 1987) an IgM anti-DNA (ss & ds) antibody derived
from a six month old non-autoimmune individual, was compared with an anti-DNA
mouse monoclonal (Kofier et a l 1985). Striking homologies were found within the
CDR2 regions of the heavy chain where 9 out of 15 residues were shared including an
identical amino acid stretch of Ser-Thr-Asn-Tyr-Asn. There was also a shared Ser-Tyr
‘construction’ at positions 31 and 32. Studies have shown that these regions are likely
Table 13
Characteristics of human IgM monoclonal anti-DNA antibodies
DNA GERMLINE D JH L GERMLINE REFERENCESANTIBODY
21/28
ORIGIN
SLE PEL
ANTIGEN
ss
VH
1
COUNTERPART
HA2 DXP’l 4 K / Dersimonian et al 19878E10 Leprosy PEL ss 1 HA2 ? 4 K / Dersimonian et al 1987IH2R SLE SPL ss 1 51P1 ? 5 K hkl37 Mannheimer-Lory et al 1991C6B2 sickle cell anemia ss > ds 2 VCE-1 DXPT 3 ? / Hoch et al 1992
TH3SPL
Leprosy PEL ss 2 VCE-1 DXPT 4 / / Dersimonian et al 198718\2 SLEPEL ss + ds 3 VH26 DXPT 5 K / Dersimonian et al 1987POP Leukemia and ss 3 VH26 DUR4 4 / / Spatz et al 1990
B19.7
periferal neuropathy PEL
Fetal EML ss (+RF) 3 VH26 7 4 / Young et al 1990III3R SLE SPL ds 3 56P1 DXPT 4 K / Mannheimer-Lory et al 1991
Kim 4.6 Healthy child ss 3 VHl,9111 DXPT 6 k / Cairns et al 1989
BEG-2tonsil Ls
Fetal liver Ls ss + ds 4 VH4.71-2 ? 5 X / Watts et al 1991IC4 Myeloma EML ds 4 VH4.71-4 7 4 K / Mannheimer-Lory et al 1991n- 1 SLE SPL ds 5 VH251 7 4 K vk328 Mannheimer-Lory et al 1991L16 Fetal liver Ls ss 6 VH6 052 3 K / Logtenberg et al 1989M Ll Fetal spleen Ls ss 6 VH6 052 4 X / Logtenberg et al 1989AlO Healthy adult PEL ss 6 VH6 052 3 K / Logtenberg et al 1989A431 Healthy adult PEL ss 6 VH6 ?052 4 X / Logtenberg et al 1989AB47 Healthy adult PEL ss (+RF) 7 ?VH7 ? 4 X / Sanz et al 1989
SUE = systemic lupus erythematosus; PEL = peripheral blood lymphocyte; ss = single stranded; SPL = splenic lymphocytes ds = double stranded; BML = bone marrow lymphocytes; RF = rheumatoid factor; Ls = lymphocytes, ? = unknown Adapted from Isenberg et al 1993
71
to form part of the antigen binding site. It is possible that hydrogen bonds occur
between the negatively charged phosphates on the backbone of DNA and the tyrosine
and asparagine residues.
Sequence analysis of an anti-DNA IgM monoclonal POP (Spatz et al. 1990)
from a patient with CLL and peripheral neuropathy has shown 96% homology with
germline gene VH26 and 99% homology to a Vic3a germline gene KV328 (Pech &
Zachau 1984). This antibody has Ser-Phe at position 31 & 32 instead of the Ser-Tyr
found in C6B2. The replacement of tyrosine with phenylalanine as described could
explain why POP monoclonal binds to ssDNA only.
B19.7 (Young et at. 1990), a 16/6 Id+ IgM anti-ssDNA of fetal origin was
compared to 18/2 because its binding to ssDNA was inhibited by the 16/6 anti-idiotype.
The results showed that antibodies with identical VH regions and expressing the same
idiotype may have different DNA binding sites. The sequences however had no
homology in the CDR3 region and the antibodies used different light chains, B 19.7
using X and 18/2 using k . These differences prevented the inhibition of ssDNA
binding by the 16/6 anti-idiotype in B19.7. Kim 4.6, an anti-ssDNA IgM monoclonal
derived from a normal healthy adult had 100% homology with the germline VH gene
segment VH9.III and 98% homology with a VL gene segment expressed in Burkitt’s
lymphoma (Cairns etal. 1989b). Kim 4.6 contains a region of 25 nucleotides which
matches the germline segment DXP’l (Ichihara et at. 1988a). The single D germline
gene DXP’ 1 has been seen in anti-DNA antibodies from different sources. The reading
frame of DXP’ 1 seen in anti-DNA antibodies encodes the amino acid sequence
YYGSGS. Other immunoglobulin heavy chains which were not selected for DNA
binding specificity did not show such a preference for the DXP’ 1 gene and the CDR3
core sequence YYGSGS was not identified once in more than fifty sequences
(Dersimonian et al. 1989). The DXP’ 1 gene product was identified in these heavy
chains but an alternative reading frame was used. However the YYGSGS CDR3
sequence is not an absolute requirement for DNA binding as several anti-DNA VH
CDR3 regions do not contain this sequence (Dersimonian et al. 1990) These results
suggest a role for the heavy chain CDR3 in DNA binding.
Ab47, an IgMX monoclonal binds ssDNA from a healthy adult showed less
than 76% homology with most VHl members (Sanz et al. 1989b). The highest
sequence identity (78%) was found to fetal cDNA sequence 51P1 (Schroeder etal.
72
1987). This antibody was found to be encoded by the single member VH7 gene family
Studies done by Logtenberg et al. (1989) revealed that fetally derived
autoantibodies encoded by the VH6 gene family display binding patterns reminiscent of
autoantibodies present in the sera of SLE patients. Nucleotide analysis revealed that the
CDR3 region is of conserved length. The VH6 encoded autoantibodies were found to
bind ssDNA. The binding was believed not to be simply charge related as these
antibodies did not bind to similar negatively charged molecules such as RNA.
Comparison of the sequences of the VH6 genes expressed by the fetal lines L16 and
MLl to germline gene 61G1 revealed 100% homology.
AlO and A431 were derived from normal adult PBMC’s differed from the
prototype VH6 germline by 6 and 3 nucleotides respectively. Eight of the nine
differences were concentrated in the CDRl and CDR2 regions each resulting in amino
acid replacements. This pattern of nucleotide substitutions suggests somatic mutation.
Further studies of these antibodies showed that AlO and A431 had undergone somatic
mutation in the absence of class switching.
Eight anti-dsDNA antibodies were described by Manheimer-Lory et at.
(1991). These antibodies carry the 31 idiotope which is encoded by the k light chain of
these monoclonals. A random sampling of V k I chains showed an average of 2.8
arginine, asparagine or glutamine residues in CDRl. The V k chains of the eight anti-
dsDNA monoclonals described contain an average of 3.75 of such residues within the
CDRl region. This data points to the importance of the VL chain in DNA binding.
A summary of the IgG monoclonal anti-DNA antibodies sequenced are
shown in table 1.4. T14, an IgG anti ss & dsDNA antibody has 94.8% homology
with VH4 member VH4.21 and has 98.6% homology with germline Vklllb member
Kv325 (Van Es etal 1991). The VH4 and Vklllb genes expressed by the monoclonal
T14 differ from their respective germline counterparts by 14 and 5 nucleotides
respectively. A non-random distribution of silent and replacement mutations were
found. Such a distribution of mutations within an antibody V region is consistent with
a process of Ig receptor-dependent stimulation and selection of mutations (Shlomchik
et a l 1987). Only two replacement mutations were found within the frame work
regions. Four of the ten replacement mutations yielded an arginine or asparagine antino
acid. The CDR3 region of T14 contained five tyrosines, one arginine and one lysine
residue.
Table 1.4
Characteristics of human IgG monoclonal anti-DNA antibodies
IgG DNA GERMLINE D JH L GE RMLINE REFERENCESANTIBODY ORIGIN SUBCLASS ANTIGEN VH COUNTERPART
H2F SLE PEL ? ds 3 VH26 ? 4 K rv Vk4 Mannheimer-Lory et32.B9 SLE PEL 3 ss + ds 3 VH26 DLR2 6 X vni X vni Winkler et al 1992
33.H11 SLE PEL 1 ss + (ds) 3 VH35.21 ? 4 x n ? Winkler et al 199233.F12 SLE PEL 1 ss + ds 3 VH3-8 DNI 3 K nib kv325 Winkler et al 199235.21 SLE PEL 2 ds > ss 3 H ll DMI/2 4 X 9 Winkler et al 199219 E8 SLE PEL 1 ds > ss 3 56P1 DHFL16 6 K ni Vg Winkler et al 1992l-2a SLE SPL 9 ds DNA 3 56P1 ? 4 K I DILpl Mannheimer-Lory etSD6 Healthy ? dsDNA 3 9 ? ? X U X U Paul et al 1992
adult PELT14 SLE PEL 3 ss + ds 4 VH4.21 DXPT 4 K nib kv325 Van Es et al 19912A4 Meyeloma EML ? ds 4 VH4.71-2 DM2 6 K I DILpl Davidson et al 1990
33.C9 SLE PEL 2 ss + ds 4 VH4.4E 9 3 K I hkl02 Davidson et al 1990KS3 SLE SPL ? ds 4 9 ? 9 ? n X U Van Es et al 1991
SLE = systemic lupus erythematosus; PEL = peripheral blood lymphocyte; ss = single stranded; SPL = splenic lymphocytes ds = double stranded; BML = bone marrow lymphocytes; Ls = lymphocytes; ? = unknown.Adapted from Isenberg et al 1993
d
74
NE-1 and NE-13 ,anti-ss/dsDNA antibodies were derived from a patient with
SLE and expressed the NE-1 idiotype (Hirabayashi et al 1993). The VH regions of
these antibodies had 100% homology with the germline gene VH4.21. Four arginine
residues were present in the CDR3 regions of both NE-1 and NE-13 pointing to the
importance of arginine residues within the CDR3 regions of VH4.21 encoded anti-
DNA antibodies Winkler e ta l (1992) produced six IgG anti-dsDNA monoclonals
from three patients with active SLE. Five used members of the VH3 gene family and
one was encoded by a VH4 family member. The high replacement to silent mutation
ratios were found within the CDR’s of both the VH and VL regions of all six
antibodies characteristic of an antigen driven mechanism. 35 replacement mutations
were seen in the CDR’s of these antibodies and 15 yielded arginine, asparagine or
lysine amino acid residues. There was a balanced distribution of these amino acid
residues over the CDR’s of both heavy and light chain suggesting a balanced
contribution of all CDR’s in DNA binding.
In conclusion there is much evidence to suggest a role of certain germline genes
in generating autoantibodies. Murine studies have shown the potential importance of
basic amino acid residues in binding to DNA. The distribution of mutations within
variable regions at least within the anti-DNA antibody population shows evidence of an
antigen driven mechanism. However, relatively few autoantibodies have been
sequenced from patients with active autoimmune disease, especially those of the
pathogenetically associated IgG isotype.
75
AIMS
Although much research has been focussed on the molecular origin of pathogenic
autoantibodies, a number of questions concerning the factors that contribute to the
generation of these autoantibodies are only beginning to be uncovered. What
contribution is made by genetic polymorphism and somatic mutation? What is the
relationship between natural and pathogenic autoantibodies? Do pathogenic
autoantibodies derive from a particular B cell subset and do they utilise the same or a
different array of VH and/or VL genes compared with the lymphocytes that respond to
foreign antigens?
The aim of the work described in this thesis was to investigate the genetic
origins of autoantibodies derived from patients with active autoimmune disease, in
order to see if there was biased variable region gene usage in the generation of
autoantibodies and if the presence of certain amino acid/ amino acid motifs in expressed
V regions correlate with autoreactivity and the expression of idiotypes elevated in the
sera of patients with autoimmune disease
76
CHAPTER TWO
MATRRTAKS AND METHODS
77
CHAPTER TWO
MATERIALS AND METHODS
2.1 MATERIALS
I) BACTERIA
The following strains of Escherichia coli were used
JM109 -recA" (recAl, endAl, gyrA9(), t/i/, hsdRM, supE44, relAl, X' Aijac-
/?rc?AB),[F’, zmD36,proAB, ktclq, lac7AM\5\)
XL 1-blue recA' ( recAX lac', endAl, gyrA96,thi, hsdR ll, supE44, r^/Al,[F’ proAB,
laclq, lacZAMlS, TnlO])
ID MONOCLONAL AUTOANTIBODY CELL LINES
The monclonal IgM autoantibodies “RT’ were derived from the splenocytes of
one SLE patient with active disease utilising the GM4672 fusion partner (Rioux et at
1993). These monoclonals were made and characterised by Dr. CT Ravirajan
(Ravirajan et at 1992).
The monclonal IgM autoantibody WRI176 was derived from the splenocytes of
a second SLE patient with active disease utilising the GM4672 fusion partner. VAR24,
an IgM autoantibody derived from the PEL of a patient with polymyositis utilising the
GM4672 fusion partner. WRI176 and VAR24 were made and characterised by Dr. R
Watts.
The IgM E3-MP0 was derived from the PEL of a patient with vasculitis. The
IgG monoclonals E3 and D5 were derived from the PEL of a third patient with active
SLE. E3-MPO, E3 and D5 were immortalised using the fusion partner CEF7 and were
made and characterised by Dr. M Ehrenstein ( Ehrenstein et al 1993).
The IgG monoclonal autoantibody UK4 was derived from the PEL of a patient
with Anti-phospholipid syndrome utilising the fusion partner CEF7. UK4 was made
and characterised by Dr S Menon.
78
Monoclonal cells lines were intially cultured and cloned in medium containing 10% fetal
calf serum. Bulk cultures for antibody production were subsequently cultured in serum
free medium HU-1 (Northumbria Biologicals UK).
m) VIRUSES
M l3 helper phage VCS-M13 used for single stranded rescue was obtained from
Stratagene, La Jolla, CA. U.S.A.
IV) D mVectors - Bluescript vector was obtained from Stratagene, La Jolla CA
Bluescript TA vector was obtained from Novagen, Madison WI
M13 vector was obtained from Amersham UK.
V) DNA PROBES
a) Human immunoglobulin constant region
IGHG3 for the identification of human IgG constant region was obtained as a
2kb Sac-I - SphI fragment cloned into pUC19 (Lefranc etal. 1986) was a kind gift
from M P Lefranc, University of Montpellier, Montpellier, France.
IGHM for identification of human IgM constant region was obtained as a 1.2kb
fragment cloned into pACYC184 (Babbitts etal. 1981) was a kind gift from T Babbitts,
Laboratory of Molecular Biology, Cambridge, UK.
b) Human immunoglobulin variable region probes
VH gene family probes were generously provided by Dr Fred Alt, Columbia
University, NY , USA (VHl, VH3, VH4, VH5, VH6.) (Berman etal. 1988) and by
DrT Honjo Kyoto Japan (VH2) (Takahashi etal. 1984).
79
VI) POLYMERASE CHAIN REACTION PRIMERS
The details of the primers used are as follows
VH9 (CAG GTG CAG CTG GTG GAG TCT GG)
Cpjh (GAC GGA ATT CTC ACA GGA GAC)
Cgs (CAG GGG GAA GAC CGA TGG)
HK14 (GAC ATC CAG ATG ACC CAG TCT CC)
HK26 (GAT ATT GTG ATG ACT CAG TCT CC)
VK3 (TGA GAA TTC TCC TGC TAC TCT GGC TG)
KC (CGA GAA TTC TAG ATC ATC AG A TGG CGG GA)
m i o (CAG TCT GTG TTG ACG CAG CCG CCC TC)
HA.20 (CAG TCT GCC CTG ACT CAG CCT GCC TC)
HX30 (TCC TAT GAG CTG ACT CAG CCA CCC TC )
JX (ACC TAG GAC GGT CAG CTT GGT CCC)
VH) ENZYMES
Restriction endonucleases were obtained from Gibco/BRL UK, New England
Biolabs, U.S.A. or Boehringer Corporation Ltd UK.
T4 DNA ligase, ribonuclease inhibitor (RNasin) and proteinase K were obtained from
Boehringer Corporation Ltd UK.
Sequenase kits were obtained from United States Biochemicals, and supplied by
Cambridge Bioscience UK.
DNA polymerase Klenow fragment was obtained from Gibco/BRL.
Taq polymerase was obtained from Cetus Emeryville CA
Vim RADIOCHEMICALS
Radiochemicals were purchased from New England Nuclear Inc., Boston
USA. These were (a-32p) dCTP (800 and 3000 Ci/mmol) and (a-3%) dATP
(lOOOCi/mmol)
80
IX) BUFFERS. SOLUTIONS AND GROWTH MEDIA
The following general solutions were used. Solutions specific to a particular
method are described in the appropriate section.
L-broth (LB) - 1% (w/v) bactotryptone, 0.5% (w/v) bacto yeast extract, 0.5% NaCl in
H2 O and adjusted to pH 7.2 with NaOH
Phosphate buffered saline (PBS) 135mM NaCl, 27mMKCl, lOmM Na2HP04 in H2 O
(SSC) 150mM NaCl, 15mM trisodium citrate.
Diethylpyrocarbonate (DEPC) treated water - 0.1% DEPC in H2 O was shaken for
twelve hours then autoclaved.
X) OTHER MATERIALS AND REAGENTS
Nylon and nitrocellulose filters - (Hybond N) Amersham International UK
X-ray film- X-omat AR (Kodak, UK) or Fuji RX film (Fuji Photofilm Co. UK)
Polaroid 667 film - Polaroid, UK
Bacterial growth media Difco Laboratories, UK
RNA markers Gibco BRL, Gaithersburg, MD
RNAzol B - Cinna Biotecx Labs Inc.
Geneclean II - Bio 101 Inc La Jolla, CA.
Lambda Hind HI molecular weight markers- Gibco BRL, Gaithersburg, MD
One kilobase molecular weight markers- Gibco BRL, Gaithersburg, MD
Random hexamers Gibco/BRL UK
AH deoxynucleotides - Pharmacia LKB Sweden
All other chemicals, solvents, and materials were obtained from one of the following:
Sigma chemical company, British Drug House (BDH) or Fisons Laboratories.
81
2.2 RNA METHODS
I) RNA ISOLATION
a) Large scale preparation
Total RNA was isolated from large numbers of mammalian cells (up to 10^) by
using a modified method of the guanidinium/CsCl method described by Chirgwin etal.
(1979). The washed cell pellet was lysed and dispersed in 5ml of lysis buffer (4M
guandinium isothiocyanate, O.IM p-mercaptoethanol, pH5). This extract was then
layered above 2.2ml 5.7M CsCl, O.IM EDTA, pH5 in Beckman SW41 polyallomer
tubes. The tubes were then centrifuged at 25K rpm for 22 hours at 17°C after which
the supernatant was removed by aspiration. The RNA pellet was briefly washed in
70% ethanol then resuspended in 400|il of ‘RNA dissolving buffer’ (lOOmM Tris-HCl
pH7.4, lOOmM EDTA, 1% SDS). After extraction with a 4:1 mixture of chloroform
and butan-l-ol, the organic phase was re-extracted with an equal volume of ‘RNA
dissolving buffer’. After combining the two aqueous phases, the RNA was ethanol
precipitated with sodium acetate, pH5.2 at -20°C overnight. The RNA was recovered
by centrifugation, washed and resuspended in H2 O and stored at -70°C. An aliquot
was removed spectrophotometric determination.
b) Small scale preparation (1)
A modification of the guanidinium/CsCI method as described by Wilkinson
(1988) was used to isolate total cellular RNA from a maximum of approximately 10
mammalian cells. The cells were washed and harvested 1ml of sterile PBS. After low
speed centrifugation in a microfuge, the PBS was removed and the cell pellet dissolved
in 0.5ml of guanidinium lysis buffer (4M guanidinium isothiocyanate, IM p-
mercaptoethanol, 25mM sodium acetate pH5.2). The lysed cells were drawn into a
syringe fitted with an 18-gauge hypodermic needle and forcefully ejected back into the
tube. This was repeated several times until the viscosity of the suspension was reduced
by shearing the liberated chromosomal DNA. The homogenate was then loaded onto a
cushion of 220pl of 5.7M CsCl in a Beckman mini ultracentrifuge tube. The tubes
were then filled to the top with guanidinium lysis buffer and then spun in a Beckman
82
TLS 55 rotor at 55K rpm, 17°C for 3 hours. After removal of the supernatant, the
tubes were inverted onto paper towelling to drain and wiped carefully to remove any
residual supernatant. The remaining pellet was resuspended in 100pi of DEPC-treated
H2 O and incubated on ice for 30 minutes. The solution of dissolved RNA was then
ethanol precipitated at -70°C for 30 minutes, recovered by centrifugation, washed, dried
and typically resuspended in 50pl DEPC-treated H2 O. An aliquot was taken for
spectrophotometric analysis.
b) Small scale preparation (2)
Total RNA extraction for very small quantities of cells was performed using the
RNAzol B RNA extraction kit according to the manufacturer’s instructions.
ID SPECTROPHOTOMETRIC ANALYSTS OF RNA
1 optical density unit at 260 nm was equivalent to 40mg/ml total RNA
im RNA ELECTROPHORESIS
Denaturing agarose slab were prepared in 1 x MEA buffer and formaldehyde
was added just before casting to a concentration of 2.2M. The RNA samples (up to
20pg of total RNA) were denatured prior to loading by mixing with sample buffer [1 x
MEA (20mM MOPs, ImM EDTA, 5mM sodium acetate in H2 O and adjusted to pH 7.2
with NaOH), 50% formamide, 2.5M formaldehyde] and heated to 65°C for 15 minutes.
The sample was mixed with 6 x loading buffer (50% glycerol, ImM EDTA, 0.4%
bromophenol blue 1 mg/ml ethidium bromide). RNA molecular weight markers were
prepared for loading according to the manufacturers instructions and were loaded in a
peripheral well. The gel was run in 1 x MEA buffer at lOOV usually until the
bromophenol blue had migrated 100mm from the wells. The RNA was visualised with
a ultraviolet transilluminator (wavelength 254nm) and photographed. The position of
the RNA ribosomal bands were measured from the wells. The position of bands of the
RNA molecular weight markers was also determined in this way.
83
TV) NORTHERN RT.OTTING
RNA transfer
The gels were blotted as described by Maniatis etal. (1982) but omitting any
treatment of the gel before transfer. 20 x SSC was used as the transfer buffer. The
fixing of RNA to Hybond N membranes (Amersham, Amersham U.K.) was carried
out by UV illumination (254nm) for 3-5 minutes with the RNA nearest to the light
source.
Staining of immobilised RNA molecular weight markers
The RNA molecular weight markers which were run on a peripheral well on the
Northern gel were blotted onto the Hybond N nylon membrane. The area
corresponding to the position of the markers was cut off from the rest of the membrane
prior to hybridisation and soaked in IM acetic acid for 10 minutes at room temperature.
The membrane was transferred to a staining solution [0.4M acetic acid, 0.4M sodium
acetate, 0.2% (w/v) methylene blue and soaked for a minimum of 10 minutes. The
membrane was destained in H2 O
Hybridisation of Northern blots
The best conditions were determined for all the radiolabelled cDNA probes used
in this work and were as follows:
The filter was prehybridised in prehybridisation buffer [5 x Denhardts, 50%
formamide, 5 x SSC, 0.1% SDS (w/v) 150mg/ml herring sperm DNA] at 42°C for a
minimum of 2 hours. The filter was then hybridised for a minimum of 12 hours at
42°C in buffer of the same composition as the prehybridisation buffer but with the
addition of the probe at 10 cpm/ml. The initial washes following hybridisation were 2
X SSC, 0.1% SDS at room temperature. These were followed by a series of washes at
65°C where the stringency was increased by lowering the ionic strength of the wash
and varying the times of the incubation. The wash conditions were determined by the
level of activity left on the filter as monitored by a Geiger counter. A typical high
stringency wash consisted of a 30 minute incubation at 65°C in 0.1 x SSC, 0.1% SDS.
After washing the filter was wrapped in Saran Wrap and autoradiographed with
84
intensifying screens at -70°C for 2-3 days.
Removal of hybridised probe from hvbond N membranes
Northern blots on Hybond N were stripped of hybridised probe using one of
the following methods:
Method one
The blot was washed in stripping buffer (0.005M Tris-HCl pHS, 0.002M
EDTA, 0.1 X Denhardts) at 65°C overnight.
Method two
A solution of 0.1% SDS was boiled and then poured over the filter and allowed
to cool to room temperature.
Method three
The filter was washed in stripping buffer two [50% formamide (v/v),
lOmMtris-HCL pH8] at 65°C for one hour. The filter was rinsed in 2 x SSC at room
temperature for 3-4 minutes
After stripping by any of the above methods, the filter was sealed in a bag
containing 2ml of 2 x SSC and stored at -20°C.
85
V) RNA SEOTJENCTNG
m RNA extraction using Nonidet P40
This method is based on that described by Griffiths and Milstein (1985)
The solutions and reagents used in this method are as follows:-
Lvsis buffer [0.14M NaCl, 1.5mM MgCl2 lOmM Tris-HCl pH 8.3, 0.5% Nonidet
p40 (v/v)]
PK buffer [ 0.3M NaCl, 0.2M Tris-HCl pH 7.4, 25mM EDTA, 2% SDS (w/v), 1.4
mg/ml proteinase K (added to solution just before use)]
Binding buffer [0.5M NaCl, lOmM Tris-HCl pH 8.3, 5mM EDTA, 0.5% (w/v) SDS]
Washing buffer [O.IM NaCl, lOmM Tris-HCl pH 8.3, 5mM EDTA, 0.5% (w/v) SDS]
Elution buffer [lOmM Tris-HCL pH 8.3, 5mM EDTA, 0.3% (w/v) SDS ]
Preparation of oligo dT cellulose
0.6g of oligo dT cellulose was prepared by washing sequentially with the
foUowingi-
Twice with 50ml O.IM NaOH, three times with 50ml H2 O, twice with 20ml of
elution buffer at 50°C and twice with 20ml of binding buffer at room temperature. The
mixture was vortexed during each wash and centrifuged to 1000 rpm to pellet the oligo
dT cellulose and the supernatant was decanted before the addition of the next wash
solution.
0.2g of oligo-dT cellulose was used for each isolation from 10 cells. After use
the oligo-dT cellulose was washed 10 ml of elution buffer at 50°C, resuspended in
binding buffer for storage at 4°C. The cellulose was pelleted and the supernatant
removed before subsequent use.
86
mRNA extraction using Nonidet P4Q
The mammalian cells were pelleted at 1500 rpm and washed once in ice-cold
sterile PBS. The cells were pelleted again and the supernatant was decanted off. The
pellet was resuspended in 22ml of lysis buffer and incubated on ice for 2 minutes. The
homogenate was centrifuged at 2500rpm for 20 minutes at 4°C and the supernatant was
decanted into a fresh tube. 22ml of PK buffer was added to the supernatant and was
incubated at 37°C for 1 hour. The homogenate was incubated at 65°C for 3 minutes to
heat inactivate the proteinase K. 3.5ml of 4M NaCl was added and the mixture was
allowed to cool to room temperature by placing the tube on ice. 1ml of oligo-dT
cellulose (loose pellet 0.2g approx.) was added to the deactivated RNA preparation.
The mixture was incubated at room temperature for 30 minutes on a Voss wheel. The
mixture was spun up to lOOOrpm and then back down quickly. The supernatant was
decanted off and the loose pellet was poured into a plastic column. The tube which
previously contained the pellet was washed with 1.5ml of binding buffer and this was
added to the column. In all, the column was washed with the following:
3 x 1 .5ml binding buffer at room temperature
3 X 1.5ml washing buffer at room temperature
3 X 1.5ml elution buffer at 50°C
(The column was allowed to run dry after each addition)
The 4.5ml of eluted material (from the elution buffer only) was collected into a
silconised glass corex tube. This material was the poly A + RNA.
500ml of 3M sodium acetate was added, mixed and then 12.5ml of ice cold ethanol was
added. The RNA was left at -20°C to precipitate. The oligo dT cellulose was washed
with 10ml of elution buffer at 50°C, resuspended in binding buffer and stored at 4°C.
The corex tube was spun at 10,000rpm for 30 minutes at 4°C (Sorvall RC5B
centrifuge). The supernatant was decanted off and the pellet was washed twice with
80% ethanol. The pellet was dried in a vacuum dessicator and then resuspended in a
suitable volume of water (the volume depended on the size of the pellet, varied from
between lO-lOOp.1). An aliquot was taken for spectrophotmetric analysis and the rest
was stored at -70°C.
87
m R N A sequencing
The following solutions were used in this method:-
Poly A+ RNA at a concentration of 2gg/ml
10 X RT buffer (0.5M Tris-HCl pH 8 ,1.5M KOI, O.IM MgCl).
0.2M DTT
d-NTP-C was made up by the mixture of the following stock solutions:-
lOmM dTTP 40pl
lOmM dGTP 40pl
lOmMdATP 40pl
50mM Tris-Cl, ImM EDTA pH8.3 160|xl
Sterile H^O 1320pl
a-32p dCTP 400Ci/mMol
Dideoxvnucleotides
ddGTP, ddATP, ddCTP, ddTTP. were prepared, each at the concentration of
5|xM
Chase
This solution was a 1:1:1:1 ratio of dGTP,dCTP, dATP, and dTTP all at ImM
concentration
Formamide stop solution
1ml deionised formamide
Img xylene cyanol
Img bromophenol blue
40ml 500mM EDTA
88
The following was mixed in an eppendorf tube by spinning:
Poly A+ RNA 2pg
10 X RT 2pl
0.2M DTT 2pl
dNTP-C 8pl
H2 O 2pl
Ipl of Ipg/pl CjiJ was added to this mixture and 4pl of the above mixture was
aliquoted into four tubes labelled A, C, G, T.
l|il of 5mM of each dideoxynucleotide was added to each tube e.g ddATP added to tube
A, ddCTP added to tube C etc. One unit of reverse transcriptase was added to each
tube. The tubes were incubated at 42°C for 15 minutes. At 15 minutes, 2pl of chase
solution was added to each tube and the tubes were incubated at 42°C for a further 15
minutes. 2pl of formamide stop solution was added to each tube and the reactions were
either frozen at -20°C until required or boiled in preparation for loading onto a standard
polyacrylamide sequencing gel and treated as for DNA dideoxy sequencing.
2.3 DNA METHODS
D ISOLATION OF PLASMID DNA
a) SMALL SCALE PLASMID PREPARATION
5ml of LB-broth (50pg/ml ampicilhn) was inoculated with a single colony from
an agar plate or 5pi of bacterial glycerol stock and incubated overnight at 37°C with
agitation. The culture was then spun down at 4000rpm for 10 minutes. The pellet was
resuspended in lOOpl of solution one (1% glucose, 25mM Tris-HCl pH 8.0, lOmM
EDTA Na2 pH 8.0) and incubated at room temperature for 5 minutes. 200pl of solution
two (0.2M NaOH, 1% SDS) was added, the tube was inverted and incubated on ice for
5 minutes. 150pl of ice cold solution three was added and after vortexing the mixture
was incubated on ice for 15 minutes.
(Solution three; solution of potassium acetate pH4.8 approx. made up as follows: To
60ml of 5M potassium acetate 11.5ml of glacial acetic acid was added followed by
89
28.5ml of H2 O. The resulting solution was then 3M with respect to potassium and 5M
with respect to acetate) The precipitate was then pelleted by centrifugation in a
microfuge at 13,000rpm for 10 minutes. The supernatant was then decanted into a
fresh tube. This was followed by a standard phenol/chloroform extraction and ethanol
sodium acetate precipitation. The DNA was typically resuspended in 20pl of H2 O
containing DNase free pancreatic RNase (20pg/ml) and vortexed briefly. The
preparation was either subjected to restriction digest analysis or stored at -20°C.
b)LARGE SCALE PLASMID PREPARATION
The method was based on the alkaline lysis method of Maniatis etal. (1982)
with modifications. Typically, 400ml of ampicillin-containing (50pg/ml) LB medium
was inoculated with bacteria containing the appropriate plasmid and grown to saturation
overnight in an orbital shaker at 37°C. The bacteria were harvested by centrifugation at
4.2K rpm for 20 minutes (4°C) in a Beckman JR3.2 rotor. The cells were resuspended
in 40ml lOmM EDTA, pH 8.0 before the addition of 80ml of freshly prepared 0.2M
NaOH, 1% (w/v) SDS and 40ml 3M potassium acetate pH 4.8, swirling gently after
each addition. After a 5 minute spin at 4.200rpm (at room temperature), the
supernatant was filtered through nylon gauze and spun together with 0.6 volume
isopropanol at 6,000rpm for 10 minutes in a Sorvall GS3 rotor. The DNA pellet was
rinsed in 70% ethanol containing lOOmM Tris-HCl pH8 and subsequently dissolved in
lOOmM Tris-HCl pH8, 2mM EDTA and purified by equilibrium gradient centrifugation
in a caesuim chloride-ethidium bromide gradient as described by Maniatis etal. (1982) .
The plasmid DNA solution recovered was adjusted to 10ml by the addition of of
lOOmM Tris-HCl pH8, 2mM EDTA and the nucleic acid was precipitated by the
addition of two volumes of absolute alcohol and spun at 3,000 rpm for 20 minutes in a
Beckman JR3.2 rotor. The DNA pellet was washed and dried and resuspended in
400pl H2O, 50|il 20x SSC and subjected to RNase A (at 40u/ml) and RNase T1 (at
200u/ml) for 60 minutes at 37°C. Following standard phenol extraction, the plasmid
DNA was precipitated with ethanol, resuspended in 200pl H2 O and a sample diluted for
spectrophotometric determination.
90
ID GEL ELECTROPHORESIS OF DNA
a) NON DENATURING AGAROSE GELS
The separation of DNA fragments and the assessment of various DNA cloning
steps was effected by use of horizontal agarose mini-gels on apparatus manufactured by
Pharmacia. The gels were made and run as described by Maniatis etal. (1982) in Tris -
borate (TBE) electrophoresis buffer (lOOmM Tris-HCL, lOOmM boric acid, 2mM
EDTA pH 8.35) ; the agarose concentration varied between 0.6% and 1.2% (w/v)
depending upon the size of the DNA of interest. Where DNA fragments were
subsequently used for ligation reactions or radiolabelling then low melting point
agarose was used in combination with Tris- acetate (TAE) electrophoresis buffer
(40mM Tris-HCl, ImM EDTA adjusted to pH 8.0 with glacial acetic acid). The DNA
samples were mixed with 6 x loading buffer (0.25% bromophenol blue, 0.25% xylene
cyanol, 15% [Ficoll type 4(X)] in H2 O ) immediately prior to loading. The gels were
typically run at 5-20V/cm for 60 minutes. The DNA was visualised with an ultraviolet
transiUuminator (wavelength 254nm) and photographed.
bl DENATURING POLYACRYLAMIDE GELS
The products of DNA sequencing reactions were analysed by electrophoresis
through 6% polyacrylamide/ urea gels. These were prepared and the samples run
exactly as described in the Amersham M13 cloning and sequencing booklet.
Furthermore, the gels were fixed and dried as described in this booklet.
im SOUTHERN BLOTTING
DNA transfer
The transfer of DNA from agarose gels to nitrocellulose or nylon membranes
was carried out as described by Maniatis etal. (1982). The fixing of DNA to Hybond
N membranes was carried out by UV illumination (254nm) for 3-5 minutes with the
DNA nearest to the light source.
Hybridisation of Southern blots.
The membranes were prehybridised in Hybaid bottles (manufacturer) for a
minimum of two hours in 6 x SSC and 1 x Denhardt’s reagent [ 0.2% (w/v) BSA,
91
0.2% (w/v) polyvinylpyrroloidene, 0.2% (w/v) Ficoll] at 65°C in a Hybaid
hybridisation oven. lOOpg/ml of heat denatured, fragmented salmon sperm DNA was
added 30 minutes before the addition of radiolabelled probe. Sufficient heat denatured
probe was added so that the final count was approximately 10 cpm/ml. The filters
were hybridised for a minimum of 12 hours before washing initially at a low stringency
of 2 X SSC, 0.1% SDS for 30 minutes at 65°C. The stringency and number of
subsequent washes were determined by the level of activity left on the filter as
monitored by a Geiger counter. The stringency of the washes was increased by
reducing the concentration of SSC; a typical high stringency wash was 0.1 x SSC,
0.1% SDS at 65°C for 30 minutes. The filters were then wrapped in Saran Wrap whilst
damp and autoradiographed. Nitrocellulose and nylon membranes were treated
identically.
IV) RADIOLABELLING OF DOUBLE STRANDED DNA FRAGMENTS
Linear, double stranded DNA fragments were radiolabelled according to the
oligo-labelling technique of Feinberg and Vogelstein (1984) using a-^^p dCTP. The
labelled probe was separated from unincorporated nucleotides by centrifugation through
a Sephadex G50 spun-column as described by Maniatis etal. (1982) except that the
Sephadex was equilibrated with 2 x SSC. The specific activity of the radiolabelled
DNA was determined using TCA (trichloroacetic acid) precipitation as described by
Maniatis gro/. (1982).
V) ISOLATION OF DNA FRAGMENTS
DNA fragments for use in ligation or radiolabelling reactions were excised from
low melting point TAE agarose gels whilst exposed to UV light but care was taken to
minimise the UV exposure. The DNA was purified from the agarose using a
Geneclean kit (Bio 101 Inc La Jolla, CA) which is based on a glassmilk extraction
procedure. The DNA purification was performed according to the manufacturer’s
instructions with the following modifications:
The gel slice was transferred to an eppendorf tube and 2.5-3 volumes of 6M
Nal were added and the mixture incubated at 55°C for ten minutes to dissolve the
92
agarose. lOp.1 of glassmilk suspension was added and the mixture incubated on a Voss
wheel for one hour at room temperature. The glass milk was pelleted in a microfuge at
13,000 rpm and the supernatant decanted off. The glassmilk pellet was washed three
times in 800pi New Wash solution. The pellet was dried briefly in a vacuum drier and
then resuspended in lOpl of sterile H2 O and incubated at 55°C for 30 minutes. The
tube was then spun in at microfuge for 5 minutes at 13,000rpm and the supernatant
pipetted off and placed in a new tube. A further 5pl of H2 O was added to the pellet and
the mixture was incubated at 55°C for 15 minutes. The supernatant was removed from
the pellet as previously described, pooled with the 10ml of supernatant isolated
previously, and stored at -20°C until required.
VI) cDNA SYNTHESIS
cDNA was made as follows:-
l(ig of total RNA at a concentration of 1 mg/ml was incubated at 65°C for two minutes
The following were added to the denatured RNA
10 X PCR buffer 2pl
RNAsin 0.5pl (20 units)
lOmM dNTPs 2pl
Random hexamers Ipl (9 OD units/ml )
Murine moloney reverse transcriptase (2(X)U/pl) 0.5pl
Sterile water 13pl
The above mixture was incubated at 37°C for one hour and stored at -20°C until
required.
93
VnyPOLYMERASE CHAIN REACTION.
POLYMERASE CHAIN REACTION
After producing cDNA using the protocol described above, amplification of the
immunoglobulin variable regions was performed. The following were added to the
20pl cDNA reaction mixture produced above
lOx PCR buffer 8pl
lOmM dNTPs 5pl
primer 1 Ipl (concentration of each primer is l|ig/ml )
primer 2 Ipl
Taq polymerase Ipl (5 units/^il)
sterile water 63pl
The conditions for the variable region amplification is dependent on the PCR primers
used and were as follows:
For VH9/CpJH (for amplification of all IgMVH regions) and VKJH/CKJH (for
amplification of all V k regions) combinations the following conditions were used:-
1 minute at 94°C
1 minute at 42°C
1 minute at 72°C
for 30 cycles with a final step of 2 minutes at 72°C
For combinations VH4/Cy (for amplification of VH4 IgG variable regions), VH4/Cp
(for amplification of VH4 IgM variable regions), KC/HK14 (for amplification of VkI
variable regions), KC/HK26 (for amplification of Vk2 variable regions), KC/Vx3 (for
amplification of Vk3 variable regions) the following conditions were usedi-
1 minute at 94°C
1 minute at 56°C
1 minute at 72°C
for 30 cycles with a final step of 2 minutes at 72°C
94
For combinations JX/HXIO (for the amplification of VA.1 variable regions), JX/HX20 (for
the amplification of VX2 variable regions) Jk/HX30 (for the amplification of VA.3
variable regions) the following conditions were used:
1 minute at 94°C
1 minute at 65°C
1 minute at 72°C
for 30 cycles with a final step of 2 minutes at 72°C
b^PURIFICATION OF PCR PRODUCTS
The PCR reactions were electrophoresed through a 1% agarose gel in TAE buffer
against a 1KB ladder (molecular weight markers )
The expected sizes of the products obtained from the various primers were between
350-400kb
The relevant bands on the gel were excised with a scalpel and the DNA purified using
the Gene Clean Kit (Bio 101 Inc La Jolla, CA) with the modifications described
previously.
VBB CLONING OF PCR PRODUCTS
PCR products were cloned into plasmid vectors by a variety of methods as
described below
Ligation of PCR products into plasmid vectors
The method of ligation was dependent upon the manner in which the DNA
fragments had been recovered:
a) Gel purified DNA fragments were mixed with vector DNA in a 5:1 molar ratio so
that the total mass did not exceed 200ng and the final reaction volume was lOpl. After
addition of ligase buffer (5mM Tris-Hcl pH 7.8, ImM MgCl2 and 2mM DTT) and ATP
to ImM, 10 units of T4 ligase were added and the reaction incubated for a minimum of
5 hours at 15°C
b) ‘In gel’ ligations were performed with DNA fragments which had been excised from
low melting point agarose gels. The gel slab containing the DNA was melted at 65°C
for approximately 10 minutes. A volume of the molten gel was then added to the vector
95
DNA so that a 5:1 molar ratio was created between the fragment and the vector but the
total mass of DNA did not exceed lOOng. The volume of the reaction was then
adjusted so that the agarose concentration was not greater than 2% (w/v) (Crouse etal.
1983). After addition of the ligase buffer and the ATP to ImM, 100 units of T4 DNA
ligase was added and the reaction incubated for a minimum of 5 hours at room
temperature. The ligation mixture was heated to 65°C for 5 minutes immediately before
its addition to competent cells.
DC TA CLONING
TA vector was either obtained from Novagen or made using the following
protocol;
Bluescript DNA (4pg EcoRV cut ) lOpl
10 X PCR buffer 2pl
20mM dTTP 2pl
Taq polymerase 4 units
sterile water 5pl
The above were incubated at 70°C for two hours.
The tailed DNA was purified by the Gene clean kit using the standard protocol or by
phenol/chloroform, chloroform extraction and ethanol precipitation.
l(X)ng was used per ligation as described in the first ligation method above.
X) PREPARATION OF E. COLI COMPETENT CELLS AND TRANSFORMATION
WITH PLASMID DNA
Stocks of JM103 competent cells were prepared using the calcium chloride
method described by Maniatis etal. (1982). Aliqouts of the competent cells were
stored at -70°C until required. The transformation procedure used was that described
by Maniatis erfir/. (1982).
Xn SELECTION OF RECOMBINANT PLASMIDS
Since all plasmids used in this work conferred ampicillin resistance to their host
cells the transformed bacteria were plated onto LB agar plates containing ampicillin
96
(50|ig/ml). AU the plasmid vectors used in this work are constructed so that a fragment
of foreign DNA inserted into the multiple cloning site will disrupt the protein coding
region of the 5’ end of the lac Z gene. This results in the failure of the plasmid to
display a-complementation activity of the lac Z gene product, p-galactosidase, with the
host bacterium. Lac+ bacteria form blue colonies in the presence of X-gal but bacterial
colonies containing recombinant plasmids have a white appearance. Thus, the plates
contained X-gal (800ng) and an inducer of the lac Z gene, IPTG (ImM), to facilitate the
selection of recombinant plasmids by ‘blue/ white selection’.
The picked white colonies were then subjected to in situ hybridisation by
protocol n described by Maniatis etal. (1982), which is an adaptation of the method of
Grunstein & Hogness (1975). This further selection procedure ensured the putative
positives represented recombinants rather than recyclized vector DNA which had lost
their p-galactosidase complementation activity by deletion. The prehybridisation and
hybridisation conditions and the subsequent washing of nitrocellulose filters are
described in the Southern blotting section. After identification of the positives, glycerol
stocks were made from the corresponding colonies on the master plate and confirmation
of the size of the DNA insert was obtained by restriction enzyme digest of plasmid
DNA made by the ‘mini prep’ method.
XIT DNA SEQUENCING
DNA sequencing was performed using the Sequenase kit supplied by United
States Biochemical Corporation, Cleveland, OH. This procedure is a modification of
DNA sequencing originally described by Sanger etal. (1977) which utilises the enzyme
Sequenase a modified form of T7 DNA polymerase. The methods which follow use
this kit according to the manufacturer’s protocol with adaptations described in the
appropriate sections.
XIID DIRECT SEQUENCING OF PCR PRODUCTS
1 pi of the relevant primer (approximately lOOng) was added to 7 pi of gene
cleaned PCR product (approximately lOOng). This mixture was heated to 100° C for
five minutes then placed on dry ice for 1 minute. The tube was centrifuged in a
microfuge for five seconds and 7.5 pi pre-made labelling mix was added to the tube.
97
The pre-made labelling mix was made up as follows (the reagents described
below were present in the USB sequencing kit):-
2pl diluted labelling mix
Ipl DIT
0.5pl 35S dATP
2pl 5 X buffer
2pl diluted sequenase enzyme
The mixture was spun for five seconds and incubated at room temperature for two
minutes. After one minute 3.5pl of pre-made labelling mix and primer was aliquoted
into the top edge of each of four tubes each containing 2.5pi of dideoxynucleotide
ddGTP, ddATP, ddTTP or ddCTP. The four tubes were spun briefly in a microfuge
and then incubated for five minutes at 37-40° C. 4pi of stop solution was added, the
tubes were spun briefly and stored on ice until ready to load, or frozen at -20° C.
Before loading, the reactions were incubated for 2-3 minutes at 80°C to denature.
XIV) DOUBLE STRANDED SEQUENCING
One positive bacterial colony was picked into 5ml of L-broth containing
lOOpg/ml ampicillin and was grown up with shaking at 37°C overnight. The plasmid
DNA was extracted from the culture using the small scale DNA extraction method.
13% PEG 8(X)0 was added to 16pl of plasmid DNA and incubated on ice for 20
minutes. The mixture was spun at 13,(X)0rpm in a microfuge for 10 minutes. The
pellet was washed in 50pi of 100% ethanol, vacuum dried and resuspended in 18pl of
sterile water. 2pl 2M NaOH was added and left at room temperature for five minutes.
8pl of 5M ammonium acetate and 112pl 100% ethanol were added and placed in an
IMS/CO2 for 10 minutes. The denatured plasmid DNA was pelleted in a
microfuge at 13,(KX)rpm for 5 minutes. The pellet was resuspended in 7pl of sterile
water and sequencing reactions were performed as the protocol described by USB
dideoxy sequencing kit
98
XV) SINGLE STRAND DNA RESCUE
One positive colony was inoculated into L broth containing lOOpg/ml ampicillin
and grown up overnight. The overnight culture was diluted 1/100 in L broth containing
lOOpg/ml ampicillin and to this VCS-M13 helper phage was added to give a final
concentration of 1/1000 (4 x 10 plaque forming units/ml approx.). This culture was
grown up for one hour at 37°C on an orbital shaker at a speed of at least 300 rpm.
After one hour, kanomycin was added to a final concentration of 50pg/ml. The culture
was then grown up at 37° C for 5-8 hours on an orbital shaker at a speed of at least
300rpm.
Phage DNA was then prepared and sequenced using the method described in the USB
sequencing kit protocol.
XVD GENE SEQUENCE ANALYSIS
The nearest germline counterparts of the variable region genes expressed in the
monoclonal autoantibodies were determined using the V BASE sequence directory,
Tomlinson et al MRC Centre for Protein Engineering, Cambridge, UK
99
CHAPTER THREE
RESULTS
100
CHAPTER THREE
RESULTS
3 .1 NORTHERN BLOTTING ANALYSIS OF MONOCLONAL AUTOANTIBODIES
The source, antigen specificity, idiotype expression and variable
heavy chain usage of monoclonal autoantibodies described in this thesis
are summarised in Table 3.1. The “RT” and “RSP” monoclonals were
made and characterised by Dr Chelliah Ravirajan (Ravirajan etal 1992).
These monoclonal autoantibodies were derived from the splenocytes of
two patients with active SLE utilising the GM4672 fusion partner (Rioux
et al 1993) D5, B3 and E3-MPO were made and characterised by Dr
Michael Ehrenstein (Ehrenstein cf a/ 1993, Ehrenstein era/ 1994). E3-
MPO was derived from the PBL of a patient with vasculitis and B3 and D5
were derived from a patient with active SLE. These monoclonal
autoantibodies were made utilising the CBF7 fusion partner. VAR 24 and
WRI176 were made and characterised by Dr Richard Watts (by kind
permission). VAR 24 was derived from the PBL of a patient with
polymyositis and WRI176 was derived from a patient with active SLE .
Both of these monoclonals were immortalised using the GM4672 fusion
partner. UK4 was made and characterised by Dr. S Menon (by kind
permission). UK4 was from a patient with Anti-phospholipid syndrome
utilising the CBF7 fusion partner.
A Northern blot analysis was performed on the monoclonal
autoantibody cell lines available using probes specific for the six human
VH gene families. The VH gene usage of the monoclonals is summarised
in Table 3.1. Northern blots hybridised with the VH gene family probes
are shown in fig 3.01-3.04. Eleven of the seventeen monoclonals used the
VH4 gene family, four used the VH3 family, one used VH2 and one used
VH5. Thus, 65% used the VH4 family, 23% used VH3. VH3 is the
Table 3.1
Antigen specificity and id expression of monoclonal autoantibodies
101
M onoclonals Sources A g binding specificity V Gene usage Id 's expressed
RT16 SLE ss+dsDNA/Sm-DpeplO VH4
RT55 SLE ssDNA/CDL/Sm-RNP/Hl VH4
RT79 SLE ssDNA/CDL VH4 9G4
RT129 SLE ss+dsDNA/CDL VH4
RT72 SLE ss+dsDNA/CDL/Sm-DpeplO VH4 RT72id
RT6 SLE Hl/Sm-RNP7T4 VH3 RT6id/16/6
RT84 SLE ssDNA/CDL VH4 RT84id/9G4
RT119 SLE Hl/H2B/Sm-Dpep/SS-B VH4
RT83 SLE Sm-D/SS-B VH5
RT121 SLE Sm-D VH4
RSP18 SLE CDL/Sm-Dpepl6 VH2
WRI176 SLE ss+dsDNA VH3 WRI176id/16/6
VAR24 Polymyositis 56KdRNP VH4
E3-M P0 Vasculitis MPO VH4 9G4
D 5(IgG ) SLE ssDNA VH4 D5id/9G4
B3 (IgG) SLE dsDNA VH3 B3id/16/6
UK4(IgG ) APS CDL VH3 16/6
102
largest human VH gene family and if random VH gene usage occurred one
one expect 48% of antibodies to be VH3 encoded. All the monoclonal
antibodies with anti-DNA specificity derived from the patient “RT” were
VH4 encoded. The VH4 family is now regarded as the second largest VH
gene family and if random usage of this VH gene family occurred one
would expect 22% of antibodies to be VH4 encoded. It appears that there
is an overexpression of the VH4 gene family in the panel of autoantibodies
tested. The fusion partner GM4672 was utilised in the immortalisation of
the “RT”, “RSP”, WRI176 and VAR 24 monoclonals and is a low level
secretor of an IgGK antibody. The VH4 encoded heavy chain of GM4672
was not detected by the VH4 probe (fig 3.03) therefore the VH4 gene bias
observed cannot be attributed to the presence of the GM4672 heavy chain.
The VH2 and VH5 families were utilised by 6% of the monoclonals each.
The expected usage of these families on the basis of their complexity is
18% and 6% respectively, therefore VH2 has also been under-represented
in the panel of monoclonal autoantibodies tested.
103
Figure 3.01
NORTHERN BLOTS HYBRIDISED WITH THE VHl
AND Cfx PROBES.
1 2 3 4 5 1 2 3 4
2 4 kb
VHl C i
Figure 3.01 A
1. 15-1 plasmid (VHl probe -positive control).
2. RSP18.
3. RT119.
4. KT83.
5. KT6
Figure 3.01 B
1.Fusion partner GM4672.
2. Kril9.
3. RT83
4.RT6
104
Figure 3.02
NORTHERN BLOTS HYBRIDISED WITH THE VH3
AND VH5 PROBES.
1 2 3 4
B
1 2
24 kb
VH3 VH5
Figure 3.02 A
1. RSP 18
2. KT119
3. RT83
4. KT6
Figure 3.02 B
1. KT6
2. RT83
105
Figure 3.03
NORTHERN BLOTS HYBRIDISED WITH THE VH4
PROBE
1 2 3 4
B
1 2 3
. .
2 4 kb
VH4 VH4
Figure 3.03 A
1. RSP 18
2. Rril9
3. KT83
4. KT6
Figure 3.03 B
1.Fusion partner GM4672
2 . Kri6
3. RT121
106
Figure 3.04
NORTHERN BLOTS HYBRIDISED WITH THE VH2
AND VH6 PROBES
24 kb
1 2 3 4 1 2 3 4
VH2 VH6
Figure 3.04 A
1. RSP18
2. KT119
3. RT83
4. KT6
Figure 3.04 B
1. RSP 18
2. RT119
3. RT83
4. RT6
107
3.2 ANTI-DNA AUTOANTIBODIES ENCODED BY THE
VH4.21 GERMLINE GENE.
An idiotope designated 9G4 is known to be a marker for
immunoglobulins which utilise a particular VH gene, the VH4.21 gene
(Thompson etal 1991). The use of the VH4.21 gene appears mandatory
for antibodies reactive against the li carbohydrate antigen on the surface of
red cells (Pascual et al 1991). These antibodies are known as cold
agglutinins (CA). The 9G4 idiotope is defined by the amino acid motif
Ala-Val-Tyr at amino acid positions 23-25 in FRl of the VH4.21 VH
segment (Potter etal 1993). The 9G4 idiotope has been expressed on anti-
DNA antibodies and has been identified in 45% of sera from patients with
SLE and in the disease lesions (Isenberg era/ 1993).
Four monoclonal autoantibodies in the panel reported were found to
be VH4.21 encoded, three anti-DNA antibodies including one of which
was of the IgG isotype, and one anti-myeloperoxidase (MPO) IgM
antibody. The anti-DNA antibodies were isolated from two SLE patients
with active disease and the anti-MPO antibody was isolated from a patient
with microscopic polyarteritis.
The nucleotide sequence of the VH region of the IgM anti-ssDNA
antibody RT79 indicated that it was identical to the VH4.21 germline gene
with no mutations (fig 3.05). The long D segment encodes 17 amino acids
and had a complex genetic origins DLR4 with some possible contributions
from DXP’l and DLR5. The encoded amino acids of the CDR3 region
were predominantly basic with five arginine residues and only one acidic
residue aspartic acid and all the arginines appear to be contributed via N
region addition. The J region was unmutated JH3
RT79 variable light chain region aligned most closely at the
nucleotide level (99% homology) to the VkII expressed gene GM607 (fig
3.06) also used by a cold agglutinin FSl (Pascual etal 1991). One of the
two nucleotide differences from GM607 was common to the CA
monoclonal autoantibody FS-1 and RT79 which were both GM607
derived. The other nucleotide change from this gene gave rise to a
108
Figure 3.05 RT 79 HEAVY CHAIN VARIABLE REGION
F R I
VH4.2IRT79
Q c a g
V g t g
Qc a g
L c t a
Qc a g
Qc a g
W t g g
G g g c
A g c a
G g g a
L c t g
L t t g
K a a g
VH4.21RT79
Pc e t
St c g
E g > g
T a c c
L c t g
St c c
Le t c
Ta c e
C t g c
A g c t
V g t c
Y t a t
G g g t
VH4.21RT79
G g g g
St o c
C DR 1 F
t t cS
• g tG
g g tY
t a cY
t a cW
t g gS
• g c
F R 2 W
t g gI
a t cR
c g cQ
c a g
VH4.21RT79
PC G C
pc c a
G g g g
K a a g
G g g g
L c t g
Eg a g
W t g g
Ia t t
G g g g
CDR 2 E
g ■ ■I
a t cN
a a t
VH4.21RT79
Hc a t
S a g t
G g g «
Sa g c
T a c c
N a a c
Yt a c
N a a c
Pc c g
S t c c
Le t c
K ■ « g
Sa g t
VH4.21RT79
P R 3 R
c g aV
g t cT
a c eI
a t aS
t e aV
g t aD
g a cT
a c gS
t c cK
a a gN
a a cQ
c a gF
t t c
VH4.21RT79
St c o
L c t g
K a a g
L c t g
Sâ g e
St c t
V g t g
Ta c e
Ag c c
A g c g
D g a c
T a c g
A g c t
VH4.21RT79
V g t g
Y t a t
Yt a c
Ct g t
A g c g
R a g a
CDR 3 V
g t a
R
c g g
R
c g a
S
t c g
G
g g t
R
a g g
V
g t a
RT79V
g t *V
g t aP
c c aA
g c tA
g c cp
c e tR
c g tN
a a tR
a g gD
g » t
JH3RT79
A g c t
F t t t
D g a t
Ia t c
W t g g
G g g c
Qc a a
G g g g
T a c a
M a t g
V g t c
Ta c e
V g t c
S S JH3 t c t t e a
RT79 ____ _______
D ASSIGNMENT
RT79 GT AC G OC GAT C G GOT AG GOT AGT AC C AGC T GC CC C T C GT A AT A G G G ATDLR4____________________ A _ _ A _ _ T T ________________________ T A T G . CDXPl G _ T _________G _ _ T TDLR5 T T A
109
Figure 3.06 RT 79 KAPPA CHAIN VARIABLE REGION
F R ID I V M T Q S F L S L P V
GM607 g a t a t t g t g a t g a c t c a g t ô t c c a e t c t o c c t g c c c g t cRT79
CDR 1T P G E P A S I S C R S S
GM607 a c c c e t g g a g a g c c g g c c t c c a t c t c c t g c m g g t c t m g tRT79 ____
Q S L L H S N G Y N Y L DGM607 c m g a g c e t c e t c c a t a g t a a t g g a t a c a a c t a t t t g g a tRT79 ____ ______________ g ____ _________________________________________________
F R 2W Y L Q K P G Q > K S P Q L L
GM607 t g g t a c c t g c a g a a g c c a g g g c a g t c t c c a c a g e t c c t gRT79 ____ __________________________________________a _ ___________________________________
CDR 2 F R 3I Y L G S N R A S G V P D
GM607 a t c t a t t t g g g t t c t a a t c g g g c c t c c g g g g t c c e t g a cRT79 ____ ________________________________________________________
R F S G S G S G T D F T L GM607 a g g t t c a g t g g c a g t g g a t e a g g c a c a g a t t t t a c a c t gRT79 ____ ____________________________________________________________________________________
K l S R V E A E D V G V Y GM607 a a a a t c a g c a g a g t g g a g g e t g a g g a t g t t g g g g t t t a tRT79 ____ ____________________________________________________________________________________
CDR 3Y C M Q A L Q T P R
GM607 t a c t g c a t g c a a g e t c t a c a a a c t c e t * * *RT79 _______ g c g a
I > L T F G Q G T R L E I K RJK5 a t c a c e t t c g g g c a a g g c a c a c g a c t g g a g a t t a a a c g a
RT79 c _ _ __________________ _ g _______________ _________ _________ _________ _________ _________
110
Figure 3.07 RT 84 HEAVY CHAIN VARIABLE REGION
F R 1Q V Q L Q Q W G A G L L K
VH4.21 c a g g t g c a g c t a c a g c a g t g g g g c g c a g g a c t g t t g a a gRT84
P S E T L S L T C A V Y G VH4.21 c e t t c g g a g a c c c t g t o c e t c a c c t g c g o t g t c t a t g g tRT84
C D R 1 F R 2G S F S G Y Y W S W I R Q
VH4.21 g g g t o c t t c m g t g g t t a c t a c t g g a g c t g g a t c c g c c a gRT84
CDR 2P P G K G L E W I G E I N
VH4.21 c c c c c a g g g a a g g g g c t g g a g t g g a t t g g g g a a a t c a a tRT84
H S G S T N Y N P S L K S VH4.21 c a t a g t g g a a g c a c c a a c t a c a a c c c g t c c e t c a a g a g tRT84 ____
F R 3R V T I S V D T S K N Q F
VH4.21 c g a g t c a c c a t a t e a g t a g a e a e g t e e a a g a a e e a g t t eRT84 ____ ____________________________________________________________________________________
S L K L S S V T A A D T A VH4.21 t e e e t g a a g e t g a g e t e t g t g a e e g e e g e g g a e a e g g e tRT84 ____ _____________________________________________________________________________
CDR 3V Y Y C A R G R R I T G
VH4.21 g t g t a t t a e t g t g e g a g aRT84 ____ ___________________________________g g c c g g c g t a t a a c t g g a
T K RRT84 a c g a a g a g g
Y> H Y Y G M D V W G K> Q G T TJH6 t a e t a e t a e g g t a t g g a e g t e t g g g g e a a a g g g a e e a e g
RT84 e _ _ ____ _________________________________________________e _ _____________________
V T V S SJH6 g t c a e e g t e t e e t e a
RT84
D R E G IO N A S S IG N M E N T
RT84 G G C C G G C G T AT A A C T G G A A C G A A G A G G C A CDM1 G _______________________ T _ C
DYAl GT
I l l
Figure 3.08 RT 84 KAPPA CHAIN VARIABLE REGION
F R 1D I Q M T Q S P S S L S A
08/ 018 g a c a t e c a g a t g a c e c a g t c t o c a t e c t e c c t g t c t g c aRT84
CDR 1S V G N R V T I T C Q > R A S
08/018 t c t g t a g g a g a c a g a g t c a c c a t e a c t t g c c a g g c g a g tRT84 ________________________________________ g__________ _________
F R 2Q D I S > R N > I Y L N > T W Y Q Q K
08/018 c a g g a c a t t a g e a a c t a t t t a a a t t g g t a t c a g c a g a a aRT84 _______________ _ a _ t _ _____ _ c _
CDR 2P G K A P K L L l Y D A S
08/018 c c a g g g a a a g c c c e t a a g e t c c t g a t e t a c g a t g c a t c cRT84 ____ _______ _____________________________________________________________________________
F R 3N > K L E T G V > D P S R F S G S > T
08/018 a a t t t g g a a a c a g g g g t c c c a t e a a g g t t c a g t g g a a g tR T 8 4 a _______________ a c
G S G T D > V F T F T 1 S S L 08/018 g g a t c t g g g a c a g a t t t t a c t t t c a c c a t e a g e a g e c t gRT84 _________ _ _ c ___ _ t _ _____ _________________________________________________
CDR 3Q P E D 1 > F A T Y Y C Q Q Y > H
08/018 c a g c e t g a a g a t a t t g c a a c a t a t t a c t g t c a a c a g t a tRT84 ________________ t _ _ ____________ g _____c _ _
D N> H L P08/018 g a t a a t e t c c e tRT84 ____ c _ _ ______________
L T F G G G T K V E l K VJK4 e t c a c t t t c g g c g g a g g g a c c a a g g t g g a g a t c a a a g t a
RT84 ___________ _______
112
replacement amino acid in FR2 and the Jk5 is unmutated apart from an
amino acid change from I (ATC) to L (CTC) at the junction with CDR3
which have been found in other J k5 sequences (Hieter etal 1982).
However there is an insertion of an arginine residue between the CDR3
and the J region which could have arisen by N region addition and this
could affect interaction with the antigen
The VH region of the anti ssDNA monoclonal RT84 had 100%
homology with the germline gene VH4.21 at the nucleotide level (fig
3.07). The CDR3 region encoded nine amino acids and arose from
contributions of Dipal and DM1 D region genes. There were three arginine
and one lysine residue within the CDR3 region of the VH region. The
framework four region was generated from JH6 with two replacement
mutations (Tÿr>His, Lys>Gln).
The VL of RT84 had 94% homology to the V kI germline gene
08/018 (fig 3.08) (Scott etal 1991). Unfortunately, the VL region of RT84
had 100% homology with the V kI encoded VL region expressed by the
fusion partner GM4672. GM4672 is a low level secretor of an IgG k
monoclonal antibody (Rioux etal 1993). The light chain was encoded by
the V kI member 08/018 (Scott eta l 1991) and the heavy chain was
encoded by the VH4 member V71-2 (Kodaira et al 1986). Fourteen
monoclonals immortalised by fusion with GM4672 were sequenced and of
these six were encoded by the V kI subgroup, and none of them were
identical to the GM4672 kappa chain (Rioux etal 1993). The GM4672
fusion partner was used in the immortalisation of the “RT”, “RSP”,
WRI176, and VAR24 monoclonal antibodies. None of these monoclonals
(apart from RT84) treated in an identical manner gave rise to the GM4672
fusion partner V k region. In fact, amplification of the GM4672 kappa
chain in preference of the V kI expressed by the donor B cell has never
been observed (M Newkirk personal communication). It is possible that
the PCR primers employed preferentially amplified the V k of GM4672 or
that RT84 light chain is encoded by the sam e V k as the fusion partner.
The heavy chain variable region of RT79 was compared to a panel
113
Figure 3.09
IgM VH4.21 ENCODED ANTI-DNA HEAVY CHAINS COMPARED WITH COLD AGGLUTININS WHICH DO NOT BIND DNA
VH4.21FS-2FS-3FS-6FS-7RT79RT84N El
F R l CDRlQ V Q L Q Q W G A G L L K P S E T L S L T C A V y G G S F S G Y Y W S
H T D
H
VH4.21FS-2FS-3FS-6FS-7RT79RT84N El
FR2W I R Q P P G K G L E W I G
CDR2E I N H S G S T N Y N P S L K S D _ _ I __________________ Y ________________________
L
VH4.21FS-2FS-3FS-6FS-7RT79RT84N El
FR3R V T I S V D T S K N Q F S L K L S S V T A A D T A V Y Y C A R
L
FS-2 E W G S G A Y Y P Y Y F D Y JH4FS-3 G T R S W Y P S D F D Y JH4FS-6 A L G D G S T E G L P D Y JH4FS-7 G R E Q W L V R G G Y F D Y JH4RT79 V R R S G R V V V P A A P R N R D A F D I JH3RT84 G R R I T G T K R H Y Y G M D V JH6NEl R H V R P R V T R MD V JH6
The amino acid motif A VY in FRl defines the 9G4 idiotope (Potter et a! 1993)
114
of VH4.21 encoded cold agglutinins which did not bind DNA (Stevenson
et al 1993) (fig 3.09). On examination of these monoclonals, there
appeared to be differences within the CDR3 regions. RT79 had five
arginines within its CDR3 region which did not appear within the non-
DNA binding cold agglutinins.
The variable heavy chain region of RT84 had 100% homology with
VH4.21 and the content of the CDR3 region also reflects the pattern found
in RT79 VH region. The CDR3 region of RT84 consisted of nine amino
acids which is considerably shorter than the CDR3 region found in RT79.
However there are some features which were common to both regions, the
most significant being the preponderance of positively charged amino acids
found in the CDR3 regions of both of these monoclonals. Interestingly
both RT79 and RT84 had an arginine couplet at the 5’ end of their CDR3
regions. It is likely that these residues arose by N addition. RT79 and
RT84 VH regions were also compared to the VH4.21 encoded anti-DNA
antibody NEl (Hirabayashi et al 1993). NEl VH region had 100%
identity with the VH4.21 germline gene and had four arginine residues
within the CDR3 region. The CDR3 regions of RT79, RT84 and NEl
contained at least one arginine couplet and apart from one aspartic acid
residues in the VH CDR3 region of RT79, there were no acidic amino acid
residues present. Interestingly, the CDR3 of the cold agglutinin FS-2
contained four tyrosine residues which are thought to aid binding to DNA
but did not bind DNA. The FR4 region in NEl and RT84 was encoded by
JH6 gene segment and thus the heavy chain VH regions of NEl and RT84
were very similar.
The VH region of the IgG anti-ssDNA antibody D5 was derived
from the VH4.21 gene (94% homology) (fig 3.10). There were 18
mutations from the germline sequence of which 11 encode replacement
amino acids. These were found both in the CDR’s and in FR3 and the R:S
ratios are 1.5 and 2.5 respectively which did not indicate any selection for
amino acid changes in the CDRs. The D segment of D5 contained
elements derived from the DXP’l gene with various additions. The
deduced fourteen amino acid sequence for the CDR3 region was again
115
Figure 3.10 D5 HEAVY CHAIN VARIABLE REGION
F R 1Q V Q L Q Q W G A G L L K
VH4.21 c a g g t g c a g c t a c a g c a g t g g g g c g c a g g a c t g t t g a a gD5
P S E T L S L T C A V Y GVH4.21 c e t t c g g a g a c c c t g t c c e t c a c c t g c g e t g t c t a t g g t
D5
C D R l F R 2G S r S > T G Y > H Y W S W I R Q
VH4.2I g g g t c c t t c m g t g g t t a c t a c t g g a g c t g g a t c c g c c a gD5 _____________ _ c _ ______ g c ____ _ t
CDR 2P P G K G L E W I G E I N
VH4.21 c c c c c a g g g a a g g g g c t g g a g t g g a t t g g g g a a a t c a a tD5
H S G S T > A N > Y Y > F N > S P S L K SVH4.21 c a t a g t g g a a g c a c c a a c t a c a a c c c g t c c e t c a a g a g t
D5 c ________ _____t g t t _ _ g_____________________________________
F R 3R V T > I I > M S V D T S K N Q F
VH4.21 c g a g t c a c c a t a t e a g t a g a c a c g t c c a a g a a c c a g t t cD5 _ t t _ _ g ----------------- ---------- ---------- ---------- --------------------- ----------
S L K> N L S > T S V T A A D T AVH4.21 t c c c t g a a g c t g a g c t c t g t g a c c g c c g c g g a c a c g g e t
D5 ____ c __________ c _______ c _______
CDR 3V Y Y> F C A R G A P N F G N
VH4.21 g t g t a t t a c t g t g c g a g aD5____ ___________ t t _______________g g c g c c c c a a a t t t c g g g a a c
Y Y K A R R G DS t a t t a t a a a g c g c g c c g a g g c
W F D P W G Q G T L V T V JHS t g g t t c g a c c c c t g g g g c c a g g g a a c c c t g g t c a c c g t cD5 ____ ____________________________________________________________________________________
S SJHS t c c t e aDS ____ _______
D ASSIGNMENT
DS G G C G C C C C A A A T T T C T T T * A A C T A T T A T A A A DXPl T A T T A_ T _ T G G __________ G_ _G T ______________ C
116
basic with two arginines, one lysine and no acidic residues and the J region
was unmutated JH5
The heavy chain variable region of D5 had marked sequence
similarity to the previously reported anti-dsDNA (T14) secreted by an
EBV-transformed B cell line from a patient with active SLE (Van Es et al
1991) (fig 3.11). T14 also utilised the VH4.21 gene segment and when
compared to D5 there were two common amino acid replacements in CDRl
and CDR2 and the CDR3s are each derived from DXP’l in the same
reading frame leading to four common amino acids. Frequent usage of this
D segment gene by IgM anti-DNA has been noted previously (Cairns etal
1989). There are other similarities between D5 and T14 in CDR3 which
are difficult to assign for example, the common Arg-Gly residues at the 3’
end of the CDR3 region. This structural similarity between two
independent selected IgG anti-DNA antibodies suggest that the VH4.21
gene segment generates this specificity in a limited manner and if arginine
residues are important that they operate from CDR3 rather than from CDRl
or CDR2. In fact D5 had an arginine doublet at the 3’ end of CDR3
whereas T14 had only a single arginine which could affect DNA binding
affinity.
D5 VL region had 95.5% homology with the VKlIIb family member
HumlGKVQ (fig 3.12) (Chen etal 1987). Of 13 mutations in the V k of
D5 nine gave rise to replacements but five were in the CDRs and four in
the framework one and two. As with the heavy chain of D5, R:S ratios in
the light chain did not indicate a selective pressure on the CDRs and a
unmutated Jk5 is used. The replacement amino acids did not show
increased basicity although the CDRl and CDR2 each had a substitution of
serine by asparagine which might be relevant to the recognition of DNA
(Eilat etal 1988).
In summary, three VH4.21 encoded anti-DNA antibodies were
analysed. The VH regions of the IgM anti-DNA antibodies RT79 and
RT84 had 100% identity with the germline gene VH4.21. Both contained
a preponderance of arginine and lysine residues in the VH CDR3 region. \
117
Figure 3.11
DS IgG HEAVY CHAIN VARIABLE REGION COMPARED WITH T14 VH REGION
FR l CD RlVH4.21 Q V Q L Q Q W G A G L L K P S E T L S L T C A V Y G G S F S G Y Y W S
D5 ________ ___________________________________________________ _ T _ H _____T14 H F
FR2 CDR2VH4.21 W I R Q P P G K G L E W I G E I N H S G S T N Y N P S L K S
D5 A Y F S __________T14 D I S
FR3VH4.21 R V T I S V D T S K N Q F S L K L S S V T A A D T A V Y Y C A R
D5 _ _ I M N _ T ____________________ F ______T14 R A
C D R 3
G A P NG W P Y
D5 G A P N F G N Y Y K A R R G JH5T14 G W P Y Y Y G A G S Y Y K R G JH4
118
RT84 had remarkable similarity to the VH4.21 encoded anti-DNA antibody
NEl. These monoclonals used quite diverse V k light chains.
The VH region D5, an IgG anti-DNA antibody, had 94% homology
with VH4.21 and also had positively charged amino acid residues in the
VH CDR3. D5 VH region had a strong similarity to the anti-DNA
antibody T14. The light chains of both D5 and T14 were VKlIIb encoded.
The CDR3 regions of RT79, NEl, D5, and T14 were derived from
contributions from the DXP’l D segment. There appeared to be no
preferential usage with respect to JH utilisation.
119
Figure 3.12 D5 KAPPA CHAIN VARIABLE REGION
F R 1E l V L T Q S P G T L S L
IGKVQ g a a a t t g t g t t g a c g c a g t c t c c a g g c a c c c t g t c t t t gD5
C DR 1S P G E > D R > G A T > S L S C R A S
IGKVQ t c t c c a g g g g a a a g a g c c a c c e t c t c c t g c a g g g c c m g tD5 ______________ _ _ t g _ _t _ _ ____________________________________
F R 2Q S y S > T S S > N Y L A W Y Q Q
IGKVQ c a g a g t g t t a g c a g c a g c t a c t t a g c c t g g t a c c a g c a g
CDR 2K P G Q> H A P R L L I V G A
IGKVQ a a a c e t g g c c a g g e t c c c a g g e t c e t c a t c t a t g g t g c aD5 ______________ c ______________ _________ _________ _________ _________ _________ _________ _________
F R 3S S > N R A T G I P D R F S G
IGKVQ t c c a g c a g g g c c a c t g g c a t c c c a g a c a g g t t c a g t g g cD5 ____ ___ a ____ ______________________________________________________________________
S G S G T D F T L T I S RIGKVQ a g t g g g t c t g g g a c a g a c t t c a c t e t c a c c a t c a g c a g a
D5 ____ ____________________________________________________________________________________
CDR 3L E P E D F A V Y Y C Q Q > H
IGKVQ c t g g a g c e t g a a g a t t t t g c a g t g t a t t a c t g t c a g c a gD5 _______ _ _ c
Y G S > T S PIGKVQ t a t g g t a g c t e a c e t
D5 _______ c _
I T F G Q G T R L E I K RJK5 a t c a c c t t c g g c c a a g g g a c a c g a c t g g a g a t t a a a c g aD5 _________ _____________________
120
3.3 AN ANTI-M YELOPEROXIDASE MONOCLONAL
ANTIBODY ENCODED BY THE VH4.21 GERMLINE GENE.
Circulating antibodies to myeloperoxidase are primarily associated
with pauci-immune glomerulonephritis and systemic vasculitis. Anti-MPO
antibodies belong to a group of autoantibodies, anti-neutrophil cytoplasmic
antibodies (ANCA) which have been suggested to play a pathogenic role in
vasculitis.
The monoclonal E3-MPO was generated from the peripheral blood
lymphocytes of a patient with microscopic polyarteritis and immortalised
using the CBF7 heteromyeloma fusion partner (Ehrenstein etal 1992).
E3-MPO was found to carry the 9G4 idiotope (Longhurst etal 1994)
which is a marker for the VH4.21 germline. E3-MPO has been shown
previously to be specific for the native form of human leucocyte
myeloperoxidase and not to an extensive panel of autoantigens (Ehrenstein
etal 1992).
The nucleotide sequence of the VH region of the IgM anti MPO
antibody E3 showed that it had 90% homology with the germline gene
VH4-21 at the nucleotide level (fig 3.13). The D region can be aligned to
two D elements DLR4 and Dl. A JH3 region with one silent mutation was
utilised.
A summary of the replacement mutations in the amino acid
sequence of E3-MPO, assuming VH4.21 as the germline gene of origin, is
shown in table 3.2. In the framework regions, a total of nine replacement
mutations were present with only two resulting in a change to a negatively
charged residues, in both cases, glutamic acid. Framework one was
highly conserved having only one replacement mutation, glutamine to
glutamic acid which resulted in the gain of a negative charge. Framework
two had two replacement mutations (Pro>Ser, Gly>Glu) which resulted in
the gain of one negative charge. Framework three (FR3) had six
replacement mutations, none of which yielded charged residues. However
a positively charged lysine residue was replaced by a uncharged serine
residue resulting in the loss of a positive charge. In CDRl there were
121
Figure 3.13 E3-MPO HEAVY CHAIN VARIABLE REGION
F R 1Q V Q L Q> E Q W O A G L L K
VH4.21 c a g g t g c a g c t a c a g c a g t g g g g c g c a g g a c t g t t g a a gE3 _____________ t __g _______ ____________ t _______ c _______________
P S E T L S L T C A V Y G VH4.21 c e t t c g g a g a c c c t g t c c e t c a c c t g c g e t g t c t a t g g t
E3 ____________________________ 1 _ _
C D R l F R 2G S F > L S G ¥ > F Y W S > T W I R Q
VH4.21 g g g t c c t t c m g t g g t t a c t a c t g g a g c t g g a t c c g c c a gE3 ______ c ___ _____________ t c _________ _________ _________ _________ _________
C D R 2P > S P G> E K G L E W I G E I N> S
VH4.21 c c c c c a g g g a a g g g g c t g g a g t g g a t t g g g g a a a t c a a tE3 t _ _ ________ a _ t _ g _
H S > R G S > ¥ T N > D ¥ N P > T S L > I K SVH4.21 c a t a g t g g a a g c a c c a a c t a c a a c c c g t c c e t c a a g a g t
E3 ---------- _ _ a _ _ t t a t --------g _ ____ _______--------------------- --------------------- ----------
F R 3R V T I S V > I D T S > Y K N Q F > V
VH4.21 c g a g t c a c c a t a t e a g t a g a c a c g t c c a a g a a c c a g t t cE3________ ____________________________a ______________ a t _______ a g _______
S L K > S L S S V T A A D T A > SVH4.21 t c c c t g a a g c t g a g c t c t g t g a c c g c c g c g g a c a c g g e t
E3 -------------- _ g t ___________ t _ _
CDR 3V Y Y C A > V R S T H R S
VH4.21 g t g t a t t a c t g t g c g a g aE3 c ____ _ t a g t a c c c a c c g g t c g
A F D V W G Q G T M V T VJH3 g c c t t t g a t g t c t g g g g c c a a g g g a c a a t g g t c a c c g t cE3 _ _ t ,_________________________________________________________________________________
S SJH3 t c t t c gE3 ____ ___ _ a
D REGION ASSIGNMENT
E3 AGT A C C C A C C G G T C GDLR4 X G _ T _DIrev _ A_ _
122
three replacement mutations resulting in no gain or loss of charge. There
were six replacement mutations within CDR2 including serine to arginine
and asparagine to aspartic acid. The CDR3 region contained five amino
acids and two were positively charged, namely histidine and arginine.
A panel of cold agglutinins were tested for anti-MPO activity and
were found to be negative (Longhurst etal 1994). The VH sequence of
E3-MPO was compared to a panel of cold agglutinins which were also
encoded by VH4.21 but did not bind to myeloperoxidase (fig 3.14). RT79
was also negative for myeloperoxidase binding and E3-MPO had no anti-
DNA binding activity (Longhurst etal 1994). A unique feature of these
monoclonals was the conserved framework one region. Analysis by Potter
etal (1993) revealed that a motif iWY within framework one defines the
9G4 idiotope. As this region was not mutated in E3-MPO and thus could
account for the expression of 9G4 by this antibody in spite of the presence
of replacement mutations in other areas of the VH region. The positively
charged residues found in RT79 VH CDR3 were not present in E3-MPO.
Myeloperoxidase is a highly basic molecule and thus it is possible that
negatively charged residues such as glutamic and aspartic acid could
enhance binding to such an antigen. However apart from a relatively
unique glutamic acid residue within framework one there did not appear to
be a preponderance of negatively charged residues within E3-MPO.
The nucleotide sequence of E3-MPO light chain variable region had
90% homology with the V k I germline gene 012 (fig 3.15) (Zachau etal
1993). A summary of the replacement mutations of E3-MPO compared
with 012 is shown in table 3.2. Framework two had three replacement
mutations, none of which yielded a change in charge. In framework three
two replacement mutations were present. A JkI gene segment was utilised
with four replacement mutations. CDRl had three replacement mutations
one of which gave rise to a gain in positive charge (Ser > Arg ). CDR2 had
four replacement mutations, most significantly alanine was replaced by
negatively charged aspartic acid which could enhance binding to positively
charged myeloperoxidase. Three replacement mutations occurred in CDR3,
123
Figure 3,14
E3-MPO HEAVY CHAIN VARIABLE REGION VERSES MPO NEGATIVE COLD AGGLUTININS
VH4.21E3-MPO
FS-1FS-2FS-3FS-5FS-6FS-7RT79
F R 1Q V Q L Q Q W ' G A G L L K P S E T L S L T C A V F G G S
E
H
C D R l F S G Y Y WS L _ _ F _ _ T
_ D _ ” _ T ” D ”
H
VH4.21E3-MPO
FS-1FS-2FS-3FS-5FS-6FS-7RT79
F R 2WI R Q P P G K G L E
S E
C D R 2E l N H S G S T N Y N P S L K S
S _ R _ Y _ D _ T II I ____________
D _ I " I _ I _” “ y
F R 3VH4.21 R V T I S V D T S K N Q F S L K L S S V T A A D T A V Y Y C A RE3-MPO I Y V S S V
FS-1 A N MN T L GFS-2 LFS-3FS-5FS-6 QFS-7RT79
C D R 3E3-MPO S T H R S A F D V W G Q G T M V T V S S JH3
FS-1 G Q R G S G S D Y R Q G A . E D I L I JH4FS-2 E W G S G A Y Y P Y Y . F D Y N L JH4FS-3 G T R S W Y P S D . F D Y L JH4FS-5 G P Y Y I D D S S G Y Y P G . . . . L JH4FS-6 A L G D G S T E G L P . . D Y R L F JH4FS-7 G R E Q W L V R G G Y F D Y L JH4RT79 V R R S G R V V V P A A P R N R D A F D I JH3
The amino acid motif AVY in FRl defined the 9G4 idiotope(Potter et al 1993)
124
Figure 3.15 E3-MPO LIGHT CHAIN VARIABLE REGION
F R 1D I Q M T Q S P S S L S A
02/012 g a c a t c c a g a t g a c c c a g t c t c c a t c c t c c c t g t c t g c aE3-MPO
CDR IS V G D R V T I T C R A S
02/012 t c t g t a g g a g a c a g a g t c a c c a t c a c t t g c c g g g c a m g tE3-MPO
F R 2Q S I S > R S > N Y > S L N W Y Q Q > G K
02/012 c a g a g c a t t a g c a g c t a t t t a a a t t g g t a t c a g c a g a a aE 3 - MP O _ _ g _ « t _ c _ ------------ _ _ c _____ _______ » g g a _____
CDR 2P > A G K A P K L > M L I Y > S A> D A S > V
02/012 c c a g g g a a a g c c c e t a a g e t c c t g a t c t a t g e t g c a t c cE3-MPO g a _ g t ___ _________ _ c _ _ a t g t _
F R 3S > T L Q > G S G V P S R F S G S
02/012 a g t t t g c a a a g t g g g g t c c c a t e a a g g t t c a g t g g c a g tE3-MPO _ c _ g g _ ____ ________________________________________________________
G S G T D F T > S L T I S S L02/012 g g a t c t g g g a c a g a t t t c a c t e t c a c c a t c a g c a g t c t g
E 3 - MP O _________________ _ _ t _ g _ _ c ______
CDR 3Q> L P E D F A T Y Y C Q Q S > G
02/012 c a a c e t g a a g a t t t t g c a a c t t a c t a c t g t c a a c a g a g tE3-MPO _ t g _______________ _ g ________ _________ _ t g _ c
Y S > N T > N P02/012 t a c a g t a c c c c
E3 - MP O a _ _ a c
%> R T F G Q G T K> R V E I K > N RE3-MPO t g g a c g t t c g g c c a a g g g a c c a a g g t g g a a a t c a a a c g t
JKl c _ _ _________________________ g _________________ _________ _ g _____ _ t _ _ a
125
but no loss or gain of charge resulted from the amino acid changes.
The distribution of replacement mutations in the heavy and light
chain variable regions of E3-MP0 assuming VH4.21 and 012 were the
germ line genes of origin suggests that there was no evidence of an antigen
driven mechanism in the generation of this antibody
The utilisation of the VH4.21 germ line gene in the generation of an
anti-MPO autoantibody shows that there could be a relationship between
anti-MPO antibodies, anti-DNA antibodies and cold agglutinins.
126
Table 3.2
REPLACEMENT MUTATIONS IN E3 MFC V REGIONS
VH VL
VH4.21 E3-MPO 02/012 E3-MPO(Germline) (Germline)
FR l Gln(P) > Glu(-ve) CDRl Ser(P) > Arg(+ve)Ser(P) > Asn(P)
CDRl Phe(NP) > Leu(NP) Tyr(P) > Ser(P)Tyr(P) > Phe(NP)Ser(P) > Thr(P) FR2 Gln(P) > Gly(P)
Pro(NP) > Ala(NP)FR2 Pro(NP) > Ser(P) Leu(NP) > Met(NP)
Gly(P) > Glu(-ve) Tyr(P) > Ser(P)
CDR2 Asn(P) > Ser(P) CDR2 Ala(NP) > Asp(-ve)Ser(P) > Arg(+ve) Ser(P) > Val(NP)Ser(P) > Tyr(P) Ser(P) > Thr(P)Asn(P) > Asp(-ve) Gln(P) > Gly(P)Pro(NP) > Thr(P)Leu(NP) > Iso(NP) FR3 Thr(P) > Ser(P)
Gln(P) > Leu(NP)FR3 VaI(NP) > Iso(NP)
Ser(P) > Tyr(P) CDR3 Ser(P) > Gly(P)Phe(NP) > Val(NP) Ser(P) > Asn(P)Lys(+ve) > Ser(P) Thr(P) > Asn(P)Ala(NP) > Ser(P)Ala(NP) > Val(NP) Jk l Trp(NP) > Arg(+ve)
Lys(+ve) > Arg(+ve)Lys(+ve) > Asn(P)
(NP) = Non polar amino acid (?) = Polar amino acid (+ve) = Positive charge (-ve) = Negative charge
127
3.4 A MONOCLONAL AUTOANTIBODY ENCODED BY THE
GERMLINE GENE VH4.22
The IgM polyreactive monoclonal autoantibody RT72 was derived
from the splenocytes of a patient wth active SLE. RT72 bound to ss/ds
DNA, cardiolipin, histones HI and H4, and SmD peptide 10 (Ravirajan et
d 1992)
The nucleotide sequence of the VH region of RT72 had 100% with
the germline gene VH4.22 (fig 3.16) (Sanz etal 1989). The CDR3 region
encoded twelve amino acids and appeared to be derived from a D-D fusion
between DXP4 and DM2. The CDR3 region contained two tyrosines, one
lysine, and one asparagine which could mediate binding to DNA and/or
cardiolipin. Unlike the VH4.21 encoded anti-DNA antibodies there were
no arginine residues present in the CDR3 region of RT72 VH. RT72
heavy chain CDR3 region also contained an aspartic acid residue which
may enhance binding to positively charged histones. The framework four
was encoded by unmutated JH5.
An anti-idiotypic reagent was produced using the RT72 monoclonal
and was found to be located on the heavy chain variable region. RT72 Id
was elevated in 12% of of SLE patients and in 50% of patients with
rheumatoid arthritis (Kalsi etal 1994). A panel of monoclonals (kindly
supplied by Jan Hillson) were tested for the expression of RT72 Id and
three 2B8, 2E7 and 3B8 were found to be positive (Kalsi etal 1994). 2B8
and 2E7 were both encoded by the VH3 member VH26 and utilised a JH4
gene segment. 3B6 utilised the VH5 member 51R1 and a JH3 gene
segment. The monoclonal 4D5 was negative for RT72Id and was encoded
by the VH3 germline gene 13-2 and a JH4 gene segment. The VH amino
acid sequences of these monoclonals are compared in figure 3.17. This
comparison showed that RT72Id was not VH gene family specific. The
VH26 derived 2B8 and 2E7 tested positive for RT72Id, however 4D5
which was also encoded by a member of the VH3 family was RT72Id
negative. This exercise illustrated the difficulty of mapping idiotypes
u s i n g l i n e a r a m i n o a c i d s e q u e n c e s .
128
Figure 3.16 RT 72 HEAVY CHAIN VARIABLE REGION
F R 1
VH4.22RT72
Eg a g
St e g
G g g c
Pe e a
G g g a
L c t g
V g t g
K a a g
Pc e t
St e g
Eg a g
T a e e
L c t g
VH4.22RT72
St c c
L e t e
T a e e
C t g e
Ta c t
V g t c
St c t
Gg g t
Yt a c
St c c
Ia t e
Sa g e
C DR S
a g t
VH4.22RT72
G g g t
Y t a c
Yt a c
W t g g
G g g c
F R 2 W
t g gI
a t cR
c g gQ
c a gP
c c cP
c c aG
g g gK
a a g
VH4.22RT72
G g g a
L c t g
Eg a g
W t g g
Ia t t
G g g g
CDR 2 S
a g tI
a t cY
t a tH
c a tS
a g tG
g g gS
a g c
VH4.22RT72
Ta c c
Yt a c
Yt a c
N a a c
Pc c g
S t c c
Le t c
K a a g
Sa g t
F R 3 R
c g aV
g t eT
a c eI
a t a
VH4.22RT72
St e a
V g t a
D g a e
T a e g
St e e
K a a g
N a a e
Qe a g
F t t e
St e e
L c t g
K a a g
L c t g
VH4.22RT72
Sa g e
St e t
V g t g
Ta e e
A g c c
A g c a
D g a e
T a e g
A g c c
V g t g
Y t a t
Yt a c
C t g t
VH4.22RT72
A g c g
R a g a
CDR 3 G
g g a
Y
t a t
Y
t a c
D
g ■ t
F
t t t
W
t g g
S
a g t
G
g g t
F
c c c
G
g g a
K
a a a
RT72N
a a c
JH5RT72
W t g g
F t t e
D g a e
Pe e e
W t g g
G g g c
Qe a g
G g g a
Ta e e
L c t g
V g t e
T a e e
V g t e
S S JH5 t c c t e
RT72 _______ _
D REGION ASSIGNMENT
RT72 GGAT A T T A C G A T T T T T G G A G T G G T C C G G A A A A A C DXP4 _ T ______________________________________DM2 C C _ _
129
Figure 3.17
RT72 VH REGION COMPARED WITH RT72 Id POSITIVE MONOCLONAL VH REGIONS
F R 1 C D R 1 F R 2RT72 E S G P G L V K P S E T L S L T C T V S G Y S I S S G Y Y W G R QP P2B8 . . . G V Q G G S R S A A F T F AMS V A2E7 . . . G V Q G G S R S A A F T F " . ” AMS V “ a3B6 . . . A E V K _ I G_ S _ Kl S _ K G _______F T _ _ V " M_4D5 • • • ^ _____0 _ G G S _ R _ S _ A A _ _ F T F _ _ . _ A M S _ V_ _ A_
C D R 2 F R 3RT72 GLE>VI G S 1 Y H S G . S T Y Y N P S L K S 2B8 V A S G G A D V G2E7 V A S G G A D V G3B6 _______M_ i I _ P G D S d I r _ S _ ~ F Q _4D5 V A S G T G D P G V G
R V T I S V D T S K N Q F S L K L_ F R _ N T L Y _ QM
F r ” n T L Y Q MQ G A _ K _ I _ T A Y _ Q WQF R N A S L Y Q M
C D R 3RT72 S S V T A A D T A V Y Y C A R G Y Y D F ^ ^ S G P G K JH52B8 N _ L R E ________________ K D V Y G G JH42 E 7 N L R ~ E K V G G D G F JH43B6 _ _ L K _ S __________________ R R V JH44D5 N L R G A A R V A A A G P Y Q L JH4
130
Figure 3.18 RT 72 KAPPA CHAIN VARIABLE REGION
F R ID I Q M > L Y Q S P S T > S P > L S A
HK102 g a c a t c c a g a t g a c c c a g t c t c e t t c o a c e c c g t c t g c aRT72 ____________________ _c _ _ _ a _____ t _ _ _ t _ _____________
CDR 1S V G D R V T I T C R A S
HK102 t c t g t a g g a g a c a g a g t c a c c a t c a c t t g c c g g g c c m g tRT72 ____ ____________________________________________________________________________________
F R 2Q S I S S W L A W Y Q Q K
HK102 c a g a g t a t t a g t a g c t g g t t g g c c t g g t a t c a g c a g a a aRT72
CDR 2P G K A P K L L I Y D > K A S
HK102 c c a g g g a a a g c c c e t a a g e t c c t g a t c t a t g a t g c c t c cRT72 a g _ _ a __ t
F R 3S L E S G V P S R F S G S
HK102 a g t t t g g a a a g t g g g g t c c c a t e a a g g t t c a g c g g c a g tRT72 ________ _ a ____ ______________________________________________________________________
G S G T E F T L T I S S LHK102 g g a t c t g g g a c a g a a t t c a c t e t c a c c a t c a g c a g c c t gRT72 ____ ____________________________________________________________________________________
C DR 3Q P D D F A T Y Y C Q Q Y
HK102 c a g c e t g a t g a t t t t g c a a c t t a t t a c t g c c a a c a g t a tRT72 ____ ____________________________________________________________________________________
N S Y SHK102 a a t a g t t a t t c tRT72 ____ _____________________
% > R T F G Q G T K V E I K R JKl t g g a c g t t c g g c c a a g g g a c c a a g g t g g a a a t c a a a c g t
RT72 c_________________________________________________________ _________ _____________________
131
The nucleotide sequence of Vk region of RT72 had 97% homology
with the VkI germline gene HK102 (fig 3.18) (Bentley etal 1980) at the
nucleotide level. There were three replacement mutations within the
framework one region. None of these replacements gave rise to amino
acids that have been cited in aiding affinity to any of the autoantigens that
react with RT72 monoclonal. However, a replacement mutation in CDR2
converted aspartic acid to lysine resulting in the removal of negative charge
and the addition of lysine, an amino acid residue which could aid binding
to ssDNA. A JkI gene segment was utilised with one replacement
mutation where tryptophan was replaced by arginine, a positively charged
amino acid which could enhance DNA binding.
The V kI gene family is associated with the 31 anti-DNA associated
idiotype (Manheimer-Lory etal 1991). This Id is present in 80% of SLE
patients with anti-DNA activity. 31 positive immunoglobulins are deposited
in the skin and kidneys of SLE patients indicating a pathogenic subset.
The expression of 31 is not confined to V kI encoded antibodies and as
there is conservation of framework two regions between members of the
VkI subset and 31 positive members of the V k3 and V k4 families, 31 is
likely to be framework encoded. RT72 V kI region was compared to a
panel of 31 Id positive light chains (fig 3 .1 9 ). It is not known if RT72
carries the 31 idiotype. However RT72 V k amino acid sequence has 100%
homology within the framework one and two with the panel of 31 positive
monoclonals reported by Manheimer-Lory et al (1 9 9 1 ). It is extremely
likely RT72 is positive for this pathogenic idiotype.
RT72 VH region was totally germline encoded and the VL region
contained six replacement mutations. RT72 carried the heavy chain
associated RT72 idiotype which has been found elevated in the sera of SLE
and RA patients. It was possible that RT72 carried a second light chain
encoded idiotype, 31, which has been associated with pathogenic anti-
DNA antibodies.
132
Figure 3.19
RT72 VK REGION COMPARED WITH 3IId POSITIVE VK REGIONS
RT72m-3Rin-2R
IC4l-2a
F R lD I Q L Y Q S P S S L S A S V G D R V T I T C
MT
C D R lR A S Q S I S S W L AQ D _ _ N Y _ N
G
R T
RT72m-3R
IC4l-2a
F R 2A ^ Y Q Q K P G K A P K L L I Y
C D R 2 K A S S L E S D N T
m-2R ________________ V A _ _ T _ Q
F R 3G V P S R F S G S G S
I
R G
RT72 G T E F T L T I S S L Q P D D F A T Y Y Cm-3R _ _ D _ _ F ____________ E _ I ________m-2R _______________________________ V __________IC4l-2a I
C D R 3 Q Q Y N S Y S
D N L P A P“ D T L P L I P
K
133
3.5 MONOCLONAL AUTOANTTRODfES ENCODED BY THE
VH4.18 GERMLINE GENE
RT55, an IgM monoclonal autoantibody was derived from the
splenocytes of a patient with active SLE. RT55 had reactivity with
ssDNA, cardiolipin, Sm-RNP, and histone HI.
RT55 VH region had 100% identity with the germline gene VH4.18
(fig 3.20) (Sanz etal 1989). The CDR3 was composed of nine amino
acids including two positively charged amino acids (histidine and lysine)
and one negatively charged amino acid (glutamic acid). The CDR3 region
was composed of elements of DHQ52 and DM2. An unmutated JH5 gene
was utilised.
The VH region of VAR24, an IgM monoclonal derived from a
polymyositis patient also had 100% homology VH4.18. VAR24 had
reactivity with 56 kD RNP but not with DNA. On comparison of RT55
and \AR24, the differences between the two VH segments were not
marked (fig 3.21). VAR24 heavy chain CDR3 region consisted of twelve
amino acids. Two of these amino acids were charged, one positively
charged arginine residue and and one negatively charged aspartic acid.
RT55 heavy chain CDR3 region consisted of nine amino acids, three were
charged residues, two positively charged, histidine and lysine and one
negatively charged glutamic acid. Therefore the basicity of RT55 VH
region is greater than that of VAR24 and this factor could influence affinity
to DNA. However VAR24 had four tyrosine residues within its CDR3
region. Tyrosine residues, thought to aid DNA binding were not able to
generate DNA affinity in VAR24.
RT55 V k region had 98% homology with the germline gene
HSIGKIO (fig 3.22) (Jaenichen etal 1984). There were two replacement
mutations in framework one changing a valine residue to aspartic acid
which led to a gain in negative charge and a tryptophan residue to
glutamine. The two other replacement mutations occurred in the CDR3
region where phenylalanine was converted to serine and proline was
134
Figure 3.20 RT 55 HEAVY CHAIN VARIABLE REGION
F R 1Q L Q L Q E S G P O L V K
VH4.18 c a g c t g c a g c t g c a g g a g t c g g g c c c a g g a c t g g t g a a g RT55
F S E T L S L T C T V S G VH4.18 c e t t c g g a g a c c c t g t c c e t c a c c t g c a c t g t c t c t g g tRT55
C D R l F R 2G S I S S S S Y Y W G W I
VH4.I8 g g c t c c a t c a g c a g t m g t m g t t a c t a c t g g g g c t g g a t cRT55
C DR 2R Q P P G K G L E W I G S
VH4.18 c g c c a g c c c c c a g g g a a g g g g c t g g a g t g g a t t g g g a g tRT55
I Y Y S G S T Y Y N P S LVH4.18 a t c t a t t a t a g t g g g a g c a c c t a c t a c a a c c c g t c c e t cRT55 ____ ____________________________________________________________________________________
F R 3K S R V T I S V D T S K N
VH4.I8 a a g a g t c g a g t c a c c a t a t c c g t a g a c a c g t c c a a g a a cRT55 ____ ____________________________________________________________________________________
Q F S L K L S S V T A A D VH4.18 c a g t t c t c c c t g a a g c t g a g c t c t g t g a c c g c c g c a g a cRT55 ____ ____________________________________________________________________________________
CDR 3T A V Y Y C A R H V P K W
VH4.18 a c g g e t g t g t a t t a c t g t g c g a g a c a t g t t c e t a a g t g gRT55 ____ _________________________________________________
E P G GRT55 g a g c c g g g c g g c
W F D P W G Q G T L V T V J H5 t g g t t c g a c c c c t g g g g c c a g g g a a c c c t g g t c a c c g t c
RT55 ____ ____________________________________________________________________________________
S SJH5 t c c t e a
RT55 ____ _______
D REGION ASSIGNMENT
C A T G T T C C T A A G T G G G A G C C G G G C G G CDHQ52 C _________ ADM2 _ A ______ AA
135
Figure 3.21
RT55 VH REGION COMPARED WITH VAR24 HEAVY CHAIN.
VH4.18RT55
VAR24
FR l CD RlQ L Q L Q E S G P G L V K P S E T L S L T C T V S G G S I S S S S Y Y W G
VH4.18RT55
VAR24
FR2W I R Q P P G K G L E W I G
CDR2S l Y Y S G S T Y Y N P S L K S
VH4.18RT55
VAR24
FR3R V T I S V D T S K N Q F S L K L S S V T A A D T A V Y Y C A R
C D R 3
RT55 H V P K W E P G G JH5VAR24 A A G Y Y D S S G Y Y R JH2
136
converted to valine. There were no replacement mutations within the Jk2
region that was utilised by RT55 but there was one silent mutation.
HSIGKIO was also the germline origin of the monoclonal RT129
which binds to ss/ds DNA ( Ravirajan etal 1992). RT129 V k region had
99% homology with HSIGKIO. These light chain sequences were
compared (fig 3.23). RT129 V k region had only one replacement mutation
within the framework one region where leucine was replaced by serine.
However, there were three replacement mutations within the JkI encoded
framework four region where tryptophan was converted to arginine,
glycine was replaced by phenylalanine and lysine was converted to
asparagine. The interesting feature in the comparison removal of a
positively charged residue within the RT129 light chain in comparison to
that in RT55. This is intriguing as RT129 has reactivity with dsDNA and
RT55 does not. One would expect that a dsDNA binder would retain
positive charge if not gain such amino acid residues via somatic mutation.
It is likely that in the case of RT129 the dsDNA binding ability is
determined by the heavy chain variable region.
The VH region of \AR24, an IgM monoclonal derived from a
polymyositis patient had 100% homology VH4.18 (fig 3.24). VAR24 had
reactivity with 56 kD RNP but not with DNA. VAR24 CDR3 region
consisted of twelve amino acids. Two of these amino acids were charged,
one positively charged arginine residue and and one negatively charged
aspartic acid. The CDR3 region appeared to be encoded by a D21/9 D
segment. An unmutated JH2 gene segment was utilised
The VX region of VAR24 was not amplified by the PGR primers
available and therefore is not shown. It is possible that the VX region of
\AR24 was encoded by a VX germline gene not yet characterised.
137
Figure 3.22 RT 55 KAPPA CHAIN VARIABLE REGION
F R 1V > D I \V>Q M T Q S P S L L S A
HSIGKIO g t c a t c t g g a t g a c c c a g t c t c c a t c c t t a e t c t c t g c aRT55 _ a c a _ ____________ _______ _______ _______ _______ _______ _______ _______
C DR 1S T G D R V T I S C R M S
HSIGKIO t c t a C a g g a g a c a g a g t c a c c a t c a g t t g t c g g a t g a g tRT55 ____ _______ _______ _______ _______ _______ _______ _______ _______ _______ _______ _______ _______
F R 2Q G I S S Y L A W Y Q Q K
HSIGKIO C a g g g c a t t a g c a g t t a t t t a g c c t g g t a t c a g c a a a a aRT55 ____ _______ _______ _______ _______ _______ _______ _______ _______ _______ _______ _______ _______
C DR 2P G K A P E L L I Y A A S
HSIGKIO c c a g g g a a a g c c c e t g a g e t c c t g a t c t a t g e t g c a t c cRT55 ____ _______ _______ _______ _______ _______ _______ _______ _______ _______ _______ _______ _______
F R 3T L Q S G V P S R F S G S
HSIGKIO a c t t t g c a a a g t g g g g t c c c a t e a a g g t t c a g t g g c a g tRT55 ____ _______ _______ _______ _______ _______ _______ _______ _______ _______ _______ _______ _______
G S G T D F T L T I S C LHSIGKIO g g a t c t g g g a c a g a t t t c a c t e t c a c c a t c a g t t g c c t g
RT55 ____ _______ _______ _______ _______ _______ _______ _______ _______ _______ _______ _______ _______
C DR 3Q S E D F A T Y Y C Q Q Y
HSIGKIO C a g t c t g a a g a t t t t g c a a c t t a t t a c t g t c a a c a g t a tRT55 ____ _______ _______ _______ _______ _______ _______ _______ _______ _______ _______ _______ _______
Y S F > S P> V HSIGKIO t a t a g t t t c c e t
RT55 _ C _ g t _
Y T F G Q G T K L E I K RJK2 t a c a c t t t t g g c c a g g g g a c c a a g c t g g a g a t c a a a c g t
RT55 _______ a
138
Figure 3.23
RT55 VK REGION COMPARED WITH RT129 UGHT CHAIN
FRlHSIGKIO V I WM T Q S P S L L S A S T G D R V T I S C
RT55 D _ Q _________________________________________RT129 S
CD RlR M S Q G I S S Y L A
HSIGKIORT55RT129
FR2 CDR2W Y Q Q K P G K A P E L L I Y A A S T L Q S
HSIGKIORT55
RT129
FR3G V P S R F S G S G S G T D F T L T I S C L Q S E D F A T Y Y C
HSIGKIORT55RT129
CDR3Q Q Y Y S F P
S V
JK2RT55
RT129
W T F G Q G T K L E I K R _ _ _ _ _
JKl
139
Figure 3.24 VAR 24 HEAVY CHAIN VARIABLE REGION
F R 1
VH4.18VAR24
Qc a g
L c t g
Qc a g
L c t g
Qc a g
Eg a g
St c g
G g g c
Pc c a
G g g a
L c t g
V g t g
K a a g
VH4.18VAR24
Pc e t
St c g
Eg a g
Ta c c
L c t g
St c c
Le t c
Ta c c
C t g c
Ta c t
V g t c
St c t
G g g t
VH4.18VAR24
G g g c
St c c
Ia t c
Sa g c
CDR 1 S
■ g tS
■ f tS
■ g tY
t a cY
t a cW
t f fG
f f
F R 2 W
t g gI
a t c
VH4.18VAR24
Re g o
Qc a g
Pc c c
Pc c a
G g g g
K a a g
G g g g
L c t g
Eg a g
W t g g
Ia t t
G g g g
CDR S
a g t
VH4.18VAR24
Ia t c
Y t a t
Y t a t
S a g t
G g g g
Sa g c
Ta c c
Yt a c
Yt a c
N a a c
Pc c g
S t c c
Le t c
VH4.18VAR24
K a a g
S a g t
F R 3 R
c g aV
g t cT
a c cI
a t aS
t c cV
g t aD
g a cT
a c gS
t c cK
a a gN
a a c
VH4.18VAR24
Qc a g
F t t c
St c c
L c t g
K a a g
L c t g
Sa g c
St c t
V g t g
Ta c c
A g c c
A g c a
D g a c
VH4.18VAR24
T a c g
A g c t
c
V g t g
Y t a t
Yt a c
C t g t
A g c g
R a g a
C DR 3 A
g c aA
g c gG
g g tY
t a cY
t a t
VAR24D
g ■ ts
a g tS
a g tG
g g tY
t a tY
t a cR
c g c
JH2VAR24
Yt a c
W t 8 g
Yt a c
F t t c
D g a t
Le t c
W t g g
G g g c
R c g t
G g g c
Ta c c
L c t g
V g t c
JH2VAR24
Ta c t
V g t c
St c c
St e a
D REGION ASSIGNMENT
VAR24 G C A G C G G G T T A C T A T G A T A G T A G T G G T T A T T A C C G C D21/9 T A T A
140
3.6 MONOCLONAL AUTOANTÏBQDIES ENCODED BY THE
VH71-2 GERMLINE GENE
RT129, an IgM polyreactive autoantibody was derived from the
splenocytes of a patient with active SLE. RT129 had reactivity with ss/ds
DNA and cardiolipin (Ravirajan etal 1992).
RT129 VH region had 95% homology to the germline gene VH71-2
(fig 3.25) (Lee etal 1987). There was one replacement mutation within
the framework one region changing valine to isoleucine. A serine residue
was replaced by a glycine residue in the CDRl region. Framework two
contained one replacement mutation converting proline to a positively
charged histidine. Asparagine was replaced by tyrosine in the CDR2
region. Two replacement mutations occurred in framework three.
Positively charged lysine was replaced by isoleucine and asparagine was
converted to serine. The CDR3 region was composed of eleven amino
acids including two positively charged amino acids arginine and lysine.
The CDR3 region appeared to be derived from the D elements DXP’4 and
DM2. The framework 4 region was composed of a JH5 gene which
contained two mutations, one gave rise to a replacement mutation where
histidine was replaced by asparagine.
Interestingly two other autoantibodies derived from this germline
gene are described in the literature and these autoantibodies are dsDNA
binders. The heavy chain variable regions of these monoclonals have
been compared to RT129 (fig 3.26). The sources of these monoclonals are
also of some interest. Two of these antibodies are of the IgM isotype,
BEG-2 was derived from fetal liver lymphocytes and binds to ss/ds DNA
(Watts etal 1991) and the IgG monoclonal 2A4 binds dsDNA and was
derived from myeloma bone marrow lymphocytes (Davidson etal 1990).
The mutations in framework one from glutamic acid to glutamine and from
valine to isoleucine are common to all of the monoclonals. It is perhaps
significant that glutamic acid is replaced by glutamine as this mutation
removes negative charge and glutamine has been sited as an important
amino acid residue in antibody/ DNA interactions (Seeman etal 1976).
141
Figure 3.25 RT 129 HEAVY CHAIN VARIABLE REGION
F R 1
HV71-2RT129
Qc a g
V g t g
Qc a g
L c t g
Qc a g
Eg a g
W t c g
G g g c
Pc c a
Gg g a
L c t g
V g t g
K a a g
HV71-2RT129
Pc e t
St c g
Eg a g
Ta c c
L c t g
St c c
Le t c
Ta c c
C t g c
Ta c t
V g t c
st c t
G g g I
HV71-2RT129
G g g c
St c c
V> I g t c a t
Sa g c
CDR 1 S
* g tG
g g tS > G ■ g t g
Yt a c
Yt a c
W t g g
S• g c
F R 2 W
t g gI
a t c
HV71-2RT129
R c g g
c
Qc a g
P > H c c c
a
Pc c a
G g g g
K a a g
Gg g a
c
L c t g
Eg a g
W t g g
Ia t t
G g g g
C DR 2 Y
t a tc
HV71-2RT129
1a t c
Y t a t
Yt a c
S a g t
G g g g
S a g c
Ta c c
N> Y a a ct _
Yt a c
N a a c
Pc c c
g
S t c c
Le t c
HV71-2RT129
K a a g
S ■ g t
F R 3 R
c g aV
g t c t
I Ta t a
St e a
V g t a
D g a c
K a c g
st c c
a
K> I a a g
t a
N> S a a c
g
HV71-2RT129
Qc a g
F t t c
St c c
L c t g
K a a g
L c t g
Sa g c
St c t
V g t g
Ta c c
t
A g c t
c
A g c g
D g a c
HV71-2RT129
T a c g
A g c c
V g I g
Y t a t
Yt a c
C t g t
A g c g
R a g a
CDR 3 G
g g a
L
t t a
L
t t a
R
c g a
F
t t t
RT129L
t * gE
g 0 gW
t g gS
t c cG
g g aK
a a a
JH5RT129
H> N c a c a
W t g g
F t t c
D g a c
Pc c c
W t g g
G g g c
Qc a g
G g g a
g
T a a c
L c t g
V g t c
Ta c c
V S SJH5 g t c t o c t e a
RT129
D REGION ASSIGNMENT
RT129 G G A T T A T T A C G A T T T T T G G A G T G G T C C G G A A A ADXP4 _ _ T A _____________________________________ TADM2 CC
142
RT129 has a unique replacement mutation with in framework two from
proline to histidine introducing a positive charge. However, BEG2 and
2A4 had common mutations in CDR2 from tyrosine to arginine and
tyrosine to threonine which did not occur in RT129. A unique amino acid
motif derived from replacement mutations occurs in 2A4 was not found in
either BEG-2 or RT129 where a stretch serine-threonine-asparagine are
converted to asparagine-isoleucine-lysine, thus introducing positive charge
which is not present in the IgM monoclonals. The CDR3 regions of these
monoclonals were quite different from each other. BEG-2 has only two
amino acids in this region (glutamine and serine) suggesting that unlike
other anti-DNA antibodies described, this particular region of the antibody
may not be making a significant contribution to DNA binding. RT129
CDR3 region was encoded by eleven amino acids which is significantly
shorter than that of 2A4 which consists of seventeen amino acids. 2A4
has two arginines and a lysine within this region and RT129 has one
arginine and one lysine. As 2A4 is an IgG anti-DNA antibody which is
the isotype most likely to be associated with pathogenesis, it is likely to
have a higher affinity for DNA than either RT129 or BEG-2. However the
presence of positively charged residues within the CDR3 region of RT129
and the fact that this monoclonal was isolated from an SLE patient with
active disease could suggest that perhaps RT129 has a higher affinity for
DNA than BEG-2 which was derived from a fetal source. It is well
established that polyreactive autoantibodies of the IgM isotype which occur
in early ontogeny are low affinity binders to autoantigens. 2A4 carries the
heavy chain associated idiotype F4. F4 was found to be highly associated
with anti-DNA antibodies of the IgG isotype (Davidson etal 1990). F4
anti-idiotype recognises heavy chains on anti-DNA antibodies in 60% of
SLE patients (Davidson etal 1990). RT129 has not been tested for F4
idiotype expression.
The light chain variable region of RT129 had 99% homology to the
germline gene HSIGKIO (fig 3.27) (Jaenichen etal 1984). RT129 V k
region had only one replacement mutation within the framework one
143
Figure 3.26
RT129 VH REGION COMPARED TO BEG-2 AND 2A4 HEAVY CHAINS
FRl CD RlVH71-2 Q V Q L Q E S G P G L V K P S E T L S L T C T V S G G S V S S G S Y Y W SBEG-2 _______________________________ Q ____ ____________________I _RT129 Q _________I _ _ ] 5 ] _____
2A4 Q I N
FR2VH71-2 W I R Q P P G K G L E W I GBEG-2 __________G ________________RT129 H ___________________
2A4 A G
CDR2Y I Y Y S G S T N Y N P S L K SR _ _ T _________________________ ~ ” _ Y _________ I I IR D T ~ " n T k
VH71-2BEG-2RT129
2A4
FR3R V T I S V D T S K N Q F S L K L S S V T A A D T A V Y Y C A R
I S
CDR3 BEG-2 Q SRT129 G L L R F L E W S G K
2A4 D S I M G E I A R G P R A K G Q G
JH5 H W F D P W G Q G . T L T V S SBEG-2 N ________________ . _____________RT129 N ________________ . _____________
2A4 Y G M V T V JH6
144
region where leucine was replaced by serine. However, there were three
replacement mutations within the JkI encoded framework four region
where tryptophan was converted to arginine, glycine was replaced by
phenylalanine and arginine was replaced by aspartic acid.
In conclusion, the germline gene VH71-2 could have an association
with anti-dsDNA binding
145
Figure 3.27 RT 129 KAPPA CHAIN VARIABLE REGION
F R 1V I W M T Q S P S L > S L S A
HSIGKIO g t c a t c t g g a t g a c c c a g t c t c c a t c c t t a e t c t c t g c aRT129 _ c c _ _ g ___________
C DR 1S T G D R V T I S C R M S
HSIGKIO t c t a c a g g a g a c a g a g t c a c c a t c a g t t g t c g g a t g a g tRT129
Q G I S S Y L A W Y Q Q K HSIGKIO C a g g g c a t t a g c a g t t a t t t a g c c t g g t a t c a g c a a a a a
RT129
CDR 2P G K A P E L L I Y A A S
HSIGKIO c c a g g g a a a g c c c e t g a g e t c c t g a t c t a t g e t g c a t c cRT129 _ ________________________________________________________________________________ __________
F R 3T L Q S G V P S R F S G S
HSIGKIO a c t t t g c a a a g t g g g g t c c c a t e a a g g t t c a g t g g c a g tRT129_______ _______ _______ _______ _______ _______ _______ _______ _______ _______ _______ _______ _______
G S G T D F T L T I S C LHSIGKIO g g a t c t g g g a c a g a t t t c a c t e t c a c c a t c a g t t g c c t g
RT129_______ _______ _______ _______ _______ _______ _______ _______ _______ _______ _______ _______ _______
CDR 3Q S E D F A T Y Y C Q Q Y
HSIGKIO c a g t c t g a a g a t t t t g c a a c t t a t t a c t g t c a a c a g t a tRT129_______ _______ _______ _______ _______ _______ _______ _______ _______ _______ _______ _______ _______
Y S F P HSIGKIO t a t a g t t t c c e t
RT129_______ _______ _______ _______
^ > R T F G> F Q G T K V E I K R> DJKl t g g a c g t t c g g c c a a g g g a c c a a g g t g g a a a t c a a a c g t
RT129 C _ ____ _______ t t __________ g a _
146
3.7 MONOCLONAL ANTI-DNA ANTIBODIES ENCODED
BY THE 56P1 GERMI.INE GENE
WRI176 VH region, an IgM anti-dsDNA monoclonal derived from
an SLE patient had 97.3% homology with the germline gene 56P1 (fig
3.28). This VH3 member was derived from a 130 day old fetus
(Schroeder etal 1987). There are three replacement mutations away from
the germline sequence 56P1 resulting in glutamine replacing glutamic acid
in framework one, glycine replacing alanine in CDRl and tryptophan
replacing serine. The replacement of glutamic acid with glutamine in
framework one resulted in the removal of negative charge and the
generation of an amino acid which could improve affinity to DNA. The
CDR3 region of WRI176 which contained three arginine residues, was
derived from contributions of DKl and DXP’l D genes. The presence of
arginines in VH CDR3 regions was described earlier in relation to anti-
DNA antibodies encoded by the VH4.21 germline gene. An unmutated
JH3 gene segment was utilised.
WRI176 VH region was compared and three other anti-DNA
antibodies encoded by 56P1 germline gene (fig 3.29). These were the
IgM III3R, the IgG l-2a (Manheimer-Lory etal 1991) and the IgG 19-E7
(Winkler etal 1992). This comparison revealed that all these antibodies
have at least one positively charged amino acid residue within the CDR3
region, although only WRI176 had arginine residues in the couplet motif
described in the VH4.21 encoded anti-DNA monoclonals. There did not
seem to be any replacement mutations which were common to two or more
of the 56P1 encoded heavy chain variable regions. There appears to be
much evidence to suggest the importance of CDR3 arginines in
antibody/DNA interactions. Interestingly, the IgM III3R also utilised the
DXP’l D gene in the generation of its CDR3 region. The 56P1 encoded
monoclonal anti-DNA antibodies compared used diverse JH genes.
The light chain variable region of WRI176 had 98% homology with
the V k 3a member KV328 (fig 3.30). Two replacement mutations
147
Figure 3.28 W RI176 HEAVY CHAIN VARIABLE REGION
F R lL V E > Q S G G G V V Q P G R
56P1 c t g g t g g a g t c t g g g g g a g g c g t g g t c c a g c e t g g g a g gW R I 1 7 6 _______ c _ _
S L R L S C A A S G F T F S6P1 t c c c t g a g a e t c t c c t g t g c a g c c t c t g g a t t c a c c t t c
W R I 1 7 6 _______ _ g
C D R l F R 2S S Y A > G M H W V R Q A P G
S6P1 a g t a g c t a t g e t a t g c a c t g g g t c c g c c a g g e t c c a g g cW R I 1 7 6 ____________ g c ________ _______ _______ _______ _______ _______ _______ _______
C D R 2K G L E W V A V I S > W Y D G
56P1 a a g g g g c t g g a g t g g g t g g c a g t t a t a t e a t a t g a t g g aW R I 1 7 6 ___________________________________ g g __________ _________
F R 3S N K Y Y A D S V K G R F
56P1 a g c a a t a a a t a c t a c g c a g a c t c c g t g a a g g g c c g a t t cWRI176 _ _ t ____________ _ t ________ _______ _______ _______ _______ _______ _______
T I S R D N S K N T L Y L56P1 a c c a t c t c c a g a g a c a a t t c c a a g a a c a c g c t g t a t c t g
WRI 1 7 6 ____ _______ _______ _______ _______ _______ _______ _______ _______ _______ _______ _______ _______
Q M N S L R A E D T A V Y 56P1 c a a a t g a a c a g c c t g a g a g e t g a g g a c a c g g e t g t g t a t
W R I 1 7 6 ____ _ _ c _______ _______ _______ _______ _______ _______
C D R 3Y C A R E R R S A M A P R
56P1 t a c t g t g c g a g a W R I 1 7 6 ____ _______ _______ g a a c g g c g g t e a g e t a t g g e e c c g a g g
A F D I W G Q G T M V T VJ H3 g e t t t t g a t a t c t g g g g c c a a g g g a c a a t g g t c a c c g t c
W R I 1 7 6 ____ _______ _______ _______ _______ _______ _______ _______ _______ _______ _______ _______
S SJH3 t c t t e a
WRI176
D REGION ASSIGNMENT
DKl C _ G G C _ A T ______WRI176 G A A C G G C G G T C A G C T A T G G C C C C G A G GD X Pl GT A T T _A ___________ T T _ G_ G G ADK4 G T G A T A T T A
148
Figure 3.29
WRI176 HEAVY CHAIN VARIABLE REGION COMPARED WITH 56P1 ENCODED ANTI-DNA HEAVY CHAINS
56P1 WRI176 _ _ Q
HI-3R 19-E7 l-2a
FR l CD RlL V E S G G G V V Q P G R S L R L S C A A S G F T F S S Y A M H
_ _ G _ __ _ L _______ G _ _ R ] S
_ g I d” S P H
FR2 CDR256P1 W V R Q A P G K G L E W V A V I S Y D G S N K Y Y A D S V K G
W R I 1 7 6 ____________________________ _ _ W _________________m-3R ____________________________ N _ K Q E V _________19-E7 _____________________________ _ _ W ________________ R _____l-2a _ _ F _ _ L E _ _ S Q " _____________
56P1 WRI 176
HI-3R 19-E7 l-2a
FR3R F T I S R D N S K N T L Y L Q M N S L R A E D T A V Y Y C A R
S A T S
CDR3WRI176 E R R S A M A P R A F D I W G Q G T M V T V T S S JH3
m-3R G R M W E R W F G E S P P F D Y W G Q G T L V T V S S JH419-E7 E I N C Y G G F H Y Y N Y G M D V W G Q G T T V T L S S JH6l-2a P G K V E K W E L P F D Y W G Q G T L V T V S S JH4
149
occurred in framework one where alanine was replaced by serine and
threonine was replaced by serine. Proline was replaced by arginine in
CDR3. An unmutated Jk5 gene segment was utilised.
All the 56P1 encoded monoclonal anti-DNA antibodies utilised V k
encoded light chains. The IgG 19-E7 used a V k3 region as did WRI176 in
the generation of its light chain.
In conclusion, there may be some evidence that some 56P1 encoded
anti-DNA antibodies contain positively charged amino acids in the VH
CDR3 regions in similar configuration to that seen in VH4.21 encoded
anti-DNA antibodies
150
Figure 330 WRI176 KAPPA CHAIN VARIABLE REGION
F R 1E l V M T Q S F A > S T > S L S V
KV328 g a a a t a g t g a t g a c g c a g t c t c c a g c c a c c c t g t c t g t gW R I 1 7 6 ____ _ _ c t _ _ t _ _
CDR 1S P G B R A T L S C R A S
KV328 t c t c c a g g g g a a a g a g c c a c c e t c t c c t g c m g g g c c m g tW R I 1 7 6 ____ ________________________________________________ ____________________________
F R 2Q S V S S N L A W Y Q Q K
KV328 c a g a g t g t t a g c a g c a a c t t a g c c t g g t a c c a g c a g a a aW R I 1 7 6 ____ _______
C D R 2P G Q A P R L L I Y G A S
KV328 c e t g g c c a g g e t c c c a g g e t c e t c a t c t a t g g t g c a t c cWRI176
F R 3T R A T G I P A R F S G S
KV328 a c c a g g g c c a c t g g c a t c c c a g c c a g g t t c a g t g g c a g tW R I 1 7 6 ____ ________________________________________________ ____________________________
G S G T E F T L T I S S LKV328 g g g t c t g g g a c a g a g t t c a c t e t c a c c a t c a g c a g c c t gWR I 1 7 6 ____ _____________________________________________________________________________
CDR 3Q S E D F A V Y Y C Q Q Y
KV328 c a g t c t g a a g a t t t t g c a g t t t a t t a c t g t c a g c a g t a tWR I 1 7 6 ____ _____________________________________________________________________________
N N W P P > RKV328 a a t a a c t g g c e t c c gWRI 1 7 6 ____ _____________________c g c
T F G Q G T R L E I K RJK5 a c c t t c g g c c a a g g g a c a c g a c t g g a g a t t a a a c g a
WRI176______________________________________________________________ ______________________
151
3.8 MONOCLONAT. AUTOANTIBODIES ENCODED BY THE
VH26 GERMLINE GENE
The VH3 germline gene VH26 is known to be closely associated
with the 16/6 idiotype which is elevated in the sera of patients with active
SLE as compared to those with inactive disease. Three monoclonal
autoantibodies in the panel available for study were VH26 encoded.
The anti histone monoclonal RT6 VH region had 98% homology
with the VH3 germline gene VH26 at the nucleotide level and 96% at the
amino acid level (fig 3.31) (Schroeder eî al 1987). There are four
replacement mutations. Arginine was replaced by threonine and serine
was converted to proline in framework one, lysine was replaced by
arginine in framework two and threonine was converted to alanine in
CDR2. The CDR3 region was encoded ten amino acids and appeared to
be derived from a D-D fusion between D1 and DKl D elements. The JH
region was encoded by an unmutated JH4. This monoclonal had no DNA
binding reactivity.
WRI 170 is one of two anti-histone antibodies reported by Tuaillon
et al (1994). This monoclonal utilised a VH3 germline gene V3-9
(Matsuda etal 1993). The CDR3 region of WRI170 was generated from
contributions from D21/9 and DIR2 . A JHl gene segment was utilised.
The VH regions of RT6 and WRI170 were compared in figure
3.32. The VH region of WRI170 appeared to be more cationic than the
VH region of RT6. The cationicity of the variable regions dependent on
the frequency of positively charged amino acid residues present. Two
arginine residues were present in the framework one region of WRI170
which did not occur in RT6. The presence of negatively charged amino
acid residues in the variable regions of murine anti-histone antibodies has
been observed. In particular a concentration of such residues were
observed in the VH CDR2 region. Neither WRI 170 nor RT6 had a
preponderance of negatively charged residues in CDR2. The CDR3 region
of WRI 170 which contained two aspartic acid residues and one glutamic
acid residue was more anionic than RT6
152
Figure 331 RT 6 HEAVY CHAIN VARIABLE REGION
F R 1
VH26RT6
Eg a g
V g t g
Qc a g
L c t g
t
L t t g
Eg a g
S > P t c t c
St c t
G g g g
G g g a
G g g c
L t t g
V g t a
VH26RT6
Qc a g
Pc e t
G g g g
G g g g
St c c
L c t g
c
R > T a g a
c
Le t c
Ct c c
C t g t
A g c a
Ag c c
St c t
VH26RT6
G g g a
F t t c
Ta c c
F t t t
S > T a g c
c
CDR 1 S
» gY
t a tA
g c cM
• t gS
a g c
F R 2 W
t g gV
g t cR
c g c
VH26RT6
Qc a g
A g c t
Pc c a
G g g g
K> R a a g
g
G g g g
L c t g
Eg a g
W t g g
V g t c
St e a
C DR 2 A
g c tI
a t t
VH26RT6
S a g t
G g g t
S a g t
G g g t
G g g t
S a g c
T > A a c ag
Yt a c
Yt a c
A g c a
Dg a c
St c c
V g t g
VH26RT6
K a a g
G g g c
F R 3 R
c g gF
t t cT
a c cI
a t cS
t c cR
a g aD
g a cN
a a tS
t c cK
a a gN
a a c
VH26RT6
Ta c g
L c t g
Y t a t
L c t g
Qc a a
M a t g
N a a c
Sa g c
L c t g
R a g a
Ag c c
Eg a g
D g a c
VH26RT6
Ta c g
V
Ag c c
A
V g t a
G
Y t a t
T
Yt a c
A
Ct g t
A g c g
R a a a
CDR 3 V
g t c
G
g g t
G
g g t
I
a t a
A
g c a
RT6 g t g g c t g g t a c g g c t
JH4RT6
D g a c
Yt a c
W t g g
G g g c
Qc a g
G g g »
Ta c c
L c t g
V g t c
Ta c c
V g t c
St c c
St e a
D REGION ASSIGNMENT
RT6 G T C G G T G G T A T A G C A G T G G C T G G T A C G G C T G A C T ACD1 T A T __________G ______ T A _ A
DKl T A _____A _____
153
Figure 3.32
RT6 HEAVY CHAIN VARIABLE REGION COMPARED WITH WRI170 VH REGION
F R I C D R 1RT6 E V Q L L E P G G G L V Q P G G S L T L C C A A S G F T F T S Y A M S
WRI 170 V M V S R R S N D P L H
F R 2 C D R 2RT6 W V R Q A P G R G L E W V S A I S G S G G S A Y Y A D S V K G
WRI170 P K S G W N S I G
F R 3RT6 R F T I S R D N S K N T L Y L Q M N S L R A E D T A V Y Y C A K
WRI 170 A S L
C D R 3RT6 V G G I A V A G T A D Y WG Q G T L V T V S S JH4
WRI 170 G P P G Y Y D S S E P S D E Y F Q JHl
154
Figure 333 RT 6 LIGHT CHAIN VARIABLE REGION
F R 1Q S A L T Q P A S V S G S
DPLll c a g t c t g c c c t g a c t c a g c e t g c c t c c g t g t c t g g g t c tRT6
C D R lP G Q S I X I S C T G T S
DPLll c e t g g a c a g t c g a t c a c c a t c t c c t g c m e t g g m a c c m g eRT6
F R2S D V G G Y N y V S W Y Q
DPLll m g t g a c g t t g g t g g t t m t m m c t m t g t c t c c t g g t a c c a aRT6 ____ ____________________________
C D R 2Q H P G K A P K L M I Y E > D
DPLl 1 c a g c a c c c a g g c a a a g c c c c c a a a e t c a t g a t t t a t g m gRT6 _ t
F R 3V S N R P S G V S N R F S
DPLll g t c m g t m m t c g g c c c t c m g g g g t t t c t a a t c g c t t c t c tRT6 ____ _______________________________ g ____ ___________________________________
G S K S G N T A S L T I S DPLll g g c t c c a a g t c t g g c a a c a c g g c c t c c c t g a c a a t c t c tRT6 _______________________________________ _ _ c
C DR 3G L Q A E D E A D Y Y C S
DPLll g g g e t c c a g g e t g a g g a c g a g g e t g a t t a t t a c t g c m g eRT6 ____ ____________________________________________________________________________________
S Y T S S S T L G W DPLll t e a t a t mem a g e a g e a g e a c t e t c • * *RT6 ------------------------------------------- o f t g g g t g g
F G G G T K > Q L T V L GJL2 t t c g g c g g a g g g a c c a a g c t g a c c g t c c t a g g tRT6 c c c c
155
Figure 334
RT6 VARIABLE LIGHT CHAIN REGION COMPARED WITH WRI170 LIGHT CHAIN
F R I C D R lD P L ll Q S A L T Q P A S V S G S P G Q S I T I S C T G T S S D V G G Y N Y V S
RT6WRI170 “ P P K P Î
F R 2 C D R 2 F R 3D P L ll > \ / Y Q Q H P G K A P K L MI Y E V S N R P S G V S N R F S G S K S
RT6 _______________________________ D ___________WRI170 H T I R G M Q R ’ P D
C D R 3D P L ll G N T A S L T I S G L Q A E D E A D Y Y C S S Y T S S S T L .
RT6 ___________________________________________ G W J X 2WRI170 R N S J X 2
156
RT6 Vk region had 99% homology to the germline gene DPLll
(fig 3.33) (Williams & Winter 1993). There was one replacement mutation
in CDR2 where glutamic acid was replaced by aspartic acid. Two amino
acids, glycine and tryptophan arose by N addition at the VWX junction. A
JX2 gene segment was utilised with one replacement mutation converting
lysine to glutamine.
RT6 VX region was compared with the VX light chain of WRI 170
(fig 3.34). It was intriguing to see that as with the comparison between the
heavy chains of these antibodies, WRI 170 light chain was more cationic
than RT6 light chain. The CDRl region of WRI 170 contained one lysine
residue in instead of the asparagine residue seen in RT6. WRI 170 had two
arginines in CDR2 where RT6 had only one. In addition RT6 CDR2
contained a negatively charged aspartic acid residue not present in
WRI170. An arginine residue was also present in the framework three
region of WRI170 in place of glutamine in RT6.
The nucleotide sequence of the heavy chain of the IgG anti-dsDNA
antibody B3 revealed that it had 93.5% homology with the germline gene
segment VH26 (fig 3.35) (Chen et al 1988). The D region could be
aligned to the D gene segments DXP’l and DHQ52. The CDR3 region
contained six amino acids including one asparagine residue. There were no
charged residues present in the CDR3 region. A summary of the
replacement mutations seen in B3 V regions is shown in table 3.3. There
was a replacement mutation in CDR2 from serine to arginine which
resulted in a gain in positive charge. Lysine was replaced by glutamine at
position 65 where a positive charge was lost. There were nine other
replacement mutations within the VH region none of which lead to any
change in charge. The framework four region consisted of an unmutated
JH4 gene segment.
The nucleotide sequence of the Vk region of the IgG anti-dsDNA
antibody B3 indicated that it had 96% homology with the germline gene
DPLll (fig 3.36) (Williams & Winter 1993). There was a total of seven
replacement mutations away from the germline sequence, these mutations
157
Figure 335 B3 HEAVY CHAIN VARIABLE REGION
F R 1E V Q L L > V E S G G G L V Q
VH26B3
g a g g t g c a g c t g t t g g
g a g t c t g g g g g a g g c t t g g t a c a g
P G G S L R L S C A A> S S GVH26
B3c e t g g g g g g t c c c t g
ca g a e t c t c c t g t g c a g c c
a g -t c t g g a
F T F S > ICDR 1 S > T Y A M S
F R 2W V R Q
VH26B3
t t c a c c t t t a g c t
a g c c
t a t g c c • t g a g c t g g g t c c g c c a g
A P G K G L E W V SCDR 2 A> T I S
VH26B3
g c t c c a g g g a a g g g g c t g g a g t g g g t c t e a g c t a
a t t a g t
G S > R G G> S S T Y Y A D S V K> QVH26
B3g g t a g t
cg g t g g t
aa g c a c a t a c t a c g c a
tg • c t c c g t g • • g
c
GF R 3
R F T I > L S R D N S K N> S TVH26
B3g g c c g g t t c a c c a t c
ct c c a g a g a c a a t t c c a a g a a c
g -a c g
c
L Y L Q M N S > T L R A E D TVH26
B3c t g t a t c t g
cc a a
ga t g a a c a c g
_ g cc t g a g a g c c g a g g a c a c g
A V Y Y C A KCDR 3
P N V G S GVH26
B3g c c g t a
ct a t t a c t g t g c g a a a c e t a a t g t g g g c ■ g t g g c
W S F D S W G Q G T L V TJH4B3
t g g t c c t t t g a c t c c t g g g g c c a g g g a a c c c t g g t c a c c
V S SJH4 g t c t c c t e aB3 ____ ______________
D REGION ASSIGNMENT
C C I A A T G T G G G C A G T G G CDXFl T C ____ G _____DHQ52 C T G _ _ A
158
Figure 336 B3 LIGHT CHAIN VARIABLE REGION
F R 1Q S A L T Q P A S V S G S
DPLll c a g t c t g c c c t g a c t c a g c e t g c c t c c g t g t c t g g g t c tB3
C DR 1P G Q S I T I S C T G T S > R
DPLll c e t g g a c a g t c g a t c a c c a t c t c c t g c a c t g g a a c c a g cB3 _ a
F R 2S > R D y G G Y N Y > F V S W Y Q
DPLll a g t g a c g t t g g t g g t t a t a a c t a t g t c t c c t g g t a c c a aB3 a c _______________ t g
CDR 2Q H P G K A P K L M I Y E
DPLl l c a g c a c c c a g g c a a a g c c c c c a a a e t c a t g a t t t a t g a gB3 _ _ a ____ ______________________________________________________________________
F R 3V S N > H R P S G V S N > T R F S
DPLl l g t c a g t a a t c g g c c c t e a g g g g t t t c t a a t c g c t t c t c tB3 ____ _______c __________________________ _ c _ ___________ _________
G> A S K S G N > S T A S L T I SDPLl l g g c t c c a a g t c t g g c a a c a c g g c c t c c c t g ac c a t c t c t
B3 _ c __________________ g _________ _________ _________ _________ _________ _________ _________ _________
CDR 3G L Q A E D E A D Y Y C S
DPLl l g g g e t c c a g g e t g a g g a c g a g g e t g a t t a t t a c t g c a g cB3 ______________________________ _ _ c ___________ _________ _________
S Y T S S S > T T RDPLl l t e a t a t a e a a g c a g c a g c a c t c . .
B3 ---------------- . . . _ c _ _ g t
V V F G G G T K L T V L GJL2 g t g g t a t t c g g c g g a g g g a c c a a g c t g a c c g t c c t a g g tB3 _ _ ____ ___________________________________
159
Figure 337
B3 LIGHT CHAIN COMPARED TO RT6 AND PVll LIGHT CHAINS
F R 1 C D R lD P L ll Q S A L T Q P A S V S G S P G Q S I T I S C T G T S S D V G G Y N Y V S
B3 ______________________________________________________ R R F _ _RT6 _____________________________________________ ] _______ ___________________
P V ll
F R 2 C D R 2 F R 3D P L ll W Y Q Q H P G K A P K L MI Y E V S N R P S G V S N R F S G S K S
B3 _________________________________ H _____ ________ T ______ A ______RT6 _______________________________ D ___________
P V ll D ” “
C D R 3D P L ll G N T A S L T I S G L Q A E D E A D Y Y C S S Y T S S S T L . .
B3 _ S __________________________________________ . _ _ T _ R . . J k 2RT6 ___________________________________________ G W J X 2
P V ll S J k 2
160
TabIe3J
REPLACEMENT MUTATIONS IN B3 V REGIONS
VH VL
VH26 B3 D P L ll B3(Germline) (Germline)
F R l Leu(NP) > Val(NP) C D R l Ser(P) > Arg(+ve)Ala(NP) > Ser(P) Ser(P) > Arg(+ve)Ser(P) > Iso(NP) Tyr(P) > Phe(NP)
C D R l Ser(P) > Thr(P) C D R 2 Asn(P) > His(+ve)
C D R 2 Ala(NP) > Thr(P) F R 3 Asn(P) > Thr(P)Ser(P) > Arg(+ve) Gly(P) > Ala(NP)Gly(P) > Ser(P) Asn(P) > Ser(P)
Lys(+ve) > Gln(P)C D R 3 Ser(P) > Thr(P)
F R 3 Iso(NP) > Leu(NP)Asn(P) > Ser(P)Ser(P) > Thr(P)
(NP) = Non polar amino acid (?) = Polar amino acid (+ve) = Positive charge (-ve) = Negative charge
161
are summarised in table 3.2 . The most significant of these was the
conversion of two serine residues to two arginine residues in CDRl
resulting in the gain of positive charge. An asparagine residue was
replaced by positively charged histidine in CDR2. A threonine residue
appears to have been deleted in CDR3. An arginine residue was encoded
by N addition at the VX-JX junction. An unmutated 3X2 gene segment was
utilised.
A polyclonal anti-idiotype was raised using B3. The idiotype was
found to be encoded on the light chain of this antibody (Ehrenstein et al
1994). B3 was expressed on IgG antibodies in the serum of 20% of SLE
patients.
The B3 light chain variable region was compared to a bank of eight
8.12 Id positive and two 8.12 Id negative antibodies, all of which utilise
the Vxll gene family. B3 had the highest homology with the monoclonal
antibody PV ll (fig 3.37). The pathogenic 8.12 Id has been found in the
kidney lesions of SLE patients ( Paul et al 1992). The PV ll IgM
monoclonal was 8.12 positive, B3Id negative, and did not not bind to
DNA. Comparing the amino acid sequence of the light chain of B3 and
PV ll there were four mutations to positively charged amino acids (three
arginines) in B3, including a couplet of arginines are present in CDRl
which were not present in the PVll. The B3 light chain sequence was also
compared to the light chain of RT6, an anti-histone antibody which was
also B3Id negative and did not bind DNA. The arginine couplet which
occurred in the CDRl region of B3 was not present in RT6. It is therefore
possible that these residues could not only be responsible for enhancing the
DNA reactivity of B3 but may also define the B3 idiotype.
The IgG monoclonal UK4 bound to cardiolipin and not DNA. The
VH region of UK4 had 93% homology with the germline gene VH26 (fig
3.38). A summary of the replacement mutations in the variable regions of
UK4 are shown in table 3.4. The most significant replacement mutations
occurred in CDRl where serine was replaced by positively charged
histidine and in CDR2 where glycine was replaced by arginine. A
162
Figure 338 UK4 HEAVY CHAIN VARIABLE REGION
UK4
UK4
UK4
UK4
UK4
UK4
F R 1E
g a gV
g t gQ
c a gL
c t gL > V t t g g
Eg a g
St c t
G g g g
G g g a
G g g c
L t t g
a
V g t a
t
Qc a g
Pc e t
8
G g g g
G g g g
St c c
L c t g
R a g a
Le t c
St c c
Ct g t
A> V g c a
t
Ag c c
S t c t
G g g a
F t t c
Ta c c
F t t t
Sa g c
CDR I S > G a g g _
Y t a t
A> w g c c t g g
M a t g
S > H a g cc a c
F R 2 W
t g gV
g t cR
c g cQ
c a g
A> V g c t
t
Pc c a
G g g g
K a a g
G g g g
L c t g
E > V g a g
t
W t g g
V g t c
St e a
CDR 2 A
g c tI
a t tS > N a g t
a
G g g t
S a g t
G g g t- - g
G> R g g t a a
S a g c
T a c a
Y > St a ca g
Yt a c
A g c a
D g a c
St c c
V g t g
K a a g
G g g c
F R 3 R
c g g a
F t t c
Ta c c
Ia t c
St c c
R a g a
D g a c
N a a t
St c c
K a a g
N a a c
Ta c g
L c t g
Y t a t
L c t g
Qc a a
M a t g
N a a c
Sa c g
L c t g
R a g a
Ag c c
Eg a g
D g a c
Ta c g
Ag c c
V g t a
Y t a t
Yt a c
C t g t
A g c g
K> R a a a
g
CDR 3 D
g a t
L
c t t
G
g g a
A
g c a
V
g t g
T
a c t
A g c g
Yt a c
Yt a c
F t t t
D g a c
Yt a c
W t g g
G g g c
Qc a g
G g g a
T> V a c c g _
L c t g
V g t c
Ta c c
V g t c
S S JH4 t c c t e aUK4 _ _ _ _
D ASSIGNMENT
G A T C T T G G A G C A G T G A C T G C G TDXP4 _ T __________DAIrev _ T _____T _______
163
Figure 3 J9
UK4VH26
VARIABLE HEAVY CHAIN COMPARED WITH ENCODED MONOCLONAL AUTO ANTIBODIES
F R l C D R lVH26 E V Q L L E S G G G L V Q P G G S L R L C C A A S G F T F S S Y A M SU K 4 ________V ____________________________________ V ______________ G _ W _ HB3 ________V _____________________________________ S I T _______
RT6 P T T
VH26UK4B3
RT6
F R 2W V R Q A P G K G L E \ \ V S
V V
C D R 2A I S G S G G S T Y Y A D S V K G_ _ N R _ _ S _____________T R _ S ___________
AQ_
VH26UK4B3
RT6
F R 3R F T I S R D N S K N T L Y L Q M N S L R A E D T A V Y Y C A K__________________________________________________________________R
L S T
C D R 3
UK4 D L G A V T A Y Y F D YAVGQG V L V T V S S JH4B3 P N V G S G \ V S F D S W G Q G T L V T V S S JH4
RT6 V G G I A V A G T A D Y W G Q G T L V T V S S JH4
164
negatively charged glutamic acid residue was replaced by valine in
framework two resulting in the removal of negative charge. Interestingly,
five replacement mutations in framework regions including one in the JH4
region of UK4 generated valine residues. Valine, a nonpolar amino acid
probably does not interact with antigen in these framework positions but
could be important in generating a suitable tertiary structure for interaction
with cardiolipin. The VH CDR3 region of UK4 contained eight amino
acids including one negatively charged aspartic acid residue. The D
segments DXP’4 and DAI appeared to contribute to the generation of UK4
VH CDR3 region. A JH4 gene segment was utilised with one replacement
mutation converting threonine to valine. The distribution of replacement
mutations in UK4 VH region did not show evidence of an antigen driven
mechanism.
The VH26 encoded antibodies RT6 and B3 were compared with the
VH region of UK4 in fig 3.39. It was interesting to note that the
replacement mutations to valine in the framework regions of UK4 did not
occur in either B3 or RT6. A mutation from serine to histidine occurred in
the CDRl region of UK4 which did not occur in B3. Both UK4 and B3
had mutations to arginine within the CDR2 region although they were
located at different positions. The CDR3 regions of UK4, B3, and RT6
were quite diverse with B3 CDR3 containing an asparagine residue not
seen in UK4 or RT6. UK4 CDR3 contained an aspartic acid residue, the
only negatively charged residue present in the CDR3 regions of these
antibodies.
The VX region of UK4 had 95% homology to the VXII germline
gene DPLll (fig 3.40) (Williams & Winter 1993). Eleven replacement
mutations away from DPLll occurred in UK4 VX, region and are
summarised in Table 3.3. The majority of the replacement mutations did
not yield a change in charge and also did not give rise to valine residues as
seen in the VH region of UK4. The most notable mutations occurred in
CDRl where serine was replaced by asparagine and in framework three
where lysine was converted to glutamic acid. Three replacement mutations
occurred within the CDR3 region of UK4 where three adjacent serine
165
Figure 3.40 UK4 LIGHT CHAIN VARIABLE REGION
F R 1S A L T Q P A S V S G S F
DPLl l t c t g c c c t g a c t c a g c e t g c c t c c g t g t c t g g g t c t c e t UK4
UK4
UK4
UK4
UK4
UK4
UK4
UK4
G g g a
Qc a g
St c g
Ia t c
Ta c c
Ia t c
St c c
C t g c
C DR 1 T > P a c tc __
G g g *
Ta c c
Sg
S > N • g t
a
D g • c
Vg t t
G g g t
G> S g g t t c _
Y t a t
N a a c
Y t a t
V g t c
St c c
F R 2 W
t g gY
t a cQ
c a aQ > L c a g _ t a
H c a c
Pc c a
G g g c
a
K> E a a a g
Ag c c
Pc c c
K a a a
Le t c
M a t g
Ia t t
Y t a t
CDR 2 E > Dg « g
t
V g t c
S > T » g t_ c c
Nm a t
Rc g g
Pc c c
St e a
t
F R 3 G
g g gV
g t tS
t c tN
a a tR
c g cF
t t cS
t c tG
g g c
St c c
K > E a a g g
St c t
G g g c
N a a c
Ta c g
Ag c c
St c c
L c t g
Ta c c
Ia t c
g
St c t
G g g g
Le t c
Qc a g
A g c t
Eg a g
D g a c
Eg a g
A g c t
D g « t
Y t a t
Yt a c
ct g c
CDR 3 S
• g cS
t e a
Y t a t
T a c a
S > N a g c
a
S > R a g c
g
S > G a g c g -
Ta c t
Le t c
V g t g
V g t a
F t t c
G g g c
G g g a
G g g g
Ta c c
K a a g
L c t g
Ta c c
V g t c
L c t a
G g g t
166
Figure 3.41
UK4 LIGHT CHAIN COMPARED TO B3 AND RT6 LIGHT CHAINS
F R l C D R lD P L ll Q S A L T Q P A S V S G S P G Q S I T I S C T G T S S D V G G Y N Y V SUK4 ______________________________________________ P N S _________B3 ________________________________________________ R R _________ F _ I
RT6
F R 2 C D R 2 F R 3D P L ll W Y Q Q H P G K A P K L MI Y E V S N R P S G V S N R F S G S K SUK4 L E _____________ D _ T _ _ _ E _B3 H _ ] T ______A ______
RT6 D
C D R 3D P L ll G N T A S L T I S G L Q A E D E A D Y Y C S S Y T S S S T L . .UK4 ______________________________________________ N R G _ _ J X 2B3 _ S _______________________ I _ _ - _ _ t I r . . J X 2
RT6 G W J X 2
167
residues were replaced by asparagine, arginine and glycine respectively.
An unmutated JX2 gene segment was utilised.
UK4 light chain was also compared with B3 and RT6 VL regions as
these monoclonals appear to be generated from the same DPLll germline
gene (fig 3.41). Interestingly, the arginine couplet seen in the CDRl
region of B3 was not seen in UK4. This is of interest as UK4 did not bind
DNA. Aspartic acid residues in CDR2 occurred at the same position in
UK4 and RT6, however, the germline amino acid at this position was
glutamic acid therefore there was no change in charge. The CDR3 regions
of UK4 and B3 contained an arginine residue which did not occur in RT6.
Overall, B3 VX was more cationic than either UK4 or RT6.
It is intriguing that three monoclonals autoantibodies with diverse
binding specificities were encoded by the same VH and Vk germline genes.
The subtle differences between these antibodies are likely to have a vast
impact on their binding specificities.
Table 3.4
REPLACEMENT MUTATIONS IN UK4 V REGIONS
168
VH VL
VH26 UK4 D P L ll UK4(Germline) (Germline)
F R l Leu(NP) > Val(NP) C D R l Thr(P) > Pro(NP)Ala(NP) > Val(NP) Ser(P) > Asn(P)
Gly(P) > Asn(P)C D R l Ser(P) > Gly(P)
Ala(NP) > Trp(NP)Ser(P) > His(+ve) F R 2 Gln(P) > Leu(NP)
Lys(+ve) > Glu(-ve)F R 2 Ala(NP) > Val(NP)
Glu(-ve) > Val(NP) C D R 2 Glu(-ve) > Asp(-ve)Ser(P) > Thr(P)
C D R 2 Ser(P) > Asn(P)Gly(P) > Arg(+ve) F R 3 Lys(+ve) > Glu(-ve)Tyr(P) > Ser(P)
C D R 3 Ser(P) > Asn(P)Ser(P) > Arg(+ve)Ser(P) > Gly(P)
F R 3 Lys(+ve) > Arg(+ve)
F R 4 Thr(P) > Val(NP)
(NP) = Non polar amino acid (?) = Polar amino acid (+ve) = Positive charge (-ve) = Negative charge
169
CHAPTER FOUR
DISCUSSION
170
DISCUSSION
4.1 VH GENE FAMILY USAGE IN AIJTOANTTBODTES.
The significance of immunoglobulin gene usage in autoimmune disease is
difficult to assess due to the lack of information about the B cell repertoire in normal
adults. Investigations of the human B cell repertoire have been confined to EBV
transformed B cell clones, neoplastic B cells and a relatively small number of
hybridoma derived monoclonal antibodies (Guillaume etal 1990, Sanz etal 1989). In
situ hybridisation can greatly increase the number of B cells surveyed (Guigou et al
1990, Zouali etal 1991) and polymerase chain reaction based analyses have increased
the number even more. Nevertheless all these methods introduce their own bias.
Neither mitogen stimulated B cells nor EBV transformed B cells used in most in situ
hybridisation studies are representative of the entire B cell population (Crain et al
1989). Huang etal (1992) surveyed the VH gene usage of the B cell repertoires of two
normal healthy adults by amplifying cDNA without using variable region gene PCR
primers. These two individuals expressed quite different repertoires which were not
random relative to genomic complexity. One individual in this study showed an over
utilisation of the VH5 gene family. All studies of variable region gene usage regardless
of the nature of the B cell source only reveal the nature of the immune response at one
particular time point and do not portray the dynamics of the immune system.
The aim of this thesis was to examine the V gene usage of a panel of hybridoma
generated monoclonal autoantibodies derived from patients with autoimmune disease.
The pattern of utilisation of VH genes in the panel of monoclonals tested was not
consistent with random usage predicted on the basis of family size. The VH4 family
was expressed in 65% of the monoclonal autoantibodies. This figure is 23% higher
than could be expected if the VH gene usage was random based on the number of
functional genes in the VH4 family (Cook etal 1994) and thus the data obtained from
this study shows an overexpression of VH4 gene family in the panel tested. Frequent
usage of the VH4 gene family has been seen in B cells that secrete autoantibodies
(Pascual & Capra 1991) and the data described here was in agreement with the
observation of others.
171
The VH3 gene family was utilised by 23% of the autoantibodies tested which is
25% below the frequency of expression predicted according to the VH3 family size
(Cook etal 1994). The VH2 and VH5 gene families were each expressed by 6% of the
autoantibodies in the panel thus VH2 also is underexpressed in the panel of
autoantibodies tested.
The VH gene usage of the IgM autoantibodies in this study is similar to those
obtained by examination of the VH gene family usage of monoclonal antibodies derived
from rheumatoid synovial tissue (Brown etal 1992). A panel of eighteen monoclonal
antibodies were examined for VH gene usage and 38.5 % in this panel where found to
be VH4 derived. This figure is just over half than that obtained in the panel described
here. However, the monoclonal antibodies studied by Brown etal (1992) were, in the
main, of unknown specificity thus there was no evidence that the antibodies in this
study were autoreactive. Only two monoclonals studied were actually found bind to
autoantigens; one VH3 encoded IgM was a rheumatoid factor and the other VH4
encoded IgM monoclonal antibody was polyreactive, binding cytoskeletal proteins and
cardiolipin. In contrast, the monoclonal antibodies examined in this study bound to a
variety of autoantigens thus pointing to a more definite role in disease pathogenesis. A
study of the repertoire of normal individuals by comparison showed that only 19% of
the hybridomas derived using the same fusion partner Spatz 4 used the VH4 gene
family thus showing no bias had been introduced during the process of
immortalisation.
The IgM monoclonal autoantibodies, with the exception of the IgM monoclonal
E3-MPO, were immortalised using the GM4672 fusion partner. It was not determined
whether a bias towards certain VH gene families had been introduced to the system by
the use of the fusion partner GM4672; however data published by Rioux etal (1993)
revealed that 22/31 of hybridomas derived from the PBL of normals and autoimmune
patients using the fusion partner GM4672 utilised the VH3 family, the remainder being
derived from VHl (6/31) and VH4 (3/31). VH gene message from GM4672, a low
secretor of a VH4 encoded IgG 2 VkI antibody, was not detectable by northern blotting
using the VH4 probe under the conditions described. Thus the bias in VH4 gene usage
can only be attributed to the VH products of the donor B cells. Thus it is unlikely that
the bias within the IgM monoclonals described here can be attributed to the GM4672
172
fusion partner. In addition, the cDNA probes used in this study did not cross hybridise
under the conditions described although it is possible that the VHl probe could cross
react with the single member VH7 gene family (Mortari etal 1992).
The IgG monoclonal autoantibodies reported here were immortalised using the
heteromyeloma fusion partner CBF7. Of four IgG monoclonals studied two used the
VH3 gene family and two used the VH4 gene family. However, due to the limited
number of clones available, it is difficult to say whether this is representative of the
situation in vivo. The majority of IgG anti-DNA antibodies reported are encoded by
members of either the VH3 or VH4 gene families ( Isenberg etal 1993). Since VH3
and VH4 are now considered to be the two largest VH gene families, it is possible that
no bias in VH gene usage is seen within autoantibodies of the IgG isotype.
Autoantibodies of the IgG isotype are thought to make a greater contribution to
pathogenesis in SLE than those of the IgM isotype and thus there may be no VH gene
family bias usage amongst the “more pathogenic” population. It is necessary to study
larger numbers of IgG antibodies from normal and autoimmune sources in order to
ascertain if VH gene family bias usage is occurring in the generation of autoantibodies
of the IgG isotype.
As previously described, there appeared to be a bias towards the VH4 gene
family usage in the panel of monoclonal autoantibodies analysed by Northern blotting.
However, although Northern blotting analysis involving VH gene family specific
probes allows the analysis of a large number of B cells, it does not permit the detection
of individual VH genes. In situations where differences in V gene utilisation had not
been seen between B cells derived from different sources such as the fetus and normal
adults (Logtenberg etal 1991) it is noteworthy that although there may appear to be
similar patterns of VH gene utilisation between populations, there is no way of
detecting qualitative differences of VH gene expression within a family. It is therefore
necessary to identify individual V genes by nucleotide sequence analysis and relate the
usage of particular V genes to the specificity of the antibodies they encode.
173
4.2 SEQUENCING METHODOLOGY
The VH4 bias found in the Northern blotting study was further examined by
sequencing the immunoglobulin variable regions to investigate whether the bias seen
involved the use of particular germline origins. Several methods were employed to
sequence the immunoglobulin variable regions of the monoclonal autoantibodies, each
with associated advantages and disadvantages. Direct RNA sequencing was difficult to
optimise and required relatively large quantities of mRNA. However, this technique
involved minimal manipulation of the mRNA template thus reducing the chances of
errors and bias associated with amplification. Direct RNA sequencing is time
consuming and thus was abandoned early in the project in favour of polymerase chain
reaction based methods. Direct PCR sequencing is a rapid method of sequencing and
was successfully employed in the analysis of immunoglobulin variable regions.
Sequencing was performed on two independent PCR products for each variable region
analysed to check for amplification errors. The disadvantage of this technique was that
there was no way of checking the clonality of the “monoclonal” antibodies that were
analysed by this method. The appearance of “background” bands in the sequence
pattern obtained on the photographic film suggested the presence of a heterogeneous
PCR product. The sequencing of RT121 heavy chain variable region produced this
kind of pattern and when the RT121 VH region was cloned and sequenced it was
discovered that there were several different VH4 genes within the PCR product.
RT121 tested positive for VH4 only with Northern blotting thus suggesting it was
monoclonal. The partial sequences obtained did not resemble any other VH4 encoded
segment handled in previous experiments suggesting that contamination was not
responsible. In light of this, direct PCR sequencing was abandoned in favour of
cloning the PCR products into plasmid vectors after which several clones were
sequenced to check for clonality as well as PCR error. The plasmid clones were
sequenced either by dsDNA sequencing or single stranded rescue. The dsDNA
sequencing technique was sometimes problematic due to the appearance of secondary
structure. In most cases this was remedied by increasing the temperature of the
termination reaction and by the use of diaza nucleotides. Single stranded rescue was
overall the most successful technique and was the most frequently used.
174
4.3 UTILISATION OF THE VH4.21 GENE TN AIJTOANTTBODTES
There are obvious difficulties in studying a heterogeneous antibody response.
Therefore it was interesting when four of the monoclonals in the panel were found to
express the 9G4 idiotope. Two IgM monoclonal autoantibodies, RT84 and RT79 were
derived from the splenocytes of one SLE patient, one IgG monoclonal autoantibody,
D5 was derived from the peripheral blood mononuclear cells of a second SLE patient
with active disease. All three of these monoclonals antibodies bound ssDNA. The
fourth monoclonal autoantibody E3-MP0 was derived from a patient with vasculitis
and bound to myeloperoxidase.
The 9G4 idiotope had been closely associated with the VH4.21 germline gene.
The relatively nonpolymorphic VH gene VH4.21, is a member of the VH4 family and
represents only 0.5-1% of the VH repertoire (Chothia eta l 1992). Probing for
utilisation of this gene was facilitated by the availability of the monoclonal anti-Id 9G4
(Stevenson etal 1986) which appears to recognise Ig with the heavy chain encoded by
VH4.2I (Thompson etal 1991). This strict correlation of idiotope expression with the
utilisation of the VH4.21 gene has been confirmed in this thesis and in a survey of
immunoglobulin protein sequences (Logtenberg etal 1992). The VH4.21 gene appears
to be mandatory in the production of cold agglutinins (CA) but became of increased
interest in relation to SLE when Van Es etal 1991 described a human IgG monoclonal
anti-DNA antibody that was encoded by VH4.21. A report by Isenberg etal (1993)
stated that the 9G4 idiotope could be identified in 45% of sera from SLE patients. The
9G4 Id also fluctuated with disease activity and was detected in SLE renal biopsies.
These data indicated that a notable proportion of anti-DNA antibodies have VH
segments encoded by the same or closely related genes and that these restricted
immunoglobulins are involved in the renal pathology found in SLE. Thus the
individual VH4 member VH4.21 may be over-represented in SLE and this gene
VH4.21 may specify the structure of many autoantibodies including anti -DNA.
The IgM anti-DNA antibodies RT79 and RT84 utilised VH4.21 in germline
configuration. On comparison with a panel of cold agglutinins which did not bind
DNA, the outstanding feature was the presence of arginine and lysine residues within
the CDR3 regions of these monoclonals which did not appear in the CDR3 regions of
the VH4.21 encoded cold agglutinins. The IgM VH4.21 encoded anti-DNA antibody
175
NEl (Hirabayashi et al 1993), also utilised VH4.21 in germline configuration and had
remarkable similarity to the heavy chain variable region of RT84. The light chain usage
of RT84, RT79 and NEl was diverse. This data points to the importance of the heavy
chain CDR3 region in VH4.21 encoded anti-DNA antibodies in determining DNA
binding.
The IgG monoclonal D5 had 94% homology with VH4.21 and also had a similar
pattern of positively charged residues within the VH CDR3 region. The IgG VH4.21
encoded anti-DNA antibody T14 had a similar pattern of positively charged residues
within the VH CDR3 region as seen in D5. Both T14 and D5 utilised Vk3 light chains.
Both the IgM and IgG VH4.21 encoded anti-DNA antibodies had positively charged
amino acid residues within their VH CDR3 regions. Further analysis of the V regions
of these antibodies using specific site directed mutagenesis experiments may allow an
assessment of how important the positively charged amino acid residues are at these
sites in DNA binding.
The nucleotide sequence of the heavy chain of the IgM monoclonal anti
myeloperoxidase antibody E3-MP0 showed 90% homology to the germline gene
VH4.21. The homology with VH4-21 was not high enough to conclusively establish
that VH4-21 was the gene of origin, however, support for VH4-21 being the germ line
gene of origin was the expression on E3-MPO of the 9G4 idiotope. As previously
described, 9G4 idiotope has also been found on anti-DNA antibodies, expressed on
immunoglobulin deposits in kidney biopsies from lupus patients and found to be raised
on circulating antibodies in other autoimmune rheumatic diseases (Isenberg et aly
1993), suggesting a clinically important role for the VH4-21 gene product in a number
of autoantibodies.
Site directed mutagenesis studies revealed that the 9G4 idiotope was defined by
an amino acid motif AVY in framework one (Potter etal 1993). Comparison of the
nucleotide sequence of E3-MP0 with a bank of VH4.21 encoded cold agglutinins
revealed a highly conserved first framework region as a common feature, and may
account for its ability to react with the 9G4 reagent. E3-MPO had only one replacement
mutation within framework one but it was near the beginning of the framework one
region. The E3-MP0 monoclonal was found to be a poor agglutinator of red cells
(Longhurst etal 1994). Other examples also suggest that the VH4.21 gene segment
176
was probably necessary but not sufficient to confer cold agglutinin activity (Pascual et
d 1991a). Therefore, the finding that E3-MPO was a poor agglutinator cannot be
completely attributed to the frequency of replacement mutations within the VH region.
The conformational epitope on MPO to which E3-MPO binds had not been defined but
it was possible that local charge may play a role as MPO was a highly positively
charged molecule.
VH4.21 has, however, been found to be over represented in B cells from
normal individuals. It accounts for 40-80 % of the VH4 genes found in fetal B cells
and 10-35% of the VH4 genes found in the normal repertoire (Pascual & Capra 1992).
The basis for the frequent use of the VH4.21 gene in the normal B cell
repertoire, the physiological role of autoantibodies secreted by these B cells and their
relationship to pathogenic CA with anti i/I specificity was unknown.
Immunofluorescent staining experiments using mAh 9G4 have shown that 3-11% of B
cells from healthy individuals express the VH4.21 gene segment (Stevenson etal
1986). The 9G4 idiotope was present in the supernatant of large panels of B cell lines
from cord blood 7.4% and second trimester fetal tissues 6.3%. The basis for this over
representation remains unclear but may be influenced by several processes. In newly
formed murine B cells the chromosomal position of VH genes influences their
rearrangement frequency resulting in an over expression of gene segments located
proximal to the JH gene locus (Rawjewsky etal 1987). This is unlikely for the
VH4.21 gene because it has been mapped >500kb upstream of JH gene locus (Matsuda
etal 1993,van der Maarel etal 1993). Alternative mechanisms include regulatory DNA
sequences affecting their rearrangement frequency and Ig receptor mediated processes
that influence the life span or cause clonal expansion of B cells (Osmond 1993). As the
VH4.21 is over represented in the normal and fetal repertoire, VH4.21 utilisation in the
autoimmune repertoire may be attributed to its ability to generate antibodies of a wide
specificity. This suggests that biased usage of VH4.21 in the generation of
autoantibodies may not be as great as was first suspected.
177
4.4 AUTOANTIBODTES ENCODED RY VH4 GENES TN GENERAL
However, 50% of the VH4 encoded monoclonal autoantibodies studied were not
encoded by VH4.21 germline gene. The VH region of the anti-ssDNA antibody RT72
had 100% homology with the germline gene VH4.22 (Sanz etal 1989). This particular
VH4.22 member had not been quoted in the literature as a common origin for anti-DNA
antibodies
The Vk region of RT72 had 97% with the germline gene HK102 (Bentley etal
1980). The polyreactive monoclonal RT55 bound to ss/ds DNA, cardiolipin, SmD
peptide 10 (Ravirajan etal 1992). The heavy chain variable region of RT55 had 100%
homology to the germline gene VH4.18 (Sanz etal 1989). The VH region of VAR24,
an IgM monoclonal derived from a polymyositis patient, also had 100% homology
VH4.18. The light chain variable region of RT55 had 98% homology with the
germline gene HSIGKIO (Jaenichen etal 1984). HSIGKIO was also the germline
origin of the monoclonal RT129 which binds to ss/ds DNA (Ravirajan etal 1992).
RT129 Vk region had 99% homology with HSIGKIO. RT129 VH region had 95%
homology with the VH4 germline gene HV71-2.
It is interesting to note that most the VH4 encoded monoclonal antibodies
sequenced had very high homology (>98%) with their respective germline origins.
This data supports the idea that auto antibodies can be encoded by variable regions with
little or no somatic mutation. All VH4 encoded monoclonals sequenced that were
derived from SLE patients bound to either ss and/or ds DNA. Analysis of the D genes
used in association with VH4 family in this group of monoclonal autoantibodies
revealed that 37% had contributions from the D segment DXP’ 1. Preferential usage of
the DXP’l D gene has been seen in the generation of other autoantibodies.
(Dersimonian etal 1989). 50% of the VH4 encoded monoclonals utilised the JH5 gene
segment, however only 18% of anti-DNA antibodies described in the literature were
encoded by JH5. Interestingly, only one VH4 encoded anti-DNA antibody utilised
JH5 and this was the fetal derived anti-dsDNA IgM monoclonal BEG-2 (Watts et al
1991). This antibody was encoded by the germline gene VH71-2 as was RT129 which
also utilised JH5 and bound dsDNA.
The human monoclonal antibodies RT129, BEG-2 and 2A4 bind dsDNA. The
IgG monoclonal antibody 2A4, was derived from myeloma bone marrow lymphocytes
178
(Davidson etal 1990), BEG-2 monoclonal was fetally derived and RT129 was isolated
from a patient with active SLE. The heavy chain variable regions of all three anti-
dsDNA antibodies were derived from the VH71-2 gene. The different characteristics
of the VH regions of these monoclonal anti-DNA antibodies were compared. The most
striking difference between the variable heavy regions of 2A4, RT129 and BEG-2
occurred within the CDR3 region. The CDR3 region of BEG-2 consisted of only two
amino acids which points to the potential of dsDNA binding being more closely
associated with the VH region VH71-2 rather than the content of the CDR3 region as
was noted in the VH4.21 encoded monoclonal anti-DNA antibodies. 2A4 was an IgG
anti-DNA antibody, the isotype most likely to be associated with pathogenesis, carried
the heavy chain associated idiotype F4. F4 is strongly associated with anti-DNA
antibodies of the IgG isotype and recognises heavy chains on anti-DNA antibodies in
60% of SLE patients (Davidson etal 1990). RT129 has not been tested for F4 idiotype
expression. BEG-2 also carries an anti-DNA associated idiotype, BEG-2p (Watts etal
1991) which is thought to be associated with its lambda encoded light chain is not
associated with anti-DNA antibodies in SLE patients. These data supports the view
that the germline gene VH71-2 has been used to produce anti-dsDNA antibodies from
different sources. Other characteristics of the autoantibodies such as idiotype
expression, light chain usage, and isotype may contribute to their pathogenic potential.
In conclusion, within the VH4 encoded autoantibodies analysed, there appeared
to be preferential utilisation of the DXP’ 1 D region gene and the JH5 gene. The light
chains of the monoclonal autoantibodies were encoded by Vk genes, 45% of which
were members of the VkI gene family. This is perhaps not surprising as VkI is the
largest of the Vk gene families.
4.5 AUTOANTIBODIES ENCODED BY 56P1 AND VH26 GENES
Four monoclonal autoantibodies in the panel available for study utilised
members of the VH3 gene family. 56P1 has been reported as the VH3 germline gene
of origin for a number of anti-DNA antibodies of both the IgM (Mannheimer-Lory et
al 1991) and IgG isotype (Mannheimer-Lory etal 1991, Winkler etal 1992). WRI176
VH region, an IgM anti-DNA monoclonal derived from an SLE patient had 97.3%
179
homology with the germline gene 56P1. This VH3 member was derived from a 130
day old fetus (Schroeder etal 1987). The light chain variable region of WRI176 had
98% homology with the Vkllla member KV328. A polyclonal anti-idiotypic reagent
was raised with WRI176 and was found to be encoded on the heavy chain variable
region. (Blanco etal 1994). Although 44% of SLE patients had elevated levels of
WRI176Id, no correlation was found with disease activity (Blanco eta l 1994).
WRI176 also carries the idiotype 16/6 (Blanco etal 1994) which had been closely
associated with the germline gene VH26 (Chen etal 1988). WRI176 VH region had
89% homology with the VH26 germline gene. This level of homology with the VH26
gene was obviously sufficient to allow this idiotype to be expressed.
Data from several laboratories have led to the conclusion that certain V genes
are over-represented in the human B cell repertoire. An example of this is the VH3
member VH26 which is associated with the 16/6 idiotype which occurs on
autoantibodies and antibacterial antibodies. VH26 can be found in 10% of EBV
transformed B cells and up to 25 % of all VH3 positive clones in libraries generated by
PGR from normal blood (Stewart etal 1992). The VH26 gene has been found in fetal
B cells (Schroeder etal 1987) in antibody response to vaccination with Haemophilias
B (Silverman etal 1991) and in B cell malignancies, multiple myeloma and CLL and in
several other autoantibodies (Spatz etal 1990, Dersimonian etal 1987, Lampman etal
1989, Young etal 1990, Guillaume etal 1990). These latter include antibodies to
platelets, cardiolipin, and neuronal antigens. Furthermore VH26 appears to be utilised
by B cells obtained from the synovium of patients with rheumatoid arthritis (Brown et
al 1991). The single VH gene VH 26 is used in up to 10% of circulating B cells in
normal individuals (Stewart etal 1993) and accounts for up to 25% of all expressed
members of the VH3 family (Stewart etal 1993)
The VH26 gene also appeared to be the germline gene of origin for three other
monoclonals in this study each with completely different origins and autoreactivity.
The VH region of the anti-histone (HI) antibody RT6 had 98% homology with the
VH26 germline gene. RT6 did not bind DNA. It is interesting to note that VH26 was
found to have complete identity with the fetally derived VH3 gene 30P1 ( Schroeder et
al 1987). This supports the idea that auto-antibodies can be encoded from the same
panel of V genes that are used in early ontogeny (Pascual and Capra 1991). The IgG
180
anti-dsDNA reactive antibody B3 was generated from a patient with active SLE
(Ehrenstein etal 1994) and the heavy chain variable region had 94% homology to the
germline gene VH26. The IgG anticardiolipin monoclonal, UK4, also had 93%
homology with VH26. It appears that as with the VH4.21 gene, the VH26 germline
gene is also highly represented in the normal repertoire and thus the bias usage of this
gene in autoimmune disease may not be as marked as was first thought. If a variable
region gene is able to encode a wide variety of specificities to foreign antigens in the
normal repertoire then it is logical that such a V gene will be capable of the same
diversity with respect to binding to autoantigens.
There are probably two copies of the VH26 gene in the germline and this may
partially account for its preponderance in the repertoire (Stewart etal 1993). There are
at least two copies of the VH3 member 56P1 in the germline (Sasso etal 1990) so the
predominanance of VH26 over 56P1 can not be explain purely on the basis of multiple
gene copies. VH26 is highly conserved and is not polymorphic (Rubinstein etal 1993)
thus it may be preferentially expressed because of the antigen binding or idiotypic
properties of the expressed protein. Interestingly the 16/6 idiotype is associated with
the VH26 gene product which is elevated in the sera of patients with active lupus. If
the 16/6 idiotype is only encoded by VH26 , one would expect the prevalance of this
also idiotype to be high in normal sera. However this is not the case suggesting that
16/6 idiotype expression seen in autoimmune disease may not originate solely on VH26
encoded autoantibodies.
An interesting feature of the VH26 encoded monoclonal autoantibodies
described here was the common utilisation of the same V^II gene family in the
generation of their light chains. The VX2 gene family is thought to encode the
pathogenic 8.12 idiotype (Paul etal 1992). This idiotype has been found in immune
complexes deposited in the kidney lesions of SLE patients. RT6, B3 and UK4 were all
encoded by the Vxn member DPLl 1.
A polyclonal anti-idiotype was raised using B3. The idiotype was found to be
encoded on the light chain of this antibody. (Ehrenstein etal 1994). B3 was expressed
on IgG antibodies in the serum of 20% of SLE patients and within this group there was
a statistically significant association of B3Id with musculo skeletal disease. This
idiotype was of particular interest as was it derived from an IgG anti-DNA monoclonal
181
antibody, the isotype associated with active disease in SLE. The B3 light chain variable
region was compared to RT6 and UK4 light chains which did not carry the B3 idiotype
and the outstanding feature was the presence of an arginine couplet in the CDRl region
which probably arose by somatic mutation. This arginine couplet could define the
B3Id. B3Id was also the only anti-DNA associated idiotype which had correlation with
a particular disease manifestation of SLE.
A high arginine content appears to be a feature of many anti-DNA antibodies
and it has been postulated that concentrations of basic amino acid residues could
explain some recurrent anti-idiotype systems. For example the arginine couplet in the
CDRl of the light chain of B3 could define the B3 idiotype. Further support of this
comes from a study by Eilat etal (1988) who showed that the VH sequences of two
anti-DNA antibodies which share an idiotype differ at VH and V k but have similar
arginine containing CDR3 regions.
4.6 AMINO ACID RESIDUES ASSOCIATED WITH BINDING TO
AUTOANTIGENS
There is much evidence for the contribution of certain amino acid residues to
binding to autoantigens such as DNA. Arginine, a positively charged amino acid, has
been suggested as important in enhancing binding to DNA in murine antibodies. The
studies of Radie etal (1993) centred on a murine VH gene segment 3H9 which occurs
in anti-DNA antibodies in combination with various light chains suggesting that it plays
a dominant role in DNA binding. When the V region of 3H9 was transfected into cells
that produced a variety of light chains all transfectants bound ssDNA. The role of
particular arginine residues was then investigated. When the CDR3 arginine at position
96 was mutated to glycine the antibody did not bind either native or denatured DNA.
The data generated in this thesis suggests that the VH4.21 gene segment may also
generate anti-DNA specificity in a limited manner and if arginine residues are important
that they operate from CDR3 rather than from CDRl or CDR2.
The RRG motif at the 3’ end of D5 has also been found in the CDR3 regions of
two mouse IgG anti-DNA antibodies (Eilat etal 1988). The Fv region of one of these
(D42) has been modelled by computer and a stretch of poly(dT) ssDNA antigen was
182
docked to a cleft in the antibody combining site formed by the various CDRs and
involving a preponderance of arginine and tyrosine residues (Eilat etal 1988). The
arginine residues may interact with either the phosphate moiety or form stable hydrogen
bonds with the heterocyclic bases of DNA (Eilat etal 1988).
The arginine couplet seen in the light chain CDRl of the IgG anti-dsDNA auto
antibody B3 could also enhance DNA binding in this antibody. Arginine residues were
also present in the VH CDR3 of WRI176. The impact of arginine residues on binding
to DNA may also depend upon the residues in their vicinity. However, the anti-DNA
monoclonal RT72 did not have arginine residues in the CDR regions of either the heavy
or the light chain, thus there must be other factors such as conformation of the antigen
binding site which play a role in generation of anti-DNA affinity.
The contribution of amino acid residues such as arginine to DNA binding and
pathogenicity was recently explored by Katz etal (1994). Site directed mutagenesis
was employed to change amino acid residues in the heavy chain of a murine pathogenic
anti dsDNA antibody. The amino acids targeted in this study were, arginines in CDR2,
CDR3 and FR3 which were replaced with serine, lysine residues, glutamic acid and
aspartic acid residues in FR3 and CDR3. An amino acid substitution of an arginine at
position 52 with serine resulted in greater binding to dsDNA. Most notably, the
replacement of an aspartic acid by glycine in CDR3 resulted in loss of dsDNA binding
ability. In in vivo experiments in mice, this particular mutant antibody which was more
cationic than the parental antibody was no longer sequested in the glomeruli. The
results of this study show that it is not yet possible to predict the amino acid residues
which are critical to DNA binding. This conclusion is also suggested from the study
described here. The role of residues such as arginine and lysine may be very different
in anti-DNA antibodies encoded by different germline variable genes. The evidence for
the role of arginine residues in DNA binding in VH4.21 encoded antibodies is strong
and studies including site directed mutagenesis will investigate this further.
Although there is evidence to suggest a role for positively charged residues in
conferring autoantibody binding to negatively charged DNA, other factors may also
contribute to binding. These arguments can be paralleled when considering the binding
of antibodies to positively charged myeloperoxidase (MPO). Negatively charged
amino acid residues in the V regions, for example could aid binding to MPO or increase
183
binding affinity. Within the VH region of E3-MP0 there were three negatively charged
residues, two were glutamic acid and occured in framework regions one and two. One
aspartic acid occured in CDR2. However, in CDR3 there were two basic residues, an
arginine and a histidine residue. Overall however, there were no unique features in the
linear amino acid sequence which would indicate that changes to negatively charged
residues were important in the binding of E3-MP0 to its antigen. Only analysis of the
tertiary structure of the antibody by computer modelling and crystallographic analysis
will reveal which amino acids were exposed at the antigen binding site and may be
involved in binding to MPO. Modification of such residues will determine those which
were key in binding to MPO.
4.7 THE PATHOGENIC POTENTIAL OF POLYREACTIVE IgM
AUTOANTIBODTES
Although some of the autoantibodies described in this thesis resemble natural
autoantibodies, they were derived from patients with active autoimmune disease and
thus their contribution to pathogenesis can not be ruled out. Many of the IgM
monoclonal autoantibodies described here also carry idiotypes found in immune
complexes in disease lesions such as 9G4 and 16/6 thus supporting their possible
contribution to pathogenesis. Evidence for this was provided by Tillman etal (1992)
who addressed this question by preparing hybridomas from a single mouse at two
different times before and after the onset of IgG anti-DNA production and the
development of autoimmune disease. In at least two cases a particular gene
rearrangement coded for an IgM anti-ssDNA antibody before the onset of disease and a
mutated version of the same gene arrangement coded for anti-dsDNA antibody later
when disease was established. Thus B cells making natural autoantibodies can be the
precursor for cells making disease associated antibodies. Although, it could be argued
that human IgM poly reactive autoantibodies have a questionable role in autoimmune
disease, they may be the foundation for the production of more pathogenic
autoantibodies generated via class switching and somatic mutation. It is possible that
the IgM monoclonals described here are related to natural antibody populations and in
fact share a great many characteristics in common, such as isotype and polyreactivity.
184
One unique characteristic appears to be the VH gene usage. Every VH gene family was
represented was in a panel of natural autoantibodies described by Pascual and Capra
(1991). However, there was a VH4 gene family bias in the panel of IgM
autoantibodies examined here. Although the striking overlap between VH and VL
genes expressed in human natural and pathogenic autoantibodies suggests a
relationship between these two categories of autoantibodies it has not been proven.
Evidence for such a relationship may be obtained by returning the gene elements
encoding pathogenic autoantibodies to their germline configuration by site directed
mutagenesis and express them in eukaryotic cells. The binding specificity of the
precursor antibody may be determined and unveil the putative natural autoantibody
activity.
4.8 POTENTIAL DYSFUNCTION IN B CELLS PRODUCING
AUTOANTIBODIES
Although high affinity potentially pathogenic IgG anti dsDNA antibodies can
arise in the absence of somatic mutation and affinity maturation the anti dsDNA
antibodies that are associated with disease in mouse models of lupus and SLE patients
have extensive somatic mutation. The oligoclonality of the response, the nature of the
amino acid changes that are generated by somatic mutation, the numbers of
replacements in CDRs and the instances of the conversion from ssDNA binding to
dsDNA binding associated with these changes suggest that the anti-dsDNA response is
both driven and selected by dsDNA or an antigen that closely resembles it in
conformation and charge. Based on available data the mechanism for the production of
potentially pathogenic autoantibodies is similar to the response of non-autoimmune
individuals to foreign antigen. This suggests that B cells producing autoantibodies in
autoimmune disease are in fact behaving “normally” and the defect in autoimmune
individuals causing disease occurs elsewhere.
There is however some evidence from studies of normal mice to suggest that
the generation of arginine residues in immunoglobulin V regions expressed in
autoimmune disease does not occur in normals. Immunisation studies of normal mice
with bacterial DNA (Gilkeson etal 1993) revealed that a significant anti-DNA response
185
could be induced in this manner and that the antibodies produced resembled some lupus
anti-DNA in their binding properties. Predominate utilisation of the J558 VH gene
family was noted. None of these antibodies had more than one arginine in the CDR3
region nor the arginines at position 100-100a which is a characteristic feature of murine
autoimmune anti-DNA antibodies. The high arginine content in the CDR3 regions is
not invariant amongst anti-dsDNA antibodies (Shlomchik etal 1990), the failure of
normal mice to generate these specificities cannot relate solely to the composition of the
VH CDR3 and its regulation. The findings suggest that the process of somatic
mutation may differ in normal and lupus mice and that in the absence of certain VH
CDR3 sequences the generation of anti-dsDNA antibodies requires mutational events
that are rare in normals. This study emphasises the unique characteristics of anti-DNA
response and the likelihood that anti-DNA production in lupus requires abnormalities
such as alternations in the B cell repertoire in addition to DNA antigen drive. It is
possible that the arginine residues seen in the variable regions of the autoantibodies in
the human could be generated by abnormal somatic processes.
4.9 THERAPY
The anti-DNA response appears to be quite diverse, and because of this it is
unlikely that approaches to treatment involving anti-idiotypic antibodies as specific
reagents against pathogenic anti-DNA antibodies will eliminate a significant proportion
of these antibodies in any single individual. However if arginines are ligands for anti-
idiotypic sera then a more promising therapy could come from blocking interaction of
arginines with DNA using smaller aromatic and charged molecules This approach has
been explored by Ben-Chetrit etal (1988). It is highly likely however that such
intervention could disrupt other essential amino acid charged related interactions within
the body and cause serious side effects. Studies of the structure and origin of
pathogenic human autoantibodies may reveal unique features which can be used to
target the B cells that produce them.
186
4.10 FURTHER WORK
Computer modelling studies performed by Kalsi etal (1994) have produced a
three dimensional model of B3. Three arginine residues appear to be spread across the
antigen binding site. Two of these arginines were on the light chain at positions
twenty-seven and fifty-six, the former being the second arginine in the “CDRl arginine
couplet”, described earlier and the latter a germline encoded CDR2 residue. The third
arginine residue was heavy chain germline encoded at position seventy-two within
framework three. A possible docked complex for dsDNA with the B3 model appears
to involve these residues which interact with the phosphate backbone of the DNA.
This model will be of immense value in targeting key residues in future site directed
mutagenesis studies which could determine the contribution each is making to the
affinity this antibody has for dsDNA.
Computer modelling studies on other monoclonal autoantibodies described in
this thesis are required in order to shed light on the importance of amino acid residues
which have been postulated as important in binding to autoantigens in murine systems.
Computer models in conjunction with mutagenesis experiments will give insight into
the mechanisms involved in binding to autoantigens such as DNA.
As previously discussed, it is feasible that some of the polyreactive IgM
autoantibodies described could undergo class switching and thus be the starting point
for higher affinity IgG pathogenic autoantibodies. This could be further investigated
by clone tracing experiments in order to see if an IgM has undergone class switching
and the frequency and nature of somatic mutations. If such a IgG clone were
expressed in an expression system, the role of class switch associated somatic
mutations in increasing affinity to autoantigen could be investigated.
The V gene usage of the monoclonal antibodies shown in this study only reflect
the state of the immune system one time point. It would be interesting to investigate
the V gene usage of an autoimmune patient over a period of time and look for possible
connections between the genetic origin of antibodies expressed, their binding
specificity and isotype relative to the disease state.
In this thesis, the origins of human autoantibodies derived from patients with
active autoimmune disease were analysed and particular immunoglobulin variable
region genes were utilised in the generation of autoantibodies. However, it is not
187
known if a potential bias occurring in the production of antibodies to autoantigens in
autoimmune individuals also occurs in the production of antibodies to exogenous
antigens within the same individual. Analysis of antibodies to exogenous antigens in
autoimmune disease would perhaps shed light on the general immune response in
autoimmunity.
In the case of the IgG autoantibodies described in this study and in fact some of
the IgM monoclonals, the assignment of germline genes of origin had been performed
by obtaining the nearest variable region germline gene match by data base analysis. To
be certain that the mutations seen in expressed immunoglobulin variable region genes
are generated by somatic mutation and are not due to polymorphism, the
immunoglobulin germline genes encoding the antibody in question should be isolated
from the patient source. This is especially relevant in the case of B3 light chain where
the presence of two potentially somatically derived arginines could play an important
role in autoantigen binding and idiotype expression. The anti-MPO antibody E3-MPO
had 90% homology to the germline gene VH4.21 which is not really high enough to be
certain that VH4.21 was the germline gene of origin. Unfortunately the patient from
which E3-MPO was derived died within days of the initial collection of PEL used to
generate this antibody so germline analysis could not be carried out in this case. The
patient from which the “RT” clones were derived is not available to donate more
material for germline gene analysis. These points illustrate some of the difficulties
arising when performing analyses involving humans rather than mice.
The fact that human autoimmune disease is controlled by drug therapy also
creates problems in assessing the true situation occurring in the human disease
associated immune response.
188
4.11 CONCLUSIONS
In conclusion, the study described here shows evidence of a VH4 gene family
bias usage in the panel of monoclonal autoantibodies available, however from these
data one cannot conclude that such a bias was consistent and occured in vivo.
Sequence analysis showed that arginine residues occur frequently in association
with particular variable region genes such as VH4.21 56P1 and DPLll expressed in
antibodies that bind to DNA, although arginines did not appear to be an absolute
requirement for DNA binding in association with other variable region genes such as
VH4.22, VH26, and VH4.18. However, autoantibodies that have very high
homology to known germline genes can bind DNA without arginine residues occuring
in their expressed Ig variable regions. This illustrates that positively charged residues
are not essential for DNA binding.
There was so little immunoglobulin variable region sequence information,
regarding human autoantibodies derived from autoimmune disease, as compared to the
mouse that drawing conclusions about the relevance of variable gene usage and
mechanisms involved in binding to autoantigens was extremely difficult However, the
study described here contributes to the limited information available concerning the
genetic origins of human autoantibodies .
189
CHAPTER FIVE
REFERENCES
190
REFERENCESA
Adderson EA, Scakelford PG, Insel RA et al. Immunoglobulin light chain variable region gene sequences for human antibodies to Heamophilus influenzae type b capsular polysaccharide are dominated by a number of Vk and Vk segments and VJ combinations. J Clin Invest 1992; 729-738
Adler MK, Baumgarten A, Hecht B, Siegel NJ. Prognostic significance of DNA binding capacity patterns in patients with lupus nephritis. Ann Rheum Dis. 1975; M: 444-450.
Agopiau MS, Boctor FN, Peter JB. False positive test results for IgM anticardiolipin antibody due to IgM rheumatoid factor.Arthritis Rheum 1988; M: 1212-1213.
Alarcon-Segovia D, Deleze M, Oria CV, Sanchez-Guerrero J, Gomez-Pacheco L. Anti-phospholipid antibodies and the antiphospholipid syndrome in SLE. A prospective analysis of 500 consecutive patients. Medicine 1989b; M: 353-365.
Alarcon-Segovia D, Sanchez-Guerrero J. Primary antiphospholipid syndrome. J Rheumatol 1989a: 16: 482-488.
Alt FW, Yancopoulos GD, Blackwell TK et al. Ordered rearrangement of immunoglobulin heavy chain variable region segments. EMBO 1984; 2: 1209-19
AltF, Baltimore B. Joining of immunoglobulin heavy chain gene segments: Implications from a chromosome with evidence of three D-JH fusions. PNAS f/SA 1982;2£: 4118-4122.
Anderson MLM, Brown L, McKenzie E, Kellow JE, Young BD.Cloning and sequence analysis of an Ig X light chain mRNA expressed in Burkitt’s lymphoma. Nucleic acid Res 1985; 13:2931-2941.
Anderson MIM, Szajnert ME, Kaplan JC, McColl L, Young BD.The isolation of a human immunoglobulin V-lambda gene from a recombinant library of chromosome twenty-two and estimation of its copy number. Nucl Acid Res 1984; 12: 6647-6661.
Andrzejewski Jr. CA, Rauch J, Lafer E, Stollar BD, Schwartz RS.Antigen binding diversity and idiotype cross reactions among hybridoma autoantibodies to DNA. J Immunol 1981; 126: 226- 231.
Antin JH, Emerson SG, Martin P. Leu 1+(CD5+) B cells. A major lymphoid subpopulation in human foetal spleen phenotypic and functional studies. J immunol 1986; 136: 505-510.
Asherson RA. A primary antiphospholipid syndrome? J Rheumatol 1988: 15: 1742-1746.
Asherson RA, Khamashta MA, Ordi-Ros J, Derhsen RH, Machin SJ. Primary antiphospholipid syndrome: Major clinical and serological features. Medicine 1989; 366-375.
191
Atkinson P, Lampman G, Furie BC et a l Homology of the NH2 terminal amino acid light sequences of the heavy and light chains of human monoclonal lupus autoantibodies containing the dominant 16/6 idiotype. J Clin Invest 1985; 2^: 1138-1143.
BBallou SP & Kushner I. Lupus patients that lack detectable anti- DNA: Clinical features and survival. Arthritis Rheum 1982; 25: 1126-1129.
Baltaro RJ, Hoffman OS, Sechler JM et a l Immunoglobulin G antineutraphil cytoplasmic antibodies are produced in the respiratory tract of patients with Wegner’s granulomatosis. Am Rev Respir Dis 1991; 143: 275-278.
Barrett KJ, Stollar BD. Relationship of human variable region heavy chain germline genes to genes encoding anti-DNA autoantibodies. J Immunol 1987: 139: 2496-2501.
Bauer TR & Blomberg B. The human X light chain Ig locus. Recharacterisation of JCX6 and identification of a function^ ICXl. J Immunol 1991; 146: 2813-2820.
Ben-Cherit E, Eilat D, Ben-Sasson SA. Specific inhibition of the DNA anti-DNA immune reaction by low molecular weight anionic compounds. Immunology 1988; 479-85
Bentley DL & Rabbitts TH. Human immunoglobulin variable region genes. DNA sequences of two Vk genes and a pseudogene Nature 1980; 28S: 730-733.
Berek C & Milstein C. Mutational drift and repertiore shift in the maturation of the immune response. Immunol Rev 1987: 96: 23- 42.
Bernstein RM, Hobbs RN, Lee DJ, Ward DJ, Hughes GR. Patterns of anti-histone antibody specificity in systemic rheumatic disease. Arthritis Rheum 1985; 2&: 285-293.
Bernstein NS, Levy S, Levy R. Activation of an excluded immunoglobulin allele in a human B lymphoma cell line. Science 1989; 244: 337-339.
Berman JE & Alt FW. Human heavy chain variable region gene diversity, organisation and expression. Int Rev Immunol 1990; 3: 203-214.
Berman JE, Mellis SJ, Pollock R, et a l Content and organisation of the human Ig VH locus : Definition of three new VH families and linkage to the CH locus. EMBO J 1988; 2: 727-738.
Bielschowsky M, Hey 1er BJ, Howie JB. Spontaneous haemolytic anaemia in mice of the NZB/W strain. Roc Univ Otago Med School 1959:37: 9-11.
192
Blanco F, Kalsi J, Isenberg DA. Analysis of antibodies to RNA in patients with systemic lupus erythematosus and other autoimmune rheumatic diseases. Clin Exp Immunol 1991: 86: 66-70.
Blanco F, Longhurst C, Watts R et al. Identification and characterisation of a new human DNA reactive monoclonal antibody and a common idiotype, W RI176 Id p. Lupus 1994; 2:15-24.
Bona CA, Finley S, Waters S, et al. Anti-immunoglobulin antibodies III. Properties of the sequential anti-idiotypic antibodies to heterologous anti-gamma globulins. Detection of die reactivity of anti-idiotype antibodies with the epitopes of Fc fragments (homobodies) and with epitopes and idiotopes (epibodies). J Exp MeJ1982;15ü: 986-999.
Bona CA. V genes encoding autoantibodies: Molecular and phenotypic characteristics. Ann Rev Imm 1988; 6: 327-358.
Brodeur PH & Riblet R. The immunoglobulin heavy chain variable region (Igh-V) locus in the mouse I. One hundred Igh-V genes comprise of seven families of homologous genes. Eur J Imm 1984; 14: 922-930.
Brown CMS, Longhurst C, Haynes G et al. Immunoglobulin heavy chain variable region genes utilisation by B cell hybridomas derived from rheumatoid synovial tissue. Clin Exp Immunol 1992; 22: 230-238.
Buluwela L & Rabbitts TH. A VH gene is located within 95 KB of the human immunoglobulin heavy chain constant region. Eur J immunol 1988a: 18: 1843-1845.
Buluwela L, Albertson DO, Sherrington P et al. The use of chromosomal translocations to study human immunoglobulin gene organisation: Mapping DH segments within 35kb of the Cp gene and identification of the new DH locus. EMBO 1988b; 7: 2003- 2010 .
Butler PIG. The folding of chromatin. CRC Crit Rev Biochem 1983; i i : 57-91.
CCairns E, Massicote H, Bell DA. Expression in systemic lupus erythematosus of an idiotype common to DNA binding and nonbinding antibodies produced by normal lymphoid cells. J Clin /Mvgff 1989a; 22: 1002-1009.
Cairns E, Kwong PC, Misener V et al. Analysis of variable region genes encoding a human anti-DNA antibody of normal origin. J Immunol 1989b; 143: 685-
Cambridge G, Williams M, Leaker B, Corbett M, Smith CR. Antimyeloperoxidase antibodies in patients with rheumatoid arthritis: Prevalence, clinical correlates and IgG subclass. Ann Rheum Dis 1994; 22: 24-29.
193
Cambridge G, Hall TJ, Leaker B. Heterogeneity of antibodies to myeloperoxidase in sera from patients with vasculitis. JASN 1991; 2: 590
Carroll P, Stafford D, Schwartz RS e ta l . Murine monoclonal anti DNA autoantibodies bind to endogenous bacteria. J Immunol 1985; 135: 1086-1090.
Carson DA, Chen PP, Fox RT et al. Rheumatoid factor and immune networks. Ann Rev Immunol 1987; 109-126.
Cepellini R, Polli E, Celeda F. DNA- reacting factor in a serum of a patient with lupus erythematosus diffusus. Proc Soc Exp Bio M ed.\951\%:. 572-574.
Cervera R, Font J, Lopez-Soto A. et al Isotype distribution of anticardiolipin antibodies in systemic lupus erythematosus; prospective analysis of a series of 100 patients. Ann Rheum Dis 1990; 42: 109-113.
Cervera R, Khamashta M, Font J et al. Anti-endothelial cell antibodies in patients with antiphospholipid syndrome. Autoimmunity \9 9 \\ 43-47.
Chambon P, Elgin S, Felsenfeld G etal. The cell nucleus. In; Molecular biology of the cell. 2nd edition New York, Garland 1989; 481-549.
Charles LA, Caidas ML, Falk RJ, Terrell RS, Jennette JC. Antibodies against granule proteins activate neutrophils in vitro. J Leukoc Biol 1991 ; 50: 539-546.
Chen PP, Fong S, Houghten RA et al. Characterisation of an epibody. An anti- idiotype that reacts with both the idiotype of rheumatoid factor (RF) and the antigen recognised by RF. J Exp
1985; 161: 323-331
Chen PP, Liu MF, Sinha SA. 16/6 idiotype positive anti-DNA antibody is encoded by a conserved VH gene with no somatic mutation. Arth & Rheum 1988 H : 1429-1431.
Chen PP, Liu MF, Glass CA. Characterisation of two immunoglobulin heavy chain genes which are homologous to human rheumatoid factors. Arth & Rheum 1989; 12: 72-76.
Chen PP, Albrandt K, Kipps.TS et al. Isolation chracterisation of human Vk3 germline genes. J Immunol 1987; 139: 1727-1733.
Chen PP, Olsen NJ, Yang PM etal. From human autoantibodies to the fetal antibody repertoire to B cell malignancy: Its a small world after all. Intern rev immunol 1990; 5: 239-251.
Cherif D & Berger R. New localisations of VH sequences by in situ hybridisation with biotinylated probes. Genes Chrom Cancer 1990; 2: 103-108.
194
Chirgwin JM, Pryzbyla AE, MacDonald RJ, Rutter WJ. Isolation of biologically active ribonucleic acid from sources enriched in ribonuclease. Biochem 1979; IS: 5294-5299.
Chothia C, Lesk AM, Gherardi E et al.Structural repertoire of the human VH segments. J Mol Biol 1992; 222: 799-817.
Chuchana P, Blancher A, Brockly ¥ e ta l Definition of the human immunoglobulin variable lambda (IGLV) gene subgroups. Eur J Immunol 1994: 20: 1317-1325.
Cohen PL & Eisenberg RA. Ipr and gld: single gene models of systemic autoimmunity and lymphoproliferative disease. Ann Rev Immunol 1991: 173: 127-136.
Cohen-Tervaert JW, Huitema MG, Hene RJ et al. Prevention of relapses in Wegener’s Granulomatosis by treatment based on antineutrophil cytoplasmic antibody titre. Lancet 1990; 336: 709- 711.
Cohen Tervaert JW, Van der Woude FJ, Fauci AS et al. Association between active Wegner’s granulomatosis and anticytoplasmic antibodies. Arch Intern Med 19S9; 149: 2461-2465.
Cook GP, Tomlinson IM, Walter G et al. A map of the human immunoglobulin VH locus completed by analysis of the telomeric region of chromosome 14q. Nat Genetics 1994\X- 162-168.
Costa O & Monier JC. Antihistone antibodies detected by ELISA and immunoblotting in SLE and RA. J Rheumatol 1986; 12: 722- 725.
Cox DW, Markovic VD, Teshisma IE. Genes for immunoglobulin heavy chains and for a j-anti trypsin are localised to specific regions of chromosome 14q. Nature (London) 1982; 297: 428-430.
Coutinho A, Grandien A, Faro-Rivas J, Mota-Santos TA. Idiotypes, tailors and networks. Ann Inst Pasteur Immunol 1988; m 599-607.
Crain MJ, Sanders SK, Butler JL, Cooper MD. Epstein Barr virus preferentially induces proliferation of primed B cells. J Immunol 1989; 142: 1543-1548.
Cronin ME, Biswas RM, Van der Straeton C, Fleisher TA, Klippel JH. IgG and IgM anticardiolipin antibodies in patients with SLE with anticardiolipin antibodies associated clinical syndromes. J Rheumatol 1988; 12: 795-798.
Crouse GF, Frishcauf A, Lehrach H. An intergrated and simplified approach to cloning into plasmids and single stranded phages. Met Enzymol 1983; 101: 78-89.
Cuisinier AM, Guigan V, Boubli L. Preferential expression of VH5 and VH6 genes in early human B cell ontogeny. Scan J Immunol 1989; 20: 493-497.
195
Cuisinier AM et al. Mechanisms that generate immunoglobulin diversity operate from the 5th week of gestation in fetal liver. Eur J Immunol 1993: 23: 110-118.
Cunningham BA, Pflumm MN, Rutishauser U, Edelman G. Subgroups of amino acid sequences in the variable regions of immunoglobulin heavy chains. Proc Natl Acad Set USA 1969: 64: 997-1003.
DDavidson A, Smith A, Katz J etal.. A cross reactive idiotype on an anti-DNA antibodies defines a H chain determinant present on almost exclusively on IgG antibodies. J Immunol 1989; 143: 174- 180.
Davidson A, Halpem R, Diamond B. Speculation on the role of somatic mutation in the generation of anti-DNA antibodies. Ann NY Acad Set 1986; 47i: 174-180.
Davidson A, Manheimer-Lory A, Aranow C et al. Molecular characterisation of a somatically mutated anti-DNA antibody bearing two systemic lupus erythematosus-related idiotypes. J Clin Invest 1990; M: 1401-1409.
Deleze M, Alarcon-Sergovia D, Oria CV et al. Hemocytopenia in SLE. Relationship to antiphospholipid antibodies. J Rheumatol 1989; 16: 926-930.
DeRemee RA. The treatment of Wegner’s granulomatosis with Trimethoprism/Sulfamethoxazole: Illusion or vision. Arth Rheum 1988; 31: 1068-1072.
Derue G, Englert H, Harris EN, Hughes GRV Fetal loss in SLE: Associating with anticardiolipin antibodies. J. Obstet. Gynaecol. Neonatal. Nurs 1985; 2: 207-209.
Dersimonian H, Schwartz RS, Barrett KJ. Relationship of human variable region heavy chain germline genes to genes encoding anti- DNA antibodies. J Immunol 1987; 139: 2496-2501.
Dersimonian H, Me Adam KWJP, Mackworth-Young C. The recurrent expression of V region segments in human IgM anti-DNA autoantibodies. J Immunol 1989: 142: 4027-4033.
Dersimonian H, Long A, Rubenstein D, Stollar BD, Schwartz RS. VH genes of human autoantibodies. Intern Rev Immunol 1990; 3: 253-264.
Dharmesh DD, Krishnan MR, Swindle JT, Marion TN. Antigen- specific induction antibodies against native mamalian DNA in non- autoimmune mice. J Immmunol 1993: 151: 1614-1626.
Dyrberg T, Peterson JS, Oldstone MBA. Immunological cross reactivity between mimicking epitopes on a virus protein and a human autoantigen depends on a single amino acid residue. Clin. Immunol. Immunopathol 1990; 54: 290-297.
196
EEdelman GM, Cunningham BA, Gall WE et al. The covalent structure of the entire y G immunoglobulin molecule. Proc Natl Acad Sci USA 1969; 62: 78-85.
Edwards RHT, Isenberg DA, Wiles CM, Young A, Snaith ML. Investigation of inflammatory myopathy. J Roy Coll Phys 1981; l i : 19-24.
Ehrenstein MR, Longhurst CM, Isenberg DA. Production and analysis of IgG monoclonal anti-DNA antibodies from systemic lupus erythematosus (SLE) patients. Clin Exp Immunol 1^3; 92: 39-45.
Ehrenstein MR, Longhurst CM, Latchman DS, Isenberg DA. Serological and genetic characterisation of a human immunoglobulin G anti-DNA idiotype. J Clin Invest 1994; 92: 1787-1797.
Eichmann K. Expression and function of idiotypes of lymphocytes. Adv Immunol 1978: 26: 195-254.
Eilat D, Webster DM, Rees AR. V regions of anti-DNA and anti- RNA autoantibodies derived from NZB/NZW FI mice. J. Immunol. 1988; 141: 1745-53.
Eilat D. Anti DNA antibodies: Problems in their study and interpretation. Clin Exp Immunol 1986: 65: 215-222.
El Roiey A, Sela O, Isenberg DA et al. The sera of patients with Klebsiella in fection contain a common anti-DNA idiotype (16/6) and polynucleotide activity. Clin Exp Immunol 1987; 62l 507-515.
Enger W & Chapel HM. Titration of antibodies against neutrophil cytoplasmic antigens is useful in monitoring disease activity in systemic vasculitides. Clin Exp Immunol 1990; S2: 244-249.
Estes D & Christian CL. The natural history of systemic lupus erythematosus by prospective analysis. Medicine 1971: 50: 85-95.
Ewert BH, Jennette JC, Faulk RJ. Anti-myeloperoxidase antibodies stimulate neutrophils to damage human endothelial cells. Kidney Int 1992; 41: 375-383.
FFaaber P, Capel PJA, Rijke GPM et al. Crossreactivity of anti- DNA antibodies with proteoglycans. Clin Exp Immunol 1984; 55: 502-508.
Fauci AS, Haynes BF, Katz P, Wolff SM. Wegener’s granulomatosis: prospective clinical and therapeutic experience with 85 patients for 21 years. Ann Int Med 1983; 76-85.
Feinberg AP & Vogelstein B. A technique for radiolabeling DNA restriction endonuclease fragments to high specificity. Annals Biochem 1984: 132: 6-13.
197
French DL, Laskov R, Scharff MD. The role of somatic hypermutation in the generation of antibody diversity. Science 1989; 244: 1152-1157.
Friou GJ. Clinical applications of lupus serum - nucleoprotein reaction using fluorescent antibody technique J Clin Invest 1957; M: 890-94
Friou GJ. The significance of the lupus globulin-nucleoprotein reaction. Ann Intern Med 1958a; 42: 866-874.
Friou GJ & Detre KD. Interaction of nuclei and globulin from lupus erythematosus serum demonstrated with fluorescent antibody. J Immunol 1958b: 80: 224-229.
GGavalchin J & Datta SK. The NZB / SWR model of lupus nephritis. II. Autoantibodies deposited in renal lesions show a distinctive and restricted idiotype diversity. J Immunol 1987: 138: 138-148.
Gathings WE, Lawton AR, Cooper M. Immunofluorescent studies of the development of the pre -B cells, B lymphocytes and immunoglobulin isotype diversity in humans. Eur J Immunol 1977; 2: 804-810.
Geha RS. Idiotypic determinants on human T cells and modulation of the human T cell responses by anti-idiotypic antibodies. J Immunol 1984: 133: 1846-1851.
Gharavi AE, Harris EN, Asherton RA, Hughes GR. Anticardiolipin antibodies: Iso type distribution and phospholipid specificity. Ann. Rheum. Dis 1987; 46: 1-6.
Gilkeson GS, Bloom DB, Pisetsky DS, Clarke SH. Molecular characterisation of anti-DNA antibodies induced in normal mice by immunisation with bacterial DNA. J Immunol 1993: 151: 1353- 1364.
Glotz D, Sollozzo M, Riley S, etal. VH genes and antigen binding analysis of hybridomas of new bom Balb/c mice. J Immunol 1988; 141: 383-390.
Graninger WB. Transcriptional overexpression of the proto- oncgene Bcl-2 in patients with systemic lupus erythematosus. Wien Klin Wochenschr 1992; 104: 205-207.
Graninger WB, Seto M, Boutain B, Goldman P, Korsmeyer SJ. Expression of Bcl-2 amd Bcl-2-Ig fusion transcripts in normal and neoplastic cells. J Clin Invest 1987; 1512-1515.
Griffiths GM & Milstein C. ‘The analysis of structural diversity’ in the ‘Antibody Response’ Hybridoma technology in the biosphere edited by Springer TA (Plenum publishing).
198
Gninstein M & Hogness DS. Colony hybridization; A method for the isolation of cloned cDNAs that contain a specific gene. PNAS USA 1975:22: 3961-3965.
Guigou V, Cuisinier AM, Tonnelle C et al. Human immunoglobulin VH and VK repertoire revaeled by in situ hybridisation. Mol Immunol 1990; 2%: 935-940.
Guillaume T, Rubenstein DB, Young F et a l Individual VH genes detected with olignucleotide probes from the complimentarity- determining regions. J Immunol 1990: 145: 1934-1945.
HHahn B & Ebling P.M. Suppression of murine lupus nephritis by administration of an anti-idiotypic antibody to anti-DNA. J Immunol 1984; 122: 1871-1901.
Hahn BH & Ebling FM. Idiotypic restriction in murine lupus: High frequency of three public idiotypes on serum IgG in nephritic NZB/NZW FI mice. J Immuno\ 1987: 138: 2110-2118.
Harris EN. Antiphospholipid antibodies. Br. J. Haematol 1990; 24: 1-9.
Harris EN, Chan JK, Asherson RA eta l. Thrombosis recurrent fetal loss and thrombocytopenia. Predictive value of the anti cardiolipin antibody test. Arch Intern Med 1986; 146: 2153-2156.
Harris EN, Gharavi AE, Hughes GR. Anti-phospholipid antibodies. Clin Rheum Pis 1985: 11: 591-609.
Harris EN, Gharavi AE, Tincani A e ta l. Affinity purified anti cardiolipin and anti-DNA antibodies. J Clin Lab Immunol 1983; 12: 155-162.
Hartman AB, Mallett CP, Sniivsappa J et al. Organ reactive autoantibodies from non-immunized adult BALB/c mice are polyreactive and express non-bias VH gene usage. Mol Immunol 1989; 26: 359-370.
Hartman AB & Rudikoff. VH genes encoding the immune response to beta-(l,6)-galactan: Somatic mutation in IgMmolecules. EMBO 1984; 1: 3023-3030.
Herron IN, He XM, Ballard DW etal. An autoantibody to ssDNA: A comparison of the three-dimentional structures of the unliganded Fab and a deoxynucleotide- Fab complex. Prot Struct Funct Genet 1991; II: 159-63.
Hieter P A, Hollis G F, Korsmeyer S J, Waldmann T A, Leder P. Clustered rearrangement of immunoglobulin \ constant region genes in man. Nature 1981: 294: 536-40.
Hirabyashi Y, Munakato Y, Takai O et a l Human B cell clones expressing lupus nephritis-associated anti-DNA idiotypes are preferentially expanded without somatic mutation. Scand J Immunol 1993; 22: 533-540.
199
Hoch S & Schwaber J. Identification and sequence of the VH gene elements encoding a human anti-DNA antibody. J Immunol 1987; 139: 1689-1693.
Huang C, Stewart AK, Schwartz RS, Stollar BD. Immunoglobulin heavy chain gene expression in peripheral blood B lymphocytes. J Clin Invest 1992; &R: 1331-1343.
Huck S, Keyeux G, Ghanem N, LeFranc MP, LePrank G. A gamma 3 hinge region probe: First specific human immunoglobulin subclass probe. FEES Letters 1985: 208: 221-230.
Hughes GR, Harris EN, Gharavi AE. The anticardiolipin syndrome. 1986; 11: 486-489.
IIchihara Y, Aba M, Yasui H, Matsuoka H, Kurosawa Y. At least five DH genes of human immunoglobulin heavy chains are encoded in 9 kilobase DNA fragments. Eur J Immunol 1988a; i£ : 649-652.
Ichihara Y, Matsuoka H, Kurosawa Y. Organisation of human immunoglobulin heavy chain diversity loci. EMBO 1988b; 2: 4141-4150.
Isenberg DA, Dudeney C, Wojnaruska F et al. Detection of cross reactive anti-DNA antibody idiotypes on tissue bound immunoglobulins from skin biopsies of lupus patients. J Immunol 1985a; m 261-264.
Isenberg DA, Ehrenstein M, Longhurst C, Kalsi T. The origin, sequence, structure and consequences of developing anti-DNA antibodies. Arth & Rheum 1994; 37: 169-180.
Isenberg DA & Collins C. Detection of crossreactive anti-DNA antibody idiotypes on renal tissue bound immunoglobulins from lupus patients. J Clin Invest 1985b; 26: 287-294.
Isenberg DA ,Shoenfeld Y, Madaio MP et al. Anti-DNA antibody idiotypes in systemic lupus erythematosus. Lancet 1984; 2: 418- 22 .
Isenberg DA, Williams W, Axford J etal. Comparison of dna anti body idiotypes in human sera; an international collaborative. J Autoimmunity 1990; 2i 339-414.
Isenberg DA, Katz D ,LePage S, etal. Independant analysis of the 16/6 idiotype lupus model. J Immunol 1991:147: 4172-4177.
Isenberg DA, Spellerberg M, Williams W, Griffiths M, Stevenson F. Identification of the 9G4 idiotype in SLE. B J Rheumatol 1993; 32i 876-882.
200
JJaenichen HR, Pech M, Lindermaier W, et al. Composite V k genes and a model of their evolution. Nucl Acid Res 1984; 12: 5249- 5263.
Jennette JC & Faulk RJ. Antineutrophil cytoplasmic antibodies and associated diseases. Am J Kidney Dis 1990; Ü : 517-529.
Jeme NK. Towards a network of the immune system. Ann Immunol (Inst Pasteur) 1974; 125c: 373-389.
KKabat EA, Wu TT, Reid-Miller M, Perry H, Gottesman KS. (1987) Sequences of immunological interest. NIH Publication Bethesda MD
Kabat EA, Wu TT, Perry H, Gottesman KS, Foeller. C. (1991) Sequences of proteins of immunological interest. US Department of health and human services, Bethesda Maryland.
Kallenberg CGM, Cohen-Tervaert JW, Van der Woude FJ et al. Autoimmunity to lysosomal enzymes: New clues to vasculitis and glomerulonephritis. Immunol Today 1992a: 12: 61-64.
Kallenberg CGM, Leontine Mulder AH, Cohen-Tervaert JW. Antineutrophil cytoplasmic antibodies: A still growing class of autoantibodies in inflammatory diseases. Am J Med 1992b; 93: 675-682.
Kallenberg CGM, Brouwer E, Weening JJ, Cohen Tervaert JW. Anti-neutrophil cytoplasmic antibodies: Current diagnostic and pathophysiological potential. Kid Int 1994: 46: 1-15.
Kalsi J, Ravirajan C, Winska-Wiloch H etal. Generation and analysis of three idiotypes associated with autoimmune disease. 1994a in press.
Kalsi J, Martin A, Hirabayashi Y et al. Function and structural analysis of the binding of three human monoclonal anti DNA antibodies. 1994b in press.
Kalunian KC, Panosian-Sahakian N, Ebling FM et al. Idiotypic characterisitics of immunoglobulins associated with systemic lupus erythematosus: Studies of antibodies deposited in the glomeruli of humans. Arth & Rheum 1989: 25: 513-522.
Katsuma M, Siegel RM, Louie DC etal. Differential effects of Bcl- 2 on T and B cells in transgenic mice. Proc Natl, Acad Sci. USA. 1991; H : 11376-11380.
Katz JB, Limpanasithikul W, Diamond B. Mutational analysis of an autoantibody: Differential binding and pathogenicity. J Exp Med 1994; IM : 925-932.
201
Kehoe JM & Capra JD. Localisation of two additional hypervariable regions in immunoglobulin heavy chains. Proc Natl Acad Sci USA 1971; M: 2019-2021.
Khamashta MA, Asherson RA, Hughes GR. Possible mechanisms of action of the antiphospholipid binding antibodies. Clin Exp Rheumatol 1989; 2: 585-589.
Kipps TJ, Tomhave E, Pratt LF. Developmentally restricted VH genes is expressed at high frequency in CD5 B cell malignancies. Proc Natl Acad Sci USA 1989 M: 5913-5917.
Kipps TJ, Tomhave E, Chen PP. Autoantibody associated J light chain variable region expressed in CLL with little or no somatic mutation, implications for etiology and immunotherapy. J Exp Med 1988a: 167: 840-852.
Kipps TJ, Robbins BA, Kusten P. Autoantibody associated crossreactive idiotypes expressed at high frequency in CLL relative to B cell lymphomas of folicullar centre cell origin. Blood 1988b; 22: 422-428.
Kirsch IR Morton CC Nakahara K Leder P. Human immunoglobulin heavy chain genes mapped to a region of translocations in malignant B lymphocytes. Science 1982; 216: 301-3.
Kishimoto T, Okajima H, Okumoto T. Nucleotide sequence of the cDNAs encoding the V-regions of the H- and L- chains of a human monoclonal antibody. Nucleic Acid Res 1989; 12: 4385.
Kocks C & Rajewsky K. Stepwise intraclonal maturation of antibody affinity through somatic hypermutation. Proc Natl Acad Sci USA 1988; Ei: 8206-8210.
Kocks C & Rajewsky K. Stable expression and somatic hypermutation of antibody V regions in B-cell developmental pathways. Annu Rev Immunol 1989; 26: 537-559.
Kodaira M, Kinashi T, Umemura T. Organisation and evolution of variable region genes of the human immunoglobulin heavy chain. J Mol Biol 1986: 190: 529-541.
Koffler RD, Schur PW, Kunkel HG. Immunological studies concerning the nephritis of systemic lupus erythematosus. J Exp Med 1961 ;12Â: 607-624.
Koffler RD, Dixon FJ, Theofilopoulos AN. The genetic origin of autoantibodies, [review] Immunol Today 1987; gD: 375-380.
Koffler RD, Noonan DJ, Levy DE etal. Genetic elements used for a murine lupus anti-DNA autoantibody are closely related to these for antibodies to exogenous antigen. J. Exp Med 1985; 161: 805- 815
Komisar JL, Yeung KY, Crawley RC, Talal N, Teale JM. IgVH family repertoire of plasma cells derived from lupus prone MRL/lpr and MRL/++ mice. J Immunol 1989; 143: 340-347.
202
Kubagawa H, Cooper MD, Carroll AJ, Burrows PD. Light chain gene expression before heavy chain gene rearrangement in pre-B- cells transformed by Epstein Barr virus. Proc Natl Acad Sci 1989; M: 2356-2360.
LLafer EM, Rauch J, Andrzejewski C et al. Polyspecific monoclonal lupus autoantibodies reactive with both polynucleotides and phospholipid. J Exp Med 1981: 153: 897-909.
Lahita RG, Chiorazzi N, Gibofsky A etal. Familial systemic lupus erythematous in males. Arthritis Rheum 1983; 26: 39-44.
Lai E, Wilson R, Hood L. Physical maps of the mouse and human immunoglobulin-like loci. Adv Immunol 1989; 46: 1-59.
Lambert PH, Dixon FJ. Pathogenesis of glomerulonephritis of NZBAVmice. J Exp Med \96%\ U l:. 507-522.
Lampman G, Furie B, Schwartz K etal. Amino acid sequence of a platelet binding human monoclonal anti-DNA antibody. Blood 1989; 24: 262-269.
Lane D, Koprowski H. Molecular recognition and the future of monoclonal antibodies. Nature 1982; 296: 200-202.
Lassoued S, Sixou L, Okman F, Pages M, Fournie A. Antineutrophil cytoplasmic antibodies and antibodies to myeloperoxidase in rheumatoid arthritis. Arthritis Rheum 1991; 34: 1069-1070.
Lautner-Rieske A Huber C Alfons M et al. The immunoglobulin K locus. Characterization of the duplicated A regions. Eur J Immunol 1992; 22: 1023-1029.
Ledford DK, Goni F, Pizzolata M, etal. Preferential association of kappa Illb light chains with monoclonal human IgM kappa autoantibodies. J Immunol 1983: 131: 1322-1325.
Lee JS, Lewis JR, Morgan AR, Mosman TR, Singh B. Monoclonal antibodies showing sequence specificity in their interaction with single stranded DNAs. Nucleic acid Res 1981; £: 701-721.
Lee KH, Matsuda F, Kinashi T et al. A novel family of variable region genes of the human immunoglobulin heavy chain. J Mol Rio/ 1987; m 761-768.
Leiber MR. The mechanism of V(D)J recombination: A balance of diversity, specificity and stability. Cell 1992; 2Ü: 873-876.
Lesavre P. Antineutrophil cytoplasmic autoantibodies antigen specificity Am J Kidney Dis 1991; 1&: 159-163.
Levitt D, Cooper MD. Mouse pre-B cells synthesize and secrete p heavy chains but not light chains.Cg//1980; I^: 617-25.
203
Levy NS, Malipiero UV, Lebecque PJ, Gearhart FJ. Early onset of somatic mutation in immunoglobulin VH genes during the primary immune response. J Exp Med 1989; 169: 2007-2019.
Livneh A, Preudhomme JI, Soloman A. Preferential expression of the SLE associated idiotype 8.12 in sera containing monoclonal immunoglobulins. J Immunol 1987a: 139: 3730-3733.
Livneh A, Halpem A, Perkins D etal. A monoclonal antibody to a cross reactive idiotype on cationic human anti-DNA antibodies expressing L light chains: a new reagent to identify a potentially differential subset. J Immunol 1987b: 138: 123-127.
Logtenberg T, Young FM, Van Es JH et al. Autoantibodies encoded by the most JH proximal human immunoglobulin heavy chain variable region. J Exp Med 1989; 170: 1347-1355.
Longhurst C, Eherenstein MR, Leaker B, Latchman D, Isenberg D. Analysis of immunoglobulin variable region genes of a human IgM anti-myeloperoxidase antibody derived from a patient with vasculitis. Kidney Int 1994; in press.
Love PE & Santoro SA. Antiphospholipid antibodies; anticardiolipin and the lupus anticoagulant in SLE and non-SLE disorders. Prevalence and clinical significance. Ann Intern Med 1990; 112: 682-98.
Ludemann J, Utecht B, Gross WL. Anti—neutrophil antibodies in Wegner’s granulomatosis recognise an elastinolytic enzyme. J ExpMed\99Q\ni: 357-362.
Ludvico CL, Zweiman B, Myers AB, Herbert J, Green PA. Predictive value of anti-DNA antibody and selected laboratory studies in systemic lupus erythematosus. J Rheumatol 1980; %: 843-9.
Lui Y-J, Joshua DE, Williams GT etal. Mechanism of antigen- driven selction in germinal centres. Nature 1989: 342: 929-931.
MMackworth-Young CG, Harris EN, Steere AC et al. Anticardiolipin antibodies in Lyme disease. Arthritis Rheum 1988; &: 1052-1056.
MaddisonPJ. ANA-negative SLE. Clin Rheum Dis 19S2, 105-119.
Mageed RA, Dearlove M, Goodall DM et al. Immunogenic and antigenic epitopes of immunoglobulins XVIII monoclonal anti idiotypes to the heavy chain of human rheumatoid factors. Rheumatol Int 1986; 173-179.
Maki R, Kearney J, Paige C, Tonegawa S. Immunoglobulin gene rearrangement in immature B cells. Science 1980: 209: 1366-69.
Male DK. Idiotypes and autoimmunity. Clin Exp Immunol 1986; M: 1-9.
204
Manheimer-Lory A, Katz JB, Pillinger M et al. Molecular characteristics of antibodies bearing an anti-DNA-associated idiotype. J Exp Med 1991: 174: 1639-1652.
Maniatis T, Fritsch EE, Sambrook J. Molecular cloning-A laboratory manual. Cold Spring Harbor Lab. Cold Spring Harbor, New York. 1982.
Manser T. The intraclonal diversity and control of antibody isotype switch recombination. Inti Immunol 1990; %: 893-902.
Manser T, Parmani-Seren M, Margolies M, Gefter ML. Somatically mutated forms of a major anti-p-azophenylarsonate. J Exp Med 1987; IM : 1456-1463.
Mathyssen G & Rabbitts T. The Structure and multiplicity of genes for the human immunoglobulin heavy chain variable region. Proc Natl Acad Sci V S A \9 m \I l: 6561-5.
Matsuda F Shin EK, Hirabayashi Y etal. Organisation of variable region segments of the human immunoglobulin heavy chain duplication of the D5 cluster within the locus and interchromosomal translocation of variable region segments. EMBO 1990; £: 2501- 2506.
Matsuda F, Shin EK, Nagaoka H. etal. Structure and physical map of 64 variable segments in the 3’ 0.8 megabase region of the human immunoglobulin heavy chain gene locus. Nature Genet 1993; 2: 88-94.
MatsudaF, Lee KH, Nakai S, etal. Dispersed localisation of D segments in the human immunoglobulin heavy chain locus. EMBO 1988;%: 1047-1051.
Matsumoto K, Yoshikai Y, Asano T et al. Defect in negative selection in Ipr donor-derived T cells differentiating in non-lpr host thymus. J Exp Med 1991: 173: 127-136.
Mcbride OW Batte y J, Hollis G et al. Localisation of human variable and constant region immunoglobulin heavy chain genes on sub-telomeric band q32 of chromosome 14. Nucl Acids Res 1982; ID: 8155-8169.
McDonnel JJ, Deane N, Platt FM et al. Bcl-2 immunoglobulin transgenic mice demonstrate extended B cell survival and follicular lymphoproliferation. Cell 1989; 79-88.
McHeyzer-Williams MG & Nossal GJV. Clonal analysis of autoantibody producing cell precursors in preimmune repertoire. J Immunol 1988: 141: 4118-4123.
McHeyzer-Williams MG,.Nossal GJV, Lai or PA. Nature 1991; 350: 502-505.
Medsger TA, Dawson WN, Masi AT. The epidemiology of polymyositis. Am J Med 1970; 4&: 715-23.
205
Meek KD, Eversole T, Capra JD. Conservation of the most JH proximal immunoglobulin VH gene segment VHVI throughout primate evolution. 1991: 146: 2434-38.
Meroni P, Harris EN, Brucato A etal. Anti-mitochrondrial type M5 and anticardiolipin antibodies in autoimmune disorders; studies on their association and cross reactivity. Clin Exp Immunol 1987: 67: 484-491.
Miescher P & Straessle R. New serological methods for the detection of LE factor. Vbx 5a/ig 1957; 2* 283-287.
Miyawaka T, Uehara T, Nibu R et al. Differential expression of apoptosis-related Fas antigen on lymphocyte subpopulations in human peripheral blood. J Immunol 1992: 149: 3753-3758.
Monestier M, Easy TM, Losman MJ, Novick KE, Muller S. Structure and binding properties of monoclonal antibodies to core histones from autoimmune mice. Mol Immunol 1993; 2Û- 1069- 1075.
Monestier M, Bonin B, Migliorini P et al. Autoantibodies of various specificities encoded by genes from the VH J558 family bind to foreign antigens and share idiotypes of antibodies specific for self and foreign antigens. J Exp Med 1987; 166: 1109-1124.
Mortari F, Newton JA, Wang JY, Schroeder HW. Human cord blood antibody repertoire frequent usage of VH7 gene family. Eur J Immunol 1992: 22: 241-245.
Muller S, Barakat S, Watts R, Jouband P, Isenberg D. Longitudinal analysis of antibodies to histones SM d peptides and ubiquitin in serum of patients with SLE, RA and tuberculosis. Clin Exp Rheumatol 1990; &: 445-453.
Murphy ED & Roths JB. Autoimmunity and lymphoproliferation: induction by mutant gene Ipr and acceleration by a male associated factor in strain BXSB. In Rose NR, Bigazzi PE, Warner NL. ed Control of autoimmune diseases. North Holland, New York: Elsevier. 1979: 207-220.
Muryoi T, Sasaki,T, Harata,N et al. Heterogeneity of anti- idiotypic antibodies to anti-DNA antibodies in humans.,Clin Exp Immunol. 1988; 2 il (1) 67-72.
NNaparstek Y, Duggan D, Schattner A. Immunochemical similarities between monoclonal anti-bacterial Waldenstroms macroglobulins and monoclonal anti DNA lupus autoantibodies. J Exp Med 1985; 161 1525-1538.
Newkirk MM, Mageed RA, Jefferis K etal. Complete amino acid sequences of variable regions of two human IgM rheumatoid factors BOR andKAS of the Wa idiotypic family, reveal restricted use of heavy and light chain variable and joining region gene segments. J Exp Med m i ; m .: 550-564.
206
Newkirk NM & Capra JD. Restricted usage of immunoglobulin variable region genes in human autoantibodies; in The Immunoglobulin gene. T. Honjo TH. Rabbitts & FW. Alt NewYork Academic Press 1989; chap 11: 203-219.
Newport JW & Forbes DJ. The nucleus structure function and dynamics. Ann Rev Biochem 1987; 535-565.
Nickerson KG, Berman J, Glickman E. Early human immunoglobulin heavy chain gene assembly in EBV transformed fetal B lines. Preferential utilisation of the most JH proximal D segment DQ52 and two unusual VH related rearrangements. J Exp M e d m 9 \m : . 1391-1403.
Nossal GJV. The molecular and cellular basis of affinity maturation in the antibody response. Cell 1992; 1-2.
O
Osmond DG. The turnover of B cell populations. Immunol Today 1993; 14: 34-37.
Pabo CO, Sauer RT. Protein-DNA recognition. Annu Rev Biochem 1984: 53: 293-321.
Pankewycz OG, Migliorini P, Madaio MP. Polyreactive autoantibodies are nephritogenic in murine lupus nephritis. J Immunol 1987; 139: 3287
Pascual V, Randen 1, Thompson K et al The complete nucleotide sequences of the heavy chain variable regions of six monospecific rheumatoid factors derived from EBV transformed B cell isolated from the synovial tissue of patients with rheumatoid arthritis: Further evidence that some autoantibodies are unmutated copies of germline genes. J Clin Invest 1990: 86: 1320-28.
Pascual V, Victor K, Spellerberg M et al. VH restriction among human cold agglutinins. J Immunol 1992: 149: 2337-2344.
Paul E, Diev AA, Livneh A, Diamond B. The anti-DNA-associated idiotype 8.12 is encoded by the VXll gene family and maps to the vicinity of L chain CDRl. J Immunol 1992; 149: 3588-3595.
Pech M & Zachau HG. Immunoglobulin genes of different subgroups are interdigitated within the V k locus. Nucleic Aci Res 1984; 12: 9229-9236.
Pech MN, Jaenichen HR, Polhenz DH et al. Organization and evolution of a gene cluster for human immunoglobulin variable region of the kappa type. J Mol Biol 1984; 176: 189-204.
Perlmutter RM, Kearney JF, Chang SP, Hood LE. Developmentally controlled expression of VH genes. Science 1985; 231: 1597-1601.
207
Pernis BG, Chiappino G, Kelus AS, Gell PGH. Cellular localisation of immunoglobulins with different allotype specificities in rabbit lymphoid tissue. J Exp Med 1965; 122: 853-75.
Polijak RJ, Amzel LM, Avey HP eta l. The three dimensional structure of the Fab’ fragment of a myeloma immunoglobulin at 2.0 A resolution. Proc Natl Acad Sci USA 1974: 71: 3440-3444.
Potter KN, Li Y, Pascual V et al. Molecular characterisation of a cross-reactive idiotype on human immunoglobulins utilizing the VH4-21 gene segment J Exp Med 1993; 178: 1419-1428.
RRabbitts TRH, Forster A, Milstein CP. Human immunoglobulin heavy chain genes: Evolutionary comparisons of C mu, C delta and C gamma genes and associated switch sequences. Nucl Acids Res 1981; 4: 4509-4524.
Radie MZ, Mascelli A, Erickson J, Shan H, Weigert M. Ig H and L chain contribution to autoimmune specificities. J. Immunol 1991; 146: 176-182.
Radie MZ, Mackle J, Erikson J et al. Residues that mediate DNA binding of autoimmune antibodies. J Immunol 1993; 150: 4966- 4977.
Radac C, Gupta SK, Gherardi E, Milstein C. Mutation and selection during the secondary response to 2-phenyl oxazalone. Proc Natl Acad Set USA 1991 ; M- 5508-5512.
Raff MC. Social controls on cell survival and cell death. Nature 1992; m 397-400.
Rajewsky K & Takemori T. Genetics, expression and function of idiotypes. Ann Rev Immunol 1983; 1: 569-607.
Rajewsky K, Forster I, Cumano A. Evolutionary and somatic selection of the antibody repertoire in the mouse. Science 1987; 238: 1088-1094.
Rauch J, Tannenbaum H, Stollar BD, Schwartz RS. Monoclonal anti-cardiolipin antibodies bind to DNA. Eur J Immunol 1984; 14: 529-534.
Rauch J, Tannenbaum H, Straaton K, Massicote H, Wild J. Human/human hybridoma autoantibodies with both anti-DNA and rheumatoid factor activities. J Clin Invest 1986; 22* 106-112.
Ravetch JV, Siebenlist U, Kosmeyer S, Waldmann T, Leder P. Structure of the human immunoglobulin p locus: Characterisation of embryonic and rearranged J and D genes. Cell 1981: 27: 583- 591.
Ravirajan C, Kalsi J, Winska Wiloch H et al. Antigen binding diversity of human hybridoma autoantibodies derived from splenocytes of patients with SLE. Lupus 1992; 1: 167-174.
208
Raz E, Brezis M, Rosemann E, Eilat D. Anti-DNA antibodies bind directly to renal antigens and induce kidney dysfunction in the isolated perfused rat kidney. J Immunol 1989; 142: 3076-3082.
Rechavi G, Ram D, Glazer L, Zakut R, Givol D. Evolutionary aspects of the immunoglobulin heavy chain variable region (VH) gene subgroups. Proc Natl Acad Sci USA 1983; BO: 283-291.
Reichlin M & MattioUi M. Correlations of a precipitin reaction to an RNA protein antigen and a low prevalence of nephritis in patients systemic lupus erythematosus. N Engl J Med 1972; 286: 908-911.
Reichlin M. Clinical and immunological significance of antibodies to Ro and La in Systemic Lupus Erythematosus. Arthritis Rheum 1982; 25: 767-771.
Reth M, Hombach J, Wienands J etal. The B-cell antigen receptor complex. Immunol Today 1991: 12: 196-201.
Rioux JD, Rauch J, Zolarsky E, Newkirk MM. Molecular characterisation of GM4672 human lymphoblastoid cell line and analysis of its use as a fusion partner in the generation of human- human hybridoma autoantibodies. Hum Antibod Hybridomas 1993; 4: 107-114.
Robbins WC, Holman HR, Deicher H, Kunkel HG. Complement fixation with cell nuclei and DNA in lupus erythematosus. Proc Soc Exp Bio Med 1957; 575-579.
Rose LM, Latchman DS, Isenberg DA. Bcl-2 expression is unaltered in unfractionated peripheral blood mononuclear cells in patients with systemic lupus erythematosus. Submitted 1994.
Royston I, Majda JA, Baird SM. Human T cell antigens defined by monoclonal antibodies. The 65KD antigen on T cells T65 is also found on CLL cells bearing surface immunoglobulin. J Immunol 1980; 125: 725-731.
Rubinstein DB„ Symann M, Stewart AK, Guillaume T. Restriction fragment length polymorphisms and single germline coding sequence in VH18/2 a duplicated gene encoding autoantibody. Mol Immunol 1993; 2Û1403-12.
SSabbaga J, Pankewycz OG, Lufft V, Schwartz RS, Madaio DP. Crossreactivity distinguishes serum and nephritogenic anti-DNA antibodies in human lupus from their natural counterpart in normal serum. J Autoimmunity 1990; 2: 215
Saito B, Uchiyama K, Nakamura I. Characterisation of a human monoclonal antibody with broad reactivity to malignant tumor cells. JNatl Cane Inst 1988; BÛ: 728-734.
Sakano H, Maki R, Kurosawa Y, Roeder W, Tonegawa S. Two types of somatic recombination are necessary for the generation of complete immunoglobulin heavy chain genes. Nature (London) 1980; 2M: 676-683.
209
Sakano H, Kurosawa Y, Weigert M, Tonegawa S. Identification and nucleotide sequence of a diversity DNA segment (D) of immunoglobulin heavy chain genes. Nature (London) 1981: 290: 562-565.
Sammaritano LR, Gharavi AE, Loakshin MD. Antiphospholipid antibody syndrome. Immunologic and clinical aspects. Sent in Arthritis Rheum 1990: 20: 81-96.
Sanger F, Milken S, Coulson AR. DNA sequencing with chain terminating inhibitors. PNAS USA 1977; 24: 5463-5467.
Sanz 1, Dang H, Takei M. VH sequence of a human anti-SM autoantibody: Evidence that autoantibodies can be unmutated copies of germline genes. J Immmunol 1989a: 142: 883-887.
Sanz 1, Wang SS, Meneses G, Fischbach M. Molecular characterisation of human Ig heavy chain DIR genes. J Immunol 1994; 152: 3958-3969.
Sanz 1. Multiple mechanisms participate in the generation of diversity of human heavy chain CDR3 regions. J Immunol 1991; 147: 1720-29.
Sanz T, Casoli P, Thomas JW etal. Nucleotide sequences of eight natural autoantibody VH regions reveals apparent restricted use of VH families. J Immunol 1989b: 142: 4054-4061.
Sanz 1, Kelly P, Williams C et a l The smaller human gene families display remarkably little polymorphism. EMBO 1989c; 3741-3748.
Sasso EH, Wiilems van Dijk K, Millner ECB. Prevalence and polymorphism of human VH3 genes. J Immunol 1990; 142: 2751- 57.
Sato T, Matsuda F, Lee KH, Shin EK, Honjo T. Physical linkage of a variable region segment and the joining region segment of the human immunoglobulin heavy chain locus. Biochem. Biophys. Res. Commun. 1988: 154: 265-71.
Savige JA, Gallicchio MC, Stockman A, Cunningham TJ, Rowley MJ. Anti-neutrophil cytoplasmic antibodies in Rheumatoid Arthritis. Clin Exp Immunol 1991: 86: 92-98.
Schable KF and Zachau HG. The variable genes of the human immunoglobulin kappa locus. Biol Chem 1993; 374: 1001-1022.
Schatz DG. V(D)J recombination. Molecular biology and regulation. Annu Rev Immunol 1992; 359-383.
Schilling J, Clevinger B, Davie JM, Hood L. Amino acid sequence of homogeneous antibodies to dextran and DNA rearrangements in heavy chain V region gene segments. Nature (London ) 1980; 283: 35-40.
210
Schlissel MS, Corcoran LM, Baltimore D. \^rus transformed pre-B cells show ordered activation but not inactivation of immunoglobulin gene rearrangement and transcription J Exp Med 1991; i l l : 711-20.
Schmitt WH, Csernok E, Gross WL. ANCA and infection. Lancet 1991: 337: 1416-1417
Schroeder HW, Walker MA, Hofker MK, etal. Physical linkage of a new human immunoglobulin VH gene family (VH6) to DH. PNAS [/^A 1988; 8196-8200.
Schroeder HW, Hillson JL, Perlmutter RW. Early restriction of the human antibody repertoire. Science 1987: 238: 791-793.
Schroeder HW & Wang JY. Preferential utilisation of the conserved immunoglobulin heavy chain variable gene segments during human fetal life. Proc. Natl Acad Sci USA 1990b: 87: 6146-6150.
Schur PH & Monroe M. Antibodies to ribonucleic acid in systemic lupus erythematosus. Proc Natl Acad Sci USA 1969; 61: 1108- 1112.
Seeman NC, Rosenberg JM, Rich A. Sequence specific recognition of double helical nucleic acids by proteins. Proc. Natl. Acad. Sci. USA 1976; 21: 1745-48.
Seigneurin JM, Guilbert B, Bourgeat MJ, Avrameas S. Polyspecific natural antibodies and autoantibodies secreted by human lymphocytes immortalised with Epstein-Barr virus. Blood 1988; 21: 581-585.
Seising E, Durdik J, Moore MW, Persiani DM. Immunoglobulin lambda genes Immunoglobulin Genes Ed T Honjo FW Alt TH Rabbitts 1989; 111-12 (London Academic).
Shan H, Shlomchik M, Weigert M. Heavy chain class switch does not terminate somatic mutation. J Exp Med 1990; 172: 531-536.
Sharon J, Gefter ML, Wtsocki LJ et al. Recurrent somatic mutations in mouse antibodies to p-azophenylarsonate increase affinity for hapten. J Immunol 1989: 142: 596-601.
Sharp GC, Irvine WS, Laroque RL et al. Association of autoantibodies to different nuclear antigens with clinical patterns of rheumatic disease and responsiveness to therapy. J Clin Invest 1971; m 350-359.
Shefner R, Kleiner G, Turken A, Papazian L, Diamond B. A novel class of anti-DNA antibodies identified in BALB/c mice J Exp Med 1991; 121: 287-296.
Shen A, Humphries C, Tucker PW. Human heavy chain V region gene family non randomly rearranges in familial CLL. Proc Natl Acad Sci USA m i \ M : 8563-8567.
Shimuzu A & Honjo T. Immunoglobulin class switching. Cell 1984; 16: 801-803.
211
Shin EK Matsuda F, Fujikura J et al. Cloning of a human immunoglobulin fragment containg both VH-D anmd D-JH rearrangements: Implication for VH-D as an intermediate to VH-D- JH formation. Eur J Immunol 1993: 23: 2365-2367.
Shin EK Matsuda F, Nagaoka H eta l. Physical map of the 3’ region of the human immunoglobulin heavy chain locus: Clustering of autoantibody-related varible segments in one halotype. E M B O J \9 9 \\m 3641-3645.
Shlomchik MJ, Aucion AH, Pistsky DS, et al. Structure and function of anti DNA autoantibodies derived from a single autoimmune mouse. Proc Natl Acad Sci USA 1987; M: 9150- 9154.
Shlomchik M, Mascelli M, Shan H et al. Anti-DNA antibodies from autoimmune mice arise by clonal expansion and somatic mutation. J Exp Med \990\ 171: 265-92.
Shoenfeld Y & Schwartz RS. Heamological manifestations of SLE. In The Clinical management of systemic lupus erythematosus (1983c). Schur P ed. Grune & Stratton New York 123-136.
Shoenfeld Y ,Isenberg DA, Rauch J et al. Idiotypic cross-reactions of monoclonal human lupus autoantibodies. J Exp Med 1983b; 158: 717-730.
Shoenfeld Y, Rauch J, Massicote H, et al. Polyspecificity of monoclonal lupus autoantibodies produced by human/human hybridomas. N EnglJ Med 1983a; 308: 414-420.
Siebenlist V, Ravetch JV, Korsmeyer S, et al. Human immunoglobulin D segments encoded in tandem multi gene families. Nature 1981: 294: 631-635.
Silverman GL & Lucas AH. Variable region diversity in human circulating antibodies specific for the capsular polysaccaride of heamophilus-influenzae type-b preferential usage of two types of VH3 heavy chains. J Clin Invest 1991; &&: 911-920.
Siminovitch KA, Misner V, Kwong PC et al. A human autoantibody is encoded by developmentally restricted heavy and light chain V regions. Autoimmunity \990\^\ 97-106.
Smeenk RJT, Brinkman K, van den Brink HG etal. Reaction patterns of monoclonal antibodies to DNA. J Immunol 1988; 140: 3786-3792.
Smith CA, Williams GT, Kingston R, Jenkinson EJ, Owen JT. Antibodies to CD3/T-cell receptor complex induce death by apoptosis in immature T cells in thymic cultures. Nature 1989; 337: 181-184.
Soloman G, Schiffenbauer J, Keiser HD etal. Use of monoclonal antibodies to identify shared idiotypes on human antibodies to native DNA from patients with systemic lupus erythematosus. Proc Natl Acad Sci USA m y , m : 850-854.
212
Sonntag D, Weingartner B, Grutzmann R. A member of a novel human DH gene family DHFL16. Nucleic Acids Res 1989; 17: 1267.
Souroujon M, White W, Scharf ME, Andre-Schwartz J. Preferential autoantibody reactivity of preimmune B cell repertoire in normal mice. J Immunol 1988: 140: 1473-1479.
Spatz LA, Wong KK, Williams M et a l Cloning and sequence analysis of the and VL regions of an anti-myelin/DNA antibody from a patient with peripheral neuropathy and chronic lymphocytic leukemia. 7. Immunol 1990: 144: 2821-2828.
Stevenson F, Longhurst C, Chapman CJ e ta l . Utilisation of the VH4-21 gene segment by anti-DNA antibodies from patients with systemic lupus erythematosus. J Autoimmu 1993; è.- 809-825.
Stevenson F, Wrightham M, Glennie MJ etal. Antibodies to shared idiotypes as agents for analysis and therapy for human B cell tumors. Blood 1986; 430-36.
Steward MW & Hay FC. Changes in imunoglobulin class and subclass of anti-DNA antibodies with increasing age in NZB/W FI hybrid mice. Clin Exp Immunol 1976; 26: 363-370.
Stewart AK, Huang C, Stollar BD, Schwartz RS. High-frequency representation of a single VH gene in the expressed human B cell repertoire. J Exp Med 1993: 177: 409-418.
Stewart AK, Huang C, Long A A, Stollar BD, Schwartz RS. VH gene representation in autoantibodies reflects the normal human B- cell repertoire. Immunol Rev 1992: 128: 101-122.
Straubinger B, Thiebe R, Huber C, Osterholzer E, Zachau HG. Two unusual human V kappa genes. Biol Chem Hoppe-Seyler 1988; 26£: 601-607
Stollar BD. Molecular analysis of anti-DNA antibodies. FASEB 1994; S: 337-342.
TTakahashi N, Noma T, Honjo T. Rearranged immunoglobulin heavy chain variable region that deletes the second complementarity determining region. Proc Natl Acad Sci USA 1984; M: 5194- 5198.
Takahashi N, Ueda S, Obata M et a l Structure of human immunoglobulin gamma genes. Implications for evolution of a gene family. Cell (Cambridge Mass ) 1982; 2R: 671-679.
Tan EM. Antibodies to nuclear antigens (ANA): Theirimmunobiology and medicine. Adv Immunol 1982; 31: 167-240.
Tillman DM, Jou NT, Hill RJ, Marion TN. Both IgM and IgG anti- DNA antibodies are the products of clonally selective B cell stimulation in (NZB x NZW) FI mice. J Exp Med 1992; 176: 761- 779.
213
Tomlinson IM, Cook GP, Carter NP, et al. Human immunoglobulin VH and D segments on chromosomes 15qll.2 and 16pll.2. Hum molec Genet \99A\2^ 853-860.
Tomlinson IM, Walter G, Marks JD, Llewelyn MB, Winter G. The repertoire of human germline VH segments reveals about fifty groups of VH segments with different hypervariable loops. J Molec Biol 1992: 227: 776-798.
Tonewaga S. Somatic generation of antibody diversity. Nature (LoW; 1983; 202: 575-581.
Trauth BC, Klas C, Peters AMJ et al. Monoclonal antibody mediated tumour regression by induction of apoptosis. Science 1989; 254: 301-304.
Trepicchio W & Barrett KJ. Eleven Mrl/lpr anti-DNA autoantibodies are encoded by genes from four VH gene families. Immunol 1987a: 138: 2323-2331.
Trepicchio W, Maruya A, Barrett KJ. The heavy chain genes of a lupus anti-DNA autoantibody are encoded in the germline of a non- autoimmune strain of mouse and conserved in strains of mice polymorphic for this gene locus. J Immunol 1987b: 139: 3139- 3145.
Triplett DA, Brandt JT, Musgrave KA, Orr CA. The relationship between lupus anticoagulants and antibodies to phospholipid. JAMA 1988; 25R: 550-54.
Tsao BP, Eblihg FM, Roman C, Panosian-Sahakian N, Calame K, Hahn BH. Structural characteristics of the variable regions of immunoglobulin genes encoding a pathogenic autoantibody in murine lupus. J Clin Invest 1990; M: 530-540.
Tuaillon N, Watts RA, Isenberg DA, Muller S. Sequence analysis and fine specificity of two human monoclonal antibodies to histone HI. Mol Immunol 1994: 31: 269-277.
VVan der Heijden RWJ Bunschoten H Pascual V et al. Nucleotide sequence of a human monoclonal anti-idiotypic antibody specific for a rabies virus-neutralising monoclonal antibody reveals extensive somatic variablilty suggestive of an antigen driven immune response. J Immunol 1990: 144: 2835-39.
Van der Maarel SK, Willems van dijk SK, Alexander CM et al. Chromosomal organisation of the human VH4 family. J Immunol 1993; m 2858.
Van Es JH, Frits HJ, Gmelig Meyling WR et al. Somatic mutations in the variable regions of a human IgG anti-double stranded DNA autoantibody suggest a role for antigen in the induction of systemic lupus erythematosus. J. Exp. Med 1991: 173: 461-470.
214
Van Es JH, Aanstoot H, Gmelig-Meyling FHJ, Derksen RHWM, Logtenberg T. A human SLE-related anti-cardiolipin / ssDNA autoantibody is encoded by a somatically-mutated variant of the developmentally-restricted 51P1 VH gene. J Immunol 1992; 149: 2234-40.
Vismara A, Meroni PL, Tincani A etal. Relationship between anticardiolipin and antiendothelial cell antibodies in SLE. Clin Exp Immunol 1988; 24: 247-53.
WWalter G Tomlinson IM, Cook GP eta l. HAPPY mapping of a YAC reveals alternative haplotypes in the human immunoglobulin VH locus. Nucl Acids Res 1993; 21: 4524-4529.
Walter MA, Surti U, Hofker MH, Cox DW. The physical organisation of the immunoglobulin heavy chain gene complex. EMBOJ 1990-, 2: 3303-3313.
Watts RA, Hillson JL, Oppliger IR et al. Sequence analysis and idiotypic relationships of BEG-2 a human fetal antibody reactive with DNA. Lupus 199V, 21: 27-35.
Watts R & Isenberg D. DNA antibody idiotypes: An analysis of their clinical connections and origins. Intern Rev Immunol 1990; i : 279-293.
Winkler TH, Fehr H, Kalden JR. Analysis of immunoglobulin variable region genes from human IgG anti-DNA hybridomas. Eur J Immunol 1992: 22: 1719-1728.
Winkler M, Kohler H, Sinoda T, Putnam FW. Macroglobulin structure: Homology of p and y heavy chains of humanimmunoglobulins. Science 1969; 163: 75-78.
Willems van Dijk K, Mortari F, Kirkham PM, Schroeder HW, Milner ECB, The human VH7 gene family consists of a small polymorphic group of six to eight gene segments dispersed throughout the VH locus. Eur J Immunol 1993: 2 3: 832-39.
Williams SC & Winter G. Cloning and sequencing of human immunoglobulin VA. gene segments. Eur J Immunol 1993; 2 1 - 1456-61.
Williams W, Zumla A, Behrens R et al. Studies of a common idiotype PR4 in autoimmune rheumatic disease. Arthritis &Rheum 198S; 11: 1097-1104.
Wilkinson M. A rapid and convenient method for isolation of nuclear, cytoplasmic and total cellular RNA. Nucl Acids Res 1988; 166: 10934
Wu TT & Kabat EA. An analysis of the sequences of the variable regions of Bence Jones and myeloma light chains and their implications for antibody complementarity. J Exp Med 1970; 132: 211-250.
215
Wysocki L, Manser T, Gefter ML. Somatic evolution of variable region structures during immune response. Proc Natl Acad Sci C/5A 1985;S1: 1847-1851.
YYamada M, Wasserman R, Reichard BA et at. Preferential utilization of specific immunoglobulin heavy chain diversity and joining segments in adult human periferal blood B lymphocytes. J E xpM ed \99 \\m :. 395-407.
Yanocopoulos GD & Alt FW. Regulation of the assembly and expression of variable region genes. Annu Rev Immunol 1986; 4: 339-368.
Yonehara S, Ishii A, Yonehara M. A cell-killing monoclonal antibody (anti-Fas) to a cell surface antigen co-downregulated with receptor for tumour necrosis factor. J Exp Med 1989; 169: 1747- 1756.
Young F, Tucker L, Rubinstein D et al. Molecular analysis of a germline encoded idiotypic marker of pathogenic human lupus autoantibodies. J. Immunol 1990 145: 2545-2553.
ZZong SQ, Naki S, Matsuda F, Lee KH, Honjo T. Human immunoglobulin D segments: Isolation of a new D segment and polymorphic deletion of the D1 segment. Immunol Lett 1988; 17: 329-333.
Zouali M &Theze J. Probing VH gene-family utilization in human periferal B cells by in situ hybridization. J Immunol 1991; 146: 2885-2864.
216
A P P E N D IX I
217
APPENDIX I
Abbreviations for amino acid
Alanine Ala A
Arginine Arg R
Asparagine Asn N
Aspartic acid Asp D
Cysteine Cys C
Glutamine Gin QGlutamic acid Glu E
Glycine Gly G
Histidine His H
Isoleucine Be I
Leucine Leu L
Lysine Lys K
Methionine Met - M
Pheylalanine Phe F
Proline Pro P
Serine Ser S
Threonine Thr T
Tryptophan Trp W
Tyrosine Tyr Y
Valine Val V