Initiation and Exacerbation of Autoimmune Demyelin
http://www.100md.com
病菌学杂志 2005年第13期
Department of Microbiology and Immunology and Interdepartmental Immunobiology Center, Northwestern University Feinberg School of Medicine, 303 E. Chicago Ave., Chicago, Illinois 60611
ABSTRACT
Epidemiological studies indicate that infectious agents are important in the pathogenesis of multiple sclerosis (MS). Our previous reports showed that the infection of SJL mice with a nonpathogenic variant of Theiler's murine encephalomyelitis virus (TMEV) engineered to express a naturally occurring Haemophilus influenzae-encoded molecular mimic (HI574-586) of an immunodominant self-myelin proteolipid protein epitope (PLP139-151) induced a rapid-onset demyelinating disease associated with the activation of PLP139-151-specific Th1 responses. The current results extend our previous findings in four critical respects. We show that disease initiation by the H. influenzae mimic is prevented by tolerance to the self PLP139-151 epitope, definitively proving the occurrence of infection-induced molecular mimicry. We demonstrate that the H. influenzae mimic epitope can be processed from the flanking sequences within the native mimic protein. We show that the H. influenzae mimic epitope only induces an immunopathologic self-reactive Th1 response and subsequent clinical disease in the context of the TMEV infection and not when administered in complete Freund's adjuvant, indicating that molecular mimicry-induced disease initiation requires virus-activated innate immune signals. Lastly, we show that the infection of SJL mice with TMEV expressing the H. influenzae mimic can exacerbate a previously established nonprogressive autoimmune disease of the central nervous system. Collectively, these findings illustrate the evolving mechanisms by which virus infections may contribute to both the initiation and exacerbation of autoimmune diseases, and they have important implications for MS pathogenesis.
INTRODUCTION
Multiple sclerosis (MS) is a human autoimmune demyelinating disease of the central nervous system (CNS) which is thought to be mediated by autoreactive T cells. Both genetic and environmental factors, such as infections, have been implicated in MS susceptibility (8, 14, 15, 29), and virus infections can induce demyelinating diseases (7, 26, 32, 34). Interestingly, both healthy individuals and MS patients possess peripheral CD4+ T cells specific for myelin antigens (25, 30). Nonetheless, the mechanism of autoreactive T-cell activation in MS remains unknown. One proposed mechanism of activation is molecular mimicry, whereby autoreactive T cells can be activated by infectious agents encoding epitopes with similar structures or sequence homology to self-tissue epitopes (10, 21). T-cell mimicry between myelin and viral peptides has recently been described for MS patients (33, 36). Molecular mimicry has also been implicated in the pathogenesis of diabetes, systemic lupus erythematosis, and human T-lymphotropic virus type 1-associated myelopathy/tropical spastic paraparesis (1, 11, 16, 19). MS patients present with multiple clinical forms, including relapsing-remitting and chronic-progressive forms of disease. Although the mechanism(s) involved in the initiation and in the relapse phase of relapsing-remitting MS is currently not fully understood, several factors, including viral infection, may be important (2, 27, 28).
We previously demonstrated (20) that a nonpathogenic variant of Theiler's murine encephalomyelitis virus (TMEV) expressing either the mouse myelin proteolipid protein (PLP) epitope, PLP139-151, or a core mimic sequence of PLP139-151 derived from Haemophilus influenzae (HI574-586) (3) induced a mild, slowly progressing form of CNS demyelinating disease. Clinical disease was associated with the cross-activation and Th1 differentiation of myelin epitope PLP139-151-specific CD4+ T cells, and the disease induced by the virus encoding the self PLP139-151 epitope could be prevented by tolerance to the homologous self peptide (22).
The present studies were designed to further understand the mechanisms of infection-induced molecular mimicry in both the initiation and progression of CNS autoimmunity. The present results extend our previous studies by illustrating four additional critical observations regarding the role of infection-induced molecular mimicry in activating CNS demyelinating disease. We demonstrate (i) that the CNS disease induced by infection with the core H. influenzae mimic-expressing virus is due to the cross-activation of autoreactive T cells, as peripheral tolerance induced to the self myelin PLP139-151 epitope specifically inhibited disease induction; (ii) that the mimic H. influenzae epitope can be naturally processed from the flanking sequences in the native H. influenzae protein and presented to cross-activate PLP139-151-specific autoreactive CD4+ Th1 cells; (iii) that molecular mimicry-induced disease initiation requires virus-activated innate immune signals, since infection with TMEV expressing the H. influenzae mimic induced disease, while the core H. influenzae mimic epitope administered in complete Freund's adjuvant (CFA) did not; and (iv) that a slowly progressing CNS autoimmune disease can be significantly exacerbated upon reinfection with the H. influenzae mimic-expressing virus. Collectively, these studies demonstrate that autoreactive T cells triggered by infection-induced molecular mimicry can both initiate and exacerbate clinical autoimmune demyelination and have important implications for the initiation and relapse phases of MS pathogenesis.
MATERIALS AND METHODS
Mice. Five- to 6-week-old female SJL mice were obtained from Harlan Sprague-Dawley (Indianapolis, Ind.). Mice were housed under barrier conditions at the National Institutes of Health-approved Northwestern University Medical School animal facilities. All protocols were approved by the Northwestern University Animal Care and Use Committee. Paralyzed mice were afforded easier access to food and water.
Construction of mimic-expressing TMEV. The cDNA encoding the BeAn strain of TMEV was modified by inserting ClaI sites and deleting 23 amino acids, as previously described (20). Genomic DNA containing the sequence coding for Haemophilus influenzae serine protease IV was obtained from the American Type Culture Collection (Manassas, Va.). ClaI sites were introduced by PCR to the H. influenzae DNA at either end of a coding sequence for 39 amino acids, HI566-604, which contained the PLP139-151 mimic sequence, HI574-586, by using the following primers: HI39A, 5' GCA CCT TAA TCG ATT TGA GCG 3'; and HI39AS, 5' CTG ACC ATC GAT TTT AGG ATC 3'. Following an enzyme restriction cut with ClaI, the 39-amino-acid-encoding piece containing the HI566-604 sequence was inserted into the ClaI site in the Cla-BeAn virus cDNA. This was designated HI39-BeAn cDNA. Infectious virus was produced as previously described (20). The PLP-BeAn, OVA-BeAn, and HI-BeAn viruses were produced as previously described (20).
Infection of SJL mice with TMEV. Mice (n = 8) were infected by the intracerebral injection of 3 x 106 PFU of either wild-type TMEV (strain BeAn 8386), HI-BeAn, PLP139-BeAn, OVA-BeAn, or Cla-BeAn and were scored at weekly intervals on a clinical scale of 0 to 5, as follows: 0, asymptomatic; 1, mild waddling gait; 2, severe waddling gait; 3, paralysis of one hindlimb; 4, total hindlimb paralysis, severe dehydration, and/or malnutrition; and 5, death. The data were plotted as mean clinical scores for each group of animals.
Peptides. The proteolipid protein (PLP) peptide PLP139-151 (HSLGKWLGHPDKF), the TMEV capsid peptide VP270-86 (WTTSQEAFSHIRIPLPH), the ovalbumin peptide OVA323-339 (ISQAVHAAHAEINEAGR), the H. influenzae peptide HI574-586 (EQLVKWLGLPAPI), the H. influenzae 30-mer peptide HI566-595 (SDTKKGAQEQLVKWLGLPAPIQKLQKELNI), and the PLP 30-mer peptide PLP130-159 (QAHSLERVCHCLGKWLGHPDKFVGITYALT) (Fig. 1a) were purchased from Peptides International (Louisville, Ky.). Their amino acid compositions were verified by mass spectrometry, and their purity was assessed by high-performance liquid chromatography.
Induction of peripheral tolerance. Tolerance was induced by the intravenous injection of 5 x 107 peptide-pulsed, 1-ethyl-3-(3-dimethylaminopropyl) carbodiimide (ECDI)-fixed syngeneic splenocytes, as described previously (13, 18, 22).
Delayed-type hypersensitivity responses. Delayed-type hypersensitivity (DTH) responses were elicited by injecting mice subcutaneously in alternate ears with 10 μg of the challenge peptides after measurements of ear thickness with a Mitutoyo model 7326 engineer's micrometer (Schlesinger's Tools, Brooklyn, N.Y.). Twenty-four hours following the peptide challenge, the ears were measured again, and differences in ear thickness (swelling) from the prechallenge thickness were expressed in units of 10–4 inches (mean ± standard errors of the means), as previously described (20).
T-cell proliferation and cytokine analysis. Spleens were removed from infected mice that had not been previously used for DTH responses (n = 2) at the indicated times following infection or immunization. T-cell proliferation and cytokine analyses were performed as described previously (20). Proliferation was determined for triplicate wells for each peptide concentration and then expressed in counts per minute. For gamma interferon (IFN-) cytokine analysis, a duplicate set of proliferation wells was used to collect supernatants at 48 and 72 h, and cytokine concentrations were determined by an enzyme-linked immunosorbent assay.
Immunohistochemistry. Mice (n = 3 per group) were anesthetized and perfused with phosphate-buffered saline (PBS) on the indicated days postinfection. Spinal cords and brains were removed by dissection, and 2- to 3-mm spinal cord blocks and brains were immediately frozen in OCT (Miles Laboratories, Elkhart, Ind.) in liquid nitrogen. Multiple 6-μm-thick cross sections from the spinal cord or brain were cut and mounted on Superfrost Plus electrostatically charged slides (Fisher, Pittsburgh, Pa.), air dried, and stored at –80°C. CNS immunohistochemistry was performed as previously described (12). The results shown are representative of the cellular infiltrates from multiple sections cut from three individual mice per group. Photomicrographs of immunostained sections from a spinal cord, cerebellum, and brain stem representative of the cellular infiltrates in the different groups were used to quantify the numbers of positive inflammatory cells per area of each tissue section. Data from photomicrographs were stored as 8-bit binary images in a grayscale format. Quantification was determined with ImageJ software v1.32j (http://rsb.info.nih.gov/ij/). Threshold values of brightness and contrast were determined for the photomicrographs and were kept constant for each sample. Prior to analysis, the minimum and maximum sizes of pixels to be counted were determined whereby sections with no positive staining gave a measurement of 0.0 pixels. The data are presented as percentages of positive pixels per area of photomicrograph.
Statistics. Clinical severity results are presented as mean group clinical scores, and the statistical difference was calculated by the Mann-Whitney nonparametric ranking test. Statistical analyses of DTH and proliferative responses and IFN- secretion were performed by using the two-tailed Student t test.
RESULTS
Peripheral tolerance to PLP139-151 inhibits HI-BeAn virus-induced autoimmune demyelinating disease. As we have previously reported (20), mice infected intracerebrally with a nonpathogenic variant of the BeAn strain of TMEV (ClaI-BeAn, which carries a 23-amino-acid deletion in the viral leader sequence) engineered to express a 30-mer peptide encompassing the encephalitogenic myelin epitope PLP139-151 (PLP-BeAn) develop a severe, rapid-onset, progressive demyelinating disease, while infection with BeAn containing the PLP139-151 epitope mimic, consisting of amino acids 574 to 586 of the Haemophilus influenzae serine protease IV protein (HI-BeAn), develop a less severe nonprogressive clinical disease (Fig. 1b). HI-BeAn induces a cross-reactive PLP139-151-specific CD4+ T-cell response, as determined by in vivo DTH and in vitro T-cell proliferative assays (Fig. 1c and d), and results in the infiltration of CD4+ T cells and activated F4/80+ macrophages/microglia into the CNS (see Fig. 5d and l) and associated focal areas of demyelination, as assessed by staining with an anti-PLP antibody (manuscript in preparation), similar to the lesions found in PLP-BeAn-infected SJL mice (20). To definitively determine whether cross-reactivity (mimicry) between PLP139-151 and HI574-586 is critical for the induction of demyelinating disease, we exposed HI-BeAn- or PLP-BeAn-infected mice (day 0) to the myelin epitope, PLP139-151, or the irrelevant OVA323-339 epitope (on days –7 and +3) to induce tolerance. As we previously reported (22), PLP139-151 tolerance inhibited clinical disease in mice infected with PLP-BeAn. Significantly, the present results show that PLP139-151 tolerance also specifically inhibits the CNS disease, as assessed by the TMEV scoring system, characterized by a mild to moderate waddling gait induced by infection with the mimic-expressing HI-BeAn virus compared to the case for mice exposed to OVA323-339 (P < 0.05) (Fig. 1b). PLP139-151-tolerant mice infected with either HI-BeAn or PLP-BeAn displayed significantly reduced HI574-586- and PLP139-151-specific DTH and CD4+ T-cell proliferative responses (P < 0.05) than those of the controls (Fig. 1c and d). The tolerance was specific to the PLP139-151 epitope, as TMEV-specific VP270-86-specific T-cell responses were unaffected in PLP139-151-tolerant mice, and the responses in OVA323-339-tolerant mice were similar to those in untreated mice. Interestingly, tolerance induced with HI574-586, as well as that induced with PLP139-151, was shown to inhibit experimental autoimmune encephalomyelitis (EAE) induced by priming with PLP139-151/CFA (Fig. 1e), illustrating the bidirectional cross-reactivity of these epitopes. These results thus provide definitive evidence that the early-onset CNS disease induced by infection with the core H. influenzae mimic-expressing virus is due to the cross-activation of PLP139-151-specific autoreactive CD4+ T cells, i.e., the data prove the occurrence of infection-induced molecular mimicry.
The HI574-586 mimic can be processed from its native protein sequence and induces demyelinating disease only in the context of virus infection. The data in Fig. 1 demonstrate that TMEV engineered to express the HI574-586 myelin mimic within the core of a 30-amino-acid sequence spanning PLP130-159 (HI-BeAn) induces the in vivo activation of cross-reactive PLP139-151-specific CD4+ T cells. To determine if the HI574-586 mimic epitope can be processed from its native flanking sequences within the mimic protein, we constructed a TMEV virus (HI39-BeAn) expressing a 39-amino-acid (HI566-604) piece of the H. influenzae protease IV protein. SJL mice infected with HI39-BeAn (Fig. 2a) developed an early-onset demyelinating disease which was less severe than that induced by PLP-BeAn but equivalent to that induced by HI-BeAn, while mice infected with TMEV expressing a 30-mer peptide encompassing the nonself OVA323-339 sequence developed a late-onset disease typical of infections with the wild-type BeAn TMEV strain (20, 22). We have previously described that the insertion of base pairs encoding the OVA317-346 30-mer peptide in either the forward or reverse direction restores the ability of the parental Cla-BeAn virus to persist in the CNS and to induce late-onset clinical disease (18). Although the molecular basis for this is unknown, it is possible that the addition of the OVA peptide coding sequence to the leader restores its original length and induces a conformational change in Cla-BeAn, which allows more efficient in vivo replication and virus persistence in mouse CNS cells. HI-BeAn- and HI39-BeAn-infected mice displayed strong CD4+ DTH and proliferative T-cell responses associated with potent IFN- secretion, which is typical of Th1 responses, upon recall with either HI574-586 or the PLP139-151 self epitope (Fig. 2b to d). The level of VP270-86-induced IFN- secretion seen in PLP139-BeAn-infected mice compared to that in mice infected with the other recombinant viruses was less than twofold higher and not consistently seen and thus was not considered significant. Thus, HI574-586 is arguably a natural PLP139-151 mimic capable of being processed from its native sequence following virus infection to activate cross-reactive PLP139-151-specific autoreactive CD4+ T cells.
To determine if the HI mimic epitope could also induce clinical disease in the context of innate immune signals induced by complete Freund's adjuvant (CFA), we immunized SJL mice with either the minimal HI574-586 mimic epitope or a 30-amino-acid peptide (HI566-595) encompassing the core HI574-586 mimic epitope in CFA. Immunization with either peptide activated both HI574-586 mimic-specific and cross-reactive self-myelin PLP139-151-specific CD4+ T cells (Fig. 3c). However, unlike the case for SJL mice infected with TMEV engineered to express either the core HI574-586 mimic epitope flanked by the native PLP sequences (HI-BeAn) or the long HI566-604 epitope (HI39-BeAn) (Fig. 2a), priming with either HI574-586 or HI566-595 in CFA failed to induce a clinical demyelinating disease (Fig. 3a). This was also true if pertussis toxin and/or lipopolysaccharide was used in conjunction with CFA as an additional innate immune stimulus (data not shown). Disease induction following HI-BeAn infection appeared to induce more potent innate immune signals than priming with the mimic peptide/CFA, as exemplified by the failure of the self PLP139-151 peptide to recall either in vivo DTH responses or in vitro IFN- production (Fig. 3b and d). In contrast, the self PLP139-151 peptide recalled potent DTH and IFN- production in mice infected with either HI-BeAn or HI39-BeAn (Fig. 2b and d). Thus, the induction of autoimmunity via molecular mimicry requires critical innate immune signals specific to the infectious agent encoding the molecular mimic to induce an efficient differentiation of cross-reactive autoreactive Th1 cells with pathological potential.
Secondary virus infection in the CNS exacerbates clinical disease in mice previously infected with HI-BeAn. Mice infected with the HI-BeAn mimic virus consistently developed a less severe clinical demyelinating disease than mice infected with PLP-BeAn (Fig. 2a) (20). To determine if subsequent virus infections could exacerbate the nonprogressive demyelinating disease induced by HI-BeAn infection, we infected SJL mice intracerebrally with HI-BeAn (day 0) and then reinfected them on day +14 with HI-BeAn, OVA-BeAn, the nonpathogenic parental virus (Cla-BeAn), or an equivalent volume of PBS. A primary infection with HI-BeAn and subsequent reinfection with HI-BeAn induced a significantly more severe clinical disease (P < 0.05) than that in mice given PBS on day +14 (Fig. 4a). Interestingly, mice infected with HI-BeAn that received a second infection with OVA-BeAn also developed an exacerbated clinical disease compared with that in PBS-injected controls (P < 0.05) (Fig. 4a). The reinfection of HI-BeAn virus-infected mice with the parental, nonpersisting Cla-BeAn virus resulted in a delayed onset of clinical disease compared with that in controls, and these mice developed a delayed exacerbated disease as well (Fig. 4a). However, as observed in multiple experiments, the clinical disease induced upon reinfection with the PLP139-151 mimic-expressing HI-BeAn virus always had an earlier onset and was significantly more severe than the disease in mice reinfected with OVA-BeAn or Cla-BeAn. Mice infected with the HI-BeAn virus followed by a second infection with OVA-BeAn, Cla-BeAn, or HI-BeAn developed HI574-586-specific CD4+ T-cell proliferative responses as well as cross-reactive PLP139-151-specific CD4+ T-cell responses (Fig. 4d and e) (P < 0.05). All mice initially infected with HI-BeAn developed virus-specific CD4+ T-cell responses which increased slightly upon secondary infection (Fig. 4c). The CD4+ T-cell responses to the virus, myelin, and the mimic peptide in these infected mice were associated with IFN--specific secretion, which is indicative of Th1-type CD4+ T-cell responses (Fig. 4f). Thus, a primary infection with a mimic-encoding virus may result in a suboptimal activation of autoreactive T cells, resulting in a mild initial clinical disease, but secondary infections with neurotropic viruses expressing the PLP139-151 mimic epitope more efficiently enhance the autoimmune response, leading to a severe clinical disease compared to that induced by non-mimic-expressing neurotropic viruses.
To further determine whether a prior infection of the CNS with a neurotropic virus (OVA-BeAn) that induces a late-onset, slowly progressing disease (Fig. 2a) could predispose mice to a more severe autoimmune disease upon secondary infection with a mimic-encoding virus, we infected mice with OVA-BeAn (day 0) and then reinfected them with HI-BeAn, OVA-BeAn, Cla-BeAn, or an equivalent volume of PBS on day +14. Mice infected with OVA-BeAn that received a second infection with HI-BeAn developed a disease with a significantly earlier onset than that in mice that received a second infection with OVA-BeAn, Cla-BeAn, or PBS (P < 0.05) (Fig. 4b). OVA-BeAn virus-infected mice that received a second infection with HI-BeAn or OVA-BeAn developed a slightly more severe disease than control mice throughout the disease course. Mice infected with the OVA-BeAn virus and reinfected with the HI-BeAn virus developed HI574-586-specific CD4+ T-cell responses and had increased PLP139-151-specific CD4+ T-cell responses compared to those of control mice (Fig. 4d and e). All OVA-BeAn virus-infected mice had virus-specific CD4+ T-cell responses, although interestingly, mice reinfected with the non-CNS-persisting Cla-BeAn virus exhibited reduced reactivities to all challenge peptides (Fig. 4c to f).
Brains and spinal cords were removed from mice (n = 3) at 60 days postinfection, and frozen sections were analyzed for their levels of CD4+ T-cell and macrophage infiltrates to assess the effects of secondary virus infections on the levels of CNS inflammatory cell infiltration. Lumbar spinal cords from HI-BeAn virus-infected mice that were reinfected with HI-BeAn displayed significantly more CD4+ T cells and F4/80+ microglia/macrophages (Fig. 5a and i) than did PBS-treated control mice (Fig. 5d and l). Mice initially infected with HI-BeAn and reinfected with either OVA-BeAn or Cla-BeAn had more CD4+ T cells (Fig. 5b and c) and macrophages (Fig. 5j and k) than did PBS-treated controls (Fig. 5d and l) but fewer of these cells than did mice reinfected with HI-BeAn (Fig. 5a and i). Lumbar spinal cords from mice infected with the OVA-BeAn virus and subsequently reinfected with HI-BeAn had increased CD4+ T cells and macrophages (Fig. 5e and m) compared to control mice (Fig. 5h and p). Cerebellum and brain stem sections from mice initially infected with HI-BeAn and reinfected with HI-BeAn displayed large numbers of infiltrating CD4+ T cells and F4/80+ cells (Fig. 5q to t). However, OVA-BeAn-infected mice that were reinfected with HI-BeAn displayed limited immune cell infiltrates in the brain (Fig. 5u to x). Overall, the levels of immune cell infiltration in the CNS correlated well with the severity of clinical disease observed for mice that received secondary infections.
DISCUSSION
Molecular mimicry is a putative mechanism for the activation of autoreactive T cells in MS, a human autoimmune demyelinating disease. However, little direct evidence exists for mimicry-induced T-cell autoimmunity using infectious agents. Previously, we described a model of molecular mimicry whereby the viral delivery of a PLP139-151 myelin-mimic epitope naturally encoded in the protease IV protein of H. influenzae, HI574-586, could induce an early-onset autoimmune demyelinating disease associated with the activation of autoreactive PLP139-151-specific CD4+ Th1 cells (20). The present experiments build upon this initial observation and illustrate four critically important conditions required for the induction and/or exacerbation of autoimmune disease by infection-induced molecular mimicry.
Firstly, the current demonstration that PLP139-151 tolerance (13, 18) significantly inhibited the incidence and severity of disease following HI-BeAn infection and specifically inhibited both HI574-587- and PLP139-151-specific CD4+ T-cell responses definitively shows the functional activation of pathological cross-reactive HI574-586 and PLP139-151 CD4+ T cells following HI-BeAn infection. Secondly, the demonstration that infection with the HI39-BeAn virus induced a similar early-onset demyelinating disease and similar activation of PLP139-151-specific CD4+ Th1 cells, as observed following infection with the HI-BeAn virus, shows that the H. influenzae mimic sequence is a natural epitope which can be processed and presented in vivo by SJL antigen-presenting cells to activate PLP139-151-specific autoreactive Th1 cells. We feel that this is a significant advance over past descriptions of pathological mimic epitopes, which uniformly employed minimal mimic sequences identified in T-cell hybridoma screens for their disease-inducing abilities and did not take into account the possibility that the mimic peptide may not be capable of being processed from its surrounding amino acid residues in the native mimic protein.
Thirdly, our data showing an induction of clinical CNS disease and autoreactive PLP139-151-specific DTH and IFN- Th1 responses following infection with a TMEV engineered to express either the core HI574-586 mimic epitope flanked by the native PLP sequences (HI-BeAn) or the long HI566-604 epitope (HI39-BeAn) (Fig. 2), but not upon priming with HI-BeAn or HI39-BeAn in CFA (Fig. 3), demonstrate that an equally important component required for the infection-induced initiation of autoimmune disease via molecular mimicry is the ability of the pathogen to trigger an appropriate innate immune response leading to an efficient expansion and differentiation of pathological autoreactive Th1 responses. This finding is consistent with our demonstration that a TMEV infection of quiescent microglial cells leads to the induction of a functional antigen presentation function concomitant with a significant upregulation of major histocompatibility complex and costimulatory molecules and an increased expression of cytokines (interleukin-12 and interleukin-23) required for Th1 differentiation (23). Furthermore, we have more recently shown that TMEV infection is a more potent stimulus to resting microglia than is the ligation of other Toll-like receptors (TLRs). Compared to the activation of TLR2 via peptidoglycan, TLR3 alone via poly(I-C), TLR4 via bacterial lipopolysaccharide, or TLR9 via CpG DNA, the infection of microglia with viable TMEV caused a more potent upregulation of innate and effector immune cytokines, chemokines, and major histocompatibility complex class II and costimulatory molecules, enabling the microglia to more efficiently present myelin antigens to CD4+ Th1 cells (24). Preliminary unpublished data indicate that TMEV is also tropic for dendritic cells, and thus it is possible that TMEV expressing the Haemophilus-encoded PLP139-151 mimic causes a more efficient activation/maturation of dendritic cells than does CFA, resulting in an enhanced activation of autoreactive Th1 cells. In addition to the more efficient induction of peripheral cross-reactive PLP139-151-specific Th1 responses by HI-BeAn infection than by HI574-586/CFA priming, the fact that TMEV can persist and replicate in the CNS, leading to the local release of inflammatory cytokines and to possible deleterious effects on blood-brain barrier integrity, may also contribute to the efficient disease induction observed following infection with the mimic-expressing virus versus that with peptide priming.
Lastly, our results demonstrate that a secondary infection with a myelin mimic-expressing neurotropic virus can exacerbate a mild preexisting CNS disease, indicating that mimicry may also trigger disease relapses in autoimmune disease. Various environmental factors have been associated with initiating MS relapses, including viral and bacterial infections (2, 9, 27, 28, 31). The infection of SJL mice with the HI-BeAn mimic virus induces a mild, slowly progressive form of clinical disease. The nonprogressive disease induced by HI-BeAn may be due to the failure of the virus to fully activate the self-reactive PLP139-151-specific T-cell population (HI574-586 has a 50% infective concentration of 385 nM for I-As, while that of PLP139-151 is 87 nM) (3), or the mimic peptide may only activate a subset of the self-reactive population (6). Significantly, mice infected with a second dose of HI-BeAn developed a severe, progressive clinical disease, consistent with the repriming of HI574-586/PLP139-151-specific CD4+ T cells. We have also recently shown that the reinfection of prediabetic mice with a virus expressing an H-2Kb-restricted mimic ligand to a lymphocytic choriomeningitis virus (self) epitope transgenically expressed on pancreatic ?-cells can accelerate the development of diabetes, but this mimic ligand-expressing arenavirus, unlike HI-BeAn, is not capable of initiating autoimmune disease (4). Interestingly, HI-BeAn-infected mice reinfected with TMEV encoding an irrelevant OVA epitope also developed a more severe clinical disease, indicating that bystander activation may also restimulate autoreactive T cells that are initially activated by HI-BeAn virus infection. This is significant, as anecdotal evidence suggests that relapse episodes in MS can be triggered by various upper respiratory virus infections (2, 9). The present study also advances previous findings in which mice primed with a plasmid coding for PLP could render mice susceptible to a mild disease following a subsequent rechallenge with CFA (35). This highlights the difficulties faced when investigating the role of viruses in MS pathogenesis, as relapses may be induced by otherwise innocuous or irrelevant viral infections.
Significantly, an initial infection of the CNS with the OVA-BeAn virus followed by a secondary infection with the HI-BeAn virus slightly accelerated the clinical disease. However, reinfection with the HI-BeAn virus did not increase the severity of autoimmune demyelinating disease as significantly as when the initial infection was with the HI-BeAn virus. Although the proliferative responses to PLP139-151 were similar between the OVA-BeAn/HI-BeAn- and HI-BeAn/HI-BeAn-infected groups, the clinical disease was more severe for the HI-BeAn/HI-BeAn-infected group (Fig. 4a). Importantly, we observed that IFN- secretion, assessed by a PLP139-151 rechallenge in vitro, was significantly higher for the HI-BeAn/HI-BeAn-infected group (Fig. 4f), indicating that the enhanced disease severity correlates with the ability of the H. influenzae mimic peptide to efficiently cross-activate pathogenic PLP139-151 Th1 cells, which are essential to the initiation and progression of CNS autoimmune disease. Thus, this suggests that molecular mimicry boosts the autoreactive Th1 cell population, resulting in a more severe autoimmune demyelinating disease. Therefore, a non-mimic-encoding pathogen may only actively participate in disease pathogenesis once autoimmune cells are activated, whether by molecular mimicry or other mechanisms, e.g., epitope spreading (17). Further evidence to support this conclusion can be drawn from the observation that preinfection with a persistent virus which expressed a nonself, nonmimic epitope (OVA-BeAn) or a transient infection with Cla-BeAn did not predispose mice to exacerbated autoimmune disease upon HI574-586 immunization (5). However, mice preinfected with HI-BeAn and subsequently immunized with the H. influenzae peptide showed an exacerbated disease compared to that after H. influenzae immunization or HI-BeAn infection alone. Furthermore, infection with HI-BeAn and immunization with an irrelevant peptide (OVA) did not exacerbate the disease. Therefore, it appears that one requirement for autoimmune disease is an infection of the target organ, i.e., the CNS, with the mimic-expressing virus.
Interestingly, there were significant differences in the numbers of inflammatory cells present in the spinal cord and brain in mice receiving secondary infections. With mice infected with HI-BeAn/HI-BeAn, we observed a significant number of CD4+ T cells and macrophages present in the spinal cord, consistent with the severe clinical disease. Furthermore, the presence of inflammatory infiltrates in the cerebellum and brain stem was only observed in mice from the HI-BeAn/HI-BeAn-infected group with severe clinical disease. Therefore, the increased clinical severity may have been due to damage in these regions of the CNS. The cerebellum is involved in the coordination of voluntary motor movement, balance, equilibrium, and muscle tone. Therefore, inflammation in these structures is likely to induce a further loss of coordination of motor movement (asynergia), movement tremors (intention tremor), staggering, wide-based walking (ataxic gait), and weak muscles (hypotonia). Overall, mice that received secondary infections had a more severe disease which correlated with more infiltrating cells in the brain and spinal cord.
Our results suggest that molecular mimicry could be an important factor in the pathogenesis of MS by expanding a population of T cells that recognize self-epitopes. These autoreactive cells may then be further expanded to different degrees by differing stimuli, including reinfections with viruses or bacteria, which may or may not encode epitope mimics of relevant autoantigens via direct and/or bystander mechanisms. The potential requirements for target organ infection and persistence of the pathogen in the initiation of autoimmune disease by infection-induced molecular mimicry require further study. Furthermore, while the identification of potential mimic epitopes from pathogens by the use of computer searches is useful, it is important to determine whether the epitopes are naturally processed and presented in the infected host and if they can stimulate pathological autoimmune responses in the context of a variety of innate immune stimuli. Although the recombinant Theiler's virus model of molecular mimicry allows the determination of various conditions required for the induction of CNS autoimmunity, we are currently testing whether the infection of SJL mice with viable wild-type H. influenzae versus a sppA-deficient mutant (lacking the protease IV protein, which contains the PLP139-151 mimic epitope) induces PLP139-151-specific T-cell responses and/or clinical disease as the definitive test of infection-induced molecular mimicry. Collectively, these studies present compelling evidence that virus-induced molecular mimicry can both induce and enhance Th1-mediated CNS autoimmune disease in a model of MS, which may have important implications for the initiation and relapse phases of MS pathogenesis.
ACKNOWLEDGMENTS
This work was supported in part by the U.S. Public Health Service, by NIH grant NS-23349, and by National Multiple Sclerosis Society (NMSS) research grant RG-3166-A-4. J. L. Croxford and J. K. Olson were supported by NMSS postdoctoral fellowships (FG-1456-A-1 and FA-1517-A-1, respectively).
These authors contributed equally to this work.
REFERENCES
Albert, L. J., and R. D. Inman. 1999. Molecular mimicry and autoimmunity. N. Engl. J. Med. 341:2068-2074.
Andersen, O., P. E. Lygner, T. Bergstrom, M. Andersson, and A. Vahlne. 1993. Viral infections trigger multiple sclerosis relapses: a prospective seroepidemiological study. J. Neurol. 240:417-422.
Carrizosa, A. M., L. B. Nicholson, M. Farzan, S. Southwood, A. Sette, R. A. Sobel, and V. K. Kuchroo. 1998. Expansion by self antigen is necessary for the induction of experimental autoimmune encephalomyelitis by T cells primed with a cross-reactive environmental antigen. J. Immunol. 161:3307-3314.
Christen, U., K. H. Edelmann, D. B. McGavern, T. Wolfe, B. Coon, M. K. Teague, S. D. Miller, M. B. Oldstone, and M. G. von Herrath. 2004. A viral epitope that mimics a self antigen can accelerate but not initiate autoimmune diabetes. J. Clin. Investig. 114:1290-1298.
Croxford, J. L., H. A. Anger, and S. D. Miller. 2005. Viral delivery of an epitope from Haemophilus infuenzae induces central nervous system autoimmune disease by molecular mimicry. J. Immunol. 174:907-917.
Croxford, J. L., J. K. Olson, and S. D. Miller. 2002. Epitope spreading and molecular mimicry as triggers of autoimmunity in the Theiler's virus induced demyelinating disease model of multiple sclerosis. Autoimmun. Rev. 1:251-260.
Dal Canto, M. C., and H. L. Lipton. 1975. Primary demyelination in Theiler's virus infection. An ultrastructural study. Lab. Investig. 33:626-637.
Ebers, G. C., A. D. Sadovnick, and N. J. Risch. 1995. A genetic basis for familial aggregation in multiple sclerosis. Canadian Collaborative Study Group. Nature 377:150-151.
Edwards, S., M. Zvartau, H. Clarke, W. Irving, and L. D. Blumhardt. 1998. Clinical relapses and disease activity on magnetic resonance imaging associated with viral upper respiratory tract infections in multiple sclerosis. J. Neurol. Neurosurg. Psychiatry 64:736-741.
Fujinami, R. S., and M. B. Oldstone. 1985. Amino acid homology between the encephalitogenic site of myelin basic protein and virus: mechanism for autoimmunity. Science 230:1043-1045.
Gran, B., B. Hemmer, M. Vergelli, H. F. McFarland, and R. Martin. 1999. Molecular mimicry and multiple sclerosis: degenerate T-cell recognition and the induction of autoimmunity. Ann. Neurol. 45:559-567.
Katz-Levy, Y., K. L. Neville, J. Padilla, S. M. Rahbe, W. S. Begolka, A. M. Girvin, J. K. Olson, C. L. Vanderlugt, and S. D. Miller. 2000. Temporal development of autoreactive Th1 responses and endogenous antigen presentation of self myelin epitopes by CNS-resident APCs in Theiler's virus-infected mice. J. Immunol. 165:5304-5314.
Kennedy, M. K., L. J. Tan, M. C. Dal Canto, V. K. Tuohy, Z. J. Lu, J. L. Trotter, and S. D. Miller. 1990. Inhibition of murine relapsing experimental autoimmune encephalomyelitis by immune tolerance to proteolipid protein and its encephalitogenic peptides. J. Immunol. 144:909-915.
Kurtzke, J. F. 1993. Epidemiologic evidence for multiple sclerosis as an infection. Clin. Microbiol. Rev. 6:382-427.
Kurtzke, J. F., and K. Hyllested. 1979. Multiple sclerosis in the Faroe Islands. I. Clinical and epidemiological features. Ann. Neurol. 5:6-21.
Levin, M. C., S. M. Lee, F. Kalume, Y. Morcos, F. C. Dohan, Jr., K. A. Hasty, J. C. Callaway, J. Zunt, D. Desiderio, and J. M. Stuart. 2002. Autoimmunity due to molecular mimicry as a cause of neurological disease. Nat. Med. 8:509-513.
Miller, S. D., C. L. Vanderlugt, W. S. Begolka, W. Pao, R. L. Yauch, K. L. Neville, Y. Katz-Levy, A. Carrizosa, and B. S. Kim. 1997. Persistent infection with Theiler's virus leads to CNS autoimmunity via epitope spreading. Nat. Med. 3:1133-1136.
Miller, S. D., R. P. Wetzig, and H. N. Claman. 1979. The induction of cell-mediated immunity and tolerance with protein antigens coupled to syngeneic lymphoid cells. J. Exp. Med. 149:758-773.
Oldstone, M. B. 1998. Molecular mimicry and immune-mediated diseases. FASEB J. 12:1255-1265.
Olson, J. K., J. L. Croxford, M. Calenoff, M. C. Dal Canto, and S. D. Miller. 2001. A virus-induced molecular mimicry model of multiple sclerosis. J. Clin. Investig. 108:311-318.
Olson, J. K., J. L. Croxford, and S. D. Miller. 2001. Virus-induced autoimmunity: potential role of viruses in initiation, perpetuation, and progression of T cell-mediated autoimmune diseases. Viral Immunol. 14:227-250.
Olson, J. K., T. N. Eagar, and S. D. Miller. 2002. Functional activation of myelin-specific T cells by virus-induced molecular mimicry. J. Immunol. 169:2719-2726.
Olson, J. K., A. M. Girvin, and S. D. Miller. 2001. Direct activation of innate and antigen presenting functions of microglia following infection with Theiler's virus. J. Virol. 75:9780-9789.
Olson, J. K., and S. D. Miller. 2004. Microglia initiate central nervous system innate and adaptive immune responses through multiple TLRs. J. Immunol. 173:3916-3924.
Olsson, T., J. Sun, J. Hillert, B. Hojeberg, H. P. Ekre, G. Andersson, O. Olerup, and H. Link. 1992. Increased numbers of T cells recognizing multiple myelin basic protein epitopes in multiple sclerosis. Eur. J. Immunol. 22:1083-1087.
Osame, M., K. Usuku, S. Izumo, N. Ijichi, H. Amitani, A. Igata, M. Matsumoto, and M. Tara. 1986. HTLV-I associated myelopathy, a new clinical entity. Lancet i:1031-1032.
Panitch, H. S. 1994. Influence of infection on exacerbations of multiple sclerosis. Ann. Neurol. 36(Suppl.):S25-S28.
Rapp, N. S., J. Gilroy, and A. M. Lerner. 1995. Role of bacterial infection in exacerbation of multiple sclerosis. Am. J. Phys. Med. Rehab. 74:415-418.
Sawcer, S., H. B. Jones, R. Feakes, J. Gray, N. Smaldon, J. Chataway, N. Robertson, D. Clayton, P. N. Goodfellow, and A. Compston. 1996. A genome screen in multiple sclerosis reveals susceptibility loci on chromosome 6p21 and 17q22. Nat. Genet. 13:464-468.
Scholz, C., K. T. Patton, D. E. Anderson, G. J. Freeman, and D. A. Hafler. 1998. Expansion of autoreactive T cells in multiple sclerosis is independent of exogenous B7 costimulation. J. Immunol. 160:1532-1538.
Sibley, W. A., C. R. Bamford, and K. Clark. 1985. Clinical viral infections and multiple sclerosis. Lancet i:1313-1315.
Smyth, J. M., B. J. Sheahan, and G. J. Atkins. 1990. Multiplication of virulent and demyelinating Semliki Forest virus in the mouse central nervous system: consequences in BALB/c and SJL mice. J. Gen. Virol. 71:2575-2583.
Tejada-Simon, M. V., Y. C. Zang, J. Hong, V. M. Rivera, and J. Z. Zhang. 2003. Cross-reactivity with myelin basic protein and human herpesvirus-6 in multiple sclerosis. Ann. Neurol. 53:189-197.
Tellez-Nagel, I., and D. H. Harter. 1966. Subacute sclerosing leukoencephalitis. I. Clinico-pathological, electron microscopic and virological observations. J. Neuropathol. Exp. Neurol. 25:560-581.
Theil, D. J., I. Tsunoda, F. Rodriguez, J. L. Whitton, and R. S. Fujinami. 2001. Viruses can silently prime for and trigger central nervous system autoimmune disease. J. Neurovirol. 7:220-227.
Wucherpfennig, K. W., and J. L. Strominger. 1995. Molecular mimicry in T cell-mediated autoimmunity: viral peptides activate human T cell clones specific for myelin basic protein. Cell 80:695-705.(J. Ludovic Croxford, Juli)
ABSTRACT
Epidemiological studies indicate that infectious agents are important in the pathogenesis of multiple sclerosis (MS). Our previous reports showed that the infection of SJL mice with a nonpathogenic variant of Theiler's murine encephalomyelitis virus (TMEV) engineered to express a naturally occurring Haemophilus influenzae-encoded molecular mimic (HI574-586) of an immunodominant self-myelin proteolipid protein epitope (PLP139-151) induced a rapid-onset demyelinating disease associated with the activation of PLP139-151-specific Th1 responses. The current results extend our previous findings in four critical respects. We show that disease initiation by the H. influenzae mimic is prevented by tolerance to the self PLP139-151 epitope, definitively proving the occurrence of infection-induced molecular mimicry. We demonstrate that the H. influenzae mimic epitope can be processed from the flanking sequences within the native mimic protein. We show that the H. influenzae mimic epitope only induces an immunopathologic self-reactive Th1 response and subsequent clinical disease in the context of the TMEV infection and not when administered in complete Freund's adjuvant, indicating that molecular mimicry-induced disease initiation requires virus-activated innate immune signals. Lastly, we show that the infection of SJL mice with TMEV expressing the H. influenzae mimic can exacerbate a previously established nonprogressive autoimmune disease of the central nervous system. Collectively, these findings illustrate the evolving mechanisms by which virus infections may contribute to both the initiation and exacerbation of autoimmune diseases, and they have important implications for MS pathogenesis.
INTRODUCTION
Multiple sclerosis (MS) is a human autoimmune demyelinating disease of the central nervous system (CNS) which is thought to be mediated by autoreactive T cells. Both genetic and environmental factors, such as infections, have been implicated in MS susceptibility (8, 14, 15, 29), and virus infections can induce demyelinating diseases (7, 26, 32, 34). Interestingly, both healthy individuals and MS patients possess peripheral CD4+ T cells specific for myelin antigens (25, 30). Nonetheless, the mechanism of autoreactive T-cell activation in MS remains unknown. One proposed mechanism of activation is molecular mimicry, whereby autoreactive T cells can be activated by infectious agents encoding epitopes with similar structures or sequence homology to self-tissue epitopes (10, 21). T-cell mimicry between myelin and viral peptides has recently been described for MS patients (33, 36). Molecular mimicry has also been implicated in the pathogenesis of diabetes, systemic lupus erythematosis, and human T-lymphotropic virus type 1-associated myelopathy/tropical spastic paraparesis (1, 11, 16, 19). MS patients present with multiple clinical forms, including relapsing-remitting and chronic-progressive forms of disease. Although the mechanism(s) involved in the initiation and in the relapse phase of relapsing-remitting MS is currently not fully understood, several factors, including viral infection, may be important (2, 27, 28).
We previously demonstrated (20) that a nonpathogenic variant of Theiler's murine encephalomyelitis virus (TMEV) expressing either the mouse myelin proteolipid protein (PLP) epitope, PLP139-151, or a core mimic sequence of PLP139-151 derived from Haemophilus influenzae (HI574-586) (3) induced a mild, slowly progressing form of CNS demyelinating disease. Clinical disease was associated with the cross-activation and Th1 differentiation of myelin epitope PLP139-151-specific CD4+ T cells, and the disease induced by the virus encoding the self PLP139-151 epitope could be prevented by tolerance to the homologous self peptide (22).
The present studies were designed to further understand the mechanisms of infection-induced molecular mimicry in both the initiation and progression of CNS autoimmunity. The present results extend our previous studies by illustrating four additional critical observations regarding the role of infection-induced molecular mimicry in activating CNS demyelinating disease. We demonstrate (i) that the CNS disease induced by infection with the core H. influenzae mimic-expressing virus is due to the cross-activation of autoreactive T cells, as peripheral tolerance induced to the self myelin PLP139-151 epitope specifically inhibited disease induction; (ii) that the mimic H. influenzae epitope can be naturally processed from the flanking sequences in the native H. influenzae protein and presented to cross-activate PLP139-151-specific autoreactive CD4+ Th1 cells; (iii) that molecular mimicry-induced disease initiation requires virus-activated innate immune signals, since infection with TMEV expressing the H. influenzae mimic induced disease, while the core H. influenzae mimic epitope administered in complete Freund's adjuvant (CFA) did not; and (iv) that a slowly progressing CNS autoimmune disease can be significantly exacerbated upon reinfection with the H. influenzae mimic-expressing virus. Collectively, these studies demonstrate that autoreactive T cells triggered by infection-induced molecular mimicry can both initiate and exacerbate clinical autoimmune demyelination and have important implications for the initiation and relapse phases of MS pathogenesis.
MATERIALS AND METHODS
Mice. Five- to 6-week-old female SJL mice were obtained from Harlan Sprague-Dawley (Indianapolis, Ind.). Mice were housed under barrier conditions at the National Institutes of Health-approved Northwestern University Medical School animal facilities. All protocols were approved by the Northwestern University Animal Care and Use Committee. Paralyzed mice were afforded easier access to food and water.
Construction of mimic-expressing TMEV. The cDNA encoding the BeAn strain of TMEV was modified by inserting ClaI sites and deleting 23 amino acids, as previously described (20). Genomic DNA containing the sequence coding for Haemophilus influenzae serine protease IV was obtained from the American Type Culture Collection (Manassas, Va.). ClaI sites were introduced by PCR to the H. influenzae DNA at either end of a coding sequence for 39 amino acids, HI566-604, which contained the PLP139-151 mimic sequence, HI574-586, by using the following primers: HI39A, 5' GCA CCT TAA TCG ATT TGA GCG 3'; and HI39AS, 5' CTG ACC ATC GAT TTT AGG ATC 3'. Following an enzyme restriction cut with ClaI, the 39-amino-acid-encoding piece containing the HI566-604 sequence was inserted into the ClaI site in the Cla-BeAn virus cDNA. This was designated HI39-BeAn cDNA. Infectious virus was produced as previously described (20). The PLP-BeAn, OVA-BeAn, and HI-BeAn viruses were produced as previously described (20).
Infection of SJL mice with TMEV. Mice (n = 8) were infected by the intracerebral injection of 3 x 106 PFU of either wild-type TMEV (strain BeAn 8386), HI-BeAn, PLP139-BeAn, OVA-BeAn, or Cla-BeAn and were scored at weekly intervals on a clinical scale of 0 to 5, as follows: 0, asymptomatic; 1, mild waddling gait; 2, severe waddling gait; 3, paralysis of one hindlimb; 4, total hindlimb paralysis, severe dehydration, and/or malnutrition; and 5, death. The data were plotted as mean clinical scores for each group of animals.
Peptides. The proteolipid protein (PLP) peptide PLP139-151 (HSLGKWLGHPDKF), the TMEV capsid peptide VP270-86 (WTTSQEAFSHIRIPLPH), the ovalbumin peptide OVA323-339 (ISQAVHAAHAEINEAGR), the H. influenzae peptide HI574-586 (EQLVKWLGLPAPI), the H. influenzae 30-mer peptide HI566-595 (SDTKKGAQEQLVKWLGLPAPIQKLQKELNI), and the PLP 30-mer peptide PLP130-159 (QAHSLERVCHCLGKWLGHPDKFVGITYALT) (Fig. 1a) were purchased from Peptides International (Louisville, Ky.). Their amino acid compositions were verified by mass spectrometry, and their purity was assessed by high-performance liquid chromatography.
Induction of peripheral tolerance. Tolerance was induced by the intravenous injection of 5 x 107 peptide-pulsed, 1-ethyl-3-(3-dimethylaminopropyl) carbodiimide (ECDI)-fixed syngeneic splenocytes, as described previously (13, 18, 22).
Delayed-type hypersensitivity responses. Delayed-type hypersensitivity (DTH) responses were elicited by injecting mice subcutaneously in alternate ears with 10 μg of the challenge peptides after measurements of ear thickness with a Mitutoyo model 7326 engineer's micrometer (Schlesinger's Tools, Brooklyn, N.Y.). Twenty-four hours following the peptide challenge, the ears were measured again, and differences in ear thickness (swelling) from the prechallenge thickness were expressed in units of 10–4 inches (mean ± standard errors of the means), as previously described (20).
T-cell proliferation and cytokine analysis. Spleens were removed from infected mice that had not been previously used for DTH responses (n = 2) at the indicated times following infection or immunization. T-cell proliferation and cytokine analyses were performed as described previously (20). Proliferation was determined for triplicate wells for each peptide concentration and then expressed in counts per minute. For gamma interferon (IFN-) cytokine analysis, a duplicate set of proliferation wells was used to collect supernatants at 48 and 72 h, and cytokine concentrations were determined by an enzyme-linked immunosorbent assay.
Immunohistochemistry. Mice (n = 3 per group) were anesthetized and perfused with phosphate-buffered saline (PBS) on the indicated days postinfection. Spinal cords and brains were removed by dissection, and 2- to 3-mm spinal cord blocks and brains were immediately frozen in OCT (Miles Laboratories, Elkhart, Ind.) in liquid nitrogen. Multiple 6-μm-thick cross sections from the spinal cord or brain were cut and mounted on Superfrost Plus electrostatically charged slides (Fisher, Pittsburgh, Pa.), air dried, and stored at –80°C. CNS immunohistochemistry was performed as previously described (12). The results shown are representative of the cellular infiltrates from multiple sections cut from three individual mice per group. Photomicrographs of immunostained sections from a spinal cord, cerebellum, and brain stem representative of the cellular infiltrates in the different groups were used to quantify the numbers of positive inflammatory cells per area of each tissue section. Data from photomicrographs were stored as 8-bit binary images in a grayscale format. Quantification was determined with ImageJ software v1.32j (http://rsb.info.nih.gov/ij/). Threshold values of brightness and contrast were determined for the photomicrographs and were kept constant for each sample. Prior to analysis, the minimum and maximum sizes of pixels to be counted were determined whereby sections with no positive staining gave a measurement of 0.0 pixels. The data are presented as percentages of positive pixels per area of photomicrograph.
Statistics. Clinical severity results are presented as mean group clinical scores, and the statistical difference was calculated by the Mann-Whitney nonparametric ranking test. Statistical analyses of DTH and proliferative responses and IFN- secretion were performed by using the two-tailed Student t test.
RESULTS
Peripheral tolerance to PLP139-151 inhibits HI-BeAn virus-induced autoimmune demyelinating disease. As we have previously reported (20), mice infected intracerebrally with a nonpathogenic variant of the BeAn strain of TMEV (ClaI-BeAn, which carries a 23-amino-acid deletion in the viral leader sequence) engineered to express a 30-mer peptide encompassing the encephalitogenic myelin epitope PLP139-151 (PLP-BeAn) develop a severe, rapid-onset, progressive demyelinating disease, while infection with BeAn containing the PLP139-151 epitope mimic, consisting of amino acids 574 to 586 of the Haemophilus influenzae serine protease IV protein (HI-BeAn), develop a less severe nonprogressive clinical disease (Fig. 1b). HI-BeAn induces a cross-reactive PLP139-151-specific CD4+ T-cell response, as determined by in vivo DTH and in vitro T-cell proliferative assays (Fig. 1c and d), and results in the infiltration of CD4+ T cells and activated F4/80+ macrophages/microglia into the CNS (see Fig. 5d and l) and associated focal areas of demyelination, as assessed by staining with an anti-PLP antibody (manuscript in preparation), similar to the lesions found in PLP-BeAn-infected SJL mice (20). To definitively determine whether cross-reactivity (mimicry) between PLP139-151 and HI574-586 is critical for the induction of demyelinating disease, we exposed HI-BeAn- or PLP-BeAn-infected mice (day 0) to the myelin epitope, PLP139-151, or the irrelevant OVA323-339 epitope (on days –7 and +3) to induce tolerance. As we previously reported (22), PLP139-151 tolerance inhibited clinical disease in mice infected with PLP-BeAn. Significantly, the present results show that PLP139-151 tolerance also specifically inhibits the CNS disease, as assessed by the TMEV scoring system, characterized by a mild to moderate waddling gait induced by infection with the mimic-expressing HI-BeAn virus compared to the case for mice exposed to OVA323-339 (P < 0.05) (Fig. 1b). PLP139-151-tolerant mice infected with either HI-BeAn or PLP-BeAn displayed significantly reduced HI574-586- and PLP139-151-specific DTH and CD4+ T-cell proliferative responses (P < 0.05) than those of the controls (Fig. 1c and d). The tolerance was specific to the PLP139-151 epitope, as TMEV-specific VP270-86-specific T-cell responses were unaffected in PLP139-151-tolerant mice, and the responses in OVA323-339-tolerant mice were similar to those in untreated mice. Interestingly, tolerance induced with HI574-586, as well as that induced with PLP139-151, was shown to inhibit experimental autoimmune encephalomyelitis (EAE) induced by priming with PLP139-151/CFA (Fig. 1e), illustrating the bidirectional cross-reactivity of these epitopes. These results thus provide definitive evidence that the early-onset CNS disease induced by infection with the core H. influenzae mimic-expressing virus is due to the cross-activation of PLP139-151-specific autoreactive CD4+ T cells, i.e., the data prove the occurrence of infection-induced molecular mimicry.
The HI574-586 mimic can be processed from its native protein sequence and induces demyelinating disease only in the context of virus infection. The data in Fig. 1 demonstrate that TMEV engineered to express the HI574-586 myelin mimic within the core of a 30-amino-acid sequence spanning PLP130-159 (HI-BeAn) induces the in vivo activation of cross-reactive PLP139-151-specific CD4+ T cells. To determine if the HI574-586 mimic epitope can be processed from its native flanking sequences within the mimic protein, we constructed a TMEV virus (HI39-BeAn) expressing a 39-amino-acid (HI566-604) piece of the H. influenzae protease IV protein. SJL mice infected with HI39-BeAn (Fig. 2a) developed an early-onset demyelinating disease which was less severe than that induced by PLP-BeAn but equivalent to that induced by HI-BeAn, while mice infected with TMEV expressing a 30-mer peptide encompassing the nonself OVA323-339 sequence developed a late-onset disease typical of infections with the wild-type BeAn TMEV strain (20, 22). We have previously described that the insertion of base pairs encoding the OVA317-346 30-mer peptide in either the forward or reverse direction restores the ability of the parental Cla-BeAn virus to persist in the CNS and to induce late-onset clinical disease (18). Although the molecular basis for this is unknown, it is possible that the addition of the OVA peptide coding sequence to the leader restores its original length and induces a conformational change in Cla-BeAn, which allows more efficient in vivo replication and virus persistence in mouse CNS cells. HI-BeAn- and HI39-BeAn-infected mice displayed strong CD4+ DTH and proliferative T-cell responses associated with potent IFN- secretion, which is typical of Th1 responses, upon recall with either HI574-586 or the PLP139-151 self epitope (Fig. 2b to d). The level of VP270-86-induced IFN- secretion seen in PLP139-BeAn-infected mice compared to that in mice infected with the other recombinant viruses was less than twofold higher and not consistently seen and thus was not considered significant. Thus, HI574-586 is arguably a natural PLP139-151 mimic capable of being processed from its native sequence following virus infection to activate cross-reactive PLP139-151-specific autoreactive CD4+ T cells.
To determine if the HI mimic epitope could also induce clinical disease in the context of innate immune signals induced by complete Freund's adjuvant (CFA), we immunized SJL mice with either the minimal HI574-586 mimic epitope or a 30-amino-acid peptide (HI566-595) encompassing the core HI574-586 mimic epitope in CFA. Immunization with either peptide activated both HI574-586 mimic-specific and cross-reactive self-myelin PLP139-151-specific CD4+ T cells (Fig. 3c). However, unlike the case for SJL mice infected with TMEV engineered to express either the core HI574-586 mimic epitope flanked by the native PLP sequences (HI-BeAn) or the long HI566-604 epitope (HI39-BeAn) (Fig. 2a), priming with either HI574-586 or HI566-595 in CFA failed to induce a clinical demyelinating disease (Fig. 3a). This was also true if pertussis toxin and/or lipopolysaccharide was used in conjunction with CFA as an additional innate immune stimulus (data not shown). Disease induction following HI-BeAn infection appeared to induce more potent innate immune signals than priming with the mimic peptide/CFA, as exemplified by the failure of the self PLP139-151 peptide to recall either in vivo DTH responses or in vitro IFN- production (Fig. 3b and d). In contrast, the self PLP139-151 peptide recalled potent DTH and IFN- production in mice infected with either HI-BeAn or HI39-BeAn (Fig. 2b and d). Thus, the induction of autoimmunity via molecular mimicry requires critical innate immune signals specific to the infectious agent encoding the molecular mimic to induce an efficient differentiation of cross-reactive autoreactive Th1 cells with pathological potential.
Secondary virus infection in the CNS exacerbates clinical disease in mice previously infected with HI-BeAn. Mice infected with the HI-BeAn mimic virus consistently developed a less severe clinical demyelinating disease than mice infected with PLP-BeAn (Fig. 2a) (20). To determine if subsequent virus infections could exacerbate the nonprogressive demyelinating disease induced by HI-BeAn infection, we infected SJL mice intracerebrally with HI-BeAn (day 0) and then reinfected them on day +14 with HI-BeAn, OVA-BeAn, the nonpathogenic parental virus (Cla-BeAn), or an equivalent volume of PBS. A primary infection with HI-BeAn and subsequent reinfection with HI-BeAn induced a significantly more severe clinical disease (P < 0.05) than that in mice given PBS on day +14 (Fig. 4a). Interestingly, mice infected with HI-BeAn that received a second infection with OVA-BeAn also developed an exacerbated clinical disease compared with that in PBS-injected controls (P < 0.05) (Fig. 4a). The reinfection of HI-BeAn virus-infected mice with the parental, nonpersisting Cla-BeAn virus resulted in a delayed onset of clinical disease compared with that in controls, and these mice developed a delayed exacerbated disease as well (Fig. 4a). However, as observed in multiple experiments, the clinical disease induced upon reinfection with the PLP139-151 mimic-expressing HI-BeAn virus always had an earlier onset and was significantly more severe than the disease in mice reinfected with OVA-BeAn or Cla-BeAn. Mice infected with the HI-BeAn virus followed by a second infection with OVA-BeAn, Cla-BeAn, or HI-BeAn developed HI574-586-specific CD4+ T-cell proliferative responses as well as cross-reactive PLP139-151-specific CD4+ T-cell responses (Fig. 4d and e) (P < 0.05). All mice initially infected with HI-BeAn developed virus-specific CD4+ T-cell responses which increased slightly upon secondary infection (Fig. 4c). The CD4+ T-cell responses to the virus, myelin, and the mimic peptide in these infected mice were associated with IFN--specific secretion, which is indicative of Th1-type CD4+ T-cell responses (Fig. 4f). Thus, a primary infection with a mimic-encoding virus may result in a suboptimal activation of autoreactive T cells, resulting in a mild initial clinical disease, but secondary infections with neurotropic viruses expressing the PLP139-151 mimic epitope more efficiently enhance the autoimmune response, leading to a severe clinical disease compared to that induced by non-mimic-expressing neurotropic viruses.
To further determine whether a prior infection of the CNS with a neurotropic virus (OVA-BeAn) that induces a late-onset, slowly progressing disease (Fig. 2a) could predispose mice to a more severe autoimmune disease upon secondary infection with a mimic-encoding virus, we infected mice with OVA-BeAn (day 0) and then reinfected them with HI-BeAn, OVA-BeAn, Cla-BeAn, or an equivalent volume of PBS on day +14. Mice infected with OVA-BeAn that received a second infection with HI-BeAn developed a disease with a significantly earlier onset than that in mice that received a second infection with OVA-BeAn, Cla-BeAn, or PBS (P < 0.05) (Fig. 4b). OVA-BeAn virus-infected mice that received a second infection with HI-BeAn or OVA-BeAn developed a slightly more severe disease than control mice throughout the disease course. Mice infected with the OVA-BeAn virus and reinfected with the HI-BeAn virus developed HI574-586-specific CD4+ T-cell responses and had increased PLP139-151-specific CD4+ T-cell responses compared to those of control mice (Fig. 4d and e). All OVA-BeAn virus-infected mice had virus-specific CD4+ T-cell responses, although interestingly, mice reinfected with the non-CNS-persisting Cla-BeAn virus exhibited reduced reactivities to all challenge peptides (Fig. 4c to f).
Brains and spinal cords were removed from mice (n = 3) at 60 days postinfection, and frozen sections were analyzed for their levels of CD4+ T-cell and macrophage infiltrates to assess the effects of secondary virus infections on the levels of CNS inflammatory cell infiltration. Lumbar spinal cords from HI-BeAn virus-infected mice that were reinfected with HI-BeAn displayed significantly more CD4+ T cells and F4/80+ microglia/macrophages (Fig. 5a and i) than did PBS-treated control mice (Fig. 5d and l). Mice initially infected with HI-BeAn and reinfected with either OVA-BeAn or Cla-BeAn had more CD4+ T cells (Fig. 5b and c) and macrophages (Fig. 5j and k) than did PBS-treated controls (Fig. 5d and l) but fewer of these cells than did mice reinfected with HI-BeAn (Fig. 5a and i). Lumbar spinal cords from mice infected with the OVA-BeAn virus and subsequently reinfected with HI-BeAn had increased CD4+ T cells and macrophages (Fig. 5e and m) compared to control mice (Fig. 5h and p). Cerebellum and brain stem sections from mice initially infected with HI-BeAn and reinfected with HI-BeAn displayed large numbers of infiltrating CD4+ T cells and F4/80+ cells (Fig. 5q to t). However, OVA-BeAn-infected mice that were reinfected with HI-BeAn displayed limited immune cell infiltrates in the brain (Fig. 5u to x). Overall, the levels of immune cell infiltration in the CNS correlated well with the severity of clinical disease observed for mice that received secondary infections.
DISCUSSION
Molecular mimicry is a putative mechanism for the activation of autoreactive T cells in MS, a human autoimmune demyelinating disease. However, little direct evidence exists for mimicry-induced T-cell autoimmunity using infectious agents. Previously, we described a model of molecular mimicry whereby the viral delivery of a PLP139-151 myelin-mimic epitope naturally encoded in the protease IV protein of H. influenzae, HI574-586, could induce an early-onset autoimmune demyelinating disease associated with the activation of autoreactive PLP139-151-specific CD4+ Th1 cells (20). The present experiments build upon this initial observation and illustrate four critically important conditions required for the induction and/or exacerbation of autoimmune disease by infection-induced molecular mimicry.
Firstly, the current demonstration that PLP139-151 tolerance (13, 18) significantly inhibited the incidence and severity of disease following HI-BeAn infection and specifically inhibited both HI574-587- and PLP139-151-specific CD4+ T-cell responses definitively shows the functional activation of pathological cross-reactive HI574-586 and PLP139-151 CD4+ T cells following HI-BeAn infection. Secondly, the demonstration that infection with the HI39-BeAn virus induced a similar early-onset demyelinating disease and similar activation of PLP139-151-specific CD4+ Th1 cells, as observed following infection with the HI-BeAn virus, shows that the H. influenzae mimic sequence is a natural epitope which can be processed and presented in vivo by SJL antigen-presenting cells to activate PLP139-151-specific autoreactive Th1 cells. We feel that this is a significant advance over past descriptions of pathological mimic epitopes, which uniformly employed minimal mimic sequences identified in T-cell hybridoma screens for their disease-inducing abilities and did not take into account the possibility that the mimic peptide may not be capable of being processed from its surrounding amino acid residues in the native mimic protein.
Thirdly, our data showing an induction of clinical CNS disease and autoreactive PLP139-151-specific DTH and IFN- Th1 responses following infection with a TMEV engineered to express either the core HI574-586 mimic epitope flanked by the native PLP sequences (HI-BeAn) or the long HI566-604 epitope (HI39-BeAn) (Fig. 2), but not upon priming with HI-BeAn or HI39-BeAn in CFA (Fig. 3), demonstrate that an equally important component required for the infection-induced initiation of autoimmune disease via molecular mimicry is the ability of the pathogen to trigger an appropriate innate immune response leading to an efficient expansion and differentiation of pathological autoreactive Th1 responses. This finding is consistent with our demonstration that a TMEV infection of quiescent microglial cells leads to the induction of a functional antigen presentation function concomitant with a significant upregulation of major histocompatibility complex and costimulatory molecules and an increased expression of cytokines (interleukin-12 and interleukin-23) required for Th1 differentiation (23). Furthermore, we have more recently shown that TMEV infection is a more potent stimulus to resting microglia than is the ligation of other Toll-like receptors (TLRs). Compared to the activation of TLR2 via peptidoglycan, TLR3 alone via poly(I-C), TLR4 via bacterial lipopolysaccharide, or TLR9 via CpG DNA, the infection of microglia with viable TMEV caused a more potent upregulation of innate and effector immune cytokines, chemokines, and major histocompatibility complex class II and costimulatory molecules, enabling the microglia to more efficiently present myelin antigens to CD4+ Th1 cells (24). Preliminary unpublished data indicate that TMEV is also tropic for dendritic cells, and thus it is possible that TMEV expressing the Haemophilus-encoded PLP139-151 mimic causes a more efficient activation/maturation of dendritic cells than does CFA, resulting in an enhanced activation of autoreactive Th1 cells. In addition to the more efficient induction of peripheral cross-reactive PLP139-151-specific Th1 responses by HI-BeAn infection than by HI574-586/CFA priming, the fact that TMEV can persist and replicate in the CNS, leading to the local release of inflammatory cytokines and to possible deleterious effects on blood-brain barrier integrity, may also contribute to the efficient disease induction observed following infection with the mimic-expressing virus versus that with peptide priming.
Lastly, our results demonstrate that a secondary infection with a myelin mimic-expressing neurotropic virus can exacerbate a mild preexisting CNS disease, indicating that mimicry may also trigger disease relapses in autoimmune disease. Various environmental factors have been associated with initiating MS relapses, including viral and bacterial infections (2, 9, 27, 28, 31). The infection of SJL mice with the HI-BeAn mimic virus induces a mild, slowly progressive form of clinical disease. The nonprogressive disease induced by HI-BeAn may be due to the failure of the virus to fully activate the self-reactive PLP139-151-specific T-cell population (HI574-586 has a 50% infective concentration of 385 nM for I-As, while that of PLP139-151 is 87 nM) (3), or the mimic peptide may only activate a subset of the self-reactive population (6). Significantly, mice infected with a second dose of HI-BeAn developed a severe, progressive clinical disease, consistent with the repriming of HI574-586/PLP139-151-specific CD4+ T cells. We have also recently shown that the reinfection of prediabetic mice with a virus expressing an H-2Kb-restricted mimic ligand to a lymphocytic choriomeningitis virus (self) epitope transgenically expressed on pancreatic ?-cells can accelerate the development of diabetes, but this mimic ligand-expressing arenavirus, unlike HI-BeAn, is not capable of initiating autoimmune disease (4). Interestingly, HI-BeAn-infected mice reinfected with TMEV encoding an irrelevant OVA epitope also developed a more severe clinical disease, indicating that bystander activation may also restimulate autoreactive T cells that are initially activated by HI-BeAn virus infection. This is significant, as anecdotal evidence suggests that relapse episodes in MS can be triggered by various upper respiratory virus infections (2, 9). The present study also advances previous findings in which mice primed with a plasmid coding for PLP could render mice susceptible to a mild disease following a subsequent rechallenge with CFA (35). This highlights the difficulties faced when investigating the role of viruses in MS pathogenesis, as relapses may be induced by otherwise innocuous or irrelevant viral infections.
Significantly, an initial infection of the CNS with the OVA-BeAn virus followed by a secondary infection with the HI-BeAn virus slightly accelerated the clinical disease. However, reinfection with the HI-BeAn virus did not increase the severity of autoimmune demyelinating disease as significantly as when the initial infection was with the HI-BeAn virus. Although the proliferative responses to PLP139-151 were similar between the OVA-BeAn/HI-BeAn- and HI-BeAn/HI-BeAn-infected groups, the clinical disease was more severe for the HI-BeAn/HI-BeAn-infected group (Fig. 4a). Importantly, we observed that IFN- secretion, assessed by a PLP139-151 rechallenge in vitro, was significantly higher for the HI-BeAn/HI-BeAn-infected group (Fig. 4f), indicating that the enhanced disease severity correlates with the ability of the H. influenzae mimic peptide to efficiently cross-activate pathogenic PLP139-151 Th1 cells, which are essential to the initiation and progression of CNS autoimmune disease. Thus, this suggests that molecular mimicry boosts the autoreactive Th1 cell population, resulting in a more severe autoimmune demyelinating disease. Therefore, a non-mimic-encoding pathogen may only actively participate in disease pathogenesis once autoimmune cells are activated, whether by molecular mimicry or other mechanisms, e.g., epitope spreading (17). Further evidence to support this conclusion can be drawn from the observation that preinfection with a persistent virus which expressed a nonself, nonmimic epitope (OVA-BeAn) or a transient infection with Cla-BeAn did not predispose mice to exacerbated autoimmune disease upon HI574-586 immunization (5). However, mice preinfected with HI-BeAn and subsequently immunized with the H. influenzae peptide showed an exacerbated disease compared to that after H. influenzae immunization or HI-BeAn infection alone. Furthermore, infection with HI-BeAn and immunization with an irrelevant peptide (OVA) did not exacerbate the disease. Therefore, it appears that one requirement for autoimmune disease is an infection of the target organ, i.e., the CNS, with the mimic-expressing virus.
Interestingly, there were significant differences in the numbers of inflammatory cells present in the spinal cord and brain in mice receiving secondary infections. With mice infected with HI-BeAn/HI-BeAn, we observed a significant number of CD4+ T cells and macrophages present in the spinal cord, consistent with the severe clinical disease. Furthermore, the presence of inflammatory infiltrates in the cerebellum and brain stem was only observed in mice from the HI-BeAn/HI-BeAn-infected group with severe clinical disease. Therefore, the increased clinical severity may have been due to damage in these regions of the CNS. The cerebellum is involved in the coordination of voluntary motor movement, balance, equilibrium, and muscle tone. Therefore, inflammation in these structures is likely to induce a further loss of coordination of motor movement (asynergia), movement tremors (intention tremor), staggering, wide-based walking (ataxic gait), and weak muscles (hypotonia). Overall, mice that received secondary infections had a more severe disease which correlated with more infiltrating cells in the brain and spinal cord.
Our results suggest that molecular mimicry could be an important factor in the pathogenesis of MS by expanding a population of T cells that recognize self-epitopes. These autoreactive cells may then be further expanded to different degrees by differing stimuli, including reinfections with viruses or bacteria, which may or may not encode epitope mimics of relevant autoantigens via direct and/or bystander mechanisms. The potential requirements for target organ infection and persistence of the pathogen in the initiation of autoimmune disease by infection-induced molecular mimicry require further study. Furthermore, while the identification of potential mimic epitopes from pathogens by the use of computer searches is useful, it is important to determine whether the epitopes are naturally processed and presented in the infected host and if they can stimulate pathological autoimmune responses in the context of a variety of innate immune stimuli. Although the recombinant Theiler's virus model of molecular mimicry allows the determination of various conditions required for the induction of CNS autoimmunity, we are currently testing whether the infection of SJL mice with viable wild-type H. influenzae versus a sppA-deficient mutant (lacking the protease IV protein, which contains the PLP139-151 mimic epitope) induces PLP139-151-specific T-cell responses and/or clinical disease as the definitive test of infection-induced molecular mimicry. Collectively, these studies present compelling evidence that virus-induced molecular mimicry can both induce and enhance Th1-mediated CNS autoimmune disease in a model of MS, which may have important implications for the initiation and relapse phases of MS pathogenesis.
ACKNOWLEDGMENTS
This work was supported in part by the U.S. Public Health Service, by NIH grant NS-23349, and by National Multiple Sclerosis Society (NMSS) research grant RG-3166-A-4. J. L. Croxford and J. K. Olson were supported by NMSS postdoctoral fellowships (FG-1456-A-1 and FA-1517-A-1, respectively).
These authors contributed equally to this work.
REFERENCES
Albert, L. J., and R. D. Inman. 1999. Molecular mimicry and autoimmunity. N. Engl. J. Med. 341:2068-2074.
Andersen, O., P. E. Lygner, T. Bergstrom, M. Andersson, and A. Vahlne. 1993. Viral infections trigger multiple sclerosis relapses: a prospective seroepidemiological study. J. Neurol. 240:417-422.
Carrizosa, A. M., L. B. Nicholson, M. Farzan, S. Southwood, A. Sette, R. A. Sobel, and V. K. Kuchroo. 1998. Expansion by self antigen is necessary for the induction of experimental autoimmune encephalomyelitis by T cells primed with a cross-reactive environmental antigen. J. Immunol. 161:3307-3314.
Christen, U., K. H. Edelmann, D. B. McGavern, T. Wolfe, B. Coon, M. K. Teague, S. D. Miller, M. B. Oldstone, and M. G. von Herrath. 2004. A viral epitope that mimics a self antigen can accelerate but not initiate autoimmune diabetes. J. Clin. Investig. 114:1290-1298.
Croxford, J. L., H. A. Anger, and S. D. Miller. 2005. Viral delivery of an epitope from Haemophilus infuenzae induces central nervous system autoimmune disease by molecular mimicry. J. Immunol. 174:907-917.
Croxford, J. L., J. K. Olson, and S. D. Miller. 2002. Epitope spreading and molecular mimicry as triggers of autoimmunity in the Theiler's virus induced demyelinating disease model of multiple sclerosis. Autoimmun. Rev. 1:251-260.
Dal Canto, M. C., and H. L. Lipton. 1975. Primary demyelination in Theiler's virus infection. An ultrastructural study. Lab. Investig. 33:626-637.
Ebers, G. C., A. D. Sadovnick, and N. J. Risch. 1995. A genetic basis for familial aggregation in multiple sclerosis. Canadian Collaborative Study Group. Nature 377:150-151.
Edwards, S., M. Zvartau, H. Clarke, W. Irving, and L. D. Blumhardt. 1998. Clinical relapses and disease activity on magnetic resonance imaging associated with viral upper respiratory tract infections in multiple sclerosis. J. Neurol. Neurosurg. Psychiatry 64:736-741.
Fujinami, R. S., and M. B. Oldstone. 1985. Amino acid homology between the encephalitogenic site of myelin basic protein and virus: mechanism for autoimmunity. Science 230:1043-1045.
Gran, B., B. Hemmer, M. Vergelli, H. F. McFarland, and R. Martin. 1999. Molecular mimicry and multiple sclerosis: degenerate T-cell recognition and the induction of autoimmunity. Ann. Neurol. 45:559-567.
Katz-Levy, Y., K. L. Neville, J. Padilla, S. M. Rahbe, W. S. Begolka, A. M. Girvin, J. K. Olson, C. L. Vanderlugt, and S. D. Miller. 2000. Temporal development of autoreactive Th1 responses and endogenous antigen presentation of self myelin epitopes by CNS-resident APCs in Theiler's virus-infected mice. J. Immunol. 165:5304-5314.
Kennedy, M. K., L. J. Tan, M. C. Dal Canto, V. K. Tuohy, Z. J. Lu, J. L. Trotter, and S. D. Miller. 1990. Inhibition of murine relapsing experimental autoimmune encephalomyelitis by immune tolerance to proteolipid protein and its encephalitogenic peptides. J. Immunol. 144:909-915.
Kurtzke, J. F. 1993. Epidemiologic evidence for multiple sclerosis as an infection. Clin. Microbiol. Rev. 6:382-427.
Kurtzke, J. F., and K. Hyllested. 1979. Multiple sclerosis in the Faroe Islands. I. Clinical and epidemiological features. Ann. Neurol. 5:6-21.
Levin, M. C., S. M. Lee, F. Kalume, Y. Morcos, F. C. Dohan, Jr., K. A. Hasty, J. C. Callaway, J. Zunt, D. Desiderio, and J. M. Stuart. 2002. Autoimmunity due to molecular mimicry as a cause of neurological disease. Nat. Med. 8:509-513.
Miller, S. D., C. L. Vanderlugt, W. S. Begolka, W. Pao, R. L. Yauch, K. L. Neville, Y. Katz-Levy, A. Carrizosa, and B. S. Kim. 1997. Persistent infection with Theiler's virus leads to CNS autoimmunity via epitope spreading. Nat. Med. 3:1133-1136.
Miller, S. D., R. P. Wetzig, and H. N. Claman. 1979. The induction of cell-mediated immunity and tolerance with protein antigens coupled to syngeneic lymphoid cells. J. Exp. Med. 149:758-773.
Oldstone, M. B. 1998. Molecular mimicry and immune-mediated diseases. FASEB J. 12:1255-1265.
Olson, J. K., J. L. Croxford, M. Calenoff, M. C. Dal Canto, and S. D. Miller. 2001. A virus-induced molecular mimicry model of multiple sclerosis. J. Clin. Investig. 108:311-318.
Olson, J. K., J. L. Croxford, and S. D. Miller. 2001. Virus-induced autoimmunity: potential role of viruses in initiation, perpetuation, and progression of T cell-mediated autoimmune diseases. Viral Immunol. 14:227-250.
Olson, J. K., T. N. Eagar, and S. D. Miller. 2002. Functional activation of myelin-specific T cells by virus-induced molecular mimicry. J. Immunol. 169:2719-2726.
Olson, J. K., A. M. Girvin, and S. D. Miller. 2001. Direct activation of innate and antigen presenting functions of microglia following infection with Theiler's virus. J. Virol. 75:9780-9789.
Olson, J. K., and S. D. Miller. 2004. Microglia initiate central nervous system innate and adaptive immune responses through multiple TLRs. J. Immunol. 173:3916-3924.
Olsson, T., J. Sun, J. Hillert, B. Hojeberg, H. P. Ekre, G. Andersson, O. Olerup, and H. Link. 1992. Increased numbers of T cells recognizing multiple myelin basic protein epitopes in multiple sclerosis. Eur. J. Immunol. 22:1083-1087.
Osame, M., K. Usuku, S. Izumo, N. Ijichi, H. Amitani, A. Igata, M. Matsumoto, and M. Tara. 1986. HTLV-I associated myelopathy, a new clinical entity. Lancet i:1031-1032.
Panitch, H. S. 1994. Influence of infection on exacerbations of multiple sclerosis. Ann. Neurol. 36(Suppl.):S25-S28.
Rapp, N. S., J. Gilroy, and A. M. Lerner. 1995. Role of bacterial infection in exacerbation of multiple sclerosis. Am. J. Phys. Med. Rehab. 74:415-418.
Sawcer, S., H. B. Jones, R. Feakes, J. Gray, N. Smaldon, J. Chataway, N. Robertson, D. Clayton, P. N. Goodfellow, and A. Compston. 1996. A genome screen in multiple sclerosis reveals susceptibility loci on chromosome 6p21 and 17q22. Nat. Genet. 13:464-468.
Scholz, C., K. T. Patton, D. E. Anderson, G. J. Freeman, and D. A. Hafler. 1998. Expansion of autoreactive T cells in multiple sclerosis is independent of exogenous B7 costimulation. J. Immunol. 160:1532-1538.
Sibley, W. A., C. R. Bamford, and K. Clark. 1985. Clinical viral infections and multiple sclerosis. Lancet i:1313-1315.
Smyth, J. M., B. J. Sheahan, and G. J. Atkins. 1990. Multiplication of virulent and demyelinating Semliki Forest virus in the mouse central nervous system: consequences in BALB/c and SJL mice. J. Gen. Virol. 71:2575-2583.
Tejada-Simon, M. V., Y. C. Zang, J. Hong, V. M. Rivera, and J. Z. Zhang. 2003. Cross-reactivity with myelin basic protein and human herpesvirus-6 in multiple sclerosis. Ann. Neurol. 53:189-197.
Tellez-Nagel, I., and D. H. Harter. 1966. Subacute sclerosing leukoencephalitis. I. Clinico-pathological, electron microscopic and virological observations. J. Neuropathol. Exp. Neurol. 25:560-581.
Theil, D. J., I. Tsunoda, F. Rodriguez, J. L. Whitton, and R. S. Fujinami. 2001. Viruses can silently prime for and trigger central nervous system autoimmune disease. J. Neurovirol. 7:220-227.
Wucherpfennig, K. W., and J. L. Strominger. 1995. Molecular mimicry in T cell-mediated autoimmunity: viral peptides activate human T cell clones specific for myelin basic protein. Cell 80:695-705.(J. Ludovic Croxford, Juli)