Introduction

Multiple sclerosis (MS) is a T-cell mediated autoimmune disease, causing central nervous system (CNS) damage and a progressive decline in neurological function. The etiopathogenesis of MS is poorly understood, but it is widely accepted to be a multifactorial disease. A genome wide study reported in 2007 confirmed that HLA class II DRB1*1501 is associated with the disease, as well as two other genetic associations—to Interleukin-7 receptor-α chain (IL7RA 5p13) and interleukin-2 receptor-α chain (IL2RA 10p15) [1]; although the relatively low odds ratio for this association, as well as only 24 % concordance rate between monozygotic twins [2] suggest a crucial role of environmental factors in the development of MS.

A number of environmental factors have been suggested, including: month of birth, vitamin D levels, smoking and viral infections [3], but the most influential environmental factor that was found is geographic latitude—at birth and early adulthood [4, 5]. This gives rise to the hypothesis that climate, and perhaps seasonality, play an important role in the pathogenesis of MS.

The purpose of this review is to assess this association in terms of disease onset and activity, and to evaluate which season-dependent factors play a role in it (Table 1).

Table 1 Studies on seasonal factors and multiple sclerosis
Full size table

Month of birth

The effect of month of birth was first proposed in 1987, as a part of a study examining different neurological diseases. It suggests that those who were born in winter months have a reduced risk of developing MS, whilst people born in spring are at a higher risk [6]. An association between month of birth and MS was also found in a large-scale population based study conducted in 2005; the association was stronger in familial cases of MS, indicating a possible epigenetic influence during gestation [7]. This finding was supported by a case-control study, that found the effect of month of birth to be a more pronounced in MS patients with HLA-DRB1 genotype [8]. The month of birth also had an effect on the phenotype of the disease (primary progressive MS vs relapsing–remitting MS) [9].

It has been hypothesized that low maternal vitamin D levels, due to reduced sun exposure in the winter months, plays a significant role in this phenomenon. Reduced vitamin D levels may affect the immunological development of the fetus in a critical time during pregnancy, resulting in an increased risk of MS [10]. A recent meta-analysis and systematic review found that the month of birth effect was latitude dependent, and only significant in places that were over 52°N [11]. It was previously suggested that at this latitude, insufficient ultraviolet B (UVB) radiation reaches the skin between October and March, leading to inadequate vitamin D synthesis during this time [12, 13]. One significant limitation of the previously discussed meta-analysis was that it only included studies in the northern hemisphere. Demonstration of an inverse relationship between month of birth and MS risk in similar latitudes of the southern hemisphere would significantly strengthen its conclusion [11].

If the month of birth effect is related to maternal vitamin D levels as mentioned above, it is plausible that the effect would be eliminated by the use of vitamin D supplementation during pregnancy. A retrospective study is needed to investigate the impact of maternal vitamin D supplementation the effect of the month of birth on MS incidence, and to establish the underlying mechanism. This could potentially lead to the use of antenatal vitamin D supplementation in those with a strong family history of MS, as the effect of month of birth is strongest in familial cases [8].

Whilst maternal vitamin D levels are the most widely accepted explanation of the month of birth effect, other factors that may play a role, include: perinatal infections, climate and variations in lifestyle.

Not all investigators agree with this conclusion. A study by Fiddes et al. [14] suggests that the association between MS and the month of birth is only mediated by several confounding variables, such a year and place of birth.

Vitamin D levels

Vitamin D is synthesized in the human body on response to exposure to sunlight, and specifically UV radiation [15]; therefore, it is easy to infer that vitamin D levels are higher during the summer months, as mentioned above [13].

Vitamin D has been previously associated with numerous autoimmune diseases, such as systemic lupus erythematosus, type I diabetes, inflammatory bowel diseases, rheumatoid arthritis, and of course, with MS [16–19].

The BENEFIT trial showed an inverse relationship between vitamin D levels and MS disease activity, progression, new active lesions, yearly increase in T2 lesion volume and yearly loss in brain volume. This study was performed in patients presenting with their first MS event, although the safety and efficacy of vitamin D therapy for relapsing remitting MS patients is still under clinical trial [20, 21]. Similar results, in regard to low levels of vitamin D and high disease activity and progression were found in another study [20]. Munger et al. [22] have demonstrated a protective effect of vitamin D intake on the risk for developing MS. Norway has low UV radiation exposure, since its latitude is 58°–71°N, which should expose its residence to high risk for MS. In a research done there, reduced incidence of MS was found in coastal fishing villages compared to inland farming areas. The interpretation of this finding was based on the assumption that coastal inhabitants have increased dietary vitamin D intake [23, 24]. This shows that dietary vitamin D can modulate the latitude-dependent risk of MS in areas with low exposure to the sunlight.

It could be concluded that vitamin D plays a crucial role as a seasonal factor of MS, although it seems that the seasonality effect could not be attributed to vitamin D alone.

Infections

Many infectious diseases have seasonal pattern of outbreaks [25, 26]; these patterns are due to seasonal changes in both the pathogens and in the host [26].

Infections have been noted to increase levels of cytokine IFN-γ [27, 28], which has a significant role in MS disease exacerbation [29–31]. Sibley et al. showed minor infections, particularly those of viral origin, frequently encountered in the winter, precipitated disease exacerbation. Nearly 30 % of MS attacks were associated with either preceding or concurrent infection. Infections were mainly of the upper respiratory and gastrointestinal systems, although no particularly antigen was isolated. Bacterial infections, in particular urinary tract infections, were not significantly documented as a disease trigger [32]. Other investigators have demonstrated the relationship between infection and exacerbation since [33–35].

Edwards et al. [35] detected a significant increase in the volume of gadolinium-diethylenetriaminepentaacetic acid (Gd-DTPA) enhancing lesions in MRI scans at the time of serologically confirmed viral infections compared to non-infection periods. The area and volume of Gd-DPTA enhancing lesions has been associated with MS disease exacerbation although the quantitation of such lesions remains undetermined [35–38]. Such findings suggest a positive connection between viral infection and MRI evidence of the destruction of the blood-brain-barrier. This could result from a viral-induced inflammatory response affecting the CNS [35].

Melatonin

Melatonin is a hormone made in the pineal gland, and it is responsible for the circadian rhythms of physiological functions. Its secretion changes according to daylight time and night length, and therefore, it changes with the seasons [39]. In recent years, there is increasing evidence that Melatonin has a great influence on the immune system, and more specifically, it seems to enhance the release of T-helper type 1 (Th1) cytokines, and might also influence Th2 cells [40, 41]. Melatonin’s influence on the pathogenesis of autoimmune diseases in general, and of MS in particular, is still controversial [42]. Polymorphisms in genes related to the melatonin pathways were found to be associated with progressive subtypes and higher disability status in Finnish MS patients [43]. Evidence for the role of melatonin as an immunologic enhancer, and its adverse effect on MS can also be found in a study that tested the impact of treatment with luzindole, a melatonin receptor antagonist; Luzindole treatment prevented the development of experimental autoimmune encephalomyelitis (EAE) in mice [44].

Evidence for the ameliorating effect of melatonin on MS can be found in a study by Farez et al. [45] in which investigators showed melatonin levels were inversely correlated with MS relapses in humans. In animal models of EAE, it was shown that melatonin treatment could improve the disease’s clinical symptoms. In another study, investigators demonstrated that oral administration of melatonin reduced the clinical severity and spinal cord inflammatory infiltration in mice with EAE [46].

Sandyk et al. investigated the association between nocturnal melatonin levels and the presence of pineal gland calcification during MS exacerbations. Abnormal melatonin levels, mostly low nocturnal levels, were found in 52.0 % of patients. Melatonin levels were found to be positively correlated with the age of onset of symptoms of the disease (p < 0.0001) and negatively correlated with the disease’s duration (p < 0.05). Pineal gland calcification was found in 96 % of subjects, suggesting a possible role of pineal failure in the disease [47].

Thereby, melatonin could be one of the factors that mediated the association between MS and seasonality, and its role in the disease should be studied further.

Seasonality and disease activity

Previous studies have demonstrated seasonal variations in disease activity (Table 1). These variations have been investigated using three main parameters of disease activity: clinical relapses, magnetic resonance imaging (MRI) findings and immunoregulatory cytokine production.

Seasonality and clinical MS relapses

Several studies have examined the seasonal variation in relapse rates, and have reported conflicting results. In 2000, Jin et al. performed a meta-analysis examining the relationship between season and mono-symptomatic optic neuritis (a known symptom of MS [48]), MS onset and MS exacerbations. The researchers concluded that there is a peak in incidence during spring and a nadir during winter months in the northern hemisphere in all three parameters, which was highest with respect to MS onset (45 % with 95 % CI 36–55 %), and lowest with respect to its relapses (10 % with 95 % CI 7–13 %) [49]. A large, global multi-center study conducted in 2014 by Spelman et al. further demonstrated a latitude dependent effect. This study involved 8411 patients in 31 centers in the northern hemisphere, and 1400 patients in 15 centers in the southern one, with a total of 32,762 MS relapses analyzed. They found a significant difference of 7.62 % between peak and trough relapse difference per 100 patients, and the mean lag time between UV radiation trough and relapse onset peak was 2.72 months, for both northern and southern hemispheres. The relationship between lag time and latitude was also assessed, and it was found that with each change of 10° in latitude (away from the equator) the lag time was reduced by a mean of 28.5 days [50].

The variation according to latitude may be mediated through the UVR exposure, and its immunomodulatory effect and it is most likely due to vitamin D synthesis. However, this raises concerns because does not fully explain the high rate of exacerbations during the spring months, as vitamin D levels are lowest during winter time [13]. Inhabitants of countries further from the equator are known to have lower vitamin D levels in general [51], and thus are more likely to reach a vitamin D deficiency sooner after the UVR deprivation. In 2010, Simpson et al. reported an inverse relationship between vitamin D status and the likelihood of MS relapse [52]. An Australian study concluded that vitamin D levels and sun exposure are independent risk factors for CNS demyelination [53].

Seasonality and brain lesions in MRI

MRI is a useful technique to evaluate the effect of seasonal factors on MS, as it is an objective measure of sub-clinical disease activity in the CNS. Unfortunately, its use is limited by its high cost and therefore studies using MRI data are scant and typically have small sample sizes.

In 2000, Auer et al. [54] demonstrated a sinusoidal variation in the number of new contrast enhancing lesions (CEL), similar to the patterns described in clinical relapse rates. However, two studies published the following year independently concluded that no seasonal variation was observed using MRI, one followed 120 MS patients for 10 months while the other followed 13 patients for 9 months [55, 56]. The lack of an acceptable inclusion criteria and poor longitudinal follow up has cast doubt over the validity of the results of previous studies.

More recently, an observational cohort study concluded that there was a strong seasonal pattern of MS activity observed using non-contrast brain MRI. Although only 44 patients were included, its design and control for confounding variables, such as treatment, strengthened its validity. A total of 939 scans were used, with a median of 22 scans per patient, and it concluded there was a two-three-fold increase in likelihood of new lesions as well as increased disease activity during the spring-summer months [57].

Mowry et al. [58] reached a conclusion that vitamin D levels are inversely associated with MS activity on brain MRI, using scans of 469 subjects, and suggested this could possibly explain the effect of seasonality. Zivadinov et al. used MRI scans, serum vitamin D metabolite levels and questionnaires of 264 MS patients in order to explore the relationship between sun exposure, vitamin D and MRI documented disease activity. They concluded that sun exposure has a direct effect on MRI measures of neurodegeneration (measured by changes in brain volume and grey matter volume), independently of vitamin D levels [59]. Although this study does not examine seasonal variation of vitamin D or UVR, but rather sun exposure throughout the patient’s life, we have included it in this review, as it demonstrates the existence of an effect of UVR that is not mediated by vitamin D, and thus highlights a need for further research in order to identify this factor.

Seasonality and immunoregulatory cytokines

The CNS damage and progressive neurological decline characteristic of MS is thought to be mediated by the Th1 immune response, of which interferon γ (IFN-γ) is the hallmark cytokine. Levels of IFN-γ are naturally elevated during the inflammatory process of the nervous system via Interleukin-12 (IL-12), a pro-inflammatory cytokine. The CD40–CD40 ligand interaction between antigen presenting cells and T-cells leads to production of IL-12 and subsequently secretion of IFN-γ. IFN-γ and IL-12 are both also released by peripheral blood mononuclear (PBMN) cells [29]. In progressive MS patients, increased levels of IFN-γ have been found to precede clinical attacks and treatment of these patients with recombinant IFN-γ induces disease exacerbation [29–31].

Balashov et al. studied the relationship between levels of immunoregulatory cytokines in progressive MS patients and the time of year. They, amongst other studies such as Killestein et al., found significantly increased IFN-γ production in these patients during autumn and winter months when compared to spring and summer months, as well as when compared to IFN-γ levels of healthy patients during autumn and winter months [29, 30]. The apex of these increased levels occurred from October to December. Expanded Disability Status Scale (EDSS—validated evaluation of neurologic impairment in MS [60]) scores of progressive MS patients seen during this period worsened by 0.03 more compared to those patients seen during the March to August period. IL-12 was studied as well to determine its role in IFN-γ secretion. With neutralization of IL-12, IFN-γ production significantly decreased in progressive MS patients during the winter and autumn months. There was no change in IFN-γ levels with IL-12 neutralization during the spring and summer months for progressive MS patients or for healthy controls [29].

Although the mechanism for this seasonal effect is unknown, it is suggested that environmental factors, such as UVR, act as the inciting causes for disease onset and exacerbation. UVR exposure is positively correlated to vitamin D activation, and vitamin D inhibits IFN-γ -therefore increased UVR exposure leads to decreased IFN-γ levels [29, 61]. Vitamin D supplementation was also previously associated with increased levels of transforming growth factor beta 1 (TGF-β1) [62].

Conclusion

MS is an autoimmune multifactorial disease whose outbreak and course are influenced by genetic and environmental factors. One important environmental factor is seasonality. The effect of seasonality itself could be explained by various factors, including vitamin D levels, melatonin, infectious diseases and the month of birth, which is most widely explained by maternal vitamin D levels as described above.

Seasonal variation on MS disease’s course has been demonstrated in various studies, and seems to have an effect on clinical manifestations of the disease, MRI brain lesions, and inflammatory cytokine levels.

Low vitamin D levels are associated to most, but not all of the environmental factors mentioned. This review article supports the theory, past research and present research of vitamin D as an immunomodulator in MS, but also of seasonality as an independent factor in the disease.