| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
*
Center for Experimental Medicine, Institute of Medical Science, University of Tokyo, Tokyo, Japan; and
Kohno Clinical Medicine Research Institute, Tokyo, Japan
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Human T cell leukemia virus type I (HTLV-I) is the causative agent of
adult T cell leukemia (6). This virus encodes a
transcriptional trans-activator, Tax, in the
env-pX region, that trans-activates transcription
from the cognate viral promoter through the 21 bp enhancer
(7). Tax also activates many cellular genes
(8), including these for cytokines (9, 10, 11, 12),
cytokine receptors (13, 14), and immediate early
transcriptional factors (15, 16) through activation of
enhancers such as the NF-
B-dependent enhancers or serum-responsive
elements (7). Previously, we found that Tg mice carrying
the HTLV-1 env-pX region with its own long terminal repeat
(LTR) promoter developed chronic inflammatory polyarthropathy at a high
incidence (17). The arthritis develops in multiple joints
as early as 4 wk of age, and at 3 mo of age, 60% (BALB/cAn background)
and 20% (C3H/HeN background) of the mice are affected. The
histopathology is very similar to that of RA in humans, showing marked
synovial and periarticular inflammation with articular erosion caused
by invasion of granulation tissues. These mice develop autoimmunity
with elevated levels of Abs against IgG, IIC, and heat shock proteins
and show IgG hypergammaglobulinemia in which agalactosylated forms of
the Ig carbohydrate chains increase (18). In this regard,
we recently found that T cells from HTLV-I env-pX
region-introduced transgenic (pX-Tg) mice were
refractory against anti-Fas Ab treatment (19). This
finding suggests that this defect may be involved in the development of
autoimmunity in these mice, because Fas-mediated apoptosis of T cells
is believed to be important in eliminating autoreactive T cells in the
periphery (20). Furthermore, various genes including
inflammatory cytokine genes such as IL-1
, IL-1ß, IL-6, TNF-,
and IFN-
genes, immediate early genes such as c-fos and
c-jun, and class I and class II MHC genes are activated in
the Tg joints (18, 21), probably because of the
transcriptional trans-activating activity of the
tax gene and also as a result of immune reactions in the
affected joints in later stage. Because similar abnormalities are also
found in RA patients, we think pX-Tg mice provide a good
disease model for RA.
Although the pathogenesis of arthritis in this Tg mouse has not been
elucidated completely, several possibilities are conceivable. One is
that Tax-induced overproduction of inflammatory cytokines as well as
immediate early gene products in the synovial cells may directly cause
synovial cell growth and bone erosion. Actually, it was shown that
overproduction of TNF- in the joints caused arthritis in Tg mice
without involvement of immune reaction (22). IL-1 also
causes arthritis when it is injected into rabbit joints
(23), indicating the arthritogenic nature of these
cytokines. Another possibility is that the autoimmunity that developed
in these Tg mice is involved in the onset of the disease, as is
suggested in RA cases (24). In this connection, it is well
known that immunization of DBA1 mice with IIC causes arthritis
(25). pX-Tg mice also have the Ab against IIC,
and immunization with IIC augments the development of arthritis,
suggesting that immune reaction against IIC may be involved in the
pathogenesis (18).
In this study, the roles of the immune cells in the development of arthritis in pX-Tg mice were examined to discriminate these possibilities. For this, we conducted reciprocal replacement of the immune system between pX-Tg and normal mice. Furthermore, we examined the incidence of arthritis in athymic trangenic mice by introducing the nu gene into pX-Tg mice. The results show that immune cells, but not synovial resident cells, are primarily responsible for the development of the disease.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The original Tg mouse was produced by injecting the pX and env region of the HTLV-I genome with its own promoter into fertilized mouse ova from (C3H/HeN x C57BL/6J)F1 intercrosses (17). Then, they were backcrossed to BALB/cAn mice, and littermates at the age of 2–3 mo of backcross generations 8 to 11 were used for the experiments. The transgene was detected by dot blot hybridization of the tail DNA (18), and the transgene-positive and -negative mice are designated pX-Tg and non-Tg mice, respectively.
Male BALB/c-nu/nu mice were obtained from Charles River Japan (Kanagawa, Japan). They were crossed to pX-Tg females, and female progenies carrying the transgene were crossed with male nude mice again. Then, pX-positive hairless (nu/nu) and hairy (nu/+) mice were examined for the development of arthritis.
All the mice were kept under specific pathogen-free conditions in environmentally controlled clean rooms in the Laboratory Animal Research Center, Institute of Medical Science, University of Tokyo (Tokyo, Japan). The experiments were conducted according to the institutional ethical guidelines for animal experiments and safety guidelines for gene manipulation experiments.
BM and spleen cell transfer
old pX-Tg and non-Tg female mice, 6–8 wk old, were used as donors of BM and spleen cells. BM cells were prepared from the femurs and tibiae and were used directly (BM cells) or after treatment with mouse anti-Thy-1.2 Ab (Serotec, Oxford, U.K.) and rabbit complement (Technically, Hornby Ontario, Canada) (T cell-depleted BM cells). Age- and sex-matched recipient mice were irradiated with 7.5 Gy of gamma-rays and, after 6 h, were injected with either 107 viable BM cells only or 107 viable spleen cells with 106 BM cells i.v.
Clinical evaluation
Arthritis was evaluated macroscopically. The joints of each paw were examined weekly for swelling and redness, and the disease severity was graded from 0 to 3 (grade 0 = no special changes; grade 1 = light swelling and/or redness of the joint; grade 2 = obvious swelling; grade 3 = fixation of the joint). The score of the four paws were totaled and used as severity score (0–12 points in individual mice).
Histopathology
All of the BM cell-transferred mice and nude mice were submitted to autopsy after inspection for the development of arthritis for 26–30 wk. The limbs were fixed in 10% neutral formalin and, after decalcification with 5% formic acid, embedded in paraffin, sectioned at 4 µm, and stained with hematoxylin and eosin.
Ab titration
Serum levels of IgG rheumatoid factor (RF) were determined by ELISA as previously described (18). Briefly, polyvinyl microtiter plates (Falcon Micro Test III, Becton Dickinson, Tokyo, Japan) were coated with 50 µl heat-denatured rabbit IgG (10 µg/ml) in PBS overnight at 4°C. For total IgG and total IgM, the plates were coated with anti-mouse Ig (DAKO, Kyoto, Japan; 1/10,000 dilution). After a washing with TBS (25 mM Tris-HCl and 140 mM NaCl, pH 7.4), the plates were blocked with 1% skim milk (Difco), 5 mM EDTA, 0.02% NaN3, TBS (blocking buffer) for 1 h at room temperature. Then, 50 µl blocking buffer-diluted mouse serum were added to each well and incubated for 1 h at room temperature followed by washing with 0.05% Tween 20-TBS. Fifty microliters of alkaline phosphatase-conjugated goat anti-mouse IgG Ab or IgM (Zymed; 1/500 dilution) in the blocking buffer was added as the second Ab and incubated at room temperature for 1 h. After a washing with Tween 20-TBS, 100 µl of 1 mg/ml p-nitrophenylphosphate (Sigma, St. Louis, MO) in 50 mM NaHCO3, 5 mM MgCl2 (pH 9.5) was added, and the optical density at 415 nm was measured by an ELISA microreader (MTP-120, Colona, Tokyo, Japan) after incubation at 37°C for 1 h. The Ab levels were calculated with the use of standard positive controls.
Northern blot hybridization and RNase protection assay
Macrophages were collected from the peritoneum after injection with 2% thioglycolate, and T cells were purified from the spleen cells with a nylon wool column. B cells were purified from adherent cell-removed spleen cells by depleting T cells with mouse anti-Thy-1.2 Ab (Serotec) and rabbit complement (Technically). Joint preparation contained the knees, ankles, and the digits from which the skin and the muscle were removed. Then, total RNA was prepared by the acid-guanidinium thiocyanate-phenol-chloroform method (26). Cytokine mRNA was analyzed by Northern blot hybridization using oligo(dT) column-purified poly(A)+ RNA (18). The pX mRNA was measured by RNase protection assay using the total RNA (17). The intensity of the bands on the autoradiogram was estimated by the BAS 2000 system (Fuji-film, Kanagawa, Japan).
Probes
Mouse IL-1 and IL-1ß cDNA (CDMmIL-1 and CDMmIL-1ß,
respectively) were provided by Dr. Tetsuo Sudo (27, 28),
and the XhoI fragments (2.0 and 1.3 kb) were used as probes.
Mouse IL-6 (moIL-6) and TNF- (moTNF-) cDNA were kindly gifted
from Dr. Takashi Yokota (29, 30), digested with
BamHI (1.2 kb) and BglI-BglII (0.7
kb), respectively, and used as probes.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Previously, we reported that pX-Tg mice develop
arthritis spontaneously. Those mice began to develop arthritis around 4
wk of age, and the incidence at 30 wk of age was 
80% (Fig. 1
). To elucidate the roles of
autoimmunity in the development of arthritis, we conducted reciprocal
reconstitution of the immune system between pX-Tg and non-Tg
mice. As shown in Fig. 1A, irradiated non-Tg mice
transferred with T cell-depleted pX-Tg BM cells
([Tg
non-Tg] mice) developed arthritis 16 wk after the
transplantation, and 50% of the mice were affected after 30 wk.
Transfer of the pX-Tg spleen cells together with the BM
cells or T cell-undepleted BM cells also showed similar results,
suggesting that BM cells, but not T cells or B cells, are responsible
for the development of this arthritis. The average severity score of
the affected joints 30 wk after the transplantation was comparable with
that of untreated pX-Tg mice (Fig. 1B). No
remarkable changes were found in the BM cell populations expressing
either CD11b, Gr1, B220, or Thy-1.2 Ags between pX-Tg mice
and wild-type mice as analyzed by FACS, excluding a possibility that
cell composition is abnormal in Tg mice (data not shown). Thus, these
results indicate that BM stem cells are abnormal in
pX-Tg mice.
|
and Table I). Bone and cartilage erosion caused by
invasion of granuloma-like cells was seen in the affected joints of the
[Tgnon-Tg] mice (Fig. 2, A--C). Overgrowth
of the synovial cells and infiltration of inflammatory cells consisting
of neutrophils, lymphocytes, and plasma cells were marked in the joints
and periarticular space (Table I). Exfoliation of the synovial lining
cells was also observed in some mice. These histopathological
observations were very similar to those seen in spontaneously developed
arthritic joints and show that pX-Tg mouse BM-derived cells
are abnormal and are responsible for the development of the
disease.
|
|
Next, we examined whether replacement of the immune cells can
prevent development of the disease in pX-Tg mice. As shown
in Fig. 1
, C and D, when BM cells from non-Tg
mice were transferred into irradiated Tg mice, the development of
arthritis was completely suppressed. The presence of T cells did not
affect the results; BM transplantation with spleen cells or T
cell-undepleted BM cells transplantation gave the similar results,
although the efficiency of the suppression appeared better in BM plus
spleen cell transfer. Interestingly, the swelling and redness of the
joint became resolved in three of six mice that had already developed
arthritis after transplantation of BM plus spleen cells. Under these
conditions, the Tg mice that were transplanted with Tg BM plus spleen
cells or only BM cells developed arthritis 100%.
Joints had no remarkable histological changes in most [non-TgTg]
mice (Fig. 2, D and E), in support of the
macroscopic observation. However, proliferation of the synovial lining
cells as well as fibroblastic cells of the periarticular tissues were
observed in some of the [non-TgTg] mice (Fig. 2F and
Table I). This incomplete suppression of the disease was observed more
frequently in BM cell-transferred mice than in BM plus spleen
cell-transferred mice (Table I). Collectively, these results
established the primary role of the BM-derived cells in the
pathogenesis of arthritis in pX-Tg mice.
The pX transgene was strongly expressed in lymphatic organs, but only weakly in the joints, after transplantation of Tg BM cells
The transgene is expressed in many organs including the joints,
spleen, thymus, and muscle in the original pX-Tg mice
(17). We examined the expression of the transgene 6 mo
after the transplantation of the Tg BM cells into irradiated non-Tg
mice to survey the fate of the transferred cells. As shown in Fig. 3, the transgene was strongly expressed
in T cells including the thymocytes and splenic T cells of
[Tgnon-Tg] mice, but the expression was not detected in those of
[non-TgTg] mice. Thus, most of the T cells seemed to be replaced
by those from donor mice. Transgene expression in macrophages was
similarly observed in [Tgnon-Tg] mice but not in [non-TgTg]
mice, indicating that this population was also replaced. Transgene
expression in B cells was also observed in [Tgnon-Tg] mice, but
not in [non-TgTg] mice. In contrast, replacement of the synovial
cells in the joints seemed to occur only partially, because the
pX expression was still observed in [non-TgTg] mouse
joints and that in the [Tgnon-Tg] mouse joints was lower than that
of Tg mice (Fig. 3). These observations suggest that BM-derived cells
such as T cells, B cells, or macrophages are responsible for the
development of arthritis, but resident cells in the joint are not
directly involved.
|
Because high levels of serum Ig and RF relevant to autoimmunity
are characteristic of pX-Tg mice, these autoimmune
parameters were determined. Total IgG levels (Fig. 4A), total IgM levels (Fig. 4B), and IgG class RF levels (Fig. 4C) in
[Tgnon-Tg] mice were significantly higher than those in non-Tg
control mice. In contrast, these Ig levels in [non-TgTg] mice
significantly reduced compared with those in arthritic Tg mice, being
nearly equal to the levels in nonarthritic Tg mice. These serological
data suggest involvement of autoimmunity in the development of
arthritis in [Tgnon-Tg] mice.
|
genes, were activated in
the affected joints of [Tgnon-Tg] and [TgTg] mice as in
arthritic Tg mice (18), compared with the nonarthritic
joints of [non-TgTg] and non-Tg mice (Fig. 5). However, their expression in the
spleen was not correlated to the arthritis. The expression of these
cytokines in the nonarthritic joints of [non-Tg(BM + S)Tg] and
[non-Tg(BM) Tg] mice was not augmented (Fig. 5), in spite of the
pX expression in the joints of these mice (Fig. 3). Thus, it
was suggested that pX-gene expression in the resident cells
in the joints is not directly involved in the activation of cytokine
expression in the affected joints.
|
To elucidate roles of T cells in the development of arthritis in
pX-Tg mice, the incidence and severity of the disease were
compared between nu/nu (athymic)- and nu/+
(euthymic)-pX-Tg mice. As shown in Table II, the incidence of arthritis was
clearly lower in homozygous nude mice than in heterozygous ones. The
histology of the nonarthritic joints was normal when examined by a
microscope (Fig. 6A). The
incidence in heterozygous mice was not significantly different between
female and male mice. Although only female nu/nu mice
developed arthritis, the difference is statistically not significant.
The average incidences in heterozygous and homozygous mice including
both female and male were 56.2 and 4.3% at 3 mo and 84.2 and 13.6% at
5 mo of age, respectively. The severity scores were also low in nude
mice compared with euthymic mice, because fewer paws were affected in
those mice. These results suggest that T cell immunity plays a crucial
role in the development of arthritis in pX-Tg mice.
|
|
, B and C). In this respect, we
found that many CD3+ T cells was present in
nu/nu mouse lymph nodes at the age of 5 mo; 30% of the
lymph node cells were T cells in nu/nu mice, in contrast to
60% in wild-type mice. However, the RF levels in
nu/nu-pX-Tg mice were not elevated even after development of
arthritis (Fig. 7). These observations
indicate that RF is not necessarily required for the development of
arthritis in these mice.
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The arthritis was not caused by the T cells that were contaminated in BM cell preparations, because we depleted T cells by treating BM cells with anti-Thy-1.2 Ab and complement before transfer of the cells. Moreover, addition of spleen cells in which large amounts of T cells were present to the BM preparation did not increase the incidence of arthritis. We also tried to induce arthritis by transferring spleen cells into nude mice, but we have never succeeded (data not shown). Thus, these observations suggest that the disease is not caused by the abnormal education of T cells in the periphery but by the disorder in the stem cells.
Under our conditions, pX gene expression was observed in T
cells, B cells, and macrophages when BM cells from Tg mice were
transplanted into lethally irradiated non-Tg mice. However, the
expression disappeared completely both in T cells, B cells, and
macrophages after transfer with non-Tg mouse BM cells into irradiated
Tg mice. These results indicate that the immune system of the recipient
mice was completely substituted with the donor cell-derived population.
Suppression of arthritis was better in [non-Tg (BM + S)Tg] mice
than in [non-Tg (BM)Tg] mice, probably because the immune system
could be reconstituted earlier by the presence of spleen cells. In this
connection, it is known that presence of T cells in donor bone marrow
inocula leads to predominant repopulation with donor lymphocytes in
recipient mice (31).
In contrast, the pX gene expression was still observed in Tg mouse joints transferred with non-Tg mouse BM cells. Because the pX gene is expressed in cells other than immune cells, including muscle and synovial cells in pX-Tg mice (17), this residual expression of the pX gene is considered to be originated from the resident cells in the joints. This pX gene expression in the joints did not cause arthritis providing the BM-derived cells were normal. Not only the histology but also the autoantibody levels and cytokine production in the joints were normal in these normal BM cell-transferred Tg mice. These findings indicate that synovial cell overgrowth in the affected joints is not the direct effect of the pX gene expression in the synovial cells. Cytokine overproduction also seems not to be caused by the resident cells in the joints, although it is quite possible that BM-derived cells such as macrophage or type A cells in the synovial tissues participate in the production of inflammatory cytokines (32).
Accordingly, these findings, together with the observations that 1) a considerably long time (>15 wk), i.e., enough to replace the immune system (33), is required before onset of the disease after transplantation of Tg BM cells, 2) the effects of the BM transplantation last for at least 26 wk, and 3) transplantation of Tg T cells only could not induce arthritis, and conversely, T cells from normal mice could not prevent development of the disease (data not shown) suggest that the autoimmunity in pX-Tg mice is not caused by a T cell aberration resulting from abnormal T cell education but by a stem cell disorder. Stem cell disorder was also suggested in autoimmune-prone mice such as MRL/MP-lpr/lpr, (NZB x NZW)F1 mice, and BXSB mice. Autoimmune diseases in these mice are also prevented from developing or are completely corrected even after their development by transplantation of normal BM cells (34, 35).
We found that development of arthritis was suppressed in athymic nu/nu-pX-Tg mice, suggesting a crucial role of T cells in the pathogenesis. These observations are consistent with the autoimmune nature of the disease. In this connection, we recently found that pX-Tg mouse T cells were refractory against Fas-mediated apoptosis, as is the case in lpr/lpr mice (19). Thus, the stem cell disorder seems to manifest at least in T cells.
However, a small but significant number of athymic nu/nu-pX-Tg mice developed arthritis. This observation suggests that the pX gene can induce arthritis without involvement of T cells. Another possibility is that extrathymically developed T cells in aged nu/nu mice are involved in the development of the disease. However, because RF levels were not elevated in nu/nu-pX-Tg mice, these T cells seem to be immunologically unresponsive. Consistent with this notion, Gutierrez-Ramos et al. (36) reported that peripherally developed T cells in nu/nu mice are anergic against self components, in spite of the defect in negative selection in the thymus. Therefore, the defects in pX-Tg-T cells could be supplementary, and the autoimmunity may only accelerate the disease.
In this context, the observation that cytokine gene expression,
including IL-1, IL-6, and TNF-, is augmented in the Tg BM
cell-transferred joints suggests that Tg BM-derived cells, such as type
A synovial cells which have some characteristics similar to those of
macrophages, distribute in the joints (32) and produce
these cytokines. We propose that this activation of inflammatory
cytokine genes by Tax in the joints may trigger the inflammation by
inducing other cytokines, chemokines, chemical mediators, and
proteases. This should cause swelling of the joint, infiltration of
neutrophils into synovial tissues, and destruction of the bony
structure by activating osteoclasts without involvement of immune
reaction. Actually, it was reported that overproduction of TNF- in
the joints caused joint destruction without infiltration of immune
cells (22). Thus, the arthritis seems to be induced by two
steps: 1) the triggering by the cytokine overproduction; and 2)
acceleration by the autoimmune reaction against synovial components.
Cytokine overproduction may also contribute to the development of
autoimmunity by enhancing recruitment of immune cells, augmenting
expression of MHC genes, and activating T cells in anergy
(37). These mechanisms may also explain why these mice
develop such a tissue-specific autoimmunity.
Our study shows that T cell-dependent autoimmunity plays crucially important roles in manifesting symptoms of arthritis. We are now examining the roles of cytokines and autoimmunity in the development of arthritis by producing cytokine knockout mice and tissue-specific tax-Tg mice.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
2 Address correspondence and reprint requests to Dr. Yoichiro Iwakura, Center for Experimental Medicine, Institute of Medical Science, University of Tokyo, Shirokanedai, Minato-ku, Tokyo 108-8639, Japan. E-mail address:
3 Abbreviations used in this paper: RA, rheumatoid arthritis; IIC, type II collagen; HTLV-I, human T cell leukemia virus type I; LTR, long terminal repeat; Tg, transgenic; pX-Tg, HTLV-I env-pX region-introduced transgenic; RF, rheumatoid factor.
Received for publication October 5, 1998. Accepted for publication August 27, 1999.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
This article has been cited by other articles:
|
|
 |
|
  Y. Kobari, Y. Misaki, K. Setoguchi, W. Zhao, Y. Komagata, K. Kawahata, Y. Iwakura, and K. Yamamoto T cells accumulating in the inflamed joints of a spontaneous murine model of rheumatoid arthritis become restricted to common clonotypes during disease progression Int. Immunol., January 1, 2004; 16(1): 131 - 138. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
···, n = 32). The
incidence and severity score of arthritis in untreated
pX-Tg are also shown (···•···,
n = 35). C and D,
[non-Tg
