вход по аккаунту


HLA-DR4 mycobacteria heat-shock proteins and rheumatoid arthritis.

код для вставкиСкачать
Arthritis & Rheumatism
Official Journal of the American College of Rheumatology
The cell membrane-associated glycoproteins
known as HLA class I1 antigens include the HLA-DR,
DQ, and DP molecules. These are dimers consisting of
an a and a p chain which fold to form the “groove” in
which peptides are bound before they can be recognized by CD4+ lymphocytes. Recently, it has
emerged that several HLA class I1 DR alleles are
associated with an increased risk of developing rheumatoid arthritis (RA) (1,2).
The potential importance of this observation
was enhanced by the discovery that although several
different DR types show this association, those which
do so share identical or similar sequences in a crucial
part of the third hypervariable region of the p chain
(2,3) (Table 1). In some populations, up to 85% of RA
patients bear HLA class I1 molecules containing one
of these related sequences. This “shared epitope”
forms part of the peptide-binding groove; therefore, its
association with RA could be attributable to an ability
to bind to-and so present to CD4+ T cells-some
crucial peptide. Moreover, some reports suggest that
the T cell response in RA is oligoclonal. That is to say,
when T cells are grown from RA lesions, there is
evidence of restricted usage of the possible alternative
From the Department of Medical Microbiology, University College and Middlesex School of Medicine, London, United
Graham Rook, MD: Reader; John McCulloch, PhD: Research Assistant.
Address reprint requests to Graham Rook, MD, Department of Medical Microbiology, University College and Middlesex
School of Medicine, 67-73 Riding House Street, London WlP 7PP,
United Kingdom.
variable regions of the p chain of the T cell receptor.
Several groups have found a disproportionate percentage of T cells expressing V,14 (ref. 4, and reviewed in
ref. 5). It is logical to postulate that if both the shape of
the peptide-presenting groove and the shape of the T
cell receptor are restricted in RA patients, then so
must be the shape of the peptide that is being recognized when disease-causing T cells interact with the
antigen-presenting cell. Arguments such as these are
used by those who regard the immunogenetic findings as
essential clues to the nature of RA, and feel that
restricted use of DR subtypes and T cell receptor V,
genes indicate that RA is an autoimmune disease, directed toward a specific unidentified “self” component.
The inadequacy of the immunogenetic
There is another side to this story, however.
Some calculations suggest that RA is only about 30%
genetic (6), the remaining factors being environmental.
Moreover, the major histocompatibility complex
(MHC), which includes the HLA genes, contributes
less than 50% of this genetic susceptibility (7). Indeed,
there must be other factors, because the DR subtypes
containing the “shared epitope” are present in approximately 45% of a Caucasoid population without
RA (7). These RA-associated, third hypervariable region sequences are absent from a majority of Greek
patients (€9,and in Israeli Jews, the association is with
“shared epitope”-containing DR1 subtypes, rather
than with DR4 (9).
Arthritis and Rheumatism, Vol. 35, No. 12 (December 1992)
Table 1. HLA-DR variants associated with increased susceptibility to rheumatoid arthritis
Sequence at position 70-74
of the third hypervariable
region of the DRP chain
DRl (0101)
DRw 14,Dw16 (1402)
Gln-Lys- Arg-Ala-Ala
DR4,Dw15 (0405)
DR4,Dw14 (0404 and 0408)
DR4,Dw4 (0401)
DrwlO (1001)
There is also great uncertainty about the significance of the evidence for oligoclonality of T cells in
RA. The V, gene products that are overrepresented in
RA lesions are not the same in all studies, and these
studies have always used very small groups of patients
(for review, see ref. 5).
These uncertainties make it clear that although
any hypothesis as to the cause of RA must be able to
accommodate the DR associations, it is equally important that such a hypothesis should explain those cases
with no such associations.
One solution is to postulate that RA is a clinical
syndrome of more than one disease. Another theory
which cannot yet be dismissed is that the class I1
associations are really due to functions of other genes
that are encoded within the MHC and are in linkage
disequilibrium with HLA class I1 alleles (for review,
see ref. 10). In this editorial we consider those current
hypotheses which suggest that RA is triggered by past
or ongoing infection, in order to see whether the
immunogenetic findings in RA help to support or
eliminate any of them.
Hypotheses based on previous or ongoing
Silman has pointed out that the epidemiology of
RA is best explained by an infectious etiology (6). An
infection could cause the disease in at least 3 different
ways. First, the disease may be due to autoimmunity.
This includes the “hit-and-run” hypothesis, whereby
there is a cross-reactive immune response which persists after the infection has gone. A related hypothesis
suggests that this cross-reactive response requires the
continued presence of the infection. The third, radically different, view is that the disease is due to a slow
bacterial infection, while the autoimmune manifestations are secondary events of little pathogenetic relevance. We now consider these possibilities.
Hypotheses in which the disease is due to autoimmunity. This type of hypothesis assumes that an
infection has caused an immune response to a micro-
bial component, which then causes autoreactivity to a
similar host component. These hypotheses exist in two
forms. First, the “hit-and-run” hypothesis assumes
that such cross-reactive autoimmunity could persist in
the absence of the infectious agent (11); that is to say,
in the absence of the relevant cross-reactive microbial
component, and of microbial factors capable of delivering “second signals” (12). This is controversial. It is
possible to evoke such responses in experimental
animals, but they tend to be self-limiting, and therefore
unlike RA.
The second variant gets over this problem by
suggesting that persistent or recurrent infection in, for
instance, the gut or bladder can maintain the crossreactive response. There is evidence that the “shared
epitope” sequences within the DRpl chains are components of epitopes recognized by monoclonal antibodies raised against HLA class I1 antigens (13).
Therefore, RA could be an autoimmune disease in
which antibody elicited by a cross-reactive antigen is
binding to the DR molecules and causing an immunologic disorder. For instance, there is a claim that
antibodies to Proteus will bind to DR4, and vice-versa
(14). A similar hypothesis is founded on the fact that
there is sequence homology between the third hypervariable region of DR4,Dw4 and a heat-shock protein
(HSP) (dnaJ) from Escherichia coli (15). Antibodies
raised to a peptide based on the third hypervariable
region of this DR4 subtype were able to recognize
d n d . Similarly, antibodies to dnaJ bound to cells
bearing DR4,Dw4, but they failed to recognize cells
bearing DRl although it is another of the “shared
epitope”-containing alleles associated with RA (15).
These authors are suggesting autoimmunity
based on antibodies, but it can be argued that the
“shared epitope” sequences could themselves be presented to T cells, and become the focus of T cellmediated autoimmunity. It is known that peptides
derived from HLA class I1 molecules can be found in
the groove of other class I1 molecules (16). We are not
aware of studies exploring this in relation to RA.
Some hypotheses of this type directly involve
the “shared epitopes” of the RA-associated DR subtypes, either as targets of cross-reactive antibodies
(15), or as essential components of any class I1
“groove” able to present a critical peptide. Such
hypotheses do not explain the cases of RA without
such class I1 associations.
One way around this problem is to postulate
that the microbial antigen which induces the autoimmunity has multiple homologies with a human component(s), so that a variety of class IIN, combinations
could lead to autoimmunity. The bacterial 65-kd HSP
is at first sight a particularly attractive possibility
because of the overwhelming evidence associating
these proteins with the rodent models, as reviewed
elsewhere (17). Moreover, since all bacteria contain
similar homologs of this protein, bowel flora can
provide a continuing cross-reactive stimulus after the
initial triggering infection has passed; indeed, we know
that bowel flora affect the rodent models (for review,
see ref. 18). Unfortunately, the evidence that lymphocytes from RA patients respond abnormally to these
proteins is conflicting and weak. On the other hand, it
is becoming clear that antigen-presenting cells can
process and present their own HSPs in the HLA class
I1 groove, and then become targets of cross-reactive
CD4+ T cells generated in response to the bacterial
homolog (19,20).
These data may merely indicate that this is a
physiologic mechanism for recognizing and eliminating stressed or transformed cells, however. Perhaps
lymphocytes from any site of inflammation are enriched for T cells which recognize autologous HSP
peptides. Moreover, the HSP hypothesis is no better
than the others at explaining the class I1 associations
seen in RA. There do not appear to be any DR4restricted T cell epitopes in this HSP, and hsp65recognizing clones from a DR4-homozygous RA patient were DP-restricted, and responded to a peptide
with limited homology to the equivalent human sequence (21). In fact, these clones will respond to the
corresponding part of the human protein, but only at a
rather high concentration. Is such responsiveness
meaningful or not?
Hypotheses in which the autoimmune manifestations are secondary events. There are numerous pathologic and serologic analogies between RA and the
known “slow bacterial infections” (22), which are
often accompanied by autoantibodies (23), rheumatoid
factor, and arthritis. The most obvious examples are
leprosy, Whipple’s disease, Lyme disease, and syphilis. In these diseases, autoimmunity and arthritis
(when they occur) are regarded as secondary to the
141 1
infection, so why should the same not be true of RA?
Do the immunogeneticfindings rule out cryptic infection
as an explanation of RA? We suggest that they do not.
First, it is clear that there are associations
between HLA class I1 and the type of leprosy which a
patient develops. The nature and strength of these
associations differ in different countries, as they do in
RA (24). The strongest associations are between DR3
and tuberculoid leprosy, and between DQ1 and lepromatous leprosy. It should be emphasized that the DR4
associations in RA may also relate to disease severity
or disease type, as much as they relate to disease
susceptibility. This kind of distinction depends entirely on whether there are subclinical forms of the
disease, and if so, whether these forms can be detected
by the researcher. In Greece, where the majority of
RA patients do not have HLA class I1 molecules that
contain the “shared epitopes” (8), the disease shows
more anti-Ro antibody formation, but less nodule
production and less vasculitis (25).
Similarly, the evidence for oligoclonal T cell
responses with restricted V, gene product usage in RA
is paralleled by data from studies of leprosy. For
instance, de Vries has found that all the DR3-restricted
Mycobacterium leprue-responsive T cell clones from
one individual used a V,5 gene product, while most of
the DR2-restricted clones used V,18 (24).
The most tantalizing immunogenetic parallel
between leprosy and RA is the relationship between
DR4 and the skin-test response to “new tuberculin,”
which is a sonicate preparation of Mycobacterium
tuberculosis (in contrast to purified protein derivative,
which is derived from culture supernatant). In both
leprosy and RA, patients who have DR4 show larger
skin-test responses to this antigen preparation (26,27).
In contrast, DR4 does not influence the response in
normal subjects or in patients with tuberculosis.
Theoretically, there are several ways in which
one can rationalize the fact that the skin-test response
to a very complex mixture of proteins, doubtless
containing thousands of epitopes potentially recognized by T cells, should be influenced in this way by
DR4 in RA and leprosy. However, none of these
hypotheses would lead us to expect RA patients to
respond abnormally to a randomly selected DR4restricted epitope on a protein not thought to crossreact with human components. In one such study
published elsewhere in this issue of Arthritis and
Rheumatism (28), no abnormality was found. The
relationship between DR4 and RA is clearly more
subtle. The main message of this brief review is that
this relationship is not only obscure, but is also, in our
present state of understanding, not particularly helpful. Immunogenetic considerations neither prove that
RA is primarily an autoimmune disease nor disprove
the alternative hvDothesis that it is an infection.
1. Stastny P, Ball EJ, Khan MA, Olsen NJ, Pincus T, Gao
X: HLA-DR4 and other genetic markers in rheumatoid
arthritis. Br J Rheumatol 27:132-138, 1988
2. Lanchbury JS: The HLA association with rheumatoid
arthritis. Clin Exp Rheumatol 10:301-304, 1992
3. Gregersen PK, Silver 3, Winchester RJ: The shared
epitope hypothesis: an approach to understanding the
molecular genetics of susceptibility to rheumatoid arthritis. Arthritis Rheum 30: 1205-1213, 1987
4. Paliard X, West SG, Lafferty JA, Clements JR, Kappler
JW, Marrack P, Kotzin BL: Evidence for the effects of
a superantigen in rheumatoid arthritis. Science 253:325329, 1991
5. Richardson BC: T cell receptor usage in rheumatic
disease. Clin Exp Rheumatol 10:271-283, 1992
6. Silman AJ: Is rheumatoid arthritis an infectious disease?
Br Med J [Clin Res] 303:20&201, 1991
7. Silman AJ: The genetic epidemiology of rheumatoid
arthritis. Clin Exp Rheumatol 10:30%312, 1992
8. Boki K, Vaughan RW, Drosos A, Moutsopoulos H,
Panayi GS, Lanchbury JSS: HLA class I1 sequence polymorphisms and susceptibility to rheumatoid arthritis in
Greece (abstract). Clin Exp Rheumatol 10:313, 1992
9. De Vries N, Ronningen KS, Tilanus MGJ, Rombouts A,
Segal R, Egeland T, Thorsby T, van de Putte LBA,
Brautbar C: Sequencing HLA-DRl in Israeli rheumatoid
arthritis patients (abstract). Clin Exp Rheumatol 10:314,
10. Charron DJ: New aspects of HLA: perspectives for rheumatoid arthritis. Clin Exp Rheumatol 10:293-296, 1992
11. Cohen IR: Autoimmunity to chaperonins in the pathogenesis of arthritis and diabetes. Annu Rev Immunol
9:567-589, 1991
12. Janeway CA: Immunogenicity signals 1,2,3 . . . . . and 0
(news). Immunol Today 10:283-286, 1989
13. Duquesnoy RJ, Marrai M, Hackbath S, Zeevi A: Serological and cellular definition of a new HLA-DR associated determinant, MCI, and its association with rheumatoid arthritis. Hum Immunol 10: 165-176, 1984
14. Ebringer A, Khalafpour S, Wilson C: Rheumatoid arthritis and Proteus: a possible aetiological association.
Rheumatol Int 9:223-228, 1989
15. Albani S, Tuckwell JE, Esparza L, Carson DA, Roudier
J: The susceptibility sequence to rheumatoid arthritis is
a cross-reactive B cell epitope shared by the Escherichia
coli heat shock protein d n d and the histocompatibility
leukocyte antigen DRB 10401 molecule. J Clin Invest
89:327-331, 1992
Rudensky AY, Preston-Hurlburt P, Hong SC, Barlow
A, Janeway CA: Sequence analysis of peptides bound to
MHC class I1 molecules. Nature 353:622-627, 1991
Winfield J, Jarjour W: Do stress proteins play a role in
arthritis and autoimmunity? Immunol Rev 121:193-220,
Rook G, Lydyard P, Stanford J: Mycobacteria and
rheumatoid arthritis. Arthritis Rheum 33:431435, 1990
Koga T, Wand-Wurttenberger A, DeBruyn J, Munk
ME, Schoel B, Kaufmann SH: T cells against a bacterial
heat shock protein recognize stressed macrophages.
Science 245: 1112-1 115, 1989
Hermann E, Lohse AW, van der Zee R, van Eden W,
Mayet WJ, Probst P, Poralla T, Meyer-zum-Buschenfelde KH, Fleischer B: Synovial fluid-derived Yersiniareactive T cells responding to human 65-kDa heat-shock
protein and heat-stressed antigen-presenting cells. Eur J
Immunol 21:2139-2143, 1991
Gaston JS, Life PF, van der Zee R, Jenner PJ, Colston
MJ, Tonks S, Bacon PA: Epitope specificity and MHC
restriction of rheumatoid arthritis synovial T cell clones
which recognize a mycobacterial 65 kDa heat shock
protein. Int Immunol 3:965-972, 1991
Rook GAW, Stanford JL: Slow bacterial infections or
autoimmunity. Immunol Today 13: 160-164, 1992
Shoenfeld Y, Isenberg DA: Mycobacteria and autoimmunity. Immunol Today 9:178-182, 1988
De Vries RR: An immunogenetic view of delayed type
hypersensitivity. Tubercle 72: 161-167, 1991
Drosos AA, Lanchbury JSS, Panayi GS, Moutsopoulos
HM: Rheumatoid arthritis in Greek and British patients:
a prospective and serological comparative study (abstract). Clin Exp Rheumatol 10:314, 1992
Ottenhoff TH, Torres P, de las Aguas JT, Fernandez R,
van Eden W, de Vries RR, Stanford JL: Evidence for an
HLA-DR4-associated immune-response gene for Mycobacterium tuberculosis: a clue to the pathogenesis of
rheumatoid arthritis. Lancet II:310-313, 1986
Bahr GM, Sattar I, Stanford JL, Shaaban MA, al Shimali B, Siddiqui Z, Gabriel M, a1 SafFar M, Shahin A,
Chugh TD, Rook GAW, Behbehani K: HLA-DR and
tuberculin tests in rheumatoid arthritis and tuberculosis.
Ann Rheum Dis 48:6348, 1987
Tan PLJ, Farmiloe S, Young J, Watson JD, Skinner
MA: Lymphocyte responses to DR4/1-restricted peptides in rheumatoid arthritis: the immunodominant T cell
epitope on the 19-kd Mycobactenum tuberculosis protein. Arthritis Rheum 35: 1419-1426, 1992
Без категории
Размер файла
402 Кб
dr4, hla, heat, protein, mycobacterium, arthritis, shock, rheumatoid
Пожаловаться на содержимое документа