Do pathogens contribute to multiple sclerosis aetiology?

Multiple sclerosis (MS) is a common neurological disease characterised by sclerotic plaques of dead and dying oligodendrocytes and neurons in the central nervous system, usually with evidence of accompanying inflammatory activity. Infectious agents might cause or contribute to this cell death, but none have yet been established as doing so. Instead, MS is regarded as predominantly an autoimmune disease, responsive to immunomodulators. However, there is compelling epidemiological and empirical evidence implicating Epstein–Barr virus (EBV), with minor support for other viruses, as contributing to the pathogenic immune response in MS. Now the recent dramatic advances in identifying the genetic risk factors for MS have provided some tantalising leads to microbe involvement. If conclusive evidence is found, vaccines, antivirals/antibiotics or pathogens themselves might prove useful therapeutics.

Multiple sclerosis is one of the most common neurological diseases of young adults. It is primarily an inflammatory disorder of the brain and spinal cord in which focal lymphocytic infiltration leads to damage of myelin and axons. It is triggered by environmental factors, especially in individuals with genetic susceptibility¹. Many of the genetic variants increasing risk have now been identified^2,3, and these point to variation in cell mediated immunity as driving susceptibility. These genes also potentially point to specific pathogens.

There is empirical and epidemiological evidence that pathogens could be environmental triggers, but the distinction between association and causation is problematic. Two likely mechanisms for causation have been described: bystander activation, where self-reactive T cells are activated due to tissue damage by a pathogen; and molecular mimicry, where T cells that recognize pathogen epitopes are also self-reactive and are activated by the pathogens (Figure 1). Alternatively, pathogens might be protective in autoimmune disease, by inducing tolerance⁴.

Figure 1.  Multiple sclerosis (MS) is considered to be caused by immune system mediated killing of the oligodendrocytes (OD) in the central nervous system. Genes affecting susceptibility are mainly expressed in antigen-presenting cells and T cells, but also in other immune cells. These genes might also affect response to infections. Infections could cause an immune response targeted on OD through bystander activation, an indiscriminate activation including to ODs; or molecular mimicry, a mistaken activation, where infectious agent and OD share similar epitopes. Infection might also tolerise immune responses.

The sudden arrival and subsequent epidemics of MS in the Faroe Islands during and after WWII was most easily explained as facilitated by the arrival of British troops and accompanying pathogens in 1943⁵. But if MS here and elsewhere was caused by an infection, the pathogen has remained elusive. From the geographic and ethnic distribution, the pathogen is likely to be ubiquitous in western countries, but also in Asia and elsewhere (because MS is described worldwide), infect children (because MS risk reflects childhood environment), cause chronic infection (because the disease is usually late onset and persistent) and is more likely in the higher latitudes (because MS prevalence increases with latitude).

The strongest candidate to date is EBV, which fulfils all these criteria⁶. Although typically 90% of the general population carry EBV, virtually 100% of people with MS do. Notably, though, other common chronic viruses, such as HHV6, chicken pox, measles and cytomegalovirus have also been described as more prevalent in MS⁷, such that exposure to these viruses might also render people susceptible to MS and the virus is not causative but associative⁸. More compellingly, late infection or severe infection with EBV increases risk of MS by up to 20-fold⁹, an increase not yet found for other pathogens. People with MS have poor CD8 responses to EBV⁶, T cells recognizing myelin proteins can be activated by EBV¹⁰, high titres of EBV antibodies are associated with relapses and paediatric MS¹¹. A prospective study of 915 individuals found the 10 who were EBV negative and later developed MS were all EBV positive before they developed MS, whereas of the 28 EBV negative controls only 35% became EBV positive over the same time period¹². High EBV antibody titres, in combination with the first MS susceptibility gene, and variant of largest effect yet known, HLA DRB1*1501, increases the risk of MS, additively and independently, not interactively¹³. Oligoclonal bands found in the CSF of many with MS can be reactive to EBV and other herpesviruses¹⁴.

EBV infects B cells through CD21⁶, and possibly CD35¹⁵. Both of these genes are not among those known to affect susceptibility to MS. The infected B cells proliferate through signaling through EBVs own analogue of human CD40. Strikingly, genetic variants of CD40 are associated with MS, and the genotype with higher expression of CD40 decreases MS risk¹⁶. Because CD40 is a costimulatory molecule required for T cell activation, it would be expected that higher expression would increase the risk of MS. A potential explanation for this paradox is that low host CD40 expression on B cells favours proliferation/survival of EBV infected B cells, using their EBV CD40. In support of this, anti-CD20 B-cell-depleting drugs such as Rituximab, which are used as a therapies to reduce EBV infection¹⁷, are effective in MS, but those targeting terminally differentiated B cells are not¹⁸. The effect of more specific depletion of EBV infected B cells on MS needs to be investigated.

Genome wide analysis studies have implicated specific viruses for some autoimmune conditions. Individuals homozygous for the Crohn’s diseases risk allele rs601338 of FUC2, the receptor for noroviruses, are protected against norovirus infection¹⁹, but are more likely to develop Crohn’s and other autoimmune diseases such as type 1 diabetes and inflammatory bowel disease²⁰, consistent with the ‘hygiene hypothesis’⁴. The MS risk gene TNFSF14 (also known as herpesvirus entry mediator ligand or LIGHT)^2,3 is an entry receptor for Herpes simplex 1 (HSV1)²¹, a neurotropic virus. A ligand for TNFSF14, TNFRSF6B, is also a MS risk gene. SLAM family proteins are used by morbilliviruses to facilitate cell entry²². These viruses then circumvent the interferon response by blocking Tyk2 signalling²³. Both SLAMF7 and TYK2 are risk genes for MS³. Measles and canine distemper virus, both implicated in the Faroe Island MS ‘epidemics’, are morbilliviruses. Further viral receptors and genes important in their infection might be in the list of MS genes. But as TNFSF14/TNFRSF6B and SLAMF7/TYK2 also have roles affecting other aspects of the immune response, further implication of a HSV1/morbillivirus contribution to MS might follow if the MS susceptibility genotype can be shown to increase tissue damage or other infectious consequences due to these viruses.

The genes affecting MS susceptibility support a role for immune response in MS, but although they, as yet, provide no smoking gun for particular viruses, the evidence that most of these genes are under recent positive selection²⁴ is consistent with the genetic variants affecting selection against pathogens. Most of these genes are also predominantly expressed in T cells or antigen-presenting cells, consistent with a role for molecular mimicry or bystander activation contributing to disease aetiology, but also with excessive reaction to self antigen independently of pathogen effect.

A ‘smoking gun’ for one environmental trigger was found. This is vitamin D. Low vitamin D has been implicated as contributing to MS development and progression from epidemiological and empirical studies²⁵. It now seems that genetic variants of the two genes regulating its activation in antigen presenting cells, CYP27B1 and CYP24A1, affect MS risk². As well as regulating DC tolerisation, vitamin D can be antimicrobial, through its upregulation of cathellicidin^6,26.

Current successful therapies for MS alter leukocyte survival, modulation or trafficking¹⁸. Further manipulation of the immune response might be beneficial if a microbial effect on MS aetiology is defined. Vaccination might be a useful strategy whether viruses such as EBV increase susceptibility (especially if late infection) or protect (especially if early infection) against MS²⁷. This might include use of the pathogen itself²⁸. To support such approaches, experimental studies are needed to determine if genes associated with MS function through their effect on infections. Novel approaches to establishing if viruses are pathogenic include sequencing of non-host DNA/RNA, especially in plaques, and amplified endogenous retroviruses. Such studies could use historic samples from the Faroe islands and elsewhere. If pathogens contribute to aetiology, even if in only a subset of patients, it is highly likely that their identification will provide significant clinical benefits.