Hit-Hit and hit-Run: viruses in the playing field of multiple sclerosis

Hit-Hit and Hit-Run: Viruses in the Playing Field of Multiple Sclerosis

I.A. Scarisbrick, PhD, and M. Rodriguez, MD

AddressDepartments of Neurology and Immunology, Mayo Medical and Gradu-ate Schools, 428 Guggenheim Building, 200 First Street SW, Rochester, MN 55905, USA.E-mail: [email protected] Neurology and Neuroscience Reports 2003, 3:265–271Current Science Inc. ISSN 1528–4042Copyright © 2003 by Current Science Inc.

IntroductionMultiple sclerosis (MS) is a common chronic demyelinat-ing disease of unknown etiology that leads to irreversibledisability. Whereas there is limited agreement concerningpotential initiating event(s), it is agreed the disease processis probably immune mediated. The extent and nature ofcentral nervous system (CNS) inflammation varies andconsists of variable degrees of CD8+ and CD4+ T and Blymphocytes, macrophages, and pathogenic antibodies atthe leading edge of white matter destruction. Genetic,immunologic, and environmental factors, such as viruses,have each been considered as possible etiologic agents [1].

Fundamental questions concerning the pathophysiologyof MS remain unanswered. First, it is unclear what event orevents initiate CNS inflammation. Two primary theoriesexist: the autoimmune and the microbial. In the autoim-mune theory, CNS white matter is intact, with the inflamma-tory insult being mediated by a destructive immune responsethat is initiated in the periphery by autoreactive T cells. In themicrobial theory, demyelination is the consequence of oneor multiple infectious agents that persist in the CNS, causingmultiple hits ("Hit-Hit"), with direct or indirect damage tomyelin and/or oligodendrocytes. Alternatively, demyelina-tion may be caused by an autoreactive immune response, thebyproduct of a peripheral microbial or viral infection, a vir-tual "Hit-Run." There is both clinical and experimental evi-dence to indicate that infection can drive destructiveimmunity and demyelination via at least four mechanisms:antigen-specific reactivity, molecular mimicry, epitopespreading, and bystander demyelination.

We review recent epidemiologic, clinicopathophysiologic,and experimental evidence of the role that infection plays inMS, and what light this evidence sheds on the potentialmechanisms initiating and driving the disease process.

Epidemiologic StudiesEpidemiologic studies support the involvement of an envi-ronmental agent, possibly infectious, in the etiology of MS,on a background of host genetic factors. These include geo-graphic association of disease susceptibility with evidenceof MS clustering [2]. For example, the prevalence of MS islow around the equator, with prevalence increasing bothdirections, north or south, with migration studies to andfrom high-risk areas suggesting that timing of exposure iscritical to future MS susceptibility. Individuals movingbefore the of age 15 years acquire the risk of the area towhich they move, whereas individuals moving after thatage keep their original risk phenotype.

It is well accepted that genetic factors play an importantrole in the pathogenesis of MS, and indeed the lifetime riskof developing MS is higher in biologic relatives of MSpatients. For example, the concordance rate is much higheramong monozygotic compared with diazygotic twins, evenin twins raised separately [3]. Ethnic predisposition alsoimplicates genetic factors. For example, relative susceptibil-ity is greater in northern Eupopeans, as compared with therelative protection of Native Americans, despite each living

Viruses have been major players in the search for the cause of multiple sclerosis (MS). In support of the viral theory is the predominance of CD8+ T cells and class-I major histo-compatibility complex in lesions, the powerful therapeutic effects of β interferons, the ease of inducing demyelination in experimental models following virus challenge, and the documented examples of several human demyelinating dis-eases conclusively demonstrated to be of viral origin. We propose two hypotheses of how viruses may cause MS. In the "Hit-Hit" hypothesis, the virus persists or may be reac-tivated in the central nervous system (CNS). Injury is the result of direct viral damage and by an attempt of the immune response to clear the infectious agent. In the "Hit-Run" hypothesis, virus infects the periphery but never enters the CNS. The virus sets up an abnormal immuno-logic milieu for subsequent autoimmunity. In both scenar-ios, knowing the inciting virus would be expected to eliminate disease if the population were vaccinated to pre-vent infection. In the treatment of patients with fully estab-lished disease, the Hit-Hit hypothesis would require that antiviral agents enter the CNS and stop replication. In the case of the Hit-Run hypothesis, treatment of patients with established disease with antiviral agents would be futile.

266 Demyelinating Disorders

in areas of high MS prevalence. The low risk of conjugal MSfurther supports data suggesting that familial risk is geneti-cally determined and not sexually transmitted [4].

Although the genes that contribute to MS susceptibilityhave not been identified, genome-wide studies haverevealed that susceptibility to MS is linked to genes in themajor hitocompatibility complex (MHC) on chromosome6. Alleles of certain class-II genes, human leukocyte antigen(HLA) DR and HLA DQ, which are known to be involvedin antigen presentation, confer the strongest degree of MSpredisposition, and this has been consistent across manypopulations. In spite of these associations, the relative riskprovided by HLA is small, increasing by threefold com-pared with the general population. In addition to suscepti-bility, genetic factors may also govern disease course andseverity [5].

ClinicopathophysiologyAlthough it had been generally accepted that similarpathophysologic mechanisms were operative in all MSpatients, recent studies indicate heterogeneity at the levelof the MS lesion [6••]. Four fundamentally different pat-terns of demyelination in acute lesions, which differbetween, but not within a given patient, have been identi-fied. Lesions differ in the extent of cellular infiltrates,antibody deposition, demyelination and remyelination,the magnitude of complement activation, and the degreeof oligodendrocyte loss. Patterns I (macrophage-associ-ated demyelination) and II (antibody-mediated demyeli-nation) resemble experimental autoimmune models ofMS, in which the toxic products of activated macrophages(I) or demyelinating antibodies (II) lead to myelindestruction. Patterns III (distal-oligodendrogliopathy)and IV (extensive oligodendrocyte loss) mirror a viral-induced or metabolic disturbance, respectively, ratherthan autoimmunity. These differing types of lesions andbroad spectrum of structural changes may reflect distinctpathogenetic mechanisms involved in different patientsubgroups. Although specific correlates have not yet beenestablished, lesion heterogeneity may explain in part theheterogeneity of MS with respect to clinical presentationand response to therapy.

There are strong arguments in favor of antigen-specificmechanisms of immune-mediated demyelination in MS.First, two groups have recently reported extensive clonalexpansion of T cells, predominantly CD8+ T cells, withinMS lesions [7••,8]. The clonal expansion of antibody-secreting B cells in the CNS and cerebrospinal fluid (CSF)of MS patients has also been reported, indicative ofrepeated exposure to the same antigen [9]. Together, theseresults support the concept of a highly focused immuneresponse driving pathogenesis in MS. Because most class-I–restricted CD8+ T-cell responses are triggered byviruses, these data provide circumstantial evidence for theviral theory of MS.

Although MS has long been thought to be a CD4+ T-cell class-II–mediated autoimmune disease, there isstrong evidence that CD8+ T cells and class-I moleculesalso play an integral role, particularly in axonal damage.Importantly, within actively demyelinating MS lesions,MHC class-I–restricted CD8+ T cells outnumber CD4+ Tcells by roughly 10-fold [7••]. MHC class-I molecules areup-regulated on neural cells during most inflammatoryand degenerative CNS diseases, and may be regulated inan activity-dependent fashion. Additionally, in vitro stud-ies have demonstrated that CD8+ cytotoxic lymphocytes(CTLs) are capable of transecting neurites in an MHCclass-I peptide-dependent fashion [10]. This evidencesupports earlier reports that Theiler's murine encephalo-myelitis virus (TMEV) infected mice, which are deficientin MHC class-I, show preservation of axons despite exten-sive demyelination [11]. Furthermore, blocking CD8+CTLs using viral-specific peptides also indicates thatCD8+ cells contribute to neuronal injury, because treatedmice exhibited improved motor function, despite wide-spread demyelination [12•].

Viruses that Have Been Proven to Induce Demyelination in Humans (Hit-Hit)The concept that viral agents may initiate MS is supportedby clear evidence that other viral infections cause CNSinflammation and demyelination in humans (Table 1)[13]. Viral-induced demyelination is most clearly associ-ated with progressive multifocal leukoencephalopathy(PML), subacute sclerosing pancencephalitis (SSPE), andhuman T-lymphotropic virus type-1 (HTLV-1)–associatedmyelopathy/tropical spastic paraparesis (HAM/TSP).SSPE is an extremely rare, fatal, chronic progressivepanencephalitis occurring 5 to 10 years after acute mea-sles virus infection. PML is caused by a ubiquitous humanvirus acquired early in life. Papopavirus (JC) is also typi-cally fatal, but predominantly affects immunocompro-mised individuals. In both SSPE and PML, the incitingvirus has been identified in oligodendroglia of affectedpatients (examples of the Hit-Hit hypothesis). TSP iscaused by the strongly neurotropic retrovirus, HTLV-1,and has considerable clinical similarity to primary pro-gressive MS. Although over 20 million people worldwideare infected with HTLV-1, only a small percentagedevelop neurologic disease. As seen in primary progres-sive MS, there is a slowly progressive spastic paraparesis,lesions in magnetic resonance imaging (MRI) scans,increased CSF immunoglobulin, and oligoclonal band-ing. Some would argue that HAM/TSP is part of a hetero-geneous MS syndrome with a clearly defined pathogen.

The association of CSF immunglobulin G (IgG) oligo-clonal bands that persist in patients unchanged through-out the course of MS provides at least circumstantialevidence of an infectious agent. Oligoclonal bands areassociated with very few CNS diseases, and those that are

Hit-Hit and Hit-Run: Viruses in the Playing Field of Multiple Sclerosis • Scarisbrick and Rodriguez 267

have been shown to be both inflammatory and infectious[14]. It is envisioned that CNS infection occurs via theblood-brain barrier and is transmitted to CNS cells byperipheral blood mononuclear cells. Neurotropic virusescan persist in the CNS, establishing a CNS-restricted antivi-ral or autoimmune process. Candidate organisms may ini-tiate MS or trigger relapses in a subset of susceptibleindividuals, or may be reactivated due to the disease itself,without contributing to symptoms.

Candidate Multiple Sclerosis Pathogens: Hit-Hit HypothesisHerpesvirusesHerpesviruses, including herpes simplex viruses (HSV),Epstein-Barr virus (EBV), and human herpesvirus 6 (HHV-6), have been described as likely pathogens in MS, in partdue to their ability to cause latent infections that periodi-cally reactivate, mirroring the often relapsing remittingcourse of MS. Additionally, some human herpesviruses canbe readily identified within the CNS, and some are knownto be capable of inducing demyelination both in humansand in experimentally infected animals.

Human herpesvirus 6 is a T-cell lymphotropic and neu-rotropic virus, generally acquired in early childhood andassociated with a high incidence of seropositivity inhealthy adults. Its association with MS was given supportby Challoner et al. [15], who demonstrated the presence ofHHV-6 DNA in over 70% of both MS and control subjectbrains, with the caveat that HHV-6 protein was identifiedonly in MS plaques. After this initial observation, severalother reports were published with confirmatory or con-

founding results. A recent longitudinal study examiningHHV-6 DNA in serum suggested that reinfection or reacti-vation of a latent HHV-6 infection might be associatedwith disease exacerbation [16]. Current data suggest ahighly neurotropic variant of HHV-6 (HHV-6A) may be ofconsiderable importance [17•,18].

Chlamydia pneumoniaeIt is clear that Chlamydia pneumoniae infection is com-monly associated with inflammatory neurologic condi-tions [19]. Initial studies showed that C. pneumoniae waspresent in a higher percentage of MS patients by poly-merase chain reaction (PCR) testing (97%) relative to con-trol subjects (19%) [20]. In addition, immunoblottingexperiments have shown that oligoclonal bands in MS CSFcan be completely adsorbed with purified C. pneumoniae[21]. However, a recent collaborative effort, in which CSFsamples of MS patients were split and sent to several differ-ent laboratories for PCR analysis, produced contradictoryfindings [22]. Kaufman et al. [22] confirmed earlier results,but three other laboratories were unable to detect C. pneu-moniae in any of the samples examined. A multicenter, col-laborative effort to detect C. pneumoniae in autopsy brainmaterial of MS patients has also been unsuccessful [23]. IfC. pneumoniae is a cause of MS, it likely would be an exam-ple of Hit-Hit.

Human endogenous retrovirusesRetroviruses, including HTLV-1, have been intensivelystudied as potential agents in the pathogenesis of MS. Theretrovirus visna, which is found in sheep, was an early ani-mal model of MS. This virus can be readily detected in the

Table 1. Viruses in the playing field of multiple sclerosis

Hit-Hit Hit-Run

Mechanism Mechanism1) Virus enters the CNS 1) Virus infects periphery2) Virus persists in the CNS 2) Virus does not persist in the CNS3) Injury by direct viral replication4) Injury mediated by the immune response to clear

infection, including toxic cytokines or pathogenic antibodies, by the development of autoimmunity via epitope spreading, and/or by the nonspecific activation of immune T cells by bystander mechanisms.

3) Injury mediated by a periperal change in the immune milieu, through the development of autoimmunity by the activation of the CNS autoreactive T cells (molecular mimicry or bystander mechanisms), or directly by toxic proinflammatory agents.

Example ExampleCNS demyelinating diseases Candidates in MS PML, SSPE, HTLV-1 (HAM/TSP)Candidates in MS

Herpesvirus (HHV-6), Chlamydia pneumoniae, HERVs (HTLV-1, MSRV)

Herpesvirus (EBV), self-limiting upper respiratory or gastrointestinal tract infections

Treatment Treatment1) Vaccination of predisposed - curative 1) Vaccination of predisposed - curative2) Antivirals for established patients - curative 2) Antivirals for established patients - futile

CNS—central nervous system; EBV—Epstein-Barr virus; HAM/TSP—human T-lymphotropic virus type-1 (HTLV-1)–associated myelopathy/tropical spastic paraparesis; HHV—human herpesvirus; MS—multiple sclerosis; MSRV—multiple sclerosis retrovirus; PML—progressive multifocal leukoencephalopathy; SSPE—subacute sclerosing pancencephalitis.


CNS of infected sheep. Another recently identified retrovi-rus, termed multiple sclerosis retrovirus (MSRV), has beenreported in MS plasma and CSF and has been culturedfrom leptomeningeal cells of an MS patient [24]. In arecent study, however, MSRV was identified in nearly halfof the CSF samples examined from MS patients at clinicalonset and in patients with other neurologic disorders, indi-cating that MSRV is not MS restricted, but may represent amarker for neurologic diseases of inflammatory origin[25]. It is important to consider that susceptibility to dis-ease resulting from infection by a given pathogen mayreflect underlying genetic predisposition. Therefore,excluding a pathogen simply because it is found in bothMS and control patient groups in similar frequency islikely to be an oversimplification. For example, some stud-ies have suggested that MSRV-positive MS patients have amore progressive disease course [26].

Candidate Multiple Sclerosis Pathogens: Hit-Run HypothesisEpstein-Barr virus is a lymphotropic herpesvirus that hasreceived considerable attention as a putative MS agent, butas for HHV-6 the results remain controversial. The virus isnot considered neurotropic. Several studies have detectedEBV antibodies in a higher percentage in MS patients serarelative to control patients [27]. A comprehensive seroepi-demiologic study, the Nurses Health Study [28•], involvingblood samples from more than 62,000 women, reportedthat higher levels of EBV antibodies were associated with afourfold increased risk of developing MS. Owing to the factthat EBV is effective in activating myeling basic protein(MBP)-specific T-cell clones, it has also has been suggestedthat periodic EBV reactivation may activate and expandself-reactive myelin-specific T cells, thereby exacerbatingdisease. There are no studies to date that have demon-strated EBV nucleic acid or protein in MS plaques, or whichhave cultured the virus from affected brains. If EBV is acause of MS, then the most likely scenario is that the virusalters the immune system in the periphery, which thentransforms with time to autoimmunity.

Current Therapeutic StrategiesCurrent therapies are mainly directed at the inflammatoryprocess that characterizes MS, in an attempt to preventclinical relapses and the irreversible damage during pro-gression. The approved treatment for long-term therapy isone of three recombinant interferon β preparations. A largebody of evidence indicates that type-1 interferons (ie, inter-feron β-1a and interferon β-1b) reduce exacerbation ratesin relapsing remitting MS (RRMS) by approximately onethird [29]. It is unclear whether β-interferons affect diseaseprogression beyond effects on relapses. Interferon β-1a wasrecently reported to also reduce relapse rates in moresevere RRMS [30]. Studies suggest some efficacy of β-inter-

ferons in secondary progressive MS (SPMS), but that theyalso may aggravate primary progressive MS (PPMS).

Although interferons are known to be immunomodu-latory, the initial rationale for use of interferons in MS wasas an antiviral agent. There is recent evidence to indicatethat the treatment efficacy of the β-interferons may indeedrelate in part to these antiviral properties. Inteferon-β wasshown to reduce HHV-6 replication in vitro, to decreaseHHV-6 cell free DNA, and to decrease serum IgM reactivityin MS patients [31]. Other antivirals have also been exam-ined in clinical trials. For example, acyclovir, targeting HSV,was examined in a placebo-controlled trial in MS patientsand found to result in a nonsignificant trend toward clini-cal benefit on relapse rate [32]. Use of a broader-spectrumantiviral, valacyclovir, which targets varicella zooster andEBV in addition to HSV, showed no differences betweenthe treatment and placebo for any of the clinical endpointsexamined [33]. However, patients with high disease activ-ity had a 68% reduction in new lesion formation.

The other approved treatment for RRMS is glatirameracetate [34]. This drug was shown to inhibit experimentalautoimmune encephalomyelitis (EAE) in rodents, and hasbeen shown to ubiquitously bind MHC class-II DR [35]. Ifviruses trigger subsequent class-II restricted CD4+ T-cell–mediated autoimmunity, then this is likely how glatirameracetate works.

Mechanisms of Viral-induced Central Nervous System DemyelinationAnimal models have provided direct evidence that viralinfection and/or autoimmune mechanisms can inciteinflammatory demyelinating disease resembling MS. Thereare two primary viral models of MS: TMEV and mouse hep-atitis virus (MHV). Each is a naturally occurring murinevirus that infects the CNS and induces demyelination ingenetically susceptible mice (Hit-Hit). Most data indicatethat demyelination is secondary to the immune responsetargeting CNS viral antigens. The other major animalmodel of MS is EAE. Similarities between MS and EAE,which is inducible by immunization with protein constitu-ents of CNS myelin such as MBP or proteolipid protein(PLP) or by adoptive transfer of CD4+ T cells specific forthese proteins, favor an autoimmune model, but do notexclude the role of an infectious agent. Notably, the induc-tion of EAE requires the use of Complete Freund's Adju-vant (CFA), which creates an artificial inflammatorymilieu. The best natural adjuvants in nature are viruses thatstimulate the immune system (Hit-Run).

There are three main mechanisms proposed to explainhow infectious agents could lead to autoimmunity. Thefirst suggests that molecular mimicry between pathogenicand self antigens leads to the activation of T cells that arecross-reactive with self-epitopes [36,37]. In the epitopespreading model, virus-specific T cells cause direct or indi-rect damage to self, with consequent autoantigen release,


resulting in de novo activation of autoreactive T cells [38].Finally, demyelination may be the result of bystandermechanisms in which nonspecific infections activateimmune T cells [39,40••]. Although there are several pos-sible mechanisms by which infectious agents may induceCNS demyelination, we propose these are initiated ineither a Hit-Hit or Hit-Run fashion.

Molecular mimicryVarious viral and bacterial peptides, showing limitedsequence homology with myelin self-peptides (HHV-6,influenza, measles, papilloma virus, and EBV), havebeen shown to activate autoreactive T-cell clones. Initialstudies showed that transgenic mice expressing virusproteins in the context of self-tissues developed autoim-mune disease following infection with the correspond-ing virus. Recently, studies have provided more directevidence that infection-induced molecular mimicrycould lead to CNS demyelination. For example, CNSinfection with a nonpathogenic variant of TMEV, con-taining a 6 amino acid PLP139-151 mimic present inHemophilus influenzae, resulted in early onset demyelina-tion and activation of T-helper 1 (Th1) cells cross-reac-tive with native PLP139-151 [41•]. These resultsdemonstrated that infection with a virus expressing amimic of self-epitope, or at least 6 amino acids identicalto a self-epitope, can induce autoreactive T cells, withpathologic potential in the absence of CFA. It is impor-tant to note that for this to occur, however, the incitingvirus must enter and then persist in the CNS, an exam-ple of demyelination in a Hit-Hit fashion.

It has been shown that viral sequence homology is nota prerequisite for activation of T cells specific for myelinantigens. T-cell receptor (TCR) cross-reactivity has beenattributed to a high degree of degeneracy in antigen recog-nition by the TCR, requiring only as few as three criticalresidues [42•]. Linking molecular mimicry to genetics, theMBP85-99 cross-reactive human TCR has been linked to atleast two alleles of the MHC class-II, DR2 haplotype[43••], a defined genetic risk factor for MS. Lang et al.[43••] found that the TCR recognized MBP85-99 in thecontext of the DRB1*1501 allele, and EBV627-641 in thecontext of the DRB5*0101 allele. This functional interac-tion between two of the MHC class-II loci in the DR2 hap-lotype could lead to increased numbers of microbialpathogen-derived peptides available for presentation to asingle cross-reactive TCR.

The potential contribution of molecular mimicry and adegenerate TCR to autoimmune disease is somewhat tem-pered by other recent studies. The capacity of HHV-6 andMBP to activate T cells isolated from MS patients and con-trol subjects was examined. T cells cross-reacting withHHV-6 and MBP were identified; however, cross-reacting Tcells were found in both groups, suggesting that such cross-reactivity may not be an important mechanism in MSpathophysiology [44]. Alternatively, other concurrent fac-

tors may render MS patients more susceptible to suchcross-reactivity, including genetic factors.

Recent studies also suggest that molecular mimicrybetween antibodies directed at clearing CNS viruses andCNS antigens may contribute to the pathogenesis of CNSautoimmune disease [45••]. Antibodies isolated fromHTLV-1–infected HAM/TSP patients were shown to cross-react with HTLV-1, to recognize a neuronal nuclear bionu-clear protein-A1 (hnRNP-A1), to specifically stain the neu-rons preferentially damaged in this disease, and to inhibitneuronal firing in vitro. Notably, hnRNP-A1 was alsofound to share significant homology to hnRNP-A2, a pro-tein known to have a critical role in the transport of MBPwithin oligodendroglial processes.

Epitope spreadingThe epitope spreading model suggests that regardless ofthe initial antigenic stimulus (Hit-Hit or Hit-Run), thespecificity of the immune response spreads to include self-epitopes, which are distinct from that initiating the inflam-matory response [38]. For example, PLP-139-151/CFAimmunization of SJL mice results in a relapsing remittingdemyelinating disease, mediated in the initial phase byPLP139-151–specific CD4+ Th1-type T cells, with the fol-lowing relapses mediated by myelin-specific Th1 cells spe-cific for other endogenous myelin epitopes, such asPLP178-191 and MBP84-104 [46]. The pathogenic signifi-cance of epitope spreading was illustrated by experimentsdemonstrating that tolerance to PLP178-191, but not toPLP139-151, was required to prevent EAE disease relapses.Additionally, T cells specific for relapse-associated epitopeswere shown to transfer disease to naive recipients. A patho-logic role for epitope spreading in virus-induced demyeli-nation has also been demonstrated. With TMEV infection,CD4+ T-cell responses are restricted to TMEV. With theonset of TMEV-induced demyelination, however, CD4+ T-cell responses to the immunodominant PLP139-151 mye-lin epitope develop, which subsequently progress toinvolve CD4 responses to a variety of additional myelinepitopes [47]. Again, it was shown that induction of toler-ance to the encephalitogenic myelin epitopes partiallyabrogated disease progression. It is important to note thatepitope spreading did not occur in TMEV-infected miceuntil there was significant demyelination as a result of adirect antiviral immune response (Hit-Hit). It has beenproposed that epitope spreading to include reactivity toaxons might explain the transition from RRMS to SPMS,whereas in PPMS, reactivity to axonal antigens may be theprimary target [48].

Bystander-mediated demyelinationAn alternative to the epitope spreading model, thebystander-mediated demyelination model proposesbystander activation of immune T cells. Notably, numer-ous studies have demonstrated the existence of myelin-spe-cific autoreactive T cells in healthy individuals, and indeed


these may be part of the normal immune repertoire. It isalso known that viral infections are associated with sys-temic immune activation, reflected by the increased pro-duction of proinflammatory cytokines, which may resultin activation and proliferation of immune T cells in abystander fashion, bypassing TCR engagement. In favor ofbystander mechanisms operating in MS, it is relatively wellaccepted that nonspecific infections, mostly upper respira-tory tract infections, are associated with disease exacerba-tions, and that these exacerbations lead to more sustainedclinical deficits [49]. Experimental evidence comes fromstudies that showed infection of young mice peripherallywith a ubiquinated PLP construct, leading to presentationof PLP by class-I MHC, followed by a later nonspecificimmunologic stimulus (CFA), resulted in 20% of the micedeveloping clinical symptoms of EAE [40••]. In a follow-up experiment, mice were infected peripherally withrecombinant vaccinia virus encoding PLP. If these micewere later given CFA, the majority of mice developed CNSinflammation. These studies demonstrate the possibility ofbystander demyelination through a mechanism of molecu-lar mimicry in a Hit-Run fashion.

Experimental evidence also exists suggesting thatbystander mechanisms may contribute to demyelination ina Hit-Hit fashion. Demyelination in response to CNS MHVinfection in LCMV gp33 TCR/RAG-deficient mice, whichcontain T cells unable to recognize MHV, occurred only inmice that had been previously primed with the now "self"gp33 [39]. Thus, CNS infection with MHV resulted in aproinflammatory milieu, capable of promoting demyelina-tion in mice with pre-existing, nonpathogenic, CNS self-reactive CD8+ T cells. Together, these studies demonstratethat it is possible that multiple infectious agents, or peri-odic reactivation of latent viruses, could expand self-reac-tive T cells and stimulate or exacerbate disease.

ConclusionsStudies in animal models and in MS patients reveal that themechanisms that ultimately lead to pathogenesis, albeit byautoimmune or infections triggers, may be diverse. Withregard to an infectious agent, potential mechanisms may beviewed in terms of an initial Hit-Hit or a Hit-Run. In the Hit-Hit scenario, demyelination can be seen as a direct conse-quence of infection, with or without the development ofautoimmunity. This hypothesis requires that the virus infectthe CNS. In terms of a Hit-Run, primary infection in theperiphery results in chronic organ-specific autoimmunity,and the inciting pathogen may never enter the CNS.Together, experimental evidence supports the concept thatinfectious agents may mediate CNS demyelination by severalmechanisms, including the induction of autoimmunity;however the identity of that agent or agents has not yet beenconclusively demonstrated.

AcknowledgmentsSupported by grants RG 3367-A-2-01 from the NationalMultiple Sclerosis Society and grant P01-NS38468 fromthe National Institutes of Health.

References and Recommended ReadingPapers of particular interest, published recently, have been highlighted as:• Of importance•• Of major importance

1. Noseworthy JH, Lucchinetti CF, Rodriguez M, Weinshenker BG: Multiple sclerosis. N Engl J Med 2000, 343:938–952.

2. Ebers GC: Natural history of multiple sclerosis. J Neurol Neu-rosurg Psychiatry 2001, 71:16–9.

3. Sadovnick A, Yee IM, Ebers GC: Recurrence risks to sibs of MS index cases: impact of consanguineous matings. Neurology 2001, 56:784–785.

4. Ebers GC, Yee IM, Sadovnick AD, Duquette P: Conjugal multi-ple sclerosis: population-based prevalence and recurrence risks in offspring. Canadian Collaborative Study Group. Ann Neurol 2000, 48:927–931.

5. Kantarci OH, de Andrade M, Weinshenker BG: Identifying dis-ease modifying genes in multiple sclerosis. J Neuroimmunol 2002, 123:144–159.

6.•• Lucchinetti C, Bruck W, Parisi J, et al.: Heterogeneity of multi-ple sclerosis lesions: implications for the pathogenesis of demyelination. Ann Neurol 2000, 47:707–717.

Established different patterns of pathology in acute demyelinating multiple sclerosis lesions, suggesting heterogeneity of disease may govern clinical course and response to therapy.

7.•• Babbe H, Roers A, Waisman A, et al.: Clonal expansions of CD8+ T cells dominate the T cell infiltrate in active multiple sclerosis lesions as shown by micromanipulation and single cell polymerase chain reaction. J Exp Med 2000, 192:393–404.

Article demonstrated that the immune response in multiple sclerosis lesions is predominated by CD8+ T cells and is likely antigen driven.

8. Jacobsen M, Cepok S, Quak E, et al.: Oligoclonal expansion of memory CD8+ T cells in the cerebrospinal fluid from multi-ple sclerosis patients. Brain 2002, 125:538–550.

9. Colombo M, Dono M, Gazzola P, et al.: Accumulation of clonally related B lymphocytes in the cerebrospinal fluid of multiple sclerosis patients. J Immunol 2000, 164:2782–2789.

10. Medana I, Martinic MA, Wekerle H, Neumann H: Transection of major histocompatibility complex class I-induced neurites by cytotoxic T lymphocytes. Am J Pathol 2001, 159:809–815.

11. Rivera-Quinones C, McGavern D, Schmelzer JD, et al.: Absence of neurological deficits following extensive demyelination in a class I-deficient murine model of multiple sclerosis. Nat Med 1998, 4:187–193.

12.• Johnson AJ, Upshaw J, Pavelko KD, et al.: Preservation of motor function by inhibition of CD8+ virus peptide-specific T cells in Theiler's virus infection. FASEB J 2001, 15:2760–2762.

Experimental evidence supporting the pathogenic role of class-I CD8+ T cells in axonal injury in an animal model of multiple sclerosis.13. Soldan SS, Jacobson S: Role of viruses in etiology and patho-

genesis of multiple sclerosis. Adv Virus Res 2001, 56:517–555.14. Gilden DH, Burgoon MP, Kleinschmidt-DeMasters BK, et al.:

Molecular immunologic strategies to identify antigens and b-cell responses unique to multiple sclerosis. Arch Neurol 2001, 58:43–48.

15. Challoner PB, Smith KT, Parker JD, et al.: Plaque-associated expression of human herpesvirus 6 in multiple sclerosis. Proc Natl Acad Sci 1995, 92:7440–7444.

16. Berti R, Brennan MB, Soldan SS, et al.: Increased detection of serum HHV-6 DNA sequences during multiple sclerosis (MS) exacerbations and correlation with parameters of MS disease progression. J Neurovirol 2002, 8:250–256.


17.• Akhyani N, Berti R, Brennan MB, et al.: Tissue distribution and variant characterization of human herpesvirus (HHV)-6: increased prevalence of HHV-6A in patients with multiple sclerosis. J Infect Dis 2000, 182:1321–1325.

This paper makes an association of the less well-characterized variant of human herpesvirus (HHV)-6, HHV-6A, with multiple sclerosis. These findings are important because HHV-6A appears to be an extremely neurotropic.18. Alvarez-Lafuente R, Martin-Estefania C, de Las Hera V, et al.:

Active human herpesvirus 6 infection in patients with multi-ple sclerosis. Arch Neurol 2002, 59:929–933.

19. Sotgiu S, Paiana A, Pugliatti M, et al.: Chlamydia pneumoniae in the cerebrospinal fluid of patients with multiple sclerosis and neurological controls. Mult Scler 2001, 7:371–374.

20. Sriram S, Stratton CW, Yao S, et al.: Chlamydia pneumoniae infection of the central nervous system in multiple sclerosis. Ann Neurol 1999, 46:6–14.

21. Yao SY, Stratton CW, Mitchell WM, Sriram S: CSF oligoclonal bands in MS include antibodies against Chlamydophila anti-gens. Neurology 2001, 56:1168–1176.

22. Kaufman M, Gaydos CA, Sriram S, et al.: Is Chlamydia pneu-monia found in spinal fluid samples from multiple sclerosis patients? Conflicting results. Mult Scler 2002, 8:289–294.

23. Hammerschlag MR, Ke Z, Lu F, et al.: Is Chlamydia pneumo-niae present in brain lesions of patients with multiple sclero-sis? J Clin Microbiol 2000, 38:4274–4276.

24. Perron H, Garson JA, Bedin F, et al.: Molecular identification of a novel retrovirus repeatedly isolated from patients with multiple sclerosis. The Collaborative Research Group on Multiple Sclero-sis. Proc Natl Acad Sci U S A 1997, 94:7583–7588.

25. Dolei A, Serra C, Mameli G, et al.: Multiple sclerosis-associated retrovirus (MSRV) in Sardinian MS patients. Neurology 2002, 58:471–473.

26. Sotgiu S, Serra C, Mameli G, et al.: Multiple sclerosis-associated retrovirus and MS prognosis. Neurology 2002, 59:1071–1073.

27. Wandinger K-P, Jabs W, Siekhaus A, et al.: Association between clinical disease activity and Epstein-Barr virus reactivation in MS. Neurology 2000, 55:178–184.

28.• Ascherio A, Munger KL, Lennette ET, et al.: Epstein-Barr virus antibodies and risk of multiple sclerosis: a prospective study. JAMA 2001, 286:3083–3088.

A large prospective, case-control study demonstrating increased risk for multiple sclerosis correlated with Epstein-Barr virus seropositivity.29. Beck RW, Chandler DL, Cole SR, et al.: Interferon beta-1a for

early multiple sclerosis: CHAMPS trial subgroup analyses. Ann Neurol 2002, 51:481–490.

30. Vermersch P, de Seze J, Stojkovic T, Hautecoeur P: Interferon beta1a (Avonex) treatment in multiple sclerosis: similarity of effect on progression of disability in patients with mild and moderate disability. J Neurol 2002, 249:184–187.

31. Hong J, Tejada-Simon MV, Rivera VM, et al.: Anti-viral proper-ties of interferon beta treatment in patients with multiple sclerosis. Mult Scler 2002, 8:237–242.

32. Lycke J, Svennerholm B, Hjelmquist E, et al.: Acyclovir treat-ment of relapsing-remitting multiple sclerosis. A random-ized, placebo-controlled, double-blind study. J Neurol 1996, 243:214–224.

33. Bech E, Lycke J, Gadeberg P, et al.: A randomized, double-blind, placebo-controlled MRI study of anti-herpes virus therapy in MS. Neurology 2002, 58:31–36.

34. Johnson KP, Brooks BR, Cohen JA, et al.: Extended use of glati-ramer acetate (Copaxone) is well tolerated and maintains its clinical effect on multiple sclerosis relapse rate and degree of disability. Neurology 2001, 57:S46–S53.

35. Hafler DA: Degeneracy, as opposed to specificity, in immuno-therapy. J Clin Invest 2002, 8:142–154.

36. Fujinami RS: Molecular mimicry that primes for autoimmu-nity which is triggered by infection. Mol Psychiatry 2002, 7:S32–S33.

37. Wucherpfennig KW: Infectious triggers for inflammatory neu-rological diseases. Nat Med 2002, 8:455–457.

38. Olson JK, Eagar TN, Miller SD: Functional activation of mye-lin-specific T cells by virus-induced molecular mimicry. J Immunol 2002, 169:2719–2726.

39. Haring JS, Pewe LL, Perlman S: Bystander CD8 T cell-mediated demyelination after viral infection of the central nervous sys-tem. J Immunol 2002, 169:1550–1555.

40.•• Theil DJ, Tsunoda I, Rodriguez F, et al.: Viruses can silently prime for and trigger central nervous system autoimmune disease. J Neurovirol 2001, 7:220–227.

Provides experimental evidence suggesting prior viral infection may set the stage for nonspecific immunologic stimuli to activate self-reac-tive T cells.41.• Olson JK, Croxford JL, Calenoff MA, et al.: A virus-induced

molecular mimicry model of multiple sclerosis. J Clin Invest 2001, 108:311–318.

This model was the first to directly demonstrate that a virus encoding a mimic of an encephalitogeic self myelin epitope could induce an autoreactive CD4+ T cell response, leading to central nervous system demyelination.42.• Hemmer B, Kondo T, Gran B, et al.: Minimal peptide length

requirements for CD4(+) T cell clones—implications for molecular mimicry and T cell survival. Int Immunol 2000, 12:375–383.

Demonstrated that as few as three critical residues may be required for T-cell receptor recognition.43.•• Lang HL, Jacobsen H, Ikemizu S, et al.: A functional and struc-

tural basis for TCR cross-reactivity in multiple sclerosis. Nat Immunol 2002, 3:940–943.

Crystal structure analysis and lymphoproliferative studies were used to demonstrate that the human myelin basic protein (MBP) T-cell receptor may cross-react with both MBP and Epstein-Barr virus in a major histocompatibility class-II DR2 restricted fashion.44. Cirone M, Cuomo L, Zompetta C, et al.: Human herpesvirus 6

and multiple sclerosis: a study of T cell cross-reactivity to viral and myelin basic protein antigens. J Med Virol 2002, 68:268–272.

45.•• Levin MC, Lee SM, Kalume F, et al.: Autoimmunity due to molecular mimicry as a cause of neurological disease. Nat Med 2002, 8:509–513.

These data demonstrate the potential importance of molecular mim-icry between pathogenic human T-cell leukemia/lymphoma virus-1 antibodies and neuronal proteins in the development of a human disease resembling multiple sclerosis.46. Vanderlugt CL, Neville KL, Nikcevich KM, et al.: Pathologic role

and temporal appearance of newly emerging autoepitopes in relapsing experimental autoimmune encephalomyelitis. J Immunol 2000, 164:670–678.

47. Katz-Levy Y, Neville KL, Padilla J, et al.: Temporal development of autoreactive Th1 responses and endogenous presentation of self myelin epitopes by central nervous system-resident APCs in Theiler's virus-infected mice. J Immunol 2000, 165:5304–5314.

48. Sadatipour BT, Greer JM, Pender MP: Increased circulating antiganglioside antibodies in primary and secondary pro-gressive multiple sclerosis. Ann Neurol 1998, 44:980–983.

49. Buljevac D, Flach HZ, Hop WC, et al.: Prospective study on the relationship between infections and multiple sclerosis exac-erbations. Brain 2002, 125:952–960.

Documents

Hit-Hit and hit-Run: viruses in the playing field of multiple sclerosis