Embed Size (px)
Citation preview
Doctorial Thesis from the Department of Immunology
The Wenner-Gren Institute, Stockholm University
MERCURY-INDUCED AUTOIMMUNITY:GENETICS AND IMMUNOREGULATION
MONIKA HANSSON
STOCKHOLM 2004
SUMMARY
The existence of immune self-tolerance allows the immune system to mount responses against infectious agents, but
not against self-molecular constitutes. Although self-tolerance is a robust phenomenon, in some individuals as well as
in experimental models, the self-tolerance breaks down and as a result, a self-destructive autoimmune disease
emerges. The underlying mechanisms for the development of autoimmune diseases are not known, but genetic, envi-
ronmental and immunological factors are suggested to be involved. In this thesis, we used murine mercury-induced
autoimmunity to test this suggestion.
In susceptible mice mercuric chloride induces a systemic autoimmune disease characterized by increased serum levels
of IgG1 and IgE, production of anti-nucleolar autoantibodies (ANolA) and formation of renal IgG deposits. In con-
trast, in resistant DBA/2 (H-2d) mice, none of these characteristics develop after exposure to mercury. By crossing and
backcrossing mercury-resistant DBA/2 mice to mercury susceptible strains, we found that the resistance was inherited
as a dominant trait in F1 hybrids and that one gene or a cluster of genes located in the H-2 loci determined the resist-
ance to ANolA production, whereas resistance to the other characteristics was found to be controlled by two or three
non-H-2 genes.
We further put forward the “cryptic peptide hypothesis” to investigate whether mercury and another xenobiotic metal
use similar pathway(s) to induce the H-2 linked production of ANolA. We found that while mercury stimulated ANolA
synthesis in all H-2 susceptible (H-2s, H-2q and H-2f) mouse strains, silver induced only ANolA responses in H-2s
and H-2q mice, but not in H-2f mice. Further studies showed that the resistance to silver-induced ANolA production in
H-2f mice was inherited as a dominant trait.
We next tested the proposition that mercury induces more adverse immunological effects in mouse strains, which are
genetically prone to develop autoimmune diseases, using tight-skin 1 mice, an animal model for human Scleroderma.
It was found that in this strain, mercury induced a strong immune activation with autoimmune characteristics, but did
not accelerate the development of dermal fibrosis, a characteristic in Tsk/1 mice.
Finally we addressed the Th1/Th2 cross-regulation paradigm by examining if a Th1-type of response could interact
with a Th2-type of response if simultaneous induced in susceptible mice. Our findings demonstrated that mercury-
induced autoimmunity (Th2-type) and collagen-induced arthritis (CIA) (Th1-type) can interact in a synergistic, antag-
onistic or additive fashion, depending on at which stage of CIA mercury is administered.
© Monika Hansson Stockholm 2004
ISBN 91-7265-813-4
Typografen Text och Bild AB,
ORIGINAL PAPERS
This thesis is based on the following papers, which will be referred to in the text by their Roman
numerals:
I. Manuchehr Abedi-Valugerdi, Monika Hansson and Göran Möller. 2001. Genetic
control of resistance to mercury-induced immune/autoimmune activation. Scand. J.
Immunol. 54: 190-197.
II. Monika Hansson and Manuchehr Abedi-Valugerdi. 2003. Xenobiotic metal-induced
autoimmunity: mercury and silver differentially induce antinucleolar autoantibody
production in susceptible H-2s, H-2q and H-2f mice. Clin. Exp. Immunol. 131: 405-
414.
III. Monika Hansson and Manuchehr Abedi-Valugerdi. Mercuric chloride induces a
strong immune activation, but does not accelerate the development of dermal fibrosis
in tight-skin 1 (Tsk1) mice. Submitted.
IV. Monika Hansson, Mounira Djerbi, Hodjattallah Rabbani, Håkan Mellstedt,
Mustapha Hassan, Joseph W DePierre and Manuchehr Abedi-Valugerdi. Interactions
between Th1-type of autoimmunity (induced by collagen), and Th2-type of autoim-
munity (induced by mercuric chloride) in DBA/1 mice. Submitted.
3
4
TABLE OF CONTENTS
ABBREVIATIONS.....................................................................................................................5
INTRODUCTION ......................................................................................................................6
TOLERANCE.............................................................................................................................8Central T-cell tolerance ..............................................................................................................8Peripheral T-cell tolerance...........................................................................................................9The role of dendritic cells in T-cell tolerance ..................................................................................9Regulatory T cells and T-cell tolerance........................................................................................10B-cell tolerance.......................................................................................................................10
AUTOIMMUNITY AND AUTOIMMUNE DISEASE ...........................................................12Genetics of autoimmune diseases ...................................................................................................................12Major histocompatibility complex (MHC) genes and susceptibility to autoimmunity..................................12The role of non-MHC genes in the development of autoimmunity ...............................................................13Gene aberration and development of autoimmunity ......................................................................................13The role of environmental factors in induction and perpetuation of autoimmunity ......................................14
MERCURY-INDUCED AUTOIMMUNITY...........................................................................15An experimental model to study the induction and development of autoimmunity ......................................15Description of the model.................................................................................................................................15Autoantibody production ................................................................................................................................15Th1/Th2 paradigm in mercury model.............................................................................................................16The role of IL-12 in regulation of mercury-induced autoimmunity...............................................................17Genetic predisposition to mercury induced autoimmunity.............................................................................18
THE PRESENT STUDY..........................................................................................................19
AIMS.........................................................................................................................................19
METHODOLOGICAL REMARKS.........................................................................................20Mercury and silver treatment (paper I-IV) .....................................................................................................20Evaluation of mercury-induced autoimmune manifestations (paper I-IV) ....................................................20H-2 genotyping (paper I-III) ...........................................................................................................................20Histological examination (paper III and IV) ..................................................................................................20Detection of antigen-specific IgG1 and IgG2a (papers III and IV) ...............................................................21Induction of CIA .............................................................................................................................................21Evaluation of the development of arthritis .....................................................................................................21
SUMMARY OF THE STUDY.................................................................................................22
RESULTS AND DISCUSSION ...............................................................................................22Paper I. ....................................................................................................................................................................22Paper II. ...................................................................................................................................................................24Paper III...................................................................................................................................................................25Paper IV. ......................................................................................................................................27
CONCLUDING REMARKS....................................................................................................30
ACKNOLEDGEMENTS..........................................................................................................31
REFERENCES .........................................................................................................................33
APPENDIX: PAPERS I-IV
ABBREVIATIONS
AFA Anti-fibrillarin antibodies
ANolA Anti-nucleolar autoantibodies
APC Antigen presenting cell
BCR B cell receptor
CII Type II collagen
CD Cluster of differentiation
CIA Collagen-induced arthritis
CTLA-4 Cytotoxic T-lymphocyte associated antigen 4
DC Dendritic cell
EAE Experimental autoimmune encephalomyelitis
H-2 Histocompatibility locus 2
IC Immune complex
IDDM Insulin-dependent diabetes mellitus
IFN Interferon
Ig Immunoglobulin
IL Interleukin
MBP Myelin basic protein
MHC Major histocompatibility complex
MS Multiple sclerosis
PD-1 Programmed cell death 1
QTL Quantitative trait locus
RA Rheumatoid arthritis
RAG Recombination activating gene
SLE Systemic lupus erythematosus
T1D Type 1 diabetes
TCR T cell receptor
TGF Transforming growth factor
Th1, Th2 T helper type 1 and type 2
TNF Tumor necrosis factor
5
INTRODUCTIONCentral to the immune systems ability to mobilize a response to an invading pathogen is its ability
to distinguish self from nonself. To efficiently protect the host from invading pathogens, both
adaptive and innate mechanisms have evolved. These mechanisms involve self-nonself discrimina-
tion and it has been suggested that this self-nonself distinction has served as a driving force in the
evolution of the complicated mechanisms responsible for the construction of T and B cell recep-
tors. How the immune system learns to discriminate between self-antigens and foreign antigens
have preoccupied scientists for decades, and different theories have been put forward.
In the beginning of 1900, Ehrlich and Morgenroth reported that animals could form antibodies
against xenogeneic and allogeneic red cells, but never against autologous cells. To explain these
phenomena Ehrlich formulated the theory of horror autotoxicus, which postulated that the organ-
ism would not normally mobilize its immunological resources to effect a destructive reaction
against its own tissue (reviewed in Silverstein and Rose 1997).
Between 1930 to early 1960, the dominant theory was an instruction theory, which offered an
explanation for the large repertoire of antibody specificities, by suggesting that the antigen act as a
template for the globulin producing machinery (reviewed in Silverstein 2002).
In 1955, with the increasing interest in tissue transplantation, immunodeficiency diseases, autoim-
munity and tolerance, Jerne put forward the natural selection theory of antibody formation. Now
the antigens purpose was to select a matching antibody and transport it to an appropriate cell,
which would be stimulated to produce more of the same. This idea did not gain much support at
the time, but catalysed the formulation of the clonal selection theory by Burnet 1957 (reviewed in
Silverstein 2002). This theory consisted of the necessary and sufficient components of a biological
theory of antibody formation: antigenic selection of pre-existing specificities; a cellular dynamic to
maintain and expand the response; and a mechanism to explain the development of immunological
tolerance. Burnet realized that a division of the world into self and nonself could not be genetically
based but rather be acquired. Therefore, he postulated that, during development of the immunolog-
ical repertoire in ontogeny, any antigen present in utero would eliminate newly formed anti-self
clonal precursors. As follows, the result of this clonal elimination an individual’s immune system
could only respond to those foreign antigens that had not been encountered during fetal develop-
ment (Silverstein and Rose 1997). However, over the years this theory has been modified, since
our knowledge of the immune system expands.
6
New theories have been put forward in the past decades. Based on the two-signal theories and the
finding that neonatal immune system, like adults can be primed to respond to foreign antigens,
Matzinger and colleagues suggested that the primary driving force for the immune system is not to
recognize self and nonself, but the need to protect against danger, the danger discrimination.
Danger signals refer to signals from stressed, damaged or necrotic cells that trigger dendritic cells,
resulting in the activation immune system (reviewed in Matzinger 1994).
Janeway and colleagues have a different view about danger discrimination. In defending the self-
nonself discrimination, they have pointed out that not only central and peripheral tolerance is
important. The danger signals that activate the immune system are co-stimulatory molecules and
cytokines induced by pathogens that are recognized by the innate immune system, and that this is
the evolutionary consequence of discriminating between non-infectious self and infectious-self
(reviewed in Janeway et al 1996, Medzhitov and Janeway 1997).
Below I will discuss some aspects of our current knowledge of tolerance and autoimmunity.
7
TOLERANCEB and T lymphocyte antigen receptors have a remarkable capacity to bind and recognize numerous
diverse foreign antigens. This immunological diversity is the result of random gene rearrange-
ments during the development of these lymphocytes. The danger of rearrangements of B- and T-
cell receptor genes is the potential risk of generation of B and T lymphocytes that express recep-
tors capable of recognizing self-antigens, which thereby might result in the development of horror
autotoxicus (Silverstein AM, 2001). To circumvent this problem the immune system has devel-
oped mechanisms to prevent the generation and/or the activation of B and T lymphocytes specific
for self-antigens. Perhaps the most important of these mechanisms is central tolerance, during of
which developing lymphocytes that express antigen receptors with high affinity for self antigens
are clonally deleted in the primary lymphoid organs, the bone marrow and thymus.
Central T-cell tolerance
In the thymus, developing T cells go through positive and negative selection events before they are
permitted to enter the periphery. These events form the basis for the self non-self discrimination of
the immune system and are collectively recognized as central tolerance. During positive selection,
thymocytes with a rearranged αβTCR will interact mainly with self-MHC molecules in the cortex.
If the interactions are of low avidity the thymocytes will down regulate RAG recombinases and
proceed to the next stage of development. If no interaction occurs the cell will continue to express
RAG and rearrange the Vα until a successful TCR is generated, but if the cell fails it will die by a
mechanism called “death by neglect”. In contrast, if the thymocytes at this cortical stage express a
TCR with high affinity for self-antigen it was thought that the cell was deleted (reviewed in
Kreuwel and Sherman 2001). However, McGargill et al. have presented evidence that a mecha-
nism of receptor editing exists for immature T cells. They showed that thymocytes with high affin-
ity for antigen presented by cortical epithelial cells, instead of being deleted continued to rearrange
the TCRα in attempt to construct a more suitable TCR (McGargill et al. 2000). If the thymocytes
survive this first step, they proceed to the medulla were they must pass additional selection crite-
ria, which is also based on the strength of the signal through the TCR. Strong interactions between
the TCR and self-peptide-MHC complexes lead to elimination of the thymocytes through apopto-
sis. This process of clonal deletion eliminates self-reactive T cells (reviewed in Kreuwel and
Sherman 2001).
However, there are some limitations of central tolerance; one is the requirement for the relevant
autoantigens to be present in the thymus. Although there is evidence that many tissue specific anti-
gens are present (Derbinski et al. 2001), many self-antigens may not access the thymus (Lo et al.
1989), whereas others are expressed later in life, after the T cell repertoire has been formed
(Matzinger 1994). There is also a substantial possibility that those lymphocytes that express recep-
tors specific for foreign antigens, may cross react with self proteins (Mason 1998).
8
Another consideration in this context is the promiscuity of TCR recognition, as a very stringent
negative selection increases the risk of dangerously narrowing the repertoire available to respond
to foreign pathogens. It might be more beneficial to let some thymocytes with a degree of self-
reactivity to escape thymic negative selection and be regulated elsewhere (reviewed in Walker and
Abbas 2002).
Peripheral T-cell tolerance
It has been shown that normal healthy individuals have self-reactive T cells in the periphery
(Lohmann et al. 1996, Semana et al. 1999), and that these T cells are more likely to express a low
affinity TCR for self-antigens (Bouneaud 2000). This suggests that there might be regulatory
mechanisms operating in the periphery that under normal conditions control these potentially
autoreactive T cells. One such mechanism is T cell ignorance of self-antigens because the antigens
are sequestered at sites not accessible to the blood/lymph borne immune system (Alferink et al.
1998). Ignorance can also be the result of low amounts of antigen, not reaching the threshold
required to trigger a T cell response (Kurts et al. 1998). Interactions of an autoreactive T cell with
self-antigens might also lead to a functional inactivation state, called anergy. Anergy was first
shown in vitro as the result of TCR ligation in the absence of co-stimulation (Jenkins and
Schwartz 1987). Later experiments have implicated that signaling through alternative receptors
rather than just lack of co-stimulation induces anergy in T cells in vivo. Indeed, it has been demon-
strated that ligation of cytotoxic T-lymphocyte-associated antigen 4 (CTLA-4) was required for
anergy in vivo. The importance of CTLA-4 in controlling the activation of T cells is evidenced by
the lethal lymphoproliferative disorder found in mice lacking this molecule (reviewed in Walker
and Abbas 2002). Another molecule that is believed to be involved in the control of T cell unre-
sponsiveness, is PD-1 (programmed cell death 1), which is expressed at high levels in anergic T
cells (Lechner et al. 2001). It is believed that PD-1 might function by causing cell cycle arrest or
inhibiting cytokine secretion (reviewed in Walker and Abbas 2002).
The role of dendritic cells in T-cell tolerance
Dendritic cells (DC) are believed to play an important role in establishing tolerance, both in the
thymus and in the periphery. In the thymus, they are localized in the medullary regions where they
present self-antigens to developing T cells and delete those that express a TCR with high affinity
for self-antigens. In the periphery, circulating immature DCs continuously sample the environ-
ment, capturing self-antigens as well as innocuous environmental proteins. It has been suggested
that proteins captured and processed by DCs in this steady state, meaning in the absence of acute
infection and inflammation, are tolerogenic, i.e., the DCs silence the corresponding antigen-specif-
ic T cell. Indeed, recent experiments have shown that antigen-loaded immature DCs silence T cells
either by deletion or by expanding regulatory T cell populations (reviewed in Steinman and
Nussenzweig 2002).
9
Regulatory T cells and T-cell tolerance
Another important factor that is involved in the regulation of self-reactive T cells in the periphery
is the presence of regulatory T cells. There are two different kinds of these cells: those, which are
produced in the thymus and express CD4 and CD25 and those which are generated after tolero-
genic encounters with antigen in the periphery. CD4+CD25+ T cells constitute approximately 10%
of peripheral murine CD4+ T cells. They express little CD45RB but constitutively express CTLA-
4. The importance of CTLA-4 is evidenced by the observation that inhibition of CTLA-4, either by
using antibodies or CTLA-4 deficient regulatory cells, reduces their suppressor functions signifi-
cantly. They have a partially anergic phenotype, evidenced by their poor proliferative response
upon TCR stimulation. In vitro studies have shown that, rather than proliferating in response to
antigen, these cells suppress the proliferation of other T cells. The suppression is mediated through
cell-cell contact, independently of cytokines such as IL-4, IL-10 and TGF-β. Although the activa-
tion is antigen-specific, once activated, these cells inhibit both CD4+ and CD8+ T cell responses
in an antigen-nonspecific manner (reviewed in Maloy and Powrie 2001, and Walker and Abbas
2002). Repetitive stimulation of T cells with alloantigens in the presence of IL-10 has been shown
to induce another set of regulatory T cells, termed TR1 cells, which partially rely on the produc-
tion of IL-10 and TGF-β for their function. There may be multiple mechanisms that the regulatory
T cells use to suppress immune responses. Studies in animal models have provided compelling
evidence for an important role for cytokines, such as IL-10 and TGF-β, in the effector function of
these cells in vivo. However, the cytokines involved vary depending on the model. The immune-
suppressive properties of TGF-β and IL-10 are most likely explained by their ability to inhibit
APC function and to mediate direct anti-proliferative effects on T cells (reviewed in Maloy and
Powrie 2001).
B-cell tolerance
There are several mechanisms that are important for generating and maintaining B cell tolerance.
In the bone marrow, developing B cells expressing a BCR with high affinity for self antigens
exhibit developmental arrest during which they continue V(D)J recombination of their Ig genes,
especially the light chain genes. This process is called receptor editing and changes the BCR anti-
gen specificity (Nemazee 2000). If the editing is successful and the strong self-reactivity is lost,
then the immature B cell can resume maturation and leave the bone marrow, if not, it will be clon-
ally deleted (reviewed in Ohashi and DeFranco 2002).
Self reactive B cells that enter the periphery can also be regulated through peripheral tolerance
mechanisms. They can either be deleted due to strong interactions with antigen or rendered anergic
due to less strong interaction with antigen. B cells with weak self specificity may also exist in a
naïve functional state, sometimes referred to as “ignorance”. This state is mainly due to low expo-
sure to antigen and/or lack of T cell help.
10
Studies in transgenic mice using the model antigen hen egg lysozyme (HEL), have demonstrated
that B cell anergy is effective in silencing B cells. This is in part due to the poor signaling through
the BCR by anergic B cells (Cooke et al. 1994, Eris et al.1994, Benschop et al. 2001), which ren-
ders them highly susceptible to Fas-mediated apoptosis when presenting antigen to specific Th
cells (Rathmell et al.1995, Rathmell et al. 1996). This is because activated Th cells express
CD40L on their surface which induces Fas expression on B cells and thus makes them susceptible
to Fas-mediated apoptosis, whereas signaling through BCR protects from apoptosis (Rothstein et
al. 1995). However, recent studies of anergic B cells recognizing double stranded (ds) DNA have
revealed that provision of T cell help induce production of anti-dsDNA autoantibody in these B
cells (reviewed in Ohashi and DeFranco 2002). Why tolerance to dsDNA was broken in these
experiments, but not in the experiments with HEL, is not clear. It could reflect experimental differ-
ences in the way T cell help was provided to the B cells, for instance it has been shown that high
levels of IL-4 protects B cells from Fas-induced apoptosis. Differences in the nature of the anti-
gens as well as variations in the states of anergy might also explain the different outcomes in these
experiments (reviewed in Ohashi and DeFranco 2002). It seems as if B cell anergy is not a single
homogenous state since different Ig transgenic mice with specificity for dsDNA have anergic B
cells with a range of phenotypes. In some cases anergic B cells can be stimulated by LPS to divide
and differentiate into antibody secreting cells, whereas in other cases, the anergic B cells are not
responsive to LPS (reviewed in Ohashi and DeFranco 2002).
11
AUTOIMMUNITY AND AUTOIMMUNE DISEASEAutoimmune disease is the consequence of an immune response directed against self, leading to
destruction or dysfunction of the targeted tissue. Traditionally autoimmune diseases have been
divided into organ-specific or systemic diseases. This division is depending on whether the
immune response is directed to a self-antigen confined to a particular organ or to a self-antigen
widely distributed in the body. Organ-specific diseases are often considered to result from defi-
ciencies in tolerance induction mechanisms, which results in failure to eliminate or deactivate self-
reactive lymphocytes. The etiology of most autoimmune diseases is still unknown. However, it has
been proposed that multiple factors including genetic, environmental and immunologic factors
contribute to the induction and development of autoimmune diseases. Moreover, there is accumu-
lating evidence that T cells play a central role in the development of various autoimmune diseases,
especially organ-specific diseases, both in humans and animals (reviewed in Van Parijs and Abbas
1998).
Genetics of autoimmune diseases
Identifying autoimmunity genes has so far been difficult, one of the major reasons for this is that
the risk ratios for developing disease if an individual susceptibility allele is present is probably
quite low. The fact that many of these alleles probably interact with other genes in the background,
as well as with environmental factors, further complicates the problem (Gregersen 1999, reviewed
in Gregersen 2003). In identical twins, the twin concordance rates for autoimmune diseases are
generally in the range of 15-30%, depending on the autoimmune disease. These relatively low con-
cordance rates provides evidence for the contribution of non-genetic factors, such as environmen-
tal factors or chance events, also contribute to the variability of disease expression in the popula-
tion (Gregersen 1999).
Major histocompatibility complex (MHC) genes and susceptibility to autoimmunity
Classical genetic studies on various autoimmune diseases have demonstrated that genetic predispo-
sition plays an important role. The most potent genetic influence on susceptibility to autoimmune
diseases is the major histocompatibility complex (MHC) (particularly MHC class II), which has
been known for many years to affect susceptibility to different diseases with autoimmune origin.
Susceptibility to a number of autoimmune diseases, including rheumatoid arthritis (RA), multiple
sclerosis (MS) and type 1 diabetes (T1D), is associated with particular alleles of MHC class II
genes such as HLA-DR and/or –DQ genes.
This association provides strong evidence for the role of MHC class II antigen presentation in
these disease processes (Todd et al. 1987, reviewed in Nepom and Erlich 1991, Wucherpfennig
and Strominger 1995).
12
It has been possible to define structural features that determine the interactions of the relevant
MHC-class II molecules and peptides in several DR-associated autoimmune diseases. For exam-
ple, susceptibility to RA is associated with the “shared epitope”, a segment of the DRβchain that is
very similar in sequence among the disease –associated subtypes of DR4 (DRB1*0401 and 0404)
and DR1 (DRB1*0101). In structural terms, this “shared epitope” primarily defines the shape and
charge of the P4 pocket (reviewed in Wucherpfennig 2001).
The role of non-MHC genes in the development of autoimmunity
Over the years, several non-MHC susceptibility loci for autoimmune disease have been identified
in humans and animal models. It has been shown that many of these loci are clustered into 12 non-
overlapping genomic regions that confer much of the susceptibility for systemic lupus erythemato-
sus (SLE), T1D, MS, experimental autoimmune encephalomyelitis (EAE), RA and Crohn´s dis-
ease ( reviewed in Morahan and Morel 2002). When interpreting these results one should bear in
mind the fact that genes with related functions tend to cluster in the genome, and that many quan-
titative trait loci (QTLs) correspond to more than one susceptibility gene. For example, in murine
autoimmune diseases, fine mapping of the NZM2410 sle loci and several of the NOD Idd loci for
insulin dependent type 1 diabetes in NOD mice have revealed that each original locus corresponds
to more than one susceptibility gene (Rev in Wanstrat and Wakeland 2001, Morel et al. 2001).
Gene aberration and development of autoimmunity
More than 40 non-MHC genes have been associated with spontaneous development of lupus-like
disease when aberrantly expressed in mice (reviewed in Kono and Theofilopoulos 2000, Wakeland
et al. 2001, Mohan 2001, Morahan and Morel 2002). These genes can be classified into at least
three functional categories: 1) molecules involved in the clearance of apoptotic cells, which if
accumulated may result in nuclear-antigen-driven expansion of autoreactive lymphocytes, (SAP,
CRP, Mertk, IgM and C1q). 2) Molecules that compromise lymphocyte apoptosis (FAS, FASL,
PTEN, PI3K, Bim and BCL-X), and which may act by inhibiting deletion of self-reactive B and T
cells. 3) Molecules that amplify or moderate lymphocyte signaling and expansion (CD19, Lyn,
Fyn, Cr2, CD22, CD45, P21, PD1, TAC, Ptp1c and Blys), and which are believed to amplify anti-
self humoral and cellular immune responses. While some of these molecules primarily affect B or
T cells, other may influence the expansion and activation of myeloid cells as well. These observa-
tions that molecules with different functional properties all lead to lupus development, suggest that
there are multiple checkpoints that control nuclear-antigen-reactive lymphocytes and that defects
in anyone of them lead to systemic autoimmunity in mice (reviewed in Raman and Mohan 2003).
The findings from forward genetic studies indicate which genes can predispose to lupus, but do
not reveal which genes that actually is responsible for spontaneous lupus development in mice and
humans.
13
To resolve this issue, several reverse genetic studies have been preformed, both in humans and in
mice. In a reverse genetic study the starting point is a well characterized disease and then by work-
ing backwards it may be possible to identify the genetic loci responsible for the disease or pheno-
type. This approach have revealed the existence of more than fifty loci predisposing to different
manifestations of lupus, including antinuclear antibodies, glomerulonephritis, splenomegaly etc in
mice (reviewed in Kono and Theofilopoulos 2000, Wakeland et al. 2001, Mohan 2001, Morahan
and Morel 2002, Raman and Mohan 2003).
In the past few years, seven human susceptibility loci have been mapped using genome scans in
human-lupus. In addition, several suggestive loci have been implicated in lupus susceptibility in
two or more scans. Moreover, loci that confer susceptibility to nephritis, thrombocytopenia,
hemolytic anemia, associated with RA, neuropsychiatric manifestations, or anti-dsDNA antibodies,
have been found in studies of phenotypically stratified lupus patients. Recent research has also
shed some light on how epistatic interactions of several different genetic loci might be required for
full-blown lupus to develop (Raman and Mohan 2003).
The role of environmental factors in induction and perpetuation of autoimmunity
Defining specific pathogenic environmental agents that may trigger the development or progres-
sion of autoimmune diseases remains the focus of increasing investigations. A factor that promotes
a given autoimmune disease may not be identical to factors that influence the severity or progres-
sion of the disease. In autoimmune connective tissue diseases, which include SLE, systemic scle-
rosis (scleroderma), Sjögren’s syndrome, RA and systemic vasculitis, exposure to several environ-
mental agents such as UV light, drugs, silica, mercuric chloride, hair products and organic solvents
have been considered as the risk factor for the development of these diseases (reviewed in D’Cruz
2000, Mayes 1999).
The mechanisms by which environmental agents in the context of susceptible genes induce or
influence the severity of the autoimmune diseases are not well understood. However, several
mechanisms including molecular mimicry, creation of cryptic self-peptides, induction of Th1/Th2
imbalance and interference with thymic selection might be involved. Regarding the last possibility,
it has been shown that environmental agents may affect thymic selection of T cells. Administration
of procainamide-hydroxylamine (PAHA) into the thymus of normal mice resulted in chromatin-
reactive T cells and autoantibodies against chromatin, a characteristic feature of human pro-
cainamide-induced lupus. In vitro culture of neonatal thymi from TCR-transgenic mice in the pres-
ence of PAHA and TCR ligand lead to enhanced positive selection of T cells expressing the trans-
genic TCR, but did not alter negative selection (Kretz-Rommel et al. 1997, Kretz-Rommel and
Rubin 2000). These results suggests that PAHA enhanced the positive selection of otherwise non-
selected low-affinity T cells specific for chromatin.
14
MERCURY-INDUCED AUTOIMMUNITY
An experimental model to study the induction and development of autoimmunity
Experimental models of chemically induced autoimmunity have been used to study the possible
associations between the environmental, genetic and immunological factors and the development
of autoimmune diseases in humans (Pieters et al. 2002, Druet 1990). Among these models, mer-
curic chloride (HgCl2)-induced autoimmunity in rodents represents a relevant and very well estab-
lished model. This is owing to the fact that mercury is an abundant environmental pollutant and
that this element has been used for thousands of years in industry and medicine (Pelletier et al.
1994).
Description of the model
In genetically susceptible mice and rats, subcutaneous injections with subtoxic doses of mercuric
chloride (1.0 mg/kg body weight) induce a systemic autoimmune disease, characterized by a T-cell
dependent polyclonal B cell activation, increased serum levels of IgG1 and IgE, production of
autoantibodies of different specificities and formation of renal IgG deposits (Mathieson 1992,
Griem and Gleichmann 1995, Enestrom and Hultman 1995, Goldman et al. 1991). These charac-
teristics appear 7-10 days after the beginning of mercury administration and a single injection may
be adequate to induce the syndrome. Although mercury is commonly administered subcutaneously,
mercury given intraperitoneally, orally, in drinking water, or aerosolized also effectively induces
similar autoimmune manifestations in susceptible animals (Mathieson 1992, Hultman and
Enestrom 1992, Hultman and Enestrom 1987). The effects of mercury are dose dependent as the
increase in serum IgE levels positively associates with dosage of mercury in susceptible rats
(Pelletier et al. 1994) and increase in serum IgG of specific autoantibodies and renal IgG immune
complexes correlate with the dosage of mercury in susceptible mice (Hultman and Enestrom
1992).
Autoantibody production
Production of autoantibodies of various specificities is one of the main characteristic of mercury-
induced autoimmunity in susceptible rats and mice. Mercury-treated rats develop antibodies
against both exogenous and endogenous antigens including antibodies against 2,4,6-trinitrophenyl
(TNP), glomerular basement membrane, double-stranded DNA (ds-DNA), thyroglobulin, collagen,
phospholipids, myeloperoxidase and laminin P1 (Hirsch et al. 1986, Pusey et al. 1990, Druet et al.
1994, Aten et al. 1995), while in susceptible mice, treatment with mercury elicits antibodies,
which are against TNP, chromatin/histone, collagen, IgG, phospholipids and nucleolar proteins
(Hultman et al. 1989, Reuter et al. 1989, Al-Balaghi et al. 1996).
15
Mercury-induced anti-laminin autoantibodies are uniquely found in rats (Druet et al. 1994, Aten et
al. 1995), whereas, anti-nucleolar autoantibodies (ANolA) are exclusively found in certain suscep-
tible mouse strains (Hultman et al. 1989, Reuter et al. 1989, Mirtcheva et al. 1989, Pietch et al.
1989, Hultman et al. 1992). Several studies have shown that mercury-induced ANolAs react with
fibrillarin, a 34–36 kDa protein (Hultman et al. 1989, Reuter et al. 1989, Takeuchi et al. 1995),
which is associated with the U3, U8, U13, U14, X and Y small nucleolar RNAs in vertebrates
(Fournier and Maxwell 1993).
Fibrillarin is also a target for autoantibodies in a subset of patients with scleroderma (Takeuchi et
al. 1995, Kasturi et al. 1995, Arnett et al. 1996). Interestingly, it has been found that murine mer-
cury-induced and spontaneous human ANolA share several similarities in binding to fibrillarin
(Takeuchi et al. 1995). This finding led to the suggestion that both mercury-induced and sponta-
neously developed ANolA interact with similar evolutionary conserved epitopes in fibrillarin
(Takeuchi et al. 1995).
A peculiar and interesting feature of mercury-induced autoimmunity in the mouse and rat is that
the polyclonal B-cell activation and autoantibody production spontaneously disappear after 2-3
weeks, despite continuous injections of mercury (Pusey et al. 1990, Al-Balaghi et al. 1996, Pietsch
et al. 1989, van Vliet et al. 1993, Bowman et al. 1984, Chalopin and Lockwood 1984, Mathieson
et al. 1991). The exact mechanism(s) for auto-regulation of mercury-induced autoantibody produc-
tion is not well understood. However, it has been proposed that in rats, this auto-regulation is
mediated by CD8+ T cells and IL-2-producing CD4+ T cells (Bowman et al. 1984, Castedo et al.
1994). It remains to elucidate whether similar types of cells are also responsible for the regulation
of mercury-induced antibody production in the mouse.
Th1/Th2 paradigm in mercury model
Mercury-induced autoimmunity is one of the autoimmune models in which Th1/Th2 imbalance
has long been considered to play a critical role (Goldman et al. 1991, Mathieson 1995). The notion
from early studies that mercury induced increases in the serum levels of IgG1 and IgE (Hirsch et
al. 1982, Pelletier et al. 1988) led to the suggestion that Th2-type T cells play an important role in
the pathogenesis of disease, as IL-4 produced by Th2 cells induces class switching to these iso-
types (reviewed in Finkelman et al. 1990). Ochel and colleagues addressed the role of this
cytokine in murine mercury-induced autoimmunity by treating susceptible A.SW mice (H-2s) with
an anti-IL-4 monoclonal antibody (mAb) prior treatment with mercury. It was observed that treat-
ment with anti-IL-4 mAb partially or completely abrogated the increases in IgG1 ANolA and IgE,
but it enhanced the production of mercury-induced IgG2a, IgG2b and IgG3 ANolA (Ochel et al.
1991). Similar results were also obtained from mercury-treated IL-4 deficient A.SW mice (H-2s)
(Bagenstose et al. 1998).
16
Taken together, these results indicate that IL-4 is required for the class switching to the IgG1 and
IgE isotypes, but is not essential for the development of ANolAs.
Attempts have been made to block the activation of the Th2-type of response seen in mercury-
induced autoimmunity. IFN-γ secreted by activated Th1 cells has been shown to play an important
role in regulating cellular immune responses and down-regulating Th2 activities (Mosmann and
Coffman 1989). Accordingly, it was shown that mercury induces IFN-γ production in the spleen
cells of mercury-resistant LEW rats (Van der Meide et al. 1995). Furthermore, in resistant B10.D2
mice (H-2d), mercury induces an IFN-γ-dependent suppression of antibody formation to sheep red
blood cells (Doth et al. 1997). To address whether IFN-γ can reverse the Th2-mediated immune
responses by mercury, recombinant IFN-γ was administered to B10.S mice prior to mercury treat-
ment. This pretreatment led to a reduction in serum IgE levels and anti-SRBC antibody levels, but
failed to prevent ANolA production and immune glomerulonephritis in B10.S mice (Doth et al.
1997). Conversely, mAb neutralization of IFN-γ in resistant B10.D2 (H-2d) mice alleviated the
mercury-induced immunosuppression, but could not convert the mice to a mercury-susceptible
phenotype (Doth et al. 1997). To more directly investigate the possible suppressive roles of the
IFN-γ in mercury-induced autoimmunity, Kono and colleagues examined the IFN-γ -deficient H-2s
mice for the susceptibility to disease. Surprisingly, they observed that mercury treatment does not
elicit ANolA or tissue lesions in these animals (Kono et al. 1998).If confirmed, this novel finding
suggests that IFN-γ rather than being suppressive play a hitherto unidentified role in breaking tol-
erance to nucleolar antigens that takes place in this model.
The role of IL-12 in regulation of mercury-induced autoimmunity
An alternative approach was pursued to modulate the response to mercury observed in A.SW (H-
2s) mice to that of a Th1 phenotype by administration of IL-12 prior to treatment with mercury.
The IL-12 cytokine induces T cell proliferation and IFN-γ production and is necessary for differen-
tiation of naive T cells into the Th1 subset and thereby is critical for the initiation of inflammatory
immune responses (Trinchieri 1993). IL-12 also contributes to certain Th1-mediated autoimmune
disorders including multiple sclerosis, insulin-dependent diabetes mellitus (IDDM) and inflamma-
tory bowel disease (Gately et al. 1998). Further, in mice susceptible to Leishmania major infec-
tion, a Th2 mediated infectious disease (Reiner and Locksley 1995) IL-12 administration can bias
toward a protective Th1 response (Heinzel et al. 1993, Sypek et al. 1993). Thus, IL-12 can serve
as a potentially important immunoregulator of disease outcome. In A.SW mice, IL-12 treatment
dramatically reduced ANolA titers of all classes and partially inhibited the mercury-induced
increase in IgG1, but had no effect on serum levels of IgE and renal Ig deposits (Bagenstose et al.
1998). Taken together, it is conceivable that Th1 and Th2 cytokines indeed play important roles in
the B-cell activation seen in mercury-induced autoimmunity. However, the question remains
whether Th1 cells participate in both the induction and regulation of this syndrome.
17
Genetic predisposition to mercury induced autoimmunity
Like in many other experimentally induced and/or spontaneously developed autoimmune diseases,
genetic factors play crucial roles in conferring susceptibility/resistance to mercury-induced autoim-
munity in both rats and mice. Genetic analysis on the F1 and backcross hybrids between suscepti-
ble and resistant animals have revealed that at least three or four genes or clusters of genes are
involved, of which one is within MHC class II region (Sapin et al. 1984, Druet et al. 1982, Sapin
et al. 1982). In mercury susceptible BN rats (RT1n), three susceptibility loci on chromosomes 9,
10 and 20 have been identified for Th2-mediated autoimmune responses (Fournie et al. 2001). The
susceptibility locus on chromosome 20, which contains the MHC region, controls the kinetics of
the IgE response and the autoantibody response to DNA. The other two loci on chromosomes 9
and 10 control the IgE response, the anti-laminin response and the deposition of IgG in glomeruli
(Fournie et al. 2001). In mercury susceptible H-2s mice (SJL, A.SW and B10.S mice), susceptibil-
ity to mercury-induced ANolA production could be mapped to the I-A loci of H-2 class II genes
(Hultman et al. 1992). It has been shown that the other H-2 class II locus (I-E) either suppresses
(Mirtcheva et al. 1989) or does not influence the mercury-induced ANolA response (Hultman et al.
1992). Determination of susceptibility to mercury-induced autoimmunity in H-2 congenic mice
and in mouse strains carrying identical H-2 genotype, but different non-H-2 background genes
have revealed that non-H-2 genes mainly control the susceptibility to and the severity of the B cell
activation, IgE production and renal IgG deposition (Hultman et al. 1993, Abedi-Valugerdi and Möller
2000). However, further studies are required to map the chromosomal localization of these genes
in the susceptible mice.
18
THE PRESENT STUDY
AIMSThe general aims of this study were to analyze the genetic susceptibility/resistance to mercury-
induced autoimmune manifestations, and to investigate how mercury-induced autoimmunity is
regulated.
We specifically aimed at investigating how resistance to murine mercury-induced autoimmunity is
inherited, regarding the four major characteristics; formation of IgG1 antibodies, increase in serum
IgE levels, production of ANolA and the development of renal IgG1 deposits (Paper I).
Furthermore, we applied the “cryptic peptide hypothesis” as a working hypothesis for metal-
induced ANolA production, to address whether mercury and silver could induce ANolA responses
in a similar fashion (Paper II).
We also tested our hypothesis that mercury confers more adverse immunological effects in individ-
uals, who are genetically or immunologically, predisposed to develop spontaneous autoimmune
disease in an animal model of Scleroderma (Paper III).
Finally, we addressed the Th1/Th2 cross-regulation paradigm by examining whether, in a suscepti-
ble mouse strain, simultaneous induction of mercury-induced autoimmunity and collagen-induced
arthritis could result in attenuation of one or both of the diseases (Paper IV).
19
METHODOLOGICAL REMARKSThe methods used in papers I-IV are described in detail in the “Material and Method” section of
each paper. However, I will give a brief description of the methods used below.
Mercury and silver treatment (paper I-IV)
All mercury-treatments in the experiments were performed according to the same protocol. The
experimental mice were injected subcutaneously (s.c.) with 0.1 ml HgCl2, 1.6 mg/kg body weight
(b.w.) every third day for 4 weeks.
In paper II the silver-treatment was performed in a similar fashion, the experimental mice received
0.1 ml AgNO3, 2.5 mg/kg b.w. s.c. every third day for 4 or 8 weeks. Control mice received sterile
0.9% NaCl s.c. in all experiments.
Evaluation of mercury-induced autoimmune manifestations (paper I-IV)
At the end of each experiment we analyzed the development of the mercury- and silver-induced
characteristics in spleens, kidneys and serum using several methods. To enumerate splenic anti-
body secreting cells we performed Protein A plaque assay. Detection of renal IgG1 deposits was
performed using a direct immunofluorescens method. For detection of serum ANolA an indirect
immunofluorescens method was used. Finally, serum IgE levels were measured with a sandwich
ELISA method.
H-2 genotyping (paper I-III)
Genomic DNA samples from were prepared from mouse tail biopsies. H-2 typing was performed
by PCR of genomic DNA, using primers, which could discriminate between different H-2 geno-
types. All amplified DNA samples were then electrophoresed through an agarose gel mix, and
after electrophoresis, the gels were stained with ethidium bromide and photographed.
Histological examination (paper III and IV)
Fixed skin (paper III) or limb (paper IV) samples were embedded in paraffin and stained with
hematoxylin and eosin (H &E) according to routine histological methods. The thickness of the
skin was measured vertically from the top of the granular layer to the junction between dermis and
subcutaneous fat on H & E stained sections using a light microscope equipped with a computer
analyzer. The limbs were examined for histological evidence of inflammation, pannus formation
and destruction of cartilage and bone
20
Detection of antigen-specific IgG1 and IgG2a (papers III and IV)
Serum levels of antigen-specific IgG1 and IgG2a were determined using a standard ELISA tech-
nique, where the antigen of interest was coated to ELISA plates, followed by incubation with
serum and subsequent detection with AP-conjugated anti-IgG1 or IgG2a antibody.
Induction of CIA
Chicken type II collagen (CII) was mixed in equal volume with complete Freund´s adjuvant
(CFA). The mice were injected intradermally (i.d.) with 0.1 ml of this emulsion at the base of the
tail. 21 days later, the animals were injected intraperitoneally (i.p.) with 100 µg CII in 0.05 M
acetic acid.
Evaluation of the development of arthritis
Following their secondary immunization with CII, the animals were assessed daily for the appear-
ance of arthritis by examining and scoring each of the forepaws and hindpaws. The severity of
arthritis in each affected paw was graded according to an established scoring system. The sum of
the scores for all four paws in each mouse was employed as an index of the overall severity and
progression of arthritis in each animal.
21
SUMMARY OF THE STUDY
RESULTS AND DISCUSSION
Paper I.
Previously it has been shown that mercury-induced autoimmune manifestations in susceptible
strains of mice are genetically controlled. The susceptibility to ANolA production has been shown
to be under the control of genes within the H-2 loci (Robinson et al. 1986, Mirtcheva et al. 1989,
Hultman et al. 1992, Hultman et al. 1993, Hultman et al. 1996, Hanley et al. 1997, Johansson et
al. 1998, Abedi-Valugerdi and Möller 2000). It has also been shown that H-2 genes in combination
with genes residing outside the H-2 loci could confer susceptibility to polyclonal B cell activation
and renal IgG deposition (Hultman et al. 1992, Hultman et al. 1993, Hultman et al. 1996, Hanley
et al. 1997, Johansson et al. 1998). The mercury-induced increase in serum levels of IgE seems to
be controlled mainly by non-H-2 genes (Abedi-Valugerdi and Möller 2000). In a study of suscepti-
bility to mercury-induced autoimmunity in several mouse strains of different H-2 genotypes it was
found that SJL and A.SW mice, both with an H-2s genotype were highly susceptible regarding all
characteristics of mercury-induced autoimmunity in mice. The NZB mice were also found to be
highly susceptible although they did not develop an ANolA response. Most other strains in this
study were classified as intermediate responders, since they were able to develop at least one of
the characteristics of mercury-induced autoimmunity. Only one of the tested mouse strains, the
DBA/2 qualified for the classification non-responder, (Abedi-Valugerdi and Möller 2000).
Since only a few studies in the past had focused on the genetics of resistance to mercury (Hanley
et al. 1997) and the existence of a highly mercury resistant mouse strain we decided to further
investigate this resistance. In this study, we used F1 hybrid crosses between the resistant DBA/2
(H-2d) mice and mercury susceptible NZB (H-2d), SJL (H-2s), A.CA (H-2f), and DBA/1 (H-2q)
as well as backcross hybrids between (DBA/2 x SJL)F1 and SJL mice to examine the genetic con-
trol of resistance to mercury-induced autoimmunity regarding four major characteristics, formation
of IgG1 antibodies, increase in serum IgE levels, production of ANolA, and development of renal
IgG1 deposits.
Our first finding was that the development of most characteristics was down-regulated in the F1
hybrids. This implies that dampening genetic factors in the DBA/2 strain can be transmitted to the
F1 progeny as dominant traits. This is in contrast to findings in the rat, where susceptibility to
mercury-induced increases of IgE and development of IC-mediated glomerulonephritis are inherit-
ed as dominant traits in crosses between susceptible and resistant strains of rats (Druet et al. 1977,
Sapin et al. 1982, Sapin et al. 1984).
22
To further investigate the inheritance of resistance to mercury-induced autoimmunity we examined
mercury-induced responses in backcross hybrids of (DBA/2 x SJL)F1 x SJL. We found that the
backcross hybrids could be categorized as non/low, intermediate and high responders, for increases
in IgG1, IgE and development of renal IC deposits, with a defined phenotypic ratio for each char-
acteristic. This suggests that resistance to any one of these characteristics is controlled by more
than one gene. Based on this suggestion and the finding that the resistance in the F1 hybrids was a
dominant trait, we concluded that a complete resistance requires heterozygosity in all genes
responsible for the phenotype.
The phenotypic ratios obtained in the backcross hybrids, suggests that a two-gene model is valid
for the resistance to increases in IgG1 and IgE (Griffiths et al. 1999). H-2 genotyping of the back-
cross hybrids revealed that genes within the H-2d loci were not involved in this resistance. Thus,
two non-MHC loci might be linked to the resistance to these characteristics. However, we did not
address the location of the genes conferring resistance but it is possible that polymorphisms in
TCR and FcγR genes are associated with resistance to increased IgG1 responses. This suggestion
is supported by the findings that in vivo induction of IgG1 is T-cell-dependent (reviewed in
Finkelman et al. 1990) and that FcγRs play a pivotal role in the regulation of B-cell-activation
(Friedman 1993). Since, B cells require cytokine-secretion, especially of IL-4 and IL-13, and liga-
tion to CD40L expressed by T cells to undergo isotype switching to IgE (Bacharier and Geha
2000), we propose that likely candidate genes involved in the resistance to increase of serum IgE
levels, might be found among genes controlling the expression of cytokines and co-stimulatory
molecules. Support for the involvement of co-stimulatory molecules in preventing mercury-
induced increases in IgG1 and IgE levels, is the report that administration of anti-B7-1 in combi-
nation with anti-B7-2 antibodies prevented increases of both IgG1 and IgE when administered to
susceptible H-2s mice (Bagenstose et al. 2002).
Evaluation of the phenotypic ratio obtained from the backcross hybrids for the formation of renal
IgG1 deposits suggests that three genes are likely to be responsible for the resistance to this char-
acteristic. Again, after genotyping we could conclude that genes within the H-2d loci were not
involved. This gains support from a previous study of mercury-treated mouse strains, expressing
H-2d on different backgrounds, where it was found that only the DBA/2 mice were resistant to
mercury-induced renal IgG1 deposits (Abedi-Valugerdi and Möller 2000). In a study by Kono and
co-workers, two genome-wide scans were performed using F2 intercrosses of DBA/2 with either
SJL or NZB mice. They reported that a single major QTL on chromosome 1, designated Hmr1,
was common to both crosses and encompassed a region containing several lupus susceptibility
loci. Hmr1 was only linked to glomerular IC deposits and not autoantibody production (Kono et
al. 2001).
23
The phenotypic ratio obtained in the backcross hybrids regarding ANolA synthesis was in agree-
ment with a one-gene model (Griffiths et al. 1999). When correlating the IgG1 ANolA production
with H-2 haplotype we found that all of the heterozygous (H-2s/d) were either non- or low respon-
ders and that a large proportion of the homozygous (H-2s/s) backcross hybrids were high respon-
ders and none of them were low responders. This supported the suggestion that one gene or a clus-
ter of genes within the H-2 loci determines the resistance to IgG1 ANolA production, and that this
resistance is mainly conferred by the H-2d haplotype. This is also in agreement with the conclu-
sion made by Hanley and co-workers, that the expression of a resistant H-2 genotype per se is able
do down-regulate ANolA responses in otherwise susceptible mice (Hanley et al. 1997). How a
mercury-resistant H-2 haplotype, despite being co-expressed with a mercury-susceptible H-2 hap-
lotype, can confer resistance to mercury induced ANolA is poorly understood. However, it has
been demonstrated in adoptive transfer experiments, that the absence of ANolA production in F1
mice was not due to a difference in thymic education or the absence of anti-fibrillarin specific T
cell help and it was suggested that the resistance was due to an intrinsic property of the haplotype
heterozygous B cells (Hanley et al. 1998). This suggestion might be applied for (DBA/2 x SJL)F1
and (DBA/2 x A.CA)F1 hybrids in which the degree of ANolA penetration was low, but it does
not explain the high degree of ANolA penetration in the (DBA/2 x DBA/1)F1 hybrids in this
study. Therefore, additional studies of how resistant H-2 haplotypes down-regulate mercury-
induced ANolA in H-2 heterozygous hybrids are required.
Paper II.
It has previously been shown that another metal, silver, can induce ANolA in mercury-susceptible
mouse strains which carry the H-2s and H-2q genotypes (Hultman et al. 1994, Hultman et al.
1995, Johansson et al. 1997). Since mercury and silver are able to form strong chemical bonds
with organic donors (Frausto da Silva and Williams1991), a general mechanism for induction of
ANolA by mercury (Griem and Gleichmann 1995) and silver (Johansson et al. 1997) was put for-
ward. It has been proposed that by having a high affinity for different organic donors, mercury and
silver and possibly other heavy metals bind tightly to several protein side-chains, forming stable
metal-protein complexes. The formation of metal-protein complexes would lead to an incomplete
protein unfolding. Further enzymatic cleavage would then result in the creation of cryptic self-pep-
tides. Presentation of these cryptic self-peptides by MHC class II to autoreactive T cells could sub-
sequently lead to the stimulation of autoreactive B cells and the production of autoantibodies
(Griem and Gleichmann 1995, Johansson et al. 1997). In this study we considered this proposition
and performed experiments to address the question whether mercury and silver induce comparable
ANolA responses in H-2 susceptible mouse strains.
The experiments were performed using H-2 congenic of B10 mouse strains, B10.S (H-2s), B10.G
(H-2q) and B10.M (H-2f), as well as mouse strains expressing H-2s, H-2q, and H-2f genotypes on
different backgrounds (A.SW, FVB/N, and A.CA respectively).
24
We found that similar to mercury, silver was able to induce ANolA/AFA in mice with H-2s and H-
2q genotypes, irrespective of their non-H-2 genes. This is consistent with findings in a previous
study (Hultman et al. 1995), and indicates that these H-2 genotypes per se are able to confer sus-
ceptibility to both silver- and mercury-induced ANolA/AFA production.
Although mercury and silver showed similarities in the induction of ANolA production in H-2s
and H-2q mice, they also exhibited two major differences. First, silver was found to be less potent
than mercury in stimulating the ANolA response in H-2s and H-2q mice, as evidenced by consis-
tently lower serum titers of ANolA. Second, and more importantly, in contrast to mercury, silver
was not able to induce ANolA responses in mice with H-2f genotype (B10.M and A.CA mice).
The unresponsiveness to silver-induced ANolA synthesis in these mice could not be due to sup-
pressive effects exerted by non-H-2 genes, as A.CA mice, which are high responders to mercury,
were completely resistant to silver.
Analysis of F1 hybrid crosses between silver susceptible A.SW (H-2s) and silver-resistant A.CA
(H-2f) mice showed that these F1 hybrids were resistant to silver but not mercury. This finding
was novel and surprising, since both H-2 haplotypes are co-expressed in the F1 hybrids (H-2s/f);
some degree of ANolA responses was expected in the silver injected animals. Since both H-2s and
H-2f haplotypes are permissive for the development of nucleolar/fibrillarin specific T cells, the
unresponsiveness to silver observed in the F1 hybrids can not be attributed to lack of specific T
cells or specific T cell help. In fact, the mercury-induced ANolA response in the F1 hybrids indi-
cates that T cells capable of activating antinucleolar/ fibrillarin specific B cells are present in these
mice. Therefore it seems likely that expression of the silver-resistant H-2f haplotype on APCs con-
fers the resistance property to silver-induced ANolA synthesis in the F1 hybrids.
Based on current knowledge about the presentation of cryptic self-determinants and on our find-
ings, we proposed that mercury and silver might use similar pathways to induce ANolA synthesis
in susceptible mice. However, they create and/or display dissimilar cryptic self-peptides, which
lead to differential production of ANolA in susceptible mouse strains. It is also possible that com-
pared to mercury, silver is less efficient in providing co-stimulating conditions (as it is less toxic
than mercury) for ANolA production in H-2 susceptible strains.
Paper III.
As it has been mentioned in the introduction section of this thesis, genetic predisposition, immune
dysregulation and exposure to environmental factors are the major prerequisites for the develop-
ment of autoimmune diseases. In the context of this statement, we have proposed that mercury as
an environmental hazard can exert more adverse immunological effects in individuals who possess
genetic or immune dysregulating predisposing factors for the development of autoimmune disease.
25
Investigators in our group have previously tested this proposition in young mice in strains which,
are genetically predisposed to develop either a systemic lupus erythematosus like [(NZB x
NZW)F1 (H-2d/z) hybrids] or an insulin dependent type1 like [nonobese diabetic (NOD) mice]
autoimmune diseases. As a support for our proposition, it was observed that in both strains of mice
mercury was able to induce a strong immune activation with autoimmune characteristics (Al-
Balaghi et al. 1996, Brenden et al. 2001). In this paper we further tested this hypothesis, using
type 1 tight-skin mice, a murine model for scleroderma (Green et al. 1976, Saito et al. 1999). We
investigated the mercury-induced immunopathological alterations in Tsk1/+ and their non-Tsk
(+/+) littermates, F1 hybrids between Tsk1/+ and B10.S (H-2s) mice, as well as in backcross
hybrids between (Tsk1/+ x B10.S)F1 and B10.S mice, in order to clarify the role of mercury in the
development of scleroderma-like disease.
Our first observation was that the number of splenic antibody producing cells in the Tsk1/+ mice
was of a significantly higher magnitude than in their non-Tsk(+/+) littermates. This is consistent
with findings that Tsk1/+ have elevated serum levels of different Ig isotypes compared to non-Tsk
littermates (Saito et al. 2002). This implies that, in addition to a defect in the fibrillin1 gene that
causes formation of collagen deposits in the skin and internal organs (Green et al. 1976, Saito et
al. 1999, Siracusa et al. 1996), Tsk1/+ mice also exhibit dysfunctional humoral immunity. It has
therefore been proposed that B cells in these mice possess a hyperactive phenotype, and that CD19
may play an important role since CD19-deficient Tsk1/+ mice do not exhibit this phenotype (Saito
et al. 2002).
A striking finding in this study was that after exposure to mercury the B cells of Tsk1/+ mice were
hyperactive and a strong immune/autoimmune response was induced, but this could not be
observed in their non-Tsk (+/+) littermates. Thus, while Tsk1/+ mice can be categorized as mercu-
ry high responders, their non-Tsk littermates can be classified as low responders.
The hallmark of mercury-induced autoimmunity in mice is the production of ANolAs that react
with fibrillarin (Hultman and Eneström 1989, Robinson et al. 1984, Mirtcheva et al. 1989,
Hultman et al. 1992, Hultman et al. 1993, Hanley et al. 1997, Abedi-Valugerdi and Möller 2000).
Fibrillarin is also a target for autoantibodies in some patients with scleroderma. These ANolA from
scleroderma patients have been shown to share several similarities with mercury-induced ANolA
(Takeuchi et al. 1995). We found in this study that despite possessing hyperactive B cells and skin
fibrosis, Tsk1/+ mice did non produce ANolA in response to mercury. Even co-expression of the
Tsk1/+ phenotype with mercury-susceptible H-2s haplotype could not evoke an ANolA response in
mercury-treated (Tsk1/+ x B10.s)F1 hybrids. This suggests that neither presence of hyperrespon-
sive B cells nor the expression of Tsk1 phenotype in Tsk1/+ mice can fulfill the requirements for
development of ANolA production.
26
One of the main characteristics of the Tsk 1 mutation is the occurrence of skin fibrosis (Green et
al. 1976, Saito et al. 1999, Siracusa et al. 1996). However, we observed that mercury, despite
being able to induce strong immune activation in Tsk1/+ mice, could not potentiate the develop-
ment of skin fibrosis. This observation may reflect two things, either that the skin fibrosis is only
weakly associated with the immune dysfunction in the Tsk1/+ mice, or that mercury differentially
affects the immune dysfunction and the development of skin fibrosis. The latter is supported by
our findings that treatment with mercury resulted in a reduction of skin thickness in the Tsk1/+
mice, and the findings that mercury and other heavy metal ions, such as cadmium and silver, have
been shown to suppress the collagen synthesis in human rheumatoid synovial cells and in human
lung fibroblasts in vitro (Goldberg et al. 1983, Chambers et al. 1998). Obviously further experi-
ments are required to verify mercury’s suppressive effects on the development of skin fibrosis in
Tsk1/+ mice.
Paper IV.
Native type II collagen (CII) in complete Freund´s adjuvant (CFA) can induce an autoimmune pol-
yarthritis, collagen induced arthritis (CIA), which resembles human rheumatoid arthritis, in sus-
ceptible animals of different species, including rats, mice and monkeys (Trentham et al. 1977, Yoo
et al. 1988, Courtenay et al. 1980). Genes encoding the H-2 control murine susceptibility to CIA,
and only mice expressing H-2q and H-2r genotypes develop polyarthritis after immunization with
CII in CFA. It has also been shown that CD4+ T cells play a central role in the development of
CIA in mice (Luross and Williams 2001, Holmdahl et al. 2002). Moreover, in CIA, Th1 type of
cells appear to play a pathogenic role, whereas effector functions of Th2 type of cells is down-reg-
ulated and also seem to protect against the development of inflammatory arthitides (Schulze-
Koops and Kalden 2001). In contrast, exposure of the same mice to mercuric chloride induces a
Th2-oriented systemic autoimmune disease (described elsewhere in this thesis). The objective of
this paper was to address the Th1/Th2 cross-regulation paradigm by examining whether in a sus-
ceptible animal, simultaneous mercury-induced autoimmunity and CIA can interact. The conse-
quence of such a reaction might be prevention and/or amelioration of one or both of these dis-
eases.
Administration of mercury prior to the induction of CIA caused only augmented production of
anti-CII IgG2a antibodies, without affecting any other characteristics of the two diseases. This
finding indicates that under certain circumstances, Th2-mediated mercury-induced autoimmunity
and Th1-type CIA can coexist without influencing one another. Furthermore, these findings sug-
gest that under inflammatory conditions caused by mercury and CIA together, mercury acts as an
additional adjuvant that preferentially enhances the induction of specific IgG2a (a Th1 related iso-
type) anti-CII antibodies associated with CIA.
27
In agreement with this suggestion, mercury has been shown to exert adjuvant effects in connection
with the specific humoral immune responses in rats immunized with ovalbumin (Watzl et al. 1999)
and in mice immunized with horse red blood cells (Roether et al. 2002).
During the induction phase of CIA, treatment with mercury resulted in an aggravation of CIA, as
well as increased mercury-induced ANolA production. This observation implies that the
immune/autoimmune activation, which occurs during the course of CIA, is also able to potentiate
the humoral immune responses induced specifically by mercury. Although further investigations
are required to clarify the exact mechanisms underlying this phenomenon, we propose that the
adjuvant properties of CFA may be involved. In agreement with this proposal is the report that
CFA increases mercury-induced immune responses in susceptible mice and can even cause a
mouse strain that does not otherwise respond to mercury to exhibit a weak response (Hultman and
Eneström 1987).
A striking finding in our study was that mercury-administration after the onset of CIA exacerbated
the severity of CIA. Increased serum levels of IgE and of anti-CII antibodies, but a substantial
reduction in ANolA titers accompanied this effect. Interestingly, during the remission phase of
CIA, this suppression is abolished and the synthesis of ANolA restored. In combination with the
observation described above, this finding suggests that while the induction and remission phases of
CIA provide permissive conditions, the effector phase of CIA suppresses the mercury-induced pro-
duction of ANolA.
In the light of differential expression of Th1 cytokines during the course of CIA, it is conceivable
that production of high levels of IFN-γ or IL-12 during the onset of the disease exerts a suppres-
sive effect on the mercury-induced production of ANolA. Furthermore, since the synthesis of IFN-
γ is essential for the development and induction of mercury-induced autoimmunity (Kono et al.
1998, Häggqvist and Hultman 2001), it is unlikely that expression of this cytokine causes the
down-regulation of ANolA. Rather, it appears highly likely that IL-12 exerts this suppressive
effect. In fact, this proposal receives strong support from the finding that administration of IL-12
to susceptible mice treated with mercury potently down-regulates the production of different iso-
types of ANolA without affecting the synthesis of IgE or the deposition of immune complexes in
the kidney (Bagenstose et al. 1998).
Finally, splenic IFN-γ/IL-4 ratios were increased in CIA mice, irrespective of whether they were
injected with mercury or not, but reduced in mice treated with mercury alone. This result further
confirms the concept that CIA is a Th1-oriented autoimmune inflammatory disease (Luross and
Williams 2001, Schulze-Koops and Kalden 2001), whereas mercury-induced autoimmunity is Th2-
mediated (Goldman et al. 1991, Eneström and Hultman 1995, Griem and Gleichmann 1995,
Mathieson 1995).
28
Moreover, injection of mercury into mice at any stage of CIA, i.e., prior to or during the induction
phase or after onset of the disease, did not reverse the increased IFN-γ/IL-4 ratio exhibited by
these animals. This finding is in agreement with the clinical symptoms observed in CIA mice treat-
ed with mercury and implies that down-regulation of the Th1-mediated autoimmunity elicited by
CIA requires a more potent stimulation of Th2 cytokine production than that caused by mercury.
Consistent with this proposal is the report that immunization with CII in incomplete Freund’s adju-
vant, a potent Th2 adjuvant (Billiau and Matthys 2001) down-regulates the synthesis of IFN-γ, as
well as the development of CIA (Mauri et al. 2003).
29
CONCLUDING REMARKS
To date, it is well accepted that for the development of autoimmune diseases, integrations of genet-
ic predisposition, environmental factors and immune dysregulation are required. Regarding the
genetic markers of susceptibility to autoimmune diseases, certain genes for MHC molecules that
both form and adjust the adaptive immune response seem to play a crucial role. Moreover, a num-
ber of non-MHC genes that are important in controlling immune responses have also been associ-
ated with the development of autoimmune diseases. It is likely that in certain humans and mice
that possess autoimmune susceptible genetic make up, exposure to certain environmental agents
such as infections and chemicals leads to an immune dysregulation as well as an inappropriate
immune response against self-components, which thereby results in organ damage and/or dysfunc-
tion. In the studies performed in this thesis, I was able to show that indeed under a certain environ-
mental condition, an immune dysregulation with autoimmune characteristics could occur in geneti-
cally susceptible, but not resistant animals. I was also able to show how resistance to this environ-
mentally induced autoimmunity is inherited. Furthermore, we could also comprehend to some
extent how two environmentally induced autoimmune diseases having opposing immune dysregu-
lations, regarding their cytokine profiles, interact in a susceptible mouse strain. However, many of
the steps involved in the pathogenesis of autoimmune diseases require further studies. Future data
will, hopefully, lead to a better understanding of the mechanisms that control the autoimmune phe-
notype and the development of new and better treatment strategies.
30
ACKNOLEDGEMENTSThis work was performed at the Department of Immunology, Wenner-Gren Institute, Stockholm
University. I would like to take the opportunity to express my sincere gratitude to family, friends
and colleagues who have helped and supported me during the years. I would especially like to
thank the following persons:
Manuchehr Abedi-Valugerdi, my supervisor, for endless support, patience and encouragements,
for always having an open mind and time to discuss science and for teaching me all I know about
working with mice.
Göran Möller, for accepting me as a student at the department.
Marita Troye-Blomberg, for all the help she has provided when I needed it the most.
Lena Israelsson, for teaching me the laboratory skills I needed, for being a good friend and for
making me feel welcome when I came to the department.
Gunilla Tillinger, our former secretary, for all “non-science” discussions, and for social skills out
of the ordinary.
The seniors, Carmen Fernandez, Klavs Berzins and Eva Sverremark-Ekström, for sharing
their knowledge of immunology and other things.
Alf Grandien, for keeping me company on the balcony in rain or sunshine, and for emphasizing
the importance of participating in the social events at the department.
My roommates, Karin Lindroth, always ready to help solve both technical and scientific prob-
lems, and for being a good cross-word puzzle partner, Anna Tjärnlund for good times and many
laughs and for not kicking me out when I have been to chatty, and Masashi Hayano, for always
being nice and forthcoming.
Ann Sjölund, Margareta Hagstedt and Gelana Yadeta, for trying to make the department run
smoothly and for providing all the help and assistance I could have asked for.
31
All the present students and all the former students at the department, Ariane Rodriguez- Munoz,
Anna-Karin Larsson, Qasi Kaleda Rahman, Jacob Minang, Ahmed Bolad, Salah Eldin
Farouk, Manijeh Vafa, Nina-Maria Vasconcelos, Halima Balugon, Shiva S. Esfahani,
Valentina Screpanti, Gun Jönsson, Eva Nordström, Ben Adu Gyan Cecilia Rietz, Mounira
Djerbi, Anki Häglind, Elisabeth Hugosson and Caroline Ekberg, for all the good times we
have shared both at the lab and outside “in the real world”.
The staff in our animal facilities, Pia Lundell, Eva Nygren and Solveig Sundberg for taking care
of my mice.
All my friends, and especially Catarina Kvarnmalm for support and encouragements when
everything felt impossible and for a memorable night at Elverket, we must do that again sometime.
My parents, Hjördis and Sten-Ove, för att ni alltid har trott att jag ska fixa det, även när jag inte
har trott det själv och för att ni alltid ställer upp när det behövs.
My brother Roland and my sisters Karin and Yvonne and their families, for training me to argue
in endless discussions at family dinners.
“Silverbågarna”, Inger and Sten, for taking interest in what I do, even though I never worked with
rats.
Evelina Silverbåge, för att du finns och ser till att jag inte tappar focus på vad som egentligen är
viktigt här i livet.
And last but not least, Thomas Silverbåge, for love and support, for always trying to understand
what I am doing, for solving all my computer problems and for printing this thesis.
32
REFERENCES
Abedi-Valugerdi M, Moller G. (2000). Contribution of H-2 and non-H-2 genes in the control of
mercury-induced autoimmunity. Int Immunol. 12: 1425-30.
al-Balaghi S, Moller E, Moller G, Abedi-Valugerdi M. (1996). Mercury induces polyclonal B cell
activation, autoantibody production and renal immune complex deposits in young (NZB x
NZW)F1 hybrids. Eur J Immunol. 26: 1519-26.
Alferink J, Tafuri A, Vestweber D, Hallmann R, Hammerling GJ, Arnold B. (1998). Control of
neonatal tolerance to tissue antigens by peripheral T cell trafficking. Science. 282: 1338-41.
Arnett FC, Reveille JD, Goldstein R, Pollard KM, Leaird K, et al.. (1996). Autoantibodies to fib-
rillarin in systemic sclerosis (scleroderma). An immunogenetic, serologic, and clinical analysis.
Arthritis Rheum. 39: 1151-60.
Aten J, Veninga A, Coers W, Sonnenberg A, Timpl R, et al.. (1995). Autoantibodies to the laminin
P1 fragment in HgCl2-induced membranous glomerulopathy. Am J Pathol. 146: 1467-80.
Bacharier LB, Geha RS. (2000). Molecular mechanisms of IgE regulation. J Allergy Clin Immunol.
105: S547-58.
Bagenstose LM, Class R, Salgame P, Monestier M. (2002). B7-1 and B7-2 co-stimulatory mole-
cules are required for mercury-induced autoimmunity. Clin Exp Immunol. 127: 12-9.
Bagenstose LM, Salgame P, Monestier M. (1998). IL-12 down-regulates autoantibody production
in mercury-induced autoimmunity. J Immunol. 160: 1612-7.
Bagenstose LM, Salgame P, Monestier M. (1998). Mercury-induced autoimmunity in the absence
of IL-4. Clin Exp Immunol. 114: 9-12.
Benschop RJ, Aviszus K, Zhang X, Manser T, Cambier JC, Wysocki LJ. (2001). Activation and
anergy in bone marrow B cells of a novel immunoglobulin transgenic mouse that is both hapten
specific and autoreactive. Immunity. 14: 33-43.
Billiau A, Matthys P. (2001). Modes of action of Freund’s adjuvants in experimental models of
autoimmune diseases. J Leukoc Biol. 70: 849-60.
33
Bouneaud C, Kourilsky P, Bousso P. (2000). Impact of negative selection on the T cell repertoire
reactive to a self-peptide: a large fraction of T cell clones escapes clonal deletion. Immunity. 13:
829-40.
Bowman C, Mason DW, Pusey CD, Lockwood CM. (1984). Autoregulation of autoantibody syn-
thesis in mercuric chloride nephritis in the Brown Norway rat. I. A role for T suppressor cells. Eur
J Immunol. 14: 464-70.
Brenden N, Rabbani H, Abedi-Valugerdi M. (2001). Analysis of mercury-induced immune activa-
tion in nonobese diabetic (NOD) mice. Clin Exp Immunol. 125: 202-10.
Castedo M, Pelletier L, Pasquier R, Druet P. (1994). Improvement of TH1 functions during the
regulation phase of mercury disease in brown Norway rats. Scand J Immunol. 39: 144-50.
Chalopin JM, Lockwood CM. (1984). Autoregulation of autoantibody synthesis in mercuric chlo-
ride nephritis in the Brown Norway rat. II. Presence of antigen-augmentable plaque-forming cells
in the spleen is associated with humoral factors behaving as auto-anti-idiotypic antibodies. Eur J
Immunol. 14: 470-5.
Chambers RC, Laurent GJ, Westergren-Thorsson G. (1998). Cadmium inhibits proteoglycan and
procollagen production by cultured human lung fibroblasts. Am J Respir Cell Mol Biol. 19: 498-
506.
Cooke MP, Heath AW, Shokat KM, Zeng Y, Finkelman FD, et al.. (1994). Immunoglobulin signal
transduction guides the specificity of B cell-T cell interactions and is blocked in tolerant self-reac-
tive B cells. J Exp Med. 179: 425-38.
Courtenay JS, Dallman MJ, Dayan AD, Martin A, Mosedale B. (1980). Immunisation against het-
erologous type II collagen induces arthritis in mice. Nature. 283: 666-8.
D’Cruz D. (2000). Autoimmune diseases associated with drugs, chemicals and environmental fac-
tors. Toxicol Lett. 112-113: 421-32.
Derbinski J, Schulte A, Kyewski B, Klein L., (2001). Promiscuous gene expression in medullary
thymic epithelial cells mirrors the peripheral self. Nat Immunol. 2: 1032-9.
Doth M, Fricke M, Nicoletti F, Garotta G, Van Velthuysen ML, et al.. (1997). Genetic differences
in immune reactivity to mercuric chloride (HgCl2):immunosuppression of H-2d mice is mediated
by interferon-gamma (IFN-gamma). Clin Exp Immunol. 109: 149-56.
34
Druet E, Guery JC, Ayed K, Guilbert B, Avrameas S, Druet P. (1994). Characteristics of polyreac-
tive and monospecific IgG anti-laminin autoantibodies in the rat mercury model. Immunology. 83:
489-94.
Druet E, Sapin C, Fournie G, Mandet C, Gunther E, Druet P. (1982). Genetic control of suscepti-
bility to mercury-induced immune nephritis in various strains of rat. Clin Immunol Immunopathol.
25: 203-12.
Druet E, Sapin C, Gunther E, Feingold N, Druet P. (1977). Mercuric chloride-induced anti-
glomerular basement membrane antibodies in the rat: genetic control. Eur J Immunol. 7: 348-51.
Druet P. (1990). Experimental models of autoimmune diseases. Curr Eye Res. 9 Suppl: 139-44.
Enestrom S, Hultman P. (1995). Does amalgam affect the immune system? A controversial issue.
Int Arch Allergy Immunol. 106: 180-203.
Eris JM, Basten A, Brink R, Doherty K, Kehry MR, Hodgkin PD. (1994). Anergic self-reactive B
cells present self antigen and respond normally to CD40-dependent T-cell signals but are defective
in antigen-receptor-mediated functions. Proc Natl Acad Sci U S A. 91: 4392-6.
Finkelman FD, Holmes J, Katona IM, Urban JF Jr, Beckmann MP, et al.. (1990). Lymphokine
control of in vivo immunoglobulin isotype selection. Annu Rev Immunol. 8: 303-33.
Fournie GJ, Cautain B, Xystrakis E, Damoiseaux J, Mas M, et al.. (2001). Cellular and genetic
factors involved in the difference between Brown Norway and Lewis rats to develop respectively
type-2 and type-1 immune-mediated diseases. Immunol Rev.184: 145-60.
Fournier MJ, Maxwell ES. (1993). The nucleolar snRNAs: catching up with the spliceosomal
snRNAs. Trends Biochem Sci. 18: 131-5.
Frausto da Silva JJR, Williams RJP. (1991). The biological chemistry of the elements: the inorgan-
ic chemistry of life, 1st edn. New York: Oxford University Press.
Fridman WH. (1993). Regulation of B-cell activation and antigen presentation by Fc receptors.
Curr Opin Immunol. 5: 355-60.
Gately MK, Renzetti LM, Magram J, Stern AS, Adorini L, et al.. (1998). The interleukin-12/inter-
leukin-12-receptor system: role in normal and pathologic immune responses. Annu Rev Immunol.
16: 495-521.
35
Goldberg RL, Kaplan SR, Fuller GC. (1983). Effect of heavy metals on human rheumatoid syn-
ovial cell proliferation and collagen synthesis. Biochem Pharmacol. 32: 2763-6.
Goldman M, Druet P, Gleichmann E. (1991). TH2 cells in systemic autoimmunity: insights from
allogeneic diseases and chemically-induced autoimmunity. Immunol Today. 12: 223-7.
Green MC, Sweet HO, Bunker LE. (1976). Tight-skin, a new mutation of the mouse causing
excessive growth of connective tissue and skeleton. Am J Pathol. 82: 493-512.
Gregersen PK. (1999). Genetics of rheumatoid arthritis: confronting complexity. Arthritis Res. 1:
37-44.
Gregersen PK. (2003). Teasing apart the complex genetics of human autoimmunity: lessons from
rheumatoid arthritis. Clin Immunol. 107: 1-9.
Griem P, Gleichmann E. (1995). Metal ion induced autoimmunity. Curr Opin Immunol. 7: 831-8.
Griffiths AJF, Gelbart VM, Miller JH, Lewontin RC, eds1999. Modern Genetic Analysis. New
York: WH Freeman, 133-37.
Haggqvist B, Hultman P. (2001). Murine metal-induced systemic autoimmunity: baseline and
stimulated cytokine mRNA expression in genetically susceptible and resistant strains. Clin Exp
Immunol. 126: 157-64.
Hanley GA, Schiffenbauer J, Sobel ES. (1997). Class II haplotype differentially regulates immune
response in HgCl2-treated mice. Clin Immunol Immunopathol. 84: 328-37.
Hanley GA, Schiffenbauer J, Sobel ES. (1998). Resistance to HgCl2-induced autoimmunity in
haplotype-heterozygous mice is an intrinsic property of B cells. J Immunol. 161: 1778-85.
Heinzel FP, Schoenhaut DS, Rerko RM, Rosser LE, Gately MK. (1993). Recombinant interleukin
12 cures mice infected with Leishmania major. J Exp Med. 177: 1505-9.
Hirsch F, Couderc J, Sapin C, Fournie G, Druet P. (1982). Polyclonal effect of HgCl2 in the rat, its
possible role in an experimental autoimmune disease. Eur J Immunol. 12: 620-5.
Hirsch F, Kuhn J, Ventura M, Vial MC, Fournie G, Druet P. (1986). Autoimmunity induced by
HgCl2 in Brown-Norway rats. I. Production of monoclonal antibodies. J Immunol. 136: 3272-6.
36
Holmdahl R, Bockermann R, Backlund J, Yamada H. (2002). The molecular pathogenesis of colla-
gen-induced arthritis in mice—a model for rheumatoid arthritis. Ageing Res Rev. 1: 135-47.
Hultman P, Enestrom S. (1987). The induction of immune complex deposits in mice by peroral
and parenteral administration of mercuric chloride: strain dependent susceptibility. Clin Exp
Immunol. 67: 283-92.
Hultman P, Enestrom S, Pollard KM, Tan EM. (1989). Anti-fibrillarin autoantibodies in mercury-
treated mice. Clin Exp Immunol. 78: 470-7.
Hultman P, Eneström S. (1989). Mercury induced B-cell activation and anti-nuclear antibodies in
mice. Clin Lab Immunol. 28:143-50.
Hultman P, Bell LJ, Enestrom S, Pollard KM., (1992). Murine susceptibility to mercury. I.
Autoantibody profiles and systemic immune deposits in inbred, congenic, and intra-H-2 recombi-
nant strains. Clin Immunol Immunopathol. 65: 98-109.
Hultman P, Enestrom S. (1992). Dose-response studies in murine mercury-induced autoimmunity
and immune-complex disease. Toxicol Appl Pharmacol. 113: 199-208.
Hultman P, Bell LJ, Enestrom S, Pollard KM. (1993). Murine susceptibility to mercury. II. autoan-
tibody profiles and renal immune deposits in hybrid, backcross, and H-2d congenic mice. Clin
Immunol Immunopathol. 68: 9-20.
Hultman P, Enestrom S, Turley SJ, Pollard KM. (1994). Selective induction of anti-fibrillarin
autoantibodies by silver nitrate in mice. Clin Exp Immunol. 96: 285-91.
Hultman P, Ganowiak K, Turley SJ, Pollard KM. (1995). Genetic susceptibility to silver-induced
anti-fibrillarin autoantibodies in mice. Clin Immunol Immunopathol. 77: 291-7.
Hultman P, Turley SJ, Enestrom S, Lindh U, Pollard KM. (1996). Murine genotype influences the
specificity, magnitude and persistence of murine mercury-induced autoimmunity. J Autoimmun. 9:
139-49.
Janeway CA Jr; Goodnow CC, Medzhitov R. (1996). Danger-pathogen on the premises!
Immunological tolerance. Curr Biol. 6(5):519-22.
Jenkins MK, Schwartz RH. (1987). Antigen presentation by chemically modified splenocytes
induces antigen-specific T cell unresponsiveness in vitro and in vivo. J Exp Med. 165: 302-19.
37
Johansson U, Hansson-Georgiadis H, Hultman P. (1997). Murine silver-induced autoimmunity: sil-
ver shares induction of antinucleolar antibodies with mercury, but causes less activation of the
immune system. Int Arch Allergy Immunol. 113: 432-43.
Johansson U, Hansson-Georgiadis H, Hultman P. (1998). The genotype determines the B cell
response in mercury-treated mice. Int Arch Allergy Immunol. 116: 295-305.
Kasturi KN, Hatakeyama A, Spiera H, Bona CA. (1995). Antifibrillarin autoantibodies present in
systemic sclerosis and other connective tissue diseases interact with similar epitopes. J Exp Med.
181: 1027-36.
Kono DH, Balomenos D, Pearson DL, Park MS, Hildebrandt B, et al.. (1998). The prototypic Th2
autoimmunity induced by mercury is dependent on IFN-gamma and not Th1/Th2 imbalance. J
Immunol. 161: 234-40.
Kono DH, Theofilopoulos AN. (2000). Genetics of systemic autoimmunity in mouse models of
lupus. Int Rev Immunol. 19: 367-87.
Kono DH, Park MS, Szydlik A, Haraldsson KM, Kuan JD, et al.. (2001). Resistance to xenobiotic-
induced autoimmunity maps to chromosome. J Immunol. 167: 2396-403.
Kretz-Rommel A, Duncan SR, Rubin RL. (1997). Autoimmunity caused by disruption of central T
cell tolerance. A murine model of drug-induced lupus. J Clin Invest. 99: 1888-96.
Kretz-Rommel A, Rubin RL. (2000). Disruption of positive selection of thymocytes causes
autoimmunity. Nat Med. 6: 298-305.
Kreuwel HT, Sherman LA. (2001). The T-cell repertoire available for recognition of self-antigens.
Curr Opin Immunol. 13: 639-43.
Kurts C, Miller JF, Subramaniam RM, Carbone FR, Heath WR. (1998). Major histocompatibility
complex class I-restricted cross-presentation is biased towards high dose antigens and those
released during cellular destruction. J Exp Med. 188: 409-14.
Lechner O, Lauber J, Franzke A, Sarukhan A, von Boehmer H, Buer J. (2001) Fingerprints of
anergic T cells.Curr Biol. 11: 587-95.
38
Lo D, Burkly LC, Flavell RA, Palmiter RD, Brinster RL. (1989). Tolerance in transgenic mice
expressing class II major histocompatibility complex on pancreatic acinar cells. J Exp Med. 170:
87-104.
Lohmann T, Leslie RD, Londei M. (1996) T cell clones to epitopes of glutamic acid decarboxylase
65 raised from normal subjects and patients with insulin dependent diabetes. J. Autoimmun. 9:
385-9.
Luross JA, Williams NA. (2001). The genetic and immunopathological processes underlying colla-
gen-induced arthritis. Immunology. 103: 407-16.
Maloy KJ, Powrie F. (2001). Regulatory T cells in the control of immune pathology. Nat Immunol.
2: 816-22.
Mason D. (1998). A very high level of crossreactivity is an essential feature of the T-cell receptor.
Immunol Today. 19: 395-404.
Mathieson PW, Stapleton KJ, Oliveira DB, Lockwood CM. (1991). Immunoregulation of mercuric
chloride-induced autoimmunity in Brown Norway rats: a role for CD8+ T cells revealed by in vivo
depletion studies. Eur J Immunol. 21: 2105-9.
Mathieson PW. (1992). Mercuric chloride-induced autoimmunity. Autoimmunity. 13: 243-7.
Mathieson PW. (1995). Mercury: god of Th2 cells? Clin Exp Immunol. 102: 229-30.
Matzinger P, (1994).Tolerance, danger, and the extended family. Annu Rev Immunol.12: 991-1045.
Mauri C, Feldmann M, Williams RO. (2003). Down-regulation of Th1-mediated pathology in
experimental arthritis by stimulation of the Th2 arm of the immune response. Arthritis Rheum. 48:
839-45.
Mayes MD. (1999). Epidemiologic studies of environmental agents and systemic autoimmune dis-
eases. Environ Health Perspect. 107 Suppl 5: 743-8.
McGargill MA, Derbinski JM, Hogquist KA. (2000). Receptor editing in developing T cells. Nat
Immunol. 1: 336-41.
Medzhitov R, Janeway CA Jr, (1997). Innate immunity: the virtue of a nonclonal system of recog-
nition. Cell. 91(3): 295-8.
39
Mirtcheva J, Pfeiffer C, De Bruijn JA, Jacquesmart F, Gleichmann E., (1989). Immunological
alterations inducible by mercury compounds. III. H-2A acts as an immune response and H-2E as
an immune ””suppression”” locus for HgCl2-induced antinucleolar autoantibodies. Eur J Immunol.
19: 2257-61.”
Mohan C. (2001). Murine lupus genetics: lessons learned. Curr Opin Rheumatol. 13: 352-60.
Morahan G, Morel L. (2002). Genetics of autoimmune diseases in humans and in animal models.
Curr Opin Immunol. 14: 803-11.
Morel L, Blenman KR, Croker BP, Wakeland EK. (2001). The major murine systemic lupus ery-
thematosus susceptibility locus, Sle1, is a cluster of functionally related genes. Proc Natl Acad Sci
U S A. 98: 1787-92.
Mosmann TR, Coffman RL. (1989). TH1 and TH2 cells: different patterns of lymphokine secre-
tion lead to different functional properties. Annu Rev Immunol. 7: 145-73.
Nemazee D. (2000). Receptor selection in B and T lymphocytes. Annu Rev Immunol. 18: 19-51.
Nepom GT, Erlich H. (1991). MHC class-II molecules and autoimmunity. Annu Rev Immunol. 9:
493-525.
Ochel M, Vohr HW, Pfeiffer C, Gleichmann E. (1991). IL-4 is required for the IgE and IgG1
increase and IgG1 autoantibody formation in mice treated with mercuric chloride. J Immunol. 146:
3006-11.
Ohashi PS, DeFranco AL. (2002). Making and breaking tolerance. Curr Opin Immunol. 14: 744-
59.
Pelletier L, Pasquier R, Guettier C, Vial MC, Mandet C, et al.. (1988). HgC12 induces T and B
cells to proliferate and differentiate in BN rats. Clin Exp Immunol. 71: 336-42.
Pelletier L., Castedo M., Bellon B,. Druet P. (1994). Mercury and autoimmunity. In
Immunotoxicology and Immunopharmacology. Eds: Dean JH. Luster MI. Munson AE. Kimber I.
New York; Raven Press. 539-552.
Pieters R, Ezendam J, Bleumink R, Bol M, Nierkens S. (2002). Predictive testing for autoimmuni-
ty. Toxicol Lett. 127: 83-91.
40
Pietsch P, Vohr HW, Degitz K, Gleichmann E. (1989). Immunological alterations inducible by
mercury compounds. II. HgCl2 and gold sodium thiomalate enhance serum IgE and IgG concen-
trations in susceptible mousestrains. Int Arch Allergy Appl Immunol. 90: 47-53.
Pusey CD, Bowman C, Morgan A, Weetman AP, Hartley B, Lockwood CM. (1990). Kinetics and
pathogenicity of autoantibodies induced by mercuric chloride in the brown Norway rat. Clin Exp
Immunol. 81: 76-82.
Raman K, Mohan C. (2003). Genetic underpinnings of autoimmunity—lessons from studies in
arthritis, diabetes, lupus and multiple sclerosis. Curr Opin Immunol. 15: 651-9.
Rathmell JC, Cooke MP, Ho WY, Grein J, Townsend SE, et al.. (1995). CD95 (Fas)-dependent
elimination of self-reactive B cells upon interaction with CD4+ T cells. Nature. 376: 181-4.
Rathmell JC, Townsend SE, Xu JC, Flavell RA, Goodnow CC. (1996). Expansion or elimination
of B cells in vivo: dual roles for CD40- and Fas (CD95)-ligands modulated by the B cell antigen
receptor. Cell. 87: 319-29.
Reiner SL, Locksley RM. (1995). The regulation of immunity to Leishmania major. Annu Rev
Immunol. 13: 151-77.
Reuter R, Tessars G, Vohr HW, Gleichmann E, Luhrmann R. (1989). Mercuric chloride induces
autoantibodies against U3 small nuclear ribonucleoprotein in susceptible mice. Proc Natl Acad Sci
U S A. 86: 237-41.
Robinson CJ, Abraham AA, Balazs T. (1984). Induction of anti-nuclear antibodies by mercuric
chloride in mice. Clin Exp Immunol. 58: 300-6.
Robinson CJ, Balazs T, Egorov IK. (1986). Mercuric chloride-, gold sodium thiomalate-, and D-
penicillamine-induced antinuclear antibodies in mice. Toxicol Appl Pharmacol. 86:159-69.
Roether S, Rabbani H, Mellstedt H, Abedi-Valugerdi M. (2002). Spontaneous downregulation of
antibody/autoantibody synthesis in susceptible mice upon chronic exposure to mercuric chloride is
not owing to a general immunosuppression. Scand J Immunol. 55: 493-502.
Rothstein TL, Wang JK, Panka DJ, Foote LC, Wang Z, et al.. (1995). Protection against Fas-
dependent Th1-mediated apoptosis by antigen receptor engagement in B cells. Nature. 374: 163-5.
41
Saito E, Fujimoto M, Hasegawa M, Komura K, Hamaguchi Y, et al.. (2002). CD19-dependent B
lymphocyte signaling thresholds influence skin fibrosis and autoimmunity in the tight-skin mouse.
J Clin Invest. 109: 1453-62.
Saito S, Kasturi K, Bona C. (1999). Genetic and immunologic features associated with scleroder-
ma-like syndrome of TSK mice. Curr Rheumatol Rep. 1: 34-7.
Sapin C, Hirsch F, Delaporte JP, Bazin H, Druet P. (1984). Polyclonal IgE increase after HgCl2
injections in BN and LEW rats: a genetic analysis. Immunogenetics. 20: 227-36.
Sapin C, Mandet C, Druet E, Gunther E, Druet P. (1982). Immune complex type disease induced
by HgCl2 in Brown-Norway rats: genetic control of susceptibility. Clin Exp Immunol. 48: 700-4.
Schulze-Koops H, Kalden JR. (2001). The balance of Th1/Th2 cytokines in rheumatoid arthritis.
Best Pract Res Clin Rheumatol. 15: 677-91.
Semana G, Gausling R, Jackson RA, Hafler DA. (1999). T cell autoreactivity to proinsulin epi-
topes in diabetic patients and healthy subjects. J Autoimmun. 12: 259-67.
Silverstein AM, Rose NR. (1997). On the mystique of the immunological self. Immunol Rev. 159:
207-18.
Silverstein AM. (2001). Autoimmunity versus horror autotoxicus: the struggle for recognition. Nat
Immunol. 2: 279-81.
Silverstein AM. (2002). The clonal selection theory: what it really is and why modern challenges
are misplaced. Nat Immunol. 3: 793-96.
Siracusa LD, McGrath R, Ma Q, Moskow JJ, Manne J, Christner PJ, Buchberg AM, Jimenez SA.
(1996). A tandem duplication within the fibrillin 1 gene is associated with the mouse tight skin
mutation. Genome Res. 6: 300-13.
Steinman RM, Nussenzweig MC. (2002). Avoiding horror autotoxicus: the importance of dendritic
cells in peripheral T cell tolerance. Proc Natl Acad Sci U S A. 99: 351-8.
Sypek JP, Chung CL, Mayor SE, Subramanyam JM, Goldman SJ, et al.. (1993). Resolution of
cutaneous leishmaniasis: interleukin 12 initiates a protective T helper type 1 immune response. J
Exp Med. 177: 1797-802.
42
Takeuchi K, Turley SJ, Tan EM, Pollard KM. (1995). Analysis of the autoantibody response to fib-
rillarin in human disease and murine models of autoimmunity. J Immunol. 154: 961-71.
Todd JA, Bell JI, McDevitt HO. (1987). HLA-DQ beta gene contributes to susceptibility and
resistance to insulin-dependent diabetes mellitus. Nature. 329: 599-604.
Trentham DE, Townes AS, Kang AH. (1977). Autoimmunity to type II collagen an experimental
model of arthritis. J Exp Med. 146: 857-68.
Trinchieri G. (1993). Interleukin-12 and its role in the generation of TH1 cells. Immunol Today.14:
335-8.
van der Meide PH, de Labie MC, Botman CA, Aten J, Weening JJ. (1995). Nitric oxide suppresses
IFN-gamma production in the spleen of mercuric chloride-exposed brown Norway rats. Cell
Immunol. 161: 195-206.
Van Parijs L, Abbas AK. (1998). Homeostasis and self-tolerance in the immune system: turning
lymphocytes off. Science. 280: 243-8.
van Vliet E, Uhrberg M, Stein C, Gleichmann E. (1993). MHC control of IL-4-dependent
enhancement of B cell Ia expression and Ig class switching in mice treated with mercuric chloride.
Int Arch Allergy Immunol.101: 392-401.
Wakeland EK, Liu K, Graham RR, Behrens TW. (2001). Delineating the genetic basis of systemic
lupus erythematosus. Immunity. 15: 397-408.
Walker LS, Abbas AK. (2002) The enemy within: keeping self-reactive T cells at bay in the
periphery. Nat Rev Immunol. 2: 11-9.
Wanstrat A, Wakeland E. (2001). The genetics of complex autoimmune diseases: non-MHC sus-
ceptibility genes. Nat Immunol. 2: 802-9.
Watzl B, Abrahamse SL, Treptow-van Lishaut S, Neudecker C, Hansch GM, et al.. (1999).
Enhancement of ovalbumin-induced antibody production and mucosal mast cell response by mer-
cury. Food Chem Toxicol. 37: 627-37.
43
Wucherpfennig KW, Strominger JL. (1995). Selective binding of self peptides to disease-associat-
ed major histocompatibility complex (MHC) molecules: a mechanism for MHC-linked susceptibil-
ity to human autoimmune diseases. J Exp Med. 181: 1597-601.
Wucherpfennig KW. (2001). Insights into autoimmunity gained from structural analysis of MHC-
peptide complexes. Curr Opin Immunol. 13: 650-6.
Yoo TJ, Kim SY, Stuart JM, Floyd RA, Olson GA, et al.. (1988). Induction of arthritis in monkeys
by immunization with type II collagen. J Exp Med. 168: 777-82.
44
