Background
Multiple sclerosis (MS) is a chronic inflammatory disease of the central nervous system (CNS) characterized by the breakdown of the insulating myelin sheath that covers the nerve axons in the CNS and subsequent degeneration of axons. The process leads most commonly to intermittent neurological symptoms followed, over time, by progressive neurological symptoms in many patients. MS affects approximately 400,000 people in the USA and more than 2.1 million people worldwide, but the incidence has increased in the last five decades, particularly in women (3.6/100,000 person-years) compared to men (2.0/100,000 person-years) (Alonso & Hernan 2008; National Multiple Sclerosis Society 2012). While the etiology of MS is not understood in detail, it is unlikely to be the result of a single causative event. Instead, converging evidence suggests that MS is caused by an abnormal autoimmune response in genetically susceptible individuals after specific environmental exposures. Thus, it is not a heritable disease in the classic sense, but a complex disease that emerges from genes interacting with other genes and genes interacting with the environment. The factors thought to mediate the risk of MS are subject to intense ongoing research and include genetic, immunologic, infectious, and environmental contributors. The aim of this chapter is to review the current data on MS risk factors, with particular emphasis on those that may be modifiable on a personal or population level.
Genes
Familial aggregation is a well-recognized phenomenon in MS, and family and twin studies have long shown evidence for a strong genetic component underlying MS. This is illustrated by the 25–30% concordance among monozygotic twins, the 5% concordance among same-sex dizygotic twins, and the 3.5% concordance among nontwin siblings (Gourraud et al. 2012). However, the inheritance of MS cannot be explained by a simple genetic model, and neither the familial recurrence rate nor twin concordance supports the presence of a Mendelian trait. Rather, susceptibility is polygenic, with each gene contributing a relatively small amount of the overall risk. More than likely, genetic heterogeneity (different susceptibilities among individuals) also exists. Additionally, epidemiological data strongly hint at a parent-of-origin effect in MS: maternal half-siblings having double the risk for MS compared to paternal half-siblings (2.35% vs. 1.31%), while the risk for MS in maternal half-siblings compared to their full siblings does not differ significantly (Gourraud et al. 2012). The mechanism of the increased risk conferred maternally remains to be elucidated, but epigenetic mechanisms such as DNA methylation or histone modification may play a role (Handel et al. 2010).
The first direct evidence for a relationship between genes and MS susceptibility came in 1972, when MS was shown to be associated with the human leukocyte antigen (HLA) on chromosome 6p21 (encoding proteins involved in presenting peptide antigens to T cells) (Gourraud et al. 2012). This association was later fine-mapped to a specific locus, HLA-DRB1 of the class II gene HLA-DRB1 (Gourraud et al. 2012). Although the HLA-DRB*1501 haplotype exerts the strongest genetic effect in MS (heterozygosity conferring an odds ratio (OR) of 2.7 and homozygosity of 6.7), the association is not straightforward. In fact, a number of HLA-DRB1 haplotypes are both positively and negatively associated with the disease, differ in magnitude of effect, and either act on their own or greatly alter risk in combination with another haplotype (Kallaur et al. 2011). For example, HLA-DRB1*08 only modestly increases MS risk, but in combination with HLA-DRB1*15, it more than doubles the risk associated with a single copy of the latter (Kallaur et al. 2011). On the other hand, HLA-DRB1*14 carries such a protective effect that it completely abrogates the increased risk of HLA-DRB1*15 (Kallaur et al. 2011). And whereas association of MS with HLA-DRB1*15 has long been known in Northern Europe, in other regions, such as Sardinia, HLA-DRB1*0301, HLA-DRB1*0405, and HLA-DRB1*1303 are more commonly associated with MS (Kallaur et al. 2011). In fact, the relative frequencies of susceptibility and protective HLA haplotypes, which vary between countries, may play important roles in determining the risk of the disease.
It has been estimated that the HLA locus accounts for 20–60% of the genetic susceptibility in MS, leaving a large portion of the genetic component of MS (still) to be explained. In 2007, the International Multiple Sclerosis Genetics Consortium (IMSGC) completed the first MS genome-wide association study (GWAS) using trios (an affected individual and both their parents) from the UK and the USA (Gourraud et al. 2012). In addition to the HLA-DRB1 region, two new risk loci were identified: the genes for interleukin-7 receptor alpha (IL-7RA) and interleukin-2 receptor alpha (IL-2RA), which have since been replicated. These genes code for the alpha chain of the IL-7 or IL-2 receptors, which promote lymphocyte growth and differentiation. MS-associated variants in the IL-2RA gene contribute to the production of soluble IL-2RA, a biomarker of peripheral inflammation. The IL-7/IL-7RA interaction is important for memory T-cell maintenance and development and proliferation and survival of B and T cells; the protective haplotype is associated with less soluble IL-7RA; the risk allele thus likely produces a change in function (Gregory et al. 2007).
The most recent GWAS data from the IMSGC demonstrate at least 102 SNPs exerting a modest effect (OR, 1.06–1.22) (Gourraud et al. 2012). Most of the loci harbor genes with pertinent immunological roles, including several genes associated with other autoimmune disorders, consistent with the autoimmune hypothesis of MS etiology. Most notably, the results of the GWAS implicate genes coding for cytokine pathways (CXCR5, IL-2RA, IL-7R, IL-7, IL-12RB1, IL-22RA2, IL-12A, IL-12B, IRF8, TNFRSF1A, TNFRSF14, TNFSF14) and for costimulatory (CD37, CD40, CD58, CD80, CD86, CLECL1) and signal transduction (CBLB, GPR65, MALT1, RGS1, STAT3, TAGAP, TYK2) molecules of immunological relevance (Gourraud et al. 2012). Of interest, at least two genes (KIF1B, GPC5) not involved in the immune system but instead with neuronal growth and repair mechanisms may also be associated with MS. These genes may influence the potential of remyelination of lesions, and their discovery gives a hint to a disturbance of repair mechanisms in addition to autoimmune processes in MS.
Still relatively little is known about how the identified MS risk variants exert their effects at the molecular and cellular levels. Their incomplete penetrance and moderate individual effects probably reflect interactions with other genes, posttranscriptional regulatory mechanisms, or significant environmental and epigenetic influences. Further genetic and functional studies are required to pinpoint the functionally relevant genes and pathways, to understand how these influence risk, and to determine if the genes themselves, or the downstream effects thereof, can be modified to alter MS risk.
The role of the environment
Genetic factors account only partially for MS susceptibility, as illustrated by the twin concordance data. Moreover, even among families, MS risk is known to be strongly influenced by location, season of birth, and the childhood environment. The environment thus appears to play an important role in setting thresholds for genetic penetrance. Further, recent increases in MS incidence are too rapid to be the result of genetic alterations and must, therefore, reflect differential exposure to environmental factors (Alonso & Hernan 2008). In particular, the rising worldwide incidence and increasing female to male preponderance have focused interest on environmental factors that may influence MS risk.
Environmental MS risk factors: The major players
All of the environmental factors involved in MS are not yet known, but accumulating evidence lends strong support to several candidates, most notably sunlight and/or vitamin D exposure, Epstein–Barr virus (EBV), and cigarette smoking (Ascherio & Munger 2007a, b), with unconfirmed or hypothetical support for obesity, diet, and altered gut microbiota as risk factors. These factors could conceivably act to alter susceptibility to MS at any point in life from conception (or even before) to the onset of disease.
Geography
The uneven geographical distribution of MS is central to understanding the role of environment. The prevalence of MS increases with distance from the equator (Ascherio & Munger 2007b) and is greater in areas with temperate rather than tropical climates. Within regions of temperate climate, MS incidence and prevalence increase with latitude. Some of these observations may be explained by the nonrandom geographic distribution of racial/ethnic groups within these risk areas, such that what appears to be a latitudinal effect may be confounded by the genetic backgrounds of those who live in the various regions (i.e., racial/ethnic groups with a higher burden of risk alleles may be those who happen to live in regions of higher prevalence). However, migration studies demonstrate that moving from a region of high to low risk, or vice versa, leads to the adoption of the risk of the new region, especially if the migration occurred at a young age (Ascherio & Munger 2007b) such that at least part of the latitudinal gradient must be due to environmental differences.
One of the strongest correlates of latitude is the duration and intensity of sunlight. Thus, it is not surprising that an inverse correlation between MS prevalence and sunlight was already noted in early ecological studies; among US veterans, the average annual hours of sunshine and the average December daily solar radiation at place of birth were strongly inversely correlated with MS (Ascherio & Munger 2007b). Furthermore, several retrospective studies have demonstrated that sun exposure during childhood and adolescence as well as outdoor activity as an occupational exposure is inversely related to MS susceptibility (Ascherio & Munger 2007b). The protective effects of sunlight are thought to be mediated by ultraviolet radiation (UVR), possibly via vitamin D (see section Vitamin D).
Migration studies and timing of environmental effect
While early migration studies suggested that migration prior to age 15 is critical to altering the risk of MS (Ascherio & Munger 2007b), more recent data suggest that the critical age period might even extend into the third decade. These intriguing findings suggest that MS risk factors may operate in childhood and beyond puberty, suggesting a more prolonged period of vulnerability (but notably also for potential intervention). There may also be transgenerational epigenetic modifications that influence MS risk, which could potentially be influenced by factors such as diet or sex hormones (Ascherio & Munger 2007b). Studies in UK migrants followed from gestation to the third decade of life suggest risk increases in the subsequent generation (Elian et al. 1990). Gestational or early life timing as a vulnerable period is also suggested by a marginally significant excess risk in dizygotic twins compared with nontwin siblings, coupled with evidence for maternal effects. More direct evidence comes from studies of month of birth in several northern countries, which have latitude-correlated increased risks for spring births and decreased risks for late fall births (Willer et al. 2005). The polarity of this distribution reverses in the southern hemisphere. Moreover, unaffected sibling controls differ in birth-month distribution from the general population as much as their affected brothers and sisters did but in the opposite direction (Willer et al. 2005). Since serum concentrations of vitamin D fluctuate in parallel with seasonal changes in exposure to ultraviolet B (UVB) light, this month of birth effect might reflect maternal end-of-winter deficiencies in vitamin D or in UVB itself. Taken together, these striking findings suggest that risk might be influenced in each of the periods of gestation, childhood, adolescence, and early adulthood. In addition to uncertainties regarding the exact timing of an exposure, it is unclear if exposure needs be discrete or prolonged. Since MS incidence peaks in early adulthood and then declines, risk cannot be determined by age-related mutations. Nevertheless, these data do not rule out a type of environmental imprinting, or that susceptibility (and resistance) could be entrained by cumulative exposures of (more than one) factors in the environment.
Vitamin D
It has become increasingly clear that vitamin D has a wide role in physiology and, importantly, also in disease. Evidence is mounting in support of vitamin D deficiency underlying risk for several autoimmune diseases. The pleiotropic actions of vitamin D, including immunomodulatory functions, lend strong support to the hypothesis that this hormone is important in the etiology of MS.
