вход по аккаунту


Triggering and exacerbation of autoimmune arthritis by the mycoplasma arthritidis superantigen mam.

код для вставкиСкачать
Objective. It has been postulated that superantigens
might play a role in the human rheumatic diseases, by
activation of self-reactive T cells or by induction of autoantibodies. The Mycoplasma arthritidis superantigen
MAM, which is derived from a naturally occurring murine arthitogenic mycoplasma, uses certain V p chains of
the murine T cell receptor (TCR) that have been proposed
to be involved in murine collagen-induced arthritis (CIA).
The present study was designed to determine whether
MAM influences the course of arthritis mediated by immunization with porcine type I1 collagen (PII).
Methods. MAM or phosphate buffered saline (PBS)
was injected locally or systemically into mice convalescing
from CIA or mice suboptimally immunized with collagen.
Results. In contrast to PBS, MAM caused an
exacerbation of arthritis in mice that were recovering
from CIA. MAM also triggered arthritis onset in mice
that had been suboptimally immunized with PI1 up to
160 days previously. Injection of MAM during the onset
From the Department of Internal Medicine, Division of Rheumatology, University of Utah School of Medicine, and the Research
Service of the Veterans Administration, Salt Lake City, Utah.
Supported by grants AR-02255 and AI-12103 from the NIH,
by a grant from the Nora Eccles Treadwell Foundation, and by the
Veterans Administration Research Service.
Barry C. Cole, PhD: Professor, Department of Internal
Medicine, Division of Rheumatology, University of Utah School of
Medicine; Marie M. Griffiths, PhD: Associate Research Professor,
Department of Internal Medicine, Division of Rheumatology, University of Utah School of Medicine, and Research Service of the
Veterans Affairs Medical Center.
Address reprint requests to Barry C. Cole, PhD, Department of Internal Medicine, Division of Rheumatology, University of
Utah School of Medicine, 50 North Medical Drive, Salt Lake City,
UT 84132.
Submitted for publication September 1 , 1992; accepted in
revised form January 12, 1993.
Arthritis and Rheumatism, Vol. 36, No. 7 (July 1993)
phase of CIA also triggered and enhanced the severity of
arthritis in mice given low doses of PII.
Conclusion. MAM can both trigger and exacerbate murine autoimmune arthritis induced by immunization with type I1 collagen. Since T cells bearing the
same Vp TCRs as are used by MAM have been found to
comprise a major portion of the activated cells in the
synovial tissue of patients with rheumatoid arthritis, it is
possible that superantigens similar to MAM may play a
role in this human disease.
In autoimmune disease, a breakdown of selftolerance leads to the generation of an immune response against a specific target antigen or antigens.
Microbial agents have long been thought to trigger
autoimmune disease by possessing epitopes that are
cross-reactive with antigens on target organs. More
recently, it has been suggested that superantigens
derived from bacteria (1,2), mycoplasma (3), or viruses (4-6) may initiate autoimmune disease by activating specific anti-self T cell clones (7-9) or by
forming a superantigen bridge that cross-links T helper
(TH) cells with preimmune B cells, thereby causing
polyclonal B cell activation and secretion of autoimmune antibodies (10,ll).
Superantigens are potent mitogens which activate T cells by a unique pathway that binds the major
histocompatibility complex (MHC) molecules on accessory cell or B lymphocyte surfaces to specific Vp
chains of the d p T cell receptor (TCR) for antigen
present on T cells. Mycoplasma arthritidis, an agent of
chronic rodent arthritis (I2), also produces a potent
superantigen, MAM, which in the presence of H-2E
molecules (3), activates T cells bearing Vp5.1, Vp6,
and Vp8 chains of the TCRaIp (8,13,14). Although the
role of MAM in chronic arthritis remains to be established, there is evidence that it contributes to lethal
toxicity and dermal necrosis induced by live M arthritidis (15). Recent work indicates that injection of
MAM into mice induces profound changes in T and B
cell functions (16).
A number of experimental autoimmune diseases
are now known to be dependent upon T cells that bear
specific Vp TCRs, irrespective of the antigen-MHC
specificity. These so-called “V region diseases” (17)
include experimental allergic encephalomyelitis and
collagen-induced arthritis (CIA) of rodents (17,18).
Some human diseases, such as multiple sclerosis (19),
autoimmune thyroiditis (20), and rheumatoid arthritis
(RA) (9,21,22), may also be dependent upon specific
Vp chain TCR-bearing T cell subsets. In the CIA
model, mice or rats immunized with heterologous or
homologous collagen develop an acute-to-chronic polyarthritis thought to be mediated by T cells that are
cross-reactive with self collagen as well as by an
antibody response to specific collagen epitopes (2327). T cells bearing the Vp8 TCRs have been shown to
play a major role in CIA (18,28), and there is some
evidence that other TCRs, such as Vp5, Vp6, and Vp7,
may also be involved, especially in mouse strains that
lack VpS (28-33).
In view of the similarity between Vp chain
usage in CIA and T cell recognition of MAM, we
examined the effect of MAM on the development of
CIA in mice. We demonstrated that MAM can trigger
or exacerbate arthritis in mice sensitized to type I1
collagen. These findings have important implications
with regard to the etiology and development of human
autoimmune diseases, especially those that may be
associated with the presence of T cells expressing
similar Vp chain segments of the TCR.
we estimate that the preparations used in this study contained 5 1 0 pg of MAM/ml. Staphylococcal enterotoxin B
(SEB) and staphylococcal enterotoxin A (SEA) were purchased from Toxin Technology (Madison, WI).
Induction of CIA and effect of MAM. Native porcine
type I1 collagen (PII) was prepared as previously described
(35). For induction of arthritis, PI1 was dissolved overnight
in 0.1N acetic acid and was then emulsified with Freund’s
complete adjuvant (FCA) (H37Ra; Difco, Detroit, MI) until
a stable emulsion formed. Two-to-four-month-old mice received 5&100 pg of PI1 emulsion intradermally in the base of
the tail, and after 3 weeks, were reinjected with 100 pg of PI1
emulsified in Freund’s incomplete adjuvant (FIA). Prior to
challenge with MAM, convalescent mice were divided into 2
groups: animals without residual arthritis, and animals with
equal degrees of residual arthritis.
In experiment 1, using (B1O.RIII x RIIIS)F, x
B1O.RIII progeny, mice convalescing from CIA at day 200
post-collagen injection were challenged by injection, into
the tail base, of MAM or mock MAM emulsified in FIA.
Experiments 2 and 3 included several different mouse F,
hybrids, and the more highly purified MAM was used. Mice
used in the second and third experiments were injected
intravenously (IV) or intraperitoneally, respectively, with
0.2 rnl of 1:35 MAM in phosphate buffered saline (PBS)
(approximately 50.05 pg MAM/mouse) or with PBS alone.
In experiment 4, B1O.RIII mice were immunized with a
single injection of 10 pg or 30 pg PI1 in FCA containing
one-half the usual dose of mycobacteria. Shortly after onset
of CIA (32 days after injection with PII), the mice were
injected IV with 0.2 ml of 1:35 MAM in PBS or with PBS
alone. In the last experiment (experiment 5 ) , mice were
immunized suboptimally with 5 pg of PI1 emulsified in 50%
FCA. The mice were challenged 16 days later with IV
injections of 0.2 ml PBS, 1:35 MAM, or SEB or SEA (5
Evaluation of arthritis. Joint swelling was scored in a
blinded manner, as previously described (36). The maximum
possible arthritis score per mouse was 47.
Antibodies to collagen. Serum samples were obtained
from the tail, and levels of IgG antibodies to mouse type I1
collagen were measured using an enzyme-linked immunosorbent assay as previously described (37).
Mice. CBA, CBA/Ca, RIIIS, BlO.RII1, C57Br, and
SWR mice were purchased from Jackson Laboratories (Bar
Harbor, ME). Inbred, F,, and test cross mice were bred in
our laboratories and were screened (Microbiological Associates, Rockville, MD) to ensure absence of viral and mycoplasma1 pathogens.
Superantigens. For the first experiment, MAM was
prepared as previously described (34), and “mock” MAM
was prepared by similar treatment of uninoculated mycoplasma medium. For experiments 2-5, the procedure included two additional ion-exchange chromatography steps
(Atkin CL, Wei S, Cole BC: manuscript in preparation). The
resulting material contained lo7 units of mitogenic activity as
previously defined (34). By comparing the activity of this
material with that of recently derived homogeneous MAM,
Previous studies established that the systemic
injection of MAM into normal mice induces a mild
toxic syndrome associated with lymphadenopathy and
splenomegaly (16,38). Arthritis could be induced with
MAM only by direct injection into joints and was transient, lasting only a few days (39). To further address
this, mice of the B1O.RIII strain, which is the most
susceptible to CIA of all of the strains used in this study
(100% incidence with 5100 wg PII), were injected IV
with 1:20 MAM. None of the animals (0/9) developed
clinical arthritis (data not shown). It is known, however,
that MAM in doses similar to those used in the present
study profoundly affects the immune system following
M A M ~
mice that had been developed to investigate the genetics of CIA susceptibility. In the second approach, we
examined the effect of MAM on the onset phase of
CIA, using the highly CIA-susceptible B1O.RIII
In the exacerbation studies, some mice had had
resolution of their arthritis, some still had residual
arthritis, and some had failed to develop arthritis. Both
MAM-injected and control groups included mice that
represented all of these stages of disease. In the first,
preliminary experiment, we used male and female
(B1O.RIII x RIIIS) X B1O.RIII test cross progeny.
Male mice of this cross developed a much more severe
chronic arthritis in response to PI1 than did females
(30). After 200 days, when only residual CIA was
present, mice were challenged with an injection, into
the tail base, of MAM emulsified in FIA or control
mock MAM (made from mycoplasma media) in FIA
(Figure 1). MAM induced a flare of arthritis only in
male mice, which was evident by 5-7 days. Control
mock MAM failed to induce arthritis in male or female
mice (Figure 1). Because of these results and other
data indicating sex differences in susceptibility to
murine CIA (30), subsequent experiments were performed using male mice only.
Each of the second and third experiments included mice from 4 crosses which differed in expression of the Vp8 family of TCR chains. The results
obtained with each of these hybrid strains are summarized in Table 1. Mice that expressed 2 or more of the
Figure 1. Effect of Mycoplasma arthritidis superantigen MAM on
collagen-induced arthritis: experiment 1 . Male and female mice from a
(BlO.RIII x RIIIS)F, x B1O.RIII test cross (H-2"') were immunized
with porcine type I1 collagen and after 200 days, when the arthritis had
largely subsided, were challenged with MAM given in a Freunds
incomplete adjuvant (FIA) emulsion by injection into the base of the
tail. Control mice received a similar injection of "mock" MAM. Male
but not female mice exhibited a significant flare of arthritis.
IV injection into mice. Thus, MAM induces a clonal
expansion of T cells bearing the Vp6 and Vp8 chains of
the TCR (40)and changes the lymphokine profile from
THI-like to T~2-likeactivities (16).
In view of studies suggesting a common Vp
TCR usage by MAM and collagen-reactive T cells
we directly examined the effect of
MAM on CIA. Two approaches were undertaken.
First, we determined whether MAM would cause an
exacerbation or flare in mice convalescing from CIA.
For these studies, we utilized a series of F, hybrid
Table 1. Development of arthritis in various mouse strains, in response to Mycoplasma arthritidis superantigen MAM*
Incidence of arthritis
Mouse hybrid, treatment
Hybrid 1, (B1O.RIII X SWR)F,
Hybrid 2, (BlO.RII1 x CBAI Ca)F,
Hybrid 3, (BlO.RII1 x CBA)F,
Hybrid 4, (RIIIS X CS7BR)F,
Vp TCRs expressed or
at murine absent by genomic deletion
H-2 gene
8.2 8.3
(+I+)§ (+I+)§ +I+I+
score at
Day of
Day of
11.6 ? 8.7 10.4 ? 4.4 17.6 ? 8.5
2.2 rt 4.4
8.3 t 8.7 18.1 2 9.7
8.3 t 4.0 16.0 t 8.9
3.0 t 0
24 t 19.7
5.8 t 1.8
11 rt 4.4
+I+ +I+
(+I+)§ (+I+)§ +I+ +I+
11.6 ? 9.8
* Mice were injected intraperitoneally or intravenously with 1:35 MAM in phosphate buffered saline (PBS) or PBS alone. TCRs = T cell
receptors; CIA = collagen-induced arthritis.
t Values are the mean ? SD of the maximum scores for each of the mice, irrespective of time, after subtraction of scores from residual CIA.
f Values are the mean rt SD number of days since MAM or PBS treatment.
0 Brackets indicate that peripheral CD4+ T cells expressing indicated Vp TCR segments are not present due to clonal deletion by MIS-laand I-E.
rz 4
120 130
Figure 2. Experiment 2. Mice of hybrids 1-3 (see Table I ) were
immunized with porcine type I1 collagen ( 0 )and after 163 days,
when the arthritis had subsided, were challenged with 0.2 ml of 1:35
Mycoplusmu arthritidis superantigen MAM or 0.2 ml of phosphate
buffered saline (PBS) given intravenously (1.V.). Mice given MAM
developed a severe arthritis which persisted for at least another 40
days. Mice given PBS failed to show an increase in arthritis ( P <
0 . 0 1 by Mann-Whitney rank sum analysis).
Vp8 TCRs homozygously or heterozygously (hybrids
1-3) exhibited a similar susceptibility to CIA, with a
cumulative incidence of 17 of 28. When the CIA had
largely subsided, the animals were challenged intravenously (experiment 2) (Figure 2) and intraperitoneally
(experiment 3) (Figure 3) with MAM in PBS or with
PBS alone. Hybrid mice of groups 1, 2, and 3 developed a mean arthritis of similar severity and a high
incidence (15 of 17 total), when challenged with MAM
(Table I). In contrast, only 3 of 12 mice showed
enhancement of disease after injection with PBS, and
the enhancement seen was very mild (Table 1). The
time to onset of arthritis (3-23 days) and the disease
severity in individual MAM-injected mice were highly
variable, thus accounting for the large standard deviations shown in Table l . Onset of arthritis flares was
more rapid in mice that still exhibited residual CIA at
the time of MAM challenge.
The time course of arthritis induction by CIA
and MAM in mouse hybrids 1-3 is summarized in
Figure 2 (experiment 2) and Figure 3 (experiment 3).
There was a rapid onset of arthritis following injection
with MAM. Furthermore, in both experiments, the
arthritis flare induced by MAM was more severe than
was the original CIA. Preliminary evidence obtained in
experiment 3 indicated that increased arthritis in response to MAM was associated with a rise in the level
of antibodies to type I1 collagen.
The responses of individual hybrid 1-3 mice to
challenge with MAM are illustrated in Figure 4. All
mice that exhibited residual CIA showed increased
disease when treated with MAM. Furthermore, mice
whose CIA had totally resolved showed renewed
activity following injection with MAM. Importantly,
MAM was also able to trigger disease in mice that had
previously shown no signs of clinical arthritis despite
immunization with PII. The 3 mice that showed slight
enhancement of disease after injection of PBS (Figure
4) all exhibited prior CIA. We attribute the increase in
disease in these animals to mild spontaneous exacerbation of arthritis, which can occur at a low incidence
in this model of chronic inflammation (30). The remaining 8 mice injected with PBS failed to develop arthritis.
Evidence has been presented that CIA is not
entirely dependent upon the VpS TCRs and that arthritis can develop in mice which lack Vp8 (28-31). We
demonstrated in the present studies that Vp8-negative
(RIIIS x C57BR)F, mice (hybrid 4) developed CIA in
a low incidence (4 of 13) (Table 1) and with less
severity than that seen with hybrid 1-3 mice (data not
shown). We also showed that MAM caused exacerbation of arthritis in 5 of 13 (RIIIS x C57BR)F, mice,
whereas PBS-treated controls failed to show flares in
disease activity (0 of 13). These results are consistent
with the fact that MAM can also activate, in vivo,
lymphocytes bearing TCRs other than Vp8.
The fourth experiment was designed to determine whether MAM could influence the early, onset
phase of CIA. The highly CIA-susceptible B 10.RIII
mouse strain was used. The animals were subopti18
8 12P
9 8
T. - - -7' . -*
E 15-
Figure 4. Arthritis scores of each mouse from hybrids 1-3 (see
Table 1 ) prior to injection of Mycoplasma arfhrifidis superantigen
MAM or phosphate buffered saline (PBS) (starting score), and
maximum arthritis scores after injection of MAM. 0 = (BlO.RII1 x
SWR)Fl mice (hybrid 1 ) ; 0 = (BiO.RII1 x CBA/Ca)F, mice (hybrid
2); x = (BIO.111 x CBA)Fl mice (hybrid 3).
mally immunized by a single injection into the tail base
of 30 pg or 10 pg PI1 emulsified in FCA using 50% of
the usual dose of mycobacteria. MAM or PBS was
injected IV 32 days later, soon after the onset of CIA.
Although mice receiving 30 pg of PI1 showed a
much more rapid onset of arthritis after injection of
MAM versus PBS, the difference was not statistically
significant, and final arthritis scores were similar in the
2 groups (data not shown). Mice immunized with 10 pg
PI1 and challenged with PBS developed a mild arthritis
of low incidence (2 of 6) (Figure 5 ) . In contrast, mice
immunized with 10 pg PI1 and challenged with MAM
developed a significantly more severe arthritis of
higher incidence (5 of 7) (Figure 5). However, there
were no significant differences in levels of antibodies
to collagen between the MAM-treated and PBStreated groups (data not shown). Since mice receiving
MAM had already begun to show the onset of CIA, the
data might also indicate that preexisting arthritis predisposes mice to an accelerated disease when given
MAM. However, as demonstrated in the present studies, mice that have not yet developed arthritis or that
have had total resolution of their arthritis can also
develop a very rapid onset of disease after injection
of MAM.
A preliminary experiment was performed to
determine the effect of superantigens other than MAM
on the development of CIA. SEB was chosen since, as
with MAM, it activates T cells bearing the Vp8 chain
segments of the TCR. SEA was also used since it fails
to activate any of the T cell subsets that have been
shown or thought to mediate CIA (i.e., T cells bearing
Vp5, 6 , 7 , 8 , and 9). In this experiment (experiment 5),
B1O.RIII mice were suboptimally immunized with 5 pg
PII. After 16 days, a period well before the mean time
until onset of CIA, mice were injected IV with PBS,
MAM, SEB, or SEA. We had previously established
that SEB and SEA in the doses used ( 5 pgmouse)
induce lymphadenopathy and splenomegaly as early as
1-2 days post-injection (Cole BC et al: unpublished
The results of experiment 5 are summarized in
Table 2. Only 1 of the control animals receiving PBS
developed arthritis, and the severity gradually increased until termination of the experiment at 31 days
post-challenge. Mice receiving either MAM or SEB
exhibited a 50% incidence of arthritis, with an early
onset time (mean 8 days). In fact, among mice treated
with MAM, 3 of 10 showed arthritis just 3 days
post-injection, and 4 of 10 SEB-treated mice had
developed arthritis by 3 days. This rapid onset of
disease mirrored that seen in the previous experiments
and summarized in Figures 1-3. Mean maximum arthritis severity for both MAM-injected and SEBinjected animals occurred at day 15. In contrast, SEA
32 dayPost
with PII:
Figure 5. Experiment 4. Mice were immunized with 10 pg of
porcine type I1 collagen (P11). At onset of collagen-induced arthritis
on day 32, all mice were challenged by intravenous injection of
Mycoplasma arthrifidis superantigen MAM or phosphate buffered
saline (PBS). MAM accelerated the onset and severity of arthritis (P
< 0.001 by Mann-Whitney rank sum analysis).
Table 2. Effect of different superantigens in BIO.KIII mice suboptimally immunized with porcine type I1 collagen
Intravenous treatment
Phosphate buffered saline,
0.2 mi
Mycoplasnin urtiiriridis
superantigen MAM 1:35,
0.2 ml
Staphylococcal enterotoxin B,
I h y of
Day of
incidence mean 2 SD mean ? SD
10.2 14.8
Staphylococcal enteroloxin A,
14.0 2 5.7
induced a late-onset arthritis (mean ? S D 14.8 2 9.0
days) occurring in only 20% of the mice and, as in
PBS-injected animals, the arthritis severity did not
peak until close to termination of the experiment.
Although the number of animals in this experiment
was insufficient for adequate statistical analysis, the
data provide preliminary evidence that MAM and
SEB, unlike SEA, trigger an early onset of CIA.
The present findings show that the superantigen
MAM, derived from Mycoplasma arthritidis, can
markedly increase the development of arthritis in mice
immunized with porcine type 11collagen. Normal mice
receiving similar injections of MAM have consistently
failed to develop arthritis (16,38). There were 3 main
observations made in this study. First, MAM caused
an exacerbation of arthritis in mice that were convalescing from CIA induced previously by immunization
with PII. This second phase of arthritis induced by
MAM was frequently more severe than the CIA obtained by immunization with PI1 alone. Second, MAM
also had the ability to trigger arthritis in animals that
had previously been immunized with PI1 but had failed
to develop clinical disease. Importantly, both exacerbation and triggering of arthritis by MAM could occur
many months (up to 200 days) after the initial immunization with collagen. Third, when MAM was injected during the onset phase of CIA into animals that
were suboptimally immunized with PII, it resulted in a
greater incidence of arthritis and greater severity of
joint involvement. Thus, MAM appeared to act synergistically with PI1 to produce a more severe disease
than that induced with PI1 alone.
The ability of MAM to trigger CIA is consistent
with observations that indicate a common Vp TCR
usage for MAM and collagen-reactive T cells. MAM
activates murine T cells bearing Vp6, 8.1, 8.2, and 8.3
TCRs (8,13), and recent work indicates additional
engagement of Vp5. 1 (14) and also of Vp 1 and Vp7 (40).
In CIA, cosegregation studies have implicated involvement of VpS and Vp6 (18,29). Evidence for direct
involvement of VpS was obtained when monoclonal
antibodies (MAb) to this gene family were shown to
suppress arthritis in BIO.RIII mice (28). TCRs other
than VpS are also involved since the anti-Vp8 MAb did
not completely inhibit BIO.RIII disease and did not
inhibit disease in D B N l mice (28). Also, antibody to
Vp5 inhibited CIA in DBA/I mice (32).
CIA can also occur in mouse strains and
crosses which lack VpS, for example, Vp8-negative
progeny from a (DBA/I x SWR)F, cross (31). The
present study showed that VpS-negative (RIIIS x
C57BR)F, mice also develop arthritis (Table I ) , and
other studies indicate that RIIIS mice which lack both
Vp6 and Vp8 develop a mild, transient CIA (30).
Recent work by Haqqi et al supports these findings
and suggests that Vp6, Vp8.2, Vp7, and Vp9 are
involved in CIA since these T cells are the ones most
commonly found in the joints and draining lymph
nodes of Bl0.Q mice with CIA (33).
A key question is whether MAM-induced enhancement of CIA is dependent upon activation of
specific collagen-reactive T cells o r is due to a nonspecific release of cytokines involved in the inflammatory
response. A preliminary experiment suggested that
triggering of CIA was in fact dependent upon activation of T cells bearing specific Vp TCRs. Thus, MAM
and SEB, both of which activate Vp8 chain-bearing T
cells, triggered an early-onset arthritis in 50% of mice
that were suboptimally immunized with PII. In contrast, mice injected with SEA, which fails to activate
any of the Vp subsets that mediate CIA, developed a
low incidence (20%) of late-occurring arthritis, the
course of which resembled that seen in the I of 10
PBS-injected control mice that developed disease.
Further support for a role of activation of specific
Vp-bearing T cells was provided by the finding that the
doses of the superantigens used, including that for
SEA, resulted in a marked lymphadenopathy and
splenomegaly, 1-2 days after IV injection into mice.
Clearly, we cannot as yet rule out a role for
nonspecific cytokine release in the development of
CIA. However, if our observations are confirmed,
it does seem that selective activation of collagenreactive T cells plays a major role in triggering the
inflammatory response. More detailed experiments are
now required to measure Vp TCR expression, cellular
responses to collagen, and cytokine profiles during the
course of MAM-induced activation of CIA.
Although our results suggest that other superantigens such as SEB appear to be capable of triggering CIA, it is relevant that MAM is derived from a
murine pathogen which itself induces a chronic proliferative arthritis (2). Whereas the role of MAM in the
chronic phase of this disease is not yet known, there is
evidence that MAM contributes to lethal toxicity and
dermal necrosis (15) due to the administration of live
M arthritidis. Furthermore, direct intraarticular injection of MAM into rat joints will also induce a transient
arthritis (39). Thus, M arthritidis-mediated chronic
arthritis may prove to be a useful natural model to
investigate the role of superantigens in microbialmediated chronic diseases.
Two main mechanisms have been proposed
whereby superantigens might play a role in the induction of autoimmune disease. First, by interacting with
the Vp chains of the TCRs of self-reactive T cells, they
may cause a clonal expansion of these cells, resulting
in a self-reactive immune cascade. This process may
increase the numbers of these self-reactive cells above
the threshold that is necessary for triggering of clinical
disease (7-9). Second, superantigens may act as a
bridge, cross-linking the Vp TCRs on THcells with the
MHC molecules on resting B cells and resulting in
polyclonal B cell activation and production of autoantibodies (10,ll).
Although immune T cells and antisera to type I1
collagen (41,42) are reported to independently induce a
mild CIA in susceptible mice, it is clear that both
self-reactive T cells and anticollagen antibodies are
required for full expression of the disease (27). We
now have evidence that the MAM superantigen can
activate both of these functions in vivo as well as in
vitro. We have recently shown that the IV injection of
MAM into appropriate mouse strains results in a
clonal expansion of T cells expressing the Vp6 and the
Vp8 TCRs (40). Lymphocytes obtained from mice
injected IV with MAM also show a change in their
lymphokine profile in that they exhibit a decrease in
production of interleukin-2 (IL-2) but an increase in
production of IL-4 and IL-6 (16). This change from
TH1-like activities to TH2-like activities would favor
development of humoral antibody responses. In fact, it
has already been shown that MAM can induce proliferation of, and immunoglobulin secretion by, B lym-
phocytes when cultured with MAM-reactive T cells
from both human (43) and murine sources (44).
New findings from our laboratory indicate that
MAM given in vivo can also result in enhanced secretion of IgG and IgM and enhanced antibody responses
to challenge with foreign antigens (16). Thus, MAM
not only expands the T cell subsets that are involved in
CIA, but can also generate a polyclonal antibody
response in vivo, which could lead to an increase in
anticollagen antibodies. Although MAM failed to induce a significant increase in the level of antibodies to
collagen in BlO.RIJ1 mice, it should be noted that
these animals are normally very high responders to
PII. Furthermore, an overall increase in antibodies to
PI1 may not have been seen since MAM may act by
selectively stimulating a more arthritogenic population
of antibodies that is restricted to certain collagen
epitopes or specific antibody isotypes. It is also possible that an additional antibody boost is, in fact, not
necessary. In this regard, we have shown that MAM
will rapidly induce a transient arthritis when injected
intraarticularly into naive rats (39). Preliminary evidence to support the notion of an increase in levels of
anticollagen antibodies is given in Figure 3. Studies to
define the respective roles of humoral versus cellular
pathways in superantigen-mediated triggering of CIA
are currently in progress.
What is the relevance of these observations to
the etiology and development of the human rheumatic
diseases? There is now evidence of oligoclonal expression of Vp TCR-bearing T cells in the joint tissues of
patients with RA (9,21,22). In one study, activated T
cells in rheumatoid synovium predominantly expressed human Vp3, Vp14, and Vp17 (22). In another,
Vp 14-bearing T cells were increased in rheumatoid
synovial fluids and decreased in the peripheral circulation (9). It may not be coincidental that these same
human Vp chains are those which share most homology (45) with murine Vp7 and Vp8 (human Vp3 and
Vp14) and murine Vp6 (human Vp17 or, alternatively,
Vp19), i.e., those murine Vp chains that are known to
react with MAM. Consistent with this is the fact that
MAM does activate human T cells (46) and that, more
specifically, it activates those bearing human Vp3 (14),
Vp17 (14,47), and Vp14 (Marrack PC: personal communication).
Superantigens may play a significant role in the
development of the flares in disease activity that are
characteristic of human RA. Such flares may be mediated by different microorganisms producing superantigens with a Vp TCR usage that is cross-reactive
with that used by autoreactive T cell clones. As seen
here for CIA, flares in RA disease activity mediated by
superantigens may occur many months after sensitization to self antigens. Identification of the Vp TCRreactive epitopes on these MAM-like molecules may
offer a new potential approach to the development of a
vaccine for the treatment of the rheumatic diseases.
In conclusion, studies of the effects of MAM on
the induction of CIA and on the development of
mycoplasma arthritis represent ideal models to define
the mechanisms by which superantigens might precipitate chronic autoimmune disease. Studies are now
needed to determine whether molecules similar to
MAM, produced exogenously by mycoplasmas and
bacteria or endogenously by viral superantigens, play
a role in the development of the human autoimmune
We thank Dr. Curtis L. Atkin and Suhua Wei for
preparation of MAM, Van R. Reese for production of
porcine collagen, and Scott Harper, Rick Harper, Diane
Kartchner, and Dr. Elsayed A. Ahmed for diligent animal
husbandry and induction and scoring of arthritis. Assays for
collagen antibodies were skillfully performed by Shawna
1. Marrack PC, Kappler JW: The staphylococcal enterotoxins and their relatives. Science 248:705-711, 1990
2. Fleischer B: Bacterial toxins as probes for the T-cell
antigen receptor. Immunol Today 10:262-264, 1989
3. Cole BC, Atkin CL: The Mycoplasma arthritidis T cell
mitogen, MAM: a model superantigen. Immunol Today
12:27 1-276, 1991
4. Frankel WN, Rudy C, Coffin JM, Huber B: Linkage of
MIS genes to endogenous mammary tumor viruses of
inbred mice. Nature 349526528, 1991
5 . Dyson PJ, Knight AM, Fairchild S, Simpson E, Tomonari K: Genes encoding ligands for deletion of V, l l T
cells cosegregate with mammary tumor virus genomes.
Nature 349531-532, 1991
6. Choi Y, Kappler JW, Marrack P: A superantigen encoded in the open reading frame of the three long
terminal repeat of mouse mammary tumor virus. Nature
350:203-207, 1991
7. White JA, Herman A, Pullen AM, Kubo R, Kappler JW,
Marrack P: The V, specific superantigen staphylococcal
enterotoxin B: stimulation of mature T cells and clonal
deletion in neonatal mice. Cell 56:27-35, 1989
8. Cole BC, Kartchner DR, Wells DJ: Stimulation of
mouse lymphocytes by a mitogen derived from Myco-
plasma arthritidis (MAM). VIII. Selective activation of
T cells expressing distinct Vp T cell receptors from
various strains of mice by the superantigen MAM. J
Immunol 144:425431, 1990
9. Paliard X, West SG, LaEerty JA, Clements JR, Kappler
JW, Marrack P, Kotzin BL: Evidence for the effects of
a superantigen in rheumatoid arthritis. Science 253:325329, 1991
10. Friedman SM, Posnett DN, Tumang JR, Cole BC, Crow
MK: A potential role for microbial superantigens in the
pathogenesis of systemic autoimmune disease. Arthritis
Rheum 34:468480, 1991
11. Mourad W, Scholl P, Diaz A, Geha R, Chatila T: The
staphylococcal toxic shock syndrome toxin 1 triggers B
cell proliferation and differentiation via major histocompatibility complex-unrestricted cognate T/B cell interaction. J Exp Med 170:2011-2022, 1989
12. Cole BC, Washburn LR, Taylor-Robinson D: Mycoplasma-induced arthritis, The Mycoplasmas. Vol. 4. Edited
by S Razin, MF Barile. New York, Academic Press,
13. Cole BC, Kartchner DR, Wells DJ: Stimulation of
mouse lymphocytes by a mitogen derived from Mycoplasma arthritidis (MAM). VIII. Selective activation of
T cells expressing distinct V, T cell receptors from
various strains of mice by the “superantigen” MAM. J
Immunol 144:425431, 1990
14. Baccala R, Smith LR, Vestberg M, Peterson PA, Cole
BC, Theofilopoulos AN: Mycoplasma arthritidis rnitogen: V, engaged in mice, rats, and humans, and requirement of HLA-DRa for presentation. Arthritis Rheum
35:434-442, 1992
15. Cole BC: The immunobiology of Mycoplasma arthritidis
and its superantigen MAM. Curr Top Microbiol Immuno1 174:107-119, 1992
16. Cole BA, Ahmed EA, Araneo BA, Shelby J, Kamerath
C, Wei S, McCall S, Atkin CR: Immunornodulation in
vivo by the Mycoplasma arthritidis superantigen MAM.
Clin Infect Dis (in press)
17. Heber-Katz E, Acha-Orbea H: The V-region disease
hypothesis: evidence from autoimmune encephalomyelitis. Irnmunol Today 10: 164-169, 1989
18. Banerjee S, Haqqi TM, Luthra HS, Stuart JM, David
CS: Possible role of V, T-cell receptor genes in susceptibility to collagen-induced arthritis of mice. J Exp Med
167:832-839, 1988
19. Wucherpfennig KW, Weiner HL, Hafler DA: T-cell
recognition of myelin basic protein. Immunol Today
12:464-465, 1991
20. Davies TF, Martin A, Concepcion ES, Graves P, Cohen
L, Ben-Nun A: Evidence of limited variability of antigen
receptors on intrathyroidal T cells in autoimmune thyroid disease. N Engl J Med 325:238-244, 1991
21. Sottini A, Imberti L, Goria R, Cattaneo R, Primi D:
Restricted expression of T-cell receptor V, but not V,
genes in rheumatoid arthritis. Eur J Immunol 21:461466, 1991
Howell MD, Diveley JP, Lundeen KA, Esty A, Winters
ST, Carlo DJ, Brostoff SW: Limited T cell receptor
0-chain heterogeneity among IL-2R+ synovial T cells
suggests a role for superantigen in rheumatoid arthritis.
Proc Natl Acad Sci U S A 88:10921-10925, 1991
Trentham DE, Townes AS, Kang AH: Autoimmunity to
type I1 collagen: an experimental model of arthritis. J
Exp Med 146:857-868, 1977
Courtenay JS, Dallman MJ, Dayan AD, Martin A,
Mosedale B: Immunization against heterologous type I1
collagen induces arthritis in mice. Nature 283:666-668,
Stuart JM, Townes AS, Kang AH: Collagen autoimmune arthritis. Annu Rev Immunol2:199-218, 1984
Wooley PH: Collagen-induced arthritis in the mouse.
Methods Enzymol 162:361-373, 1988
Seki N, Sudo Y, Yoshioka T, Sugihara S, Fujitsu T,
Sakuma S, Ogawa T, Hamaoka T, Senoh H , Fijiward H:
Type I1 collagen-induced murine arthritis. I. Induction
and perpetuation of arthritis require synergy between
humoral and cell-mediated immunity. J Immunol 140:
1477-1484, 1988
Moder KG, Luthra HS, Griffiths MM, David CS: Prevention of collagen-induced arthritis in mice by deletion
of T cell receptor Vp8 bearing T cells with monoclonal
antibodies. Br J Rheumatol 32:26-30, 1993
Banerjee S, Behlke MA, Dungeon G, Loh DY, Stuart
JM, Luthra HS, David CS: Vp6 gene of T cell receptor
may be involved in type I1 collagen induced arthritis in
mice (abstract). FASEB J 2:2120, 1988
Griffiths MM, Cole BC, It0 J, Harper DS, Anderson GD,
Cannon GW, Luthra HS, David CS: T-cell receptors and
collagen induced in H-2 mice. Submitted for publication
Spinella DG, Jeffers JR, Reife RA, Stuart JM: The role
of C5 and T-cell receptor V, genes in susceptibility to
collagen-induced arthritis. Immunogenetics 34:23-27,
Chiocchia G, Boissier M-C, Fournier C: Therapy against
murine collagen-induced arthritis with T-cell receptor
V, -specific antibodies. Eur J Immunol 21:2899-2905,
Haqqi TM, Anderson GD, Banerjee S, David CS: Restricted heterogeneity in T-cell receptor v, gene usage
in the lymph nodes and arthritic joints of mice. Proc Natl
Acad Sci U S A 89:1253-1255, 1992
Atkin CL, Cole BC, Sullivan GJ, Washburn LR, Wiley
BB: Stimulation of mouse lymphocytes by a mitogen
derived from Mycoplasma arthritidis. V. A small basic
protein from culture supernatants is a potent T cell
mitogen. J Immunol 137:1581-1589, 1986
Griffiths MM, Eichwald EJ, Martin JH, Smith CB,
DeWitt CW: Immunogenetic control of experimental
type I1 collagen-induced arthritis. I. Susceptibility and
resistance among inbred strains of rats. Arthritis Rheum
24:78 1-789, 198 1
Cole BC, Washburn LR: Evaluation of arthritogenic
properties of mycoplasmas for small laboratory animals,
Methods in Mycoplasmology. Vol. 2. New York, Academic Press, 1983
Griffiths MM, DeWitt CW: Immunogenetic control of
experimental collagen-induced arthritis in rats. 11. ECIA
susceptibility and immune response to type I1 collagen
(calf) are linked to RT1. J Immunogenet 8:463-470, 1981
Cole BC, Wells DJ: Immunosuppressive properties of
the Mycoplasma arthritidis T cell mitogen (MAM) in
vivo: inhibition of proliferative responses to T cell
mitogens. Infect Immun 58:22&236, 1990
Cannon GW, Cole BC, Ward JR, Smith JL, Eichwald
EJ: Arthritogenic effects of Mycoplasma arthritidis
T-cell mitogen in rats. J Rheumatol 14:735-741, 1988
Cole BC, Balderas RA, Ahmed EA, Kono D, Theofilopoulos AN: Genornic composition and allelic polymorphisms influence V, usage by the Mycoplasma arthritidis superantigen. J Immunol 150:3291-3299, 1993
Holmdahl R, Andersson M, Goldschmidt TJ, Gustafsson K, Jansson L , Mo JA: Type I1 collagen autoimmunity in animals and provocations leading to arthritis.
Immunol Rev 118:193-232, 1990
Stuart JM, Townes AS, Kang AH: Collagen autoimmune arthritis. Annu Rev Immunol 2:199-218, 1984
Tumang JR, Posnett DN, Cole BC, Crow MK, Friedman
SM: Helper T cell dependent human B cell differentiation mediated by a mycoplasmal superantigen bridge. J
Exp Med 171:2153-2158, 1990
Tumang JR, Cherniack EP, Gietl DM, Cole BC, Russo
C, Crow MK, Friedman SM: T helper cell-dependent,
microbial superantigen-induced murine B cell activation: polyclonal and antigen-specific antibody responses. J Immunol 147:432-438, 1991
Wilson RK, Lai E , Concannon P, Barth RK, Hood LE:
Structure, organization and polymorphism of murine
and human T-cell receptor a and /3 chain gene families.
Immunol Rev 101:149-172, 1988
Cole BC, Washburn LR, Sullivan GJ, Ward JR: Specificity of a Mycoplasma mitogen for lymphocytes from
human and various animal hosts. Infect Immun 36:662666, 1982
Friedman SM, Crow MK, Tumang JR, Tumang M, Xu
Y , Hodtstev AS, Cole BC, Posnett DN: Characterization of human T cells reactive with the Mycoplasma
arthritidis-derived superantigen (MAM): generation of a
monoclonal antibody against V, 17, the T cell receptor
gene product expressed by a large fraction of MAMreactive human T cells. J Exp Med 174:891-900, 1991
Без категории
Размер файла
910 Кб
exacerbation, mam, arthritis, triggering, mycoplasma, autoimmune, arthritidis, superantigen
Пожаловаться на содержимое документа