| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
*
Amgen Institute and Ontario Cancer Institute, Departments of Medical Biophysics and Immunology, University of Toronto, Toronto, Ontario, Canada;
Skirball Institute of Biomolecular Medicine, Department of Pathology, New York University School of Medicine, New York, NY 10016; and
Howard Hughes Medical Institute, Center for Cancer Research, and Department of Biology, Massachusetts Institute of Technology, Cambridge, MA 02139
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Ag-induced experimental autoimmune encephalomyelitis (EAE) in mice mimics human MS and has been extensively studied to uncover its pathogenetic mechanisms. For instance, EAE can be induced in H-2u PL/J mice following immunization with myelin basic protein (MBP) in CFA. The presence of CD4+ T cells and T cell activation by APCs are necessary and sufficient conditions for the development of experimentally induced EAE (5, 6). In addition to immunization-triggered EAE, spontaneous EAE has been found to develop in MBP1–17 peptide-specific TCR transgenic mice (TgMBP+) of a RAG-1-deficient background (7).
Ag-specific activation of T lymphocytes requires two signals, one by the TCR and a second by costimulatory molecules (8, 9, 10). In the best-studied costimulatory pathway, CD28 expressed on T cells interacts with the B7 counterreceptors (CD80 and CD86) expressed on APC (9, 10). Since TCR ligation in the absence of CD28 costimulation renders T cells anergic (11) and CD80/CD86 overexpression in peripheral organs can mediate autoimmune disease (12, 13), the idea has been put forward that potentially autoaggressive T lymphocytes normally remain inactive because tissue cells cannot provide the costimulation necessary for T cell activation. Therapeutic interventions targeting the CD28/B7 costimulatory pathway have been successfully attempted in several autoimmune disease models, including Ag-induced EAE (14, 15, 16). However, the mechanisms that are involved in the initiation and pathogenesis of spontaneous EAE are elusive.
To address the role of the CD28 costimulatory pathway in the pathogenesis of spontaneous EAE, we introduced a CD28-null mutation (17) into TgMBP+RAG-1-/- mice (7). We provide the first genetic evidence that CD28 costimulation is crucial for the development of spontaneous EAE. Our results also suggest that CD28 does not regulate immunological anergy but rather adjusts the threshold for the activation of autoaggressive T cells.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Transgenic mice (TgMBP+) bearing a TCR directed
against MBP were generated using 
and ß TCR sequences from an
encephalomyelitogenic T cell clone and have been previously described
(7). These animals were maintained in a RAG-1-deficient (18) and
H-2u haplotype background
(TgMBP+RAG-1-/-). The CD28-null mutation,
also previously described (17), was backcrossed for six generations
into TgMBP+RAG-1-/- mice that develop
spontaneous EAE, and into PL/J mice that develop EAE following
immunization with MBP. PL/J mice were purchased from The Jackson
Laboratory (Bar Harbor, ME). Only littermate mice were analyzed in all
experiments. Mice were kept under pathogen-free conditions at the
Ontario Cancer Institute Animal Facility (Toronto, Canada) in
accordance with institutional guidelines.
Immunization and serology
For the experimental induction of EAE, CD28-deficient PL/J mice were immunized with whole MBP (Sigma, St. Louis, MO) in CFA at doses of 200, 400, and 600 µg per mouse on day 0, followed by i.v. administration of pertussis toxin (500 ng per mouse; List Biochemical Research, Campbell, CA) on day 0 and day 2 (6). The development of EAE was followed for up to 100 days in MBP-immunized CD28-/-PL/J and control CD28+/-PL/J littermate mice, and for up to 300 days in CD28-/-TgMBP+RAG-1-/- and control CD28+/-TgMBP+RAG-1-/- littermates that developed spontaneous EAE. EAE severity was graded as follows (6): level 1, limp tail; level 2, partial hind leg paralysis; level 3, complete hind leg paralysis; level 4, front leg weakness; level 5, moribund/death. Mice reaching grade 4 or 5 of disease severity were sacrificed according to the ethical guidelines of the Canadian Medical Research Council. At different time points following MBP immunization of PL/J mice, the presence of MBP-specific autoantibodies was determined from serum samples. MBP-specific IgG and IgM autoantibody titers were determined by ELISA using intact MBP as Ag (6).
Abs and flow cytometry
Single cell suspensions of thymocytes, spleen cells, and
mesenteric lymph node cells were prepared as described (7), resuspended
in immunofluorescence staining buffer (PBS, 4% FCS, 0.1%
NaN3), and incubated for 30 min with the appropriate mAbs.
The following mAbs were used: anti-CD3
(phycoerythrin
(PE)-labeled), anti-CD4 (PE-labeled), anti-CD8
(FITC-labeled), anti-CD45RB (FITC-labeled), anti-TCRß
(FITC-, or PE-labeled), anti-B220 (FITC-labeled), anti-CD28
(PE-labeled), anti-CD69 (FITC-labeled), anti-CD25
(FITC-labeled), anti-CD44 (PE-labeled), and anti-Vß8.1/8.2
(PE-labeled). All mAbs were purchased from PharMingen (San Diego, CA).
All staining combinations were as indicated in the figure legends.
Biotinylated Abs were visualized using Streptavidin-RED670 (Life
Sciences, St. Petersburg, FL). Samples were analyzed by flow cytometry
using a FACScan (Becton Dickinson, Mountain View, CA), and 10,000
events were gated by forward and side scatter.
T cell proliferation
T cells were negatively enriched from lymph nodes of
CD28+/-TgMBP+RAG-1-/- and
CD28-/-TgMBP+RAG-1-/- mice
using affinity columns (R&D Systems, Minneapolis, MN). Purified
(>95%) T cells (2 x 105/well) were cultured in
triplicate in round-bottom 96-well plates (Costar, Cambridge, MA) in
freshly prepared Iscove’s Modified Dulbecco’s Medium (10% FCS,
10-5 M 2-ME). T cells were activated using
CD28+/-TgMBP+RAG-1-/- splenic
APC (2 x 105/well; inactivated using 2000 rad
-irradiation) pulsed with different concentrations of the specific
acetyl-MBP1–17 peptide (ASQKRPSQRSKYLATAS). T cells were
cultured for 60 h, pulsed with 1 µCi [3H]thymidine
per well and harvested 12 h later. T cell culture supernatants
were assayed in triplicate for IL-2 by ELISA (QuantikineM kit; R&D
Systems). In the case of MBP-induced EAE, CD28-deficient PL/J and
control CD28+/-PL/J littermate mice were immunized with
whole MBP (200 µg per mouse), as described above. Two weeks
later, lymph node T cells were obtained and cultured in presence
of different concentrations of whole MBP, followed by
[3H]thymidine pulsing.
Statistical analysis
Analyses of incidence (EAE-free survival) were conducted using a
multiple-sample survival analysis (extension of Gehan’s generalized
Wilcoxon test), by which a score is first assigned to each survival
time using Mantel’s procedure (19). Next, a 
-square value is
computed based on the sums for each group of this score. EAE severity
between groups was compared using ANOVA. Differences were considered to
be statistically significant at p < 0.05.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
MBP-specific TCR transgenic mice express a Vß8V4TCR on T
cells, and these cells develop into CD4+ Th cells during
thymocyte selection (7). Lymph nodes from
TgMBP+RAG-1-/- mice contain >90%
CD4+TgMBP+ T lymphocytes (Fig. 1
) and lack B lymphocytes due to the
absence of recombinase (RAG) activity (7). Thus,
TgMBP+RAG-1-/- mice harbor a single Ag
receptor reactive to the MBP1–17 peptide. Analysis of
lymph node cells, thymocytes, splenocytes, and PBL from
CD28-/-TgMBP+RAG-1-/- mice
showed that the absence of CD28 did not affect the development or
numbers of CD4+TgMBP+ T lymphocytes (Fig. 1, and data not shown). Similarly, expression levels of the transgenic
TCRß-chain, CD3 coreceptors, CD45 and the activation markers CD69,
CD25, and CD44 were not altered in the absence of CD28 expression (Fig. 1, and data not shown).
|
It has been previously shown that in a RAG-1-/-
background, all TgMBP+ mice develop EAE (7). Similarly,
100% of
CD28+/+TgMBP+RAG-1-/- and
85% of CD28+/-TgMBP+RAG-1-/-
mice developed spontaneous EAE in our animal colony (Fig. 2a). The affected
CD28+/+ and
CD28+/-TgMBP+RAG-1-/- mice
developed disease of similar severity (Fig. 2b), suggesting
that CD28 gene dosage influences the incidence but not the severity of
the disease. In contrast to CD28-expressing mice, 90% of
CD28-/-
TgMBP+RAG-1-/- mice were free of disease
through a 300-day follow-up period (p = 0.00000
as compared with CD28+/- and CD28+/+
mice). However, two
CD28-/-TgMBP+RAG-1-/- mice
that were affected developed full severity EAE with disease scores of 5
(Fig. 2b; see Material and Methods for scoring
criteria). These results show that CD28 costimulation in T cells is
crucial for the pathogenesis and initiation of spontaneous EAE in
vivo.
|
It has been previously shown that TCR ligation in the absence of
CD28 costimulation renders T cells anergic (11). Accordingly, CTLA4-Ig,
which blocks the interaction between CD28 and its ligands CD80/CD86,
has been used in the treatment of autoimmune diseases to inhibit T cell
activation and induce immmunological tolerance in vivo (14, 15). To
investigate the possibility that CD28-deficiency abrogates the
pathogenesis of spontaneous EAE by inducing anergy and T cell
unresponsiveness, we analyzed the proliferative response and IL-2
production of
CD4+TgMBP+RAG-1-/- T cells in
vitro. Surprisingly, CD28-deficient
CD4+TgMBP+RAG-1-/- cells were
able to proliferate in response to MBP1–17 peptide in a
dose-dependent fashion, albeit at lower levels than T cells from
CD28+/-TgMBP+RAG-1-/- mice (Fig. 3a). Moreover, IL-2
production by
CD28+/-CD4+TgMBP+RAG-1-/-
and
CD28-/-CD4+TgMBP+RAG-1-/-
T cells was similar and also concentration-dependent (Fig. 3b). These results indicate that CD28 plays a role in
controlling the activation of autoaggressive TgMBP+ T
cells, depending on the concentration of the autoantigen.
|
Most studies of EAE involve Ag immunization-induced models, in
which T cell immune responses are not monoclonal but rather develop
against several epitopes (5, 20). The CD28-dependency in the
CD4+MBP-specific TCR transgenic EAE model could be thus
limited to the MBP1–17 epitope. To exclude this
possibility, we introduced the CD28 mutation into PL/J mice that
develop severe EAE following immunization with MBP (1, 6). Immunization
with whole MBP (200 µg/mouse) triggered the development of EAE in
75% of CD28+/- PL/J mice within a 100-day follow-up
period (Fig. 4a). The diseased
mice developed severe EAE as determined by the EAE score (Fig. 4b). As occurred in the case of spontaneous EAE in
CD28-/-TgMBP+RAG1-/- mice, the
majority (90%) of CD28-/- PL/J mice were protected from
disease (Fig. 4, a and b). Interestingly, even at
low doses of MBP immunization (200 µg/mouse), CD28-/-
PL/J mice produced MBP-specific IgG and IgM autoAbs at titers similar
to those in CD28+/- PL/J mice (Fig. 4c).
However, MBP-specific in vitro T cell proliferation (Fig. 4d) and IL-2 production (data not shown) were not detectable
in CD28-/- PL/J mice. These data show that CD28
costimulation during T cell activation plays a crucial role in the
induction of both spontaneous and Ag-induced EAE.
|
, a and b, and
data not shown). However, following immunization with 400 µg
MBP/mouse (data not shown) or 600 µgMBP/mouse (Fig. 5, a
and b), CD28-/- PL/J mice developed EAE with
similar incidence and severity as CD28+/- PL/J mice. These
data show that, at high Ag doses, EAE can be induced in the absence of
CD28 expression.
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Full T cell activation requires a costimulatory signal in addition to TCR ligation, which is provided primarily by interactions between CD28 on T cells and B7 (CD80, CD86) counterreceptors on APCs. CD28 signaling allows T cell survival and long-term proliferation after antigenic challenge in vivo (21). Failure to activate the costimulatory signal results in T cell unresponsiveness or clonal anergy (11, 22), and, therefore, peripheral tolerance and control of autoimmunity are thought to rely on the absence of costimulatory signaling. Indeed, the CD28 counterreceptors CD80 and CD86 are expressed on inflammatory and glial cells in the CNS of diseased EAE mice as well as in CNS lesions from MS patients, and it has been shown that glial cells can function as APCs in vitro (23, 24, 25, 26). Development of EAE and relapses can be partially blocked by treatment with Abs and CTLA-4 fusion proteins that block CD28 costimulation (14, 15, 27, 28, 29). Alternatively, the course of EAE can be altered by treatments that block CTLA-4 function, leading to exacerbation of the disease (30, 31, 32). Furthermore, Lin et al. (33) have recently shown that blocking CTLA-4 during cardiac allograft in CD28-deficient mice accelerates the acute rejection, clearly demonstrating that CTLA-4 can exert its negative regulatory function independent of CD28. These data demonstrate the importance of costimulatory pathways in preventing the development of an autoimmune process.
However, costimulatory pathways other than CD28/B7/CTLA-4 also contribute to T cell activation in vivo, and that might be an explanation for why higher doses of Ag can overcome CD28-dependence for EAE development in MBP-immunized mice. Indeed, it has been demonstrated that EAE development is abrogated in CD40 ligand (CD40L)-deficient mice, showing the role of the CD40-CD40L costimulatory pathway in the control of T cell activation in vivo (34). Furthermore, in a PLP peptide-induced model of EAE, the treatment with anti-CD40L Ab protected mice when low doses of immunogen were used, while at high doses, T cell activation and disease occurred even with anti-CD40L treatment, albeit in this case the disease was in a mild form (35).
Intruigingly, in our study, CD4+ T cells isolated from CD28-/-TgMBP+RAG-1-/- mice still proliferated in response to the MBP1–17 peptide in vitro, and immunization with high doses of the autoantigen induced EAE at high prevalence and severity in CD28-/- PL/J mice. These results indicate that CD28 does not regulate immunological anergy in vivo but rather adjusts the threshold for the activation of MBP-specific T cells. Recently, it has been shown that in the absence of CD28, continued stimulation of the Ag receptor alone, either through prolonged viral replication or repeated peptide injection, prevents the induction of anergy and generates a functional CD8+ T cell response in vivo (36). By contrast, transient stimulation of the Ag receptor in the absence of CD28 costimulation induces anergy in CD8+ T cells (36). Furthermore, CD28 surface expression reduces the number of TCRs that need to be triggered for T cell activation and allows activation of T cells by low-affinity ligands (37). These effects have been described in viral-specific CD8+ T cells. However, it is possible that CD28 costimulation is required to lower the threshold for TCR activation of autoaggressive CD4+ T cells to trigger an autoimmune response. This hypothesis is in line with the finding that spontaneous and MBP-induced EAE are abrogated in the absence of CD28, but functional autoaggressive T cells are still present in CD28-/- mice.
Based on functional data from CD28-/-TgMBP+RAG-1-/- and CD28-/- PL/J mice, we propose the following model for CD28-regulated EAE. Low numbers of naive autoaggressive T cells cross the blood-brain barrier and encounter myelin sheath-specific Ags that might be presented at low density by the microglia within the healthy CNS. In this case, CD28 expression is clearly required for the activation of resting, autoreactive T cells. Hypothetically, CD28 costimulation lowers the threshold for effective Ag receptor-mediated stimulation, while CD28-deficiency results in a short-lived, low-avidity TCR engagement by CD4+TgMBP+ T cells within the CNS. Thus, CD28-/- T cells are unable to trigger an autoimmune response and cannot activate the cytokine circuits responsible for the attraction of mononuclear cells and the establishment of an inflammatory response.
Current understanding of the requirements for T cell activation unveils costimulatory pathways as potential targets for therapeutic intervention in MS, as well as other autoimmune diseases. Particularly, CD28 pathway presents itself as an attractive target, and our results provide the first genetic evidence that CD28 costimulation is crucial in the pathogenesis of spontaneous EAE.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
2 Address correspondence and reprint requests to Dr. Josef M. Penninger, Amgen Institute, 620 University Avenue, Suite 706, Toronto, M5G 2C1, Canada. E-mail address:
3 Abbreviations used in this paper: MS, multiple sclerosis; CNS, central nervous system; EAE, experimental autoimmune encephalomyelitis; MBP, myelin basic protein; PE, phycoerythrin.
Received for publication October 20, 1998. Accepted for publication January 25, 1999.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
leads to autoimmunity in transgenic mice. Proc. Natl. Acad. Sci. USA 91:5138. and lipopolysaccharide and downregulation by interleukin-10, prostaglandin E2 and cyclic AMP- elevating agents. J. Neuroimmunol. 72:83.[Medline]
-activated SJL/J astrocytes. J. Immunol. 160:4271.This article has been cited by other articles:
|
|
 |
|
  C. M. Krawczyk, R. G. Jones, A. Atfield, K. Bachmaier, S. Arya, B. Odermatt, P. S. Ohashi, and J. M. Penninger Differential Control of CD28-Regulated In Vivo Immunity by the E3 Ligase Cbl-b J. Immunol., February 1, 2005; 174(3): 1472 - 1478. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
D. Li, I. Gal, C. Vermes, M.-L. Alegre, A. S. F. Chong, L. Chen, Q. Shao, V. Adarichev, X. Xu, T. Koreny, et al. Cutting Edge: Cbl-b: One of the Key Molecules Tuning CD28- and CTLA-4-Mediated T Cell Costimulation J. Immunol., December 15, 2004; 173(12): 7135 - 7139. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
R. E. Jones, D. Bourdette, N. Moes, A. Vandenbark, A. Zamora, and H. Offner Epitope Spreading Is Not Required for Relapses in Experimental Autoimmune Encephalomyelitis J. Immunol., February 15, 2003; 170(4): 1690 - 1698. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 G. C. Furtado, M. A. C. de Lafaille, N. Kutchukhidze, and J. J. Lafaille Interleukin 2 Signaling Is Required for CD4+ Regulatory T Cell Function J. Exp. Med., September 16, 2002; 196(6): 851 - 857. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. Zhang, T. Bardos, D. Li, I. Gal, C. Vermes, J. Xu, K. Mikecz, A. Finnegan, S. Lipkowitz, and T. T. Glant Cutting Edge: Regulation of T Cell Activation Threshold by CD28 Costimulation Through Targeting Cbl-b for Ubiquitination J. Immunol., September 1, 2002; 169(5): 2236 - 2240. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
Q. Ye, B. Press, S. Kissler, X.-F. Yang, L. Lu, C. H. Bassing, B. P. Sleckman, M. Jansson, V. Panoutsakopoulou, L. A. Trimble, et al. T Cell Costimulation through CD28 Depends on Induction of the Bcl-x{gamma} Isoform: Analysis of Bcl-x{gamma}-deficient Mice J. Exp. Med., July 1, 2002; 196(1): 87 - 95. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. A. Williams, J. T. Tan, A. B. Adams, M. M. Durham, N. Shirasugi, J. K. Whitmire, L. E. Harrington, R. Ahmed, T. C. Pearson, and C. P. Larsen Characterization of Virus-Mediated Inhibition of Mixed Chimerism and Allospecific Tolerance J. Immunol., November 1, 2001; 167(9): 4987 - 4995. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
K. Pechhold, N. B. Patterson, C. Blum, C. L. Fleischacker, B. O. Boehm, and D. M. Harlan Low Dose Streptozotocin-Induced Diabetes in Rat Insulin Promoter-mCD80-Transgenic Mice Is T Cell Autoantigen-Specific and CD28 Dependent J. Immunol., February 15, 2001; 166(4): 2531 - 2539. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. Winer, I. Astsaturov, R. K. Cheung, L. Gunaratnam, V. Kubiak, M. A. Cortez, M. Moscarello, P. W. O'Connor, C. McKerlie, D. J. Becker, et al. Type I Diabetes and Multiple Sclerosis Patients Target Islet Plus Central Nervous System Autoantigens; Nonimmunized Nonobese Diabetic Mice Can Develop Autoimmune Encephalitis J. Immunol., February 15, 2001; 166(4): 2831 - 2841. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
P. B. Silver, K. S. Hathcock, C.-C. Chan, B. Wiggert, and R. R. Caspi Blockade of Costimulation Through B7/CD28 Inhibits Experimental Autoimmune Uveoretinitis, But Does Not Induce Long-Term Tolerance J. Immunol., November 1, 2000; 165(9): 5041 - 5047. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
K. Saegusa, N. Ishimaru, K. Yanagi, N. Haneji, M. Nishino, M. Azuma, I. Saito, and Y. Hayashi Autoantigen-Specific CD4+CD28low T Cell Subset Prevents Autoimmune Exocrinopathy in Murine Sjogren's Syndrome J. Immunol., August 15, 2000; 165(4): 2251 - 2257. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. T. Chang, B. M. Segal, and E. M. Shevach Role of Costimulation in the Induction of the IL-12/IL-12 Receptor Pathway and the Development of Autoimmunity J. Immunol., January 1, 2000; 164(1): 100 - 106. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J.-X. Gao, H. Zhang, X.-F. Bai, J. Wen, X. Zheng, J. Liu, P. Zheng, and Y. Liu Perinatal Blockade of B7-1 and B7-2 Inhibits Clonal Deletion of Highly Pathogenic Autoreactive T Cells J. Exp. Med., April 15, 2002; 195(8): 959 - 971. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
