|
|
||||||||

Departments of
* Neurology and
Microbiology and Immunology, University of Rochester School of Medicine and Dentistry, Rochester, NY 14642
| Abstract |
|---|
|
|
|---|
| Introduction |
|---|
|
|
|---|
Such observations have lead to the conclusion that autoreactive T cells must be tightly regulated and/or inhibited from differentiating into autoimmune effector cells in the peripheral immune systems of healthy individuals (13). Several laboratories have recently demonstrated that mature T cells specific for nominal Ags enter a state of tolerance following cognate interactions with resting or unactivated dendritic cells (14, 15, 16, 17). A similar mechanism might be employed during homeostasis to maintain peripheral tolerance against self Ags. Bone marrow-derived APCs continually scavenge self Ags that are shed as a consequence of physiological cell turnover (18, 19, 20). Since these APCs remain in a resting state in the absence of danger signals (21), the cross-presentation of self peptides under steady state conditions could result in the generation of a subpopulation of tolerized autoreactive T cells within the normal T cell repertoire. The question then arises: what factors lead to the conversion of tolerized autoreactive T cells into autoimmune effector cells in patients with autoimmune diseases such as MS?
In the studies mentioned above the activation state of APCs was a pivotal factor in deciding the fate of the T cell population targeted for tolerization. In each case coadministration of an agonist Ab specific for the macrophage/dendritic cell receptor, CD40, subverted tolerance induction and resulted in T cell priming. Based on these findings we posed the hypothesis that APC-activating agents could disrupt endogenous immunoregulatory processes, thereby allowing tolerized autoreactive T cells to regain their biological functions and mediate autoimmune disease. This concept is intriguing in light of the fact that autoimmune diseases such as MS are more likely to relapse in the setting of infectious illness, which is typically associated with APC activation (22, 23, 24, 25, 26, 27, 28). According to our theory, APCs that bear autoantigenic peptides (or molecular mimics) and are activated by "danger signals" in the course of an infection could mediate the conversion of bystander autoreactive T cells from a tolerized into an effector phenotype. Consistent with this scenario, several researchers have found that peripheral T cell tolerance was inhibited by infection of laboratory animals with bacterial, parasitic, or viral agents (29, 30). Furthermore, pretreatment of mice with bacterially derived DNA exacerbates viral and autoimmune models of MS (31).
We were particularly interested in the role of ligands for CD40 and Toll-like receptors (TLRs) in reversing tolerance against self Ags. Microbial DNA, the ligand for TLR-9, is released during the course of infections and could result in widespread APC activation (32, 33). Furthermore, microbial exposures result in the expansion and activation of anti-microbial T cells that could license APCs via CD40-CD40 ligand interactions (34, 35, 36). Engagement of either CD40 or TLR-9 on macrophages and dendritic cells results in the production of IL-12 (37, 38, 39, 40, 41, 42), a cytokine that we as well as others have shown plays a critical role in autoimmune pathogenesis (43, 44, 45, 46, 47, 48, 49, 50). Therefore, we questioned whether CD40 or TLR-9 triggering in combination with antigenic challenge would restore the pathogenic functions of tolerized autoreactive T cells by an IL-12-dependent pathway.
To test our hypothesis we used a classic protocol for Ag-specific immunosuppression (51, 52). Naive SJL mice were injected with a peptide fragment of PLP (PLP139151) by the i.p. route. Mice treated in this fashion are unable to mount proliferative or cytokine recall responses following subsequent challenge with a normally immunogenic form of PLP139151 (i.e., a s.c. inoculum of peptide in CFA) and are protected from EAE. Here we report that that myelin-reactive T cells from tolerized donors are converted into pathogenic effector cells that proliferate, secrete cytokines, and transfer EAE following reactivation in the presence of agonist anti-CD40, but not control, Abs. Moreover, a CpG-containing oligonucleotide (ODN) that simulates bacterial DNA and binds TLR-9 was also effective in reversing tolerance. The disease-promoting activities of both anti-CD40 and CpG ODNs were at least partially dependent upon the induction of IL-12. Furthermore, stimulation of tolerized autoreactive T cells with IL-12 alone was sufficient to restore encephalitogenicity.
Our findings have important implications regarding the roles of environmental agents, including infectious microbes, in the pathogenesis of autoimmune diseases such as MS. Furthermore, they suggest that agents that block CD40 and TLR-9 and/or neutralize IL-12 might be useful in preventing autoimmune exacerbations triggered during infections or following other environmental exposures that result in APC activation.
| Materials and Methods |
|---|
|
|
|---|
Female SJL mice were obtained from the National Cancer Institute (Frederick, MD) at 68 wk of age and were housed in a pathogen-free facility. Mice were 812 wk of age when the experiments were initiated.
Peptides and oligonucleotides
PLP139151 (HSLGKWLGHPDKF) was synthesized and purified by HPLC by Macromolecular Resources (Fort Collins, CO). ODN were purchased from Operon Technologies (Alameda, CA) and were phosphorothioate-modified to increase their resistance to nuclease degradation. The sequences were as follows: CpG ODN, ATAATCGACGTTCAAGCAAG; and control (non-CpG) ODN, ATAATAGAGCTTCAAGCAAG (53). The LPS content of the ODNs was <1 ng LPS/mg DNA as measured by a Limulus amebocyte assay.
Induction of tolerance and active immunization
For induction of Ag-specific tolerance, mice were injected i.p. with PLP139151 (300 µg) emulsified in IFA (1/1). Control mice were mock-tolerized with PBS alone in IFA. Two weeks later all mice were challenged with PLP139151 (100 µg) in CFA (with 5 mg/ml heat-killed mycobacteria) by s.c. injection at four sites over the flanks. In experiments involving active immunization, Bordetella pertussis toxin (List Laboratories, Campbell, CA) was injected i.p. (2 ng/mouse) on days 0 and 2 postchallenge. Animals were examined daily by a blinded examiner for signs of EAE and were rated for severity of clinical signs on a five-point scale as previously described (48).
Generation of PLP139151-specific cell lines for cell transfer and cytokine analysis
Draining lymph nodes (LNs; inguinal and axillary) were removed
1014 days following s.c. challenge of mock- or
PLP139151-tolerized mice and processed as
previously described (48). Cells were resuspended in RPMI
1640 containing 10% FCS and standard supplements and were cultured in
24-well plates (2 ml/well) with or without
PLP139151 (50 µg/ml). Where specified, the
following reagents were added: CpG ODN (100 nM), control ODN (100 nM),
rIL-12 (20 ng/ml; gift from S. Wolf, Genetics Institute, Cambridge,
MA), anti-CD40 mAb (hybridoma FGK45, (54); 10
µg/ml); anti-IL-12 mAb (C17.8; 10 µg/ml; gift from G.
Trichieri, Wistar Institute, Philadelphia, PA), and Con A (2
µg/ml; Sigma-Aldrich, St. Louis, MO). The following mAbs were
purchased from BD PharMingen (San Diego, CA) and added at 10 µg/ml
where indicated: rat IgG2a, agonist anti-CD28, blocking
anti-CTLA-4, and anti-CD3
.
Supernatants were collected after 48 h of culture for cytokine quantification. At 96 h the cells were harvested, washed extensively, resuspended in PBS, and injected into naive syngeneic SJL mice (50 x 106 cells in 0.1 cc i.p.). Recipients were examined daily and rated for the severity of neurological deficits as described above.
Proliferation assays
LN cells (LNC; 4 x 105 in 0.2 ml) were cultured with or without PLP139151 (50 µg/ml) in triplicate for 4 days in 96-well round-bottom plates (Costar, Cambridge, MA). Wells were pulsed for the final 16 h of culture with 1 µCi [3H]TdR (Amersham, Arlington Heights, IL), and incorporated radioactivity was measured using a Betaplate scintillation counter (Wallac, Gaithersburg, MD).
Cytokine ELISA
IFN-
, IL-4, and IL-10 were quantified using a sandwich ELISA
technique based on noncompeting pairs of Abs as previously described
(45).
| Results |
|---|
|
|
|---|
Consistent with previous reports (51, 52), female SJL
mice injected with high dose PLP139151 in IFA
i.p. were protected from EAE upon active immunization with the same
peptide 2 wk later. In repeated experiments control mice experienced
EAE at an incidence of 85100%, compared with 0% of PLP-tolerized
mice. Draining LNC from PLP-tolerized, as opposed to mock-tolerized,
mice failed to mount lymphoproliferative or IFN-
recall responses to
PLP ex vivo or to transfer EAE to naive syngeneic recipients (Table I
). They also failed to produce
detectable amounts of IL-10 and IL-4, indicating that they had not
defaulted to a Tr1 or Th2 pathway (data not shown). Tolerance was
selective for the PLP-specific subset, since the same LNC mounted
robust proliferative and cytokine responses when stimulated with an
agonist Ab against CD3 or Con A (Table I
; our unpublished
observations).
|
CD28 agonist mAbs, or
CTLA-4 blocking mAbs
Criss-cross experiments revealed that CD4+ T
cells from PLP-tolerized donors fail to proliferate or produce
cytokines even when stimulated with control T-depleted splenocytes. On
the other hand, APCs from tolerized mice are competent in reactivating
Ag-specific CD4+ T cells from control immunized
mice (our unpublished observations). Hence, we concluded that
tolerance to PLP represents an intrinsic property of the T cell
population and is not secondary to a deficiency of immunocompetent
APCs. In previous experiments we as well as others found that TCR
transgenic T cells parked in syngeneic recipients are not clonally
deleted consequent to i.p. injection of high doses of target Ag in IFA
(our unpublished observations and Ref. 51). By
analogy, we speculated that polyclonal PLP-specific T cells persisted
in our tolerized mice, but were functionally inactive. We considered
the possibility that introduction of Ag in the absence of microbial
adjuvants resulted in Ag-specific T cell anergy as a consequence of
underexpression of costimulatory molecules on APCs
(55, 56, 57). However, lymph node cells from tolerized mice
remained unresponsive when cultured with
PLP139151 in the presence of
anti-CD28 agonist Abs or IL-2 (Fig. 1
). Consistent with previous reports
(58), ex vivo challenge of tolerized lymph node cells with
Ag and a blocking Ab against the T cell inhibitory molecule, CTLA-4,
also failed to restore proliferative or cytokine responses (Fig. 1
).
|
CD40
agonist mAbs reverses tolerance and restores encephalitogenicitySince T cell growth factors and costimulatory ligands did not restore the biological functions of tolerized PLP-specific T cells, we tested the ability of APC-activating agents to reverse tolerance. Other researchers had demonstrated that agonist anti-CD40 Abs prevent Ag-specific T cell tolerance against soluble peptides, apoptotic cells, and tumor Ags when administered simultaneously with the tolerogen (15, 16, 59). In the current manuscript we expand upon these findings by evaluating the role of anti-CD40 in reversing well-established tolerance against PLP, a candidate autoantigen in MS.
Naive SJL mice were pretreated with either
PLP139151 or vehicle in IFA i.p. before
challenge with PLP139151 in CFA s.c. On day 10
postchallenge draining LNs were harvested and pooled from each group,
separated into single-cell suspensions, and cultured with either an
agonist mAb against CD40 (FGK45) (54) or isotype-matched
control rat IgG2a in the presence or the absence of Ag. PLP-tolerized
LNC costimulated with FGK45, but not control IgG, exhibited
PLP139151-specific proliferation and IFN-
production that approximated that of mock-tolerized LNC (Fig. 2
, A and B). More
significantly, anti-CD40 stimulation converted the previously
tolerized LNC into encephalitogenic effectors that transferred moderate
to severe EAE to 100% of naive syngeneic recipients (Fig. 2
C).
|
Dendritic cells and macrophages up-regulate MHC and costimulatory
molecules and produce inflammatory/Th1-polarizing cytokines following
exposure to CpG ODNs, which are known to act through a TLR-9 dependent
pathway (32, 33, 46, 48, 53). We questioned whether APC
activation through this alternative route would also be effective in
reversing tolerance. PLP-tolerized LNC were challenged with or without
Ag and either an immunostimulatory CpG-ODN or a control ODN. Similar to
the results obtained with anti-CD40, tolerized LNC stimulated with
CpG-ODN, but not control ODN, mounted robust PLP-specific proliferation
and cytokine responses and transferred a severe form of EAE to naive
mice (Fig. 3
).
|
Reversal of tolerance was fully (FGK45) or partially (CpG ODN)
dependent on the induction of IL-12 (Figs. 2
and 3
, respectively).
Therefore, we questioned whether IL-12 stimulation itself would be
sufficient to restore proliferative and cytokine capacities and
encephalitogenicity. As demonstrated in Fig. 4
, LNC from PLP-tolerized mice regained
full biological activities upon reactivation with Ag and IL-12
alone.
|
We next investigated the effects of APC-activating agents on tolerance against PLP peptide in vivo. Mice were injected i.p. with three doses of either FGK45 or isotype-matched control IgG between days 48 posttolerization. All mice were challenged with PLP in CFA on day 7. Some of the experimental animals were followed long term for disease course; other were sacrificed on day 17 for LN harvest and cytokine/proliferation assays.
LNC removed from PLP-tolerized mice treated with FGK45, but not those
from mice treated with control IgG, proliferated and secreted
substantial amounts of IL-3 and IFN-
in response to antigenic
challenge in vitro (Fig. 5
, A
and B; our unpublished observations). Anti-CD40
treatment also resulted in a significant incidence of EAE in tolerized
mice (Fig. 5
C).
|
| Discussion |
|---|
|
|
|---|
Microbial infection is the environmental factor most strongly associated with exacerbations of MS as well as other autoimmune diseases (26, 61, 62, 63). Furthermore, mice that are transgenic for a myelin-specific TCR develop spontaneous EAE when housed in a nonsterile facility, but not in a sterile, specific pathogen-free facility (64). In this paper we demonstrate that activation of APCs via TLR-9 and CD40, events that are likely to occur during the course of an infection, reverses tolerance against myelin Ags and breaks resistance to EAE. We have found that the bacterial cell wall component, LPS, has a similar effect to CpG ODNs in our experimental system (our unpublished observations). This suggests that in the setting of infection, multiple TLR pathways could play a role in restoring autoimmune responses. Our experiments were designed to examine T cell tolerance and its reversal on a population level, which is most relevant regarding the clinical status of the host. In future experiments we will determine whether the restoration of biological functions by CpG ODN or FGK45 affects PLP-tolerized T cells uniformly or reflects the outgrowth of a subpopulation of T cells that possibly escaped tolerance induction in the first place.
In previous reports CD40 triggering prevented the induction of
tolerance against viral, xenogenic, and tumor Ags cells
(14, 15, 16, 17). In each case the combination of anti-CD40
with an otherwise tolerizing stimulus converted T cell tolerization
into T cell priming. In other studies coadministration of IL-12 alone
or in combination with anti-CTLA-4 prevented peripheral T cell
tolerance induced by i.v. peptides (65, 66). Here we
expand upon these findings by demonstrating that stimulation of LNC
with either anti-CD40 or rIL-12 can reverse tolerance after it has
been firmly established against a candidate autoantigen in adult
wild-type mice. Furthermore, to our knowledge this paper constitutes
the first report that engagement of TLR-9 with CpG ODNs can also
reverse tolerance. Not surprisingly, the
tolerance-breaking/disease-promoting effects of both anti-CD40 and
CpG ODNs are IL-12 dependent (Figs. 2
and 3
).
While stimulation of tolerized LNC with CpG or FGK45 was sufficient to reverse tolerance, the introduction of Ag emulsified in CFA by the s.c. route was not. In addition, systemic administration of CpG ODNs at the time of either tolerization and/or challenge did not restore PLP-specific LNC responses or increase susceptibility to EAE. Similarly, substitution of CFA with CpG ODN in IFA during in vivo challenge did not overcome tolerance (our unpublished observations). These apparently inconsistent findings might be explained by differences in the efficiency with which bacterial DNA products are delivered to the relevant APC depending upon the route and mode of administration. For example, s.c. injected CFA or CpG ODN in IFA has a depot effect that could restrict the APC population exposed to Ag and microbial DNA products to Langerhans cells and dermal macrophages. In addition, the rate at which bacterial products and Ags are leaked from the depot might not be ideal for activating those APCs that are capable of breaking tolerance. On the other hand, the half-life of systemically injected CpG ODN and its distribution in various body compartments are unknown. By contrast, the in vitro culture system that we used in our studies optimizes T cell/APC interactions and TLR-9/CD40 signaling. We would argue that the latter experimental paradigm more closely simulates the microenvironment of secondary lymphoid tissues draining an infectious focus.
While our findings provide an alternative explanation for the association between infection and autoimmunity, they are not incompatible with the concept of molecular mimicry. Based on our theory, exacerbation of MS or EAE in the setting of infectious illness is dependent on the temporal and spatial concurrence of at least three events. 1) APCs, most likely dendritic cells, scavenge and process endogenous myelin Ags (that are normally expressed in secondary lymphoid tissues (7, 8, 9, 10, 11, 12)) or their molecular mimics (that are introduced during the course of the infection). 2) These APCs are activated by TLR ligands released by the invading pathogens or by CD40 ligand-expressing T cells generated in response to the microbes. 3) While still in an activated state, these APCs come into contact with tolerized myelin-reactive T cells present in the local microenvironment and mediate their conversion to functional autoimmune effector cells. Our results cast further light on the mechanisms linking autoimmune exacerbations with infectious illnesses. They also support the use of reagents that deactivate APCs and/or neutralize IL-12 for the treatment of MS disease activity, particularly in the setting of infectious or other immunostimulatory environmental exposures.
| Footnotes |
|---|
2 Address correspondence and reprint requests to Dr. Benjamin M. Segal, Department of Neurology/Neuroimmunology, University of Rochester School of Medicine, 601 Elmwood Avenue, Box 605, Rochester, NY 14642. E-mail address: benjamin_segal{at}urmc.rochester.edu ![]()
3 Abbreviations used in this paper: EAE, experimental autoimmune encephalomyelitis; LN, lymph node; LNC, LN cell; MS, multiple sclerosis; ODN, oligonucleotide; PLP, proteolipid protein; TLR, Toll-like receptor. ![]()
Received for publication May 22, 2002. Accepted for publication July 1, 2002.
| References |
|---|
|
|
|---|
cell death promotes priming of diabetogenic CD8 T lymphocytes. J. Immunol. 168:1466.
production is dependent on macrophage secretion of IL-12. Clin. Immunol. Immunopathol. 84:185.[Medline]
production by stimulation of interleukin-12 and tumor necrosis factor-
. Cell Immunol. 167:72.[Medline]
. Proc. Natl. Acad. Sci. USA 93:2879.
-producing cells and increased expression of macrophage-associated inducible nitric oxide synthase as immunomodulatory mechanisms. Am. J. Pathol. 148:375.[Abstract]
heavy chain class switching. Immunity 5:319.[Medline]
This article has been cited by other articles:
![]() |
M. A. Kroenke, T. J. Carlson, A. V. Andjelkovic, and B. M. Segal IL-12- and IL-23-modulated T cells induce distinct types of EAE based on histology, CNS chemokine profile, and response to cytokine inhibition J. Exp. Med., July 7, 2008; 205(7): 1535 - 1541. [Abstract] [Full Text] [PDF] |
||||
![]() |
A. M. Krieg Antiinfective Applications of Toll-like Receptor 9 Agonists Proceedings of the ATS, July 1, 2007; 4(3): 289 - 294. [Abstract] [Full Text] [PDF] |
||||
![]() |
J. P. Sanderson, D. J. Naisbitt, J. Farrell, C. A. Ashby, M. J. Tucker, M. J. Rieder, M. Pirmohamed, S. E. Clarke, and B. K. Park Sulfamethoxazole and Its Metabolite Nitroso Sulfamethoxazole Stimulate Dendritic Cell Costimulatory Signaling J. Immunol., May 1, 2007; 178(9): 5533 - 5542. [Abstract] [Full Text] [PDF] |
||||
![]() |
C. Bartholdy, S. O. Kauffmann, J. P. Christensen, and A. R. Thomsen Agonistic Anti-CD40 Antibody Profoundly Suppresses the Immune Response to Infection with Lymphocytic Choriomeningitis Virus J. Immunol., February 1, 2007; 178(3): 1662 - 1670. [Abstract] [Full Text] [PDF] |
||||
![]() |
S. E. Cabbage, E. S. Huseby, B. D. Sather, T. Brabb, D. Liggitt, and J. Goverman Regulatory T Cells Maintain Long-Term Tolerance to Myelin Basic Protein by Inducing a Novel, Dynamic State of T Cell Tolerance J. Immunol., January 15, 2007; 178(2): 887 - 896. [Abstract] [Full Text] [PDF] |
||||
![]() |
L. Visser, M.-J. Melief, D. van Riel, M. van Meurs, E. A. Sick, S. Inamura, J. J. Bajramovic, S. Amor, R. Q. Hintzen, L. A. Boven, et al. Phagocytes Containing a Disease-Promoting Toll-Like Receptor/Nod Ligand Are Present in the Brain during Demyelinating Disease in Primates Am. J. Pathol., November 1, 2006; 169(5): 1671 - 1685. [Abstract] [Full Text] [PDF] |
||||
![]() |
R. D. Pawar, P. S. Patole, M. Wornle, and H.-J. Anders Microbial nucleic acids pay a Toll in kidney disease Am J Physiol Renal Physiol, September 1, 2006; 291(3): F509 - F516. [Abstract] [Full Text] [PDF] |
||||
![]() |
O. Saitoh and Y. Nagayama Regulation of Graves' Hyperthyroidism with Naturally Occurring CD4+CD25+ Regulatory T Cells in a Mouse Model Endocrinology, May 1, 2006; 147(5): 2417 - 2422. [Abstract] [Full Text] [PDF] |
||||
![]() |
E. D. Ponomarev, L. P. Shriver, and B. N. Dittel CD40 Expression by Microglial Cells Is Required for Their Completion of a Two-Step Activation Process during Central Nervous System Autoimmune Inflammation J. Immunol., February 1, 2006; 176(3): 1402 - 1410. [Abstract] [Full Text] [PDF] |
||||
![]() |
S. Gregoire, A. S. Bergot, C. Feraudet, C. Carnaud, P. Aucouturier, and M. B. Rosset The Murine B Cell Repertoire Is Severely Selected against Endogenous Cellular Prion Protein J. Immunol., November 15, 2005; 175(10): 6443 - 6449. [Abstract] [Full Text] [PDF] |
||||
![]() |
T R D J Radstake, A W T van Lieshout, P L C M van Riel, W B van den Berg, and G J Adema Dendritic cells, Fc{gamma} receptors, and Toll-like receptors: potential allies in the battle against rheumatoid arthritis Ann Rheum Dis, November 1, 2005; 64(11): 1532 - 1538. [Abstract] [Full Text] [PDF] |
||||
![]() |
M.-G. de Goer de Herve, D. Durali, T.-A. Tran, G. Maigne, F. Simonetta, P. Leclerc, J.-F. Delfraissy, and Y. Taoufik Differential effect of agonistic anti-CD40 on human mature and immature dendritic cells: the Janus face of anti-CD40 Blood, October 15, 2005; 106(8): 2806 - 2814. [Abstract] [Full Text] [PDF] |
||||
![]() |
A. Nogai, V. Siffrin, K. Bonhagen, C. F. Pfueller, T. Hohnstein, R. Volkmer-Engert, W. Bruck, C. Stadelmann, and T. Kamradt Lipopolysaccharide Injection Induces Relapses of Experimental Autoimmune Encephalomyelitis in Nontransgenic Mice via Bystander Activation of Autoreactive CD4+ Cells J. Immunol., July 15, 2005; 175(2): 959 - 966. [Abstract] [Full Text] [PDF] |
||||
![]() |
Y. Li, J. S. Heuser, S. D. Kosanke, M. Hemric, and M. W. Cunningham Protection against Experimental Autoimmune Myocarditis Is Mediated by Interleukin-10-Producing T Cells that Are Controlled by Dendritic Cells Am. J. Pathol., July 1, 2005; 167(1): 5 - 15. [Abstract] [Full Text] [PDF] |
||||
![]() |
H-J Anders A Toll for lupus Lupus, June 1, 2005; 14(6): 417 - 422. [Abstract] [PDF] |
||||
![]() |
|