Review Assessment of evidence for a protective role of vitamin D in multiple sclerosis Heather E.C. Hanwell a,b,c,���, Brenda Banwell c,d a Department of Nutritional Sciences, Faculty of Medicine, University of Toronto, Toronto, Ontario, Canada b Department of Pathology and Laboratory Medicine, Mount Sinai Hospital, Toronto, Canada c Neurosciences and Mental Health, Research Institute, The Hospital for Sick Children, Toronto, Ontario, Canada d Division of Neurology, Department of Paediatrics, The Hospital for Sick Children, Toronto, Ontario, Canada a b s t r a c t a r t i c l e i n f o Article history: Received 17 November 2009 Received in revised form 5 July 2010 Accepted 26 July 2010 Available online 30 July 2010 Keywords: Vitamin D Multiple Sclerosis Hill's Criteria Evidence for a role of vitamin D insufficiency in determining risk in Multiple Sclerosis (MS) is supported by studies in both pediatric- and adult-onset patients. The potential role of vitamin D in modulating MS disease activity is an area of active clinical trials research, and the possibility of primary disease prevention with vitamin D supplementation in early life is an emerging concept. With Sir Austin Bradford Hill's criteria as a framework, the present review assesses the evidence for a causal relationship between vitamin D insufficiency and the pathobiology of MS, and discusses rationale for future clinical trials with vitamin D. �� 2010 Elsevier B.V. All rights reserved. 1. Introduction Although MS has been reported in most world regions, prevalence varies between different ethnic groups and across diverse geograph- ical regions, supporting both genetic and environmental contributions to MS biology [1���4]. The prevalence of MS is greater in areas with temperate rather than tropical climates, it increases with distance from the equator and is inversely associated with average ambient UVB [5���10]. The striking difference in prevalence of MS and some other autoimmune diseases as a function of latitude has implicated vitamin D status as a determinant of risk. The potential role of vitamin D in several autoimmune diseases, particularly MS, has been the subject of many manuscripts and reviews [11���30]. While it is best known for its role in calcium homeostasis and bone mineralization, vitamin D is also involved in modulating immune function and cell proliferation, differentiation, and apoptosis [31]. In vitro and animal models of immune cell behaviour and central nervous system inflammation have demonstrated a pro-inflammatory impact of vitamin D insufficiency and an anti-inflammatory role for vitamin D supplementation. At present, the totality of evidence for a protective role of vitamin D in MS has been deemed strong enough by some to warrant recommending vitamin D supplementation to people with MS and to individuals considered at high risk for MS [12]. Other investigators advocate large primary prevention population-based studies or randomized controlled Phase II and III studies in MS patients [19,26,32]. The present review will provide a brief outline of vitamin D metabolism, discuss the evidence for a causal relationship between impaired vitamin D status and MS and whether this evidence is sufficient to establish causality, and will propose concepts important in determining the therapeutic role for vitamin D in MS. 2. Vitamin D metabolism In humans, cholecalciferol (vitamin D3) is produced in the skin following exposure of 7-dehydrocholesteol to ultraviolet B (UVB) radiation. Vitamin D3 can also be obtained from the diet it is naturally present in oily fish and egg yolks and, in some countries, is added to foods such as milk, margarine, yoghurt, orange juice, and cereal. Estimating dietary intake of vitamin D is challenging for several reasons: Variation in mandatory fortification rules means that, between countries, different foods are fortified with varying amounts of vitamin D discretionary fortification results in only certain brands or types of those foods containing vitamin D in some countries and the amount of vitamin D naturally present in some foods may vary dramatically. For instance, natural vitamin D in animal-derived food products may vary with the season [33], the vitamin D content of the animals' diet [34], or other aspects of the animals' environment [33,35]. Vitamin supplements may contain either vitamin D3 or ergocalciferol (vitamin D2) and concentrations generally range from 50 IU in multivitamins to 1000 IU or more in products containing only Biochimica et Biophysica Acta 1812 (2011) 202���212 ��� Corresponding author. Mount Sinai Hospital, Department of Pathology and Laboratory Medicine, 600 University Ave., Toronto, Ontario, Canada M5G 1X5. Tel.: +1 416 586 4800, ext. 2726 fax: +1 416 586 8628. E-mail address: heather.hanwell@gmail.com (H.E.C. Hanwell). 0925-4439/$ ��� see front matter �� 2010 Elsevier B.V. All rights reserved. doi:10.1016/j.bbadis.2010.07.017 Contents lists available at ScienceDirect Biochimica et Biophysica Acta journal homepage: www.elsevier.com/locate/bbadis

vitamin D vitamin D2 is also present in some mushrooms, is added to some nut milks and is generally considered less bioactive than vitamin D3 [36���38]. Following either cutaneous synthesis or ingestion, vitamin D is transported to the liver bound to the vitamin D binding protein (VDBP, also known as group-specific component of serum or Gc-globulin) [39]. Vitamin D is metabolized to 25-hydroxyvitamin D3 [25(OH)D] by the hepatic cytochrome P450 mixed-function oxidases (CYP) CYP2R1 (microsomal) and CYP27A1 (mitochondrial) [40]. The concentration of the 25(OH)D metabolite in the serum represents vitamin D obtained from both UVB-catalyzed synthesis and diet, and is the accepted biomarker for vitamin D nutritional status [41,42]. The 25(OH)D metabolite is further hydroxylated by renal CYP27B1 to 1,25- dihydroxyvitamin D [1,25(OH)2D calcitriol], the most bioactive of the naturally derived vitamin D metabolites. Vitamin D signaling is mediated by calcitriol binding to the vitamin D receptor (VDR), which forms a nuclear heterodimer with the retinoid X receptor. This complex is capable of binding to genomic vitamin D response elements (VDRE), modulating expression of a variety of genes. Renal-derived calcitriol circulates bound to VDBP and acts as a potent hormone targeting bone, kidneys and the intestines to modulate calcium homeostasis. Numerous extra-renal tissues also activate vitamin D to calcitriol for local regulation of multiple biological processes including immunological recognition of self [43,44]. Calcitriol is regulated, in part, through a biofeedback loop in which the calcitriol-induced gene, CYP24A1, encodes an enzyme that initiates the catabolism and clearance of vitamin D-related metabolites via hydroxylation of carbon 24. 3. Assessment of evidence for vitamin D in MS In 1965, Sir Austin Bradford Hill proposed a set of viewpoints to aid in assessing the evidence for a causal relationship (Panel 1) [45]. Hill's criteria are arguably most appropriate for assessing evidence of causality under simplistic models of cause and effect whereby a specific outcome is attributed to a single causal agent. The criteria do not sufficiently capture the complexity of the relationship between causal complexes comprised of environmental and genetic risk factors that may be variably necessary or sufficient to induce a heterogeneous disease such as MS [46]. Nevertheless, the criteria do provide a generally well-rounded structure for a critical evaluation of evidence for causality. 4. Assessing the evidence for a relationship between vitamin D status and MS: The Bradford Hill criteria 4.1. Strength The strength of an association can be defined as the magnitude of difference in the risk, odds, or severity of a disease outcome based on variations in exposure to the factor of interest. A strong association supports a causal relationship between two entities. However, a weak association does not necessarily negate a causal relationship, particularly if the association occurs only in certain contexts. How strong are the links between MS and vitamin D status���as defined by circulating 25(OH)D���or determinants of vitamin D status such as dietary intake of vitamin D, or sun exposure? 4.1.1. Vitamin D status in utero Several studies have demonstrated a month of birth effect in MS cohorts. In Northern Sardinia���a region with very high MS incidence��� an excess of spring births was observed in MS cases (29.4%) relative to their unaffected siblings (22.1%, P=0.008) and to the general population (24.6%, P=0.036) [47]. Pooled month of birth data from MS patients in Canada, Denmark, Great Britain and Sweden (n=42,045) demonstrated an excess of MS cases born in May (odds ratio (OR) 1.10, 95% confidence interval (CI) 1.07 to 1.13) and fewer than expected births in November (OR 0.91, 95% CI 0.87 to 0.95) [48]. Overall, the risk of MS in those born in May was 13% higher than for those born in November (95% CI 5% to 22%). Given the low ambient sunlight in winter months in the countries studied, these results could be interpreted to suggest that low serum 25(OH)D during pregnancy or low vitamin D in the breast milk during first few months post-birth influence subsequent MS risk [49,50]. 4.1.2. Childhood sun exposure and MS risk Four studies have demonstrated that high sun exposure in childhood is related to a decreased risk of MS. In a case���control study (n=126 MS and 272 controls) from Tasmania, high sun exposure between the ages of 6 and 15 years was associated with a decreased risk of MS (OR 0.31, 95% CI 0.16 to 0.59) even after adjustment for skin pigmentation and smoking status prior to MS diagnosis [3]. Furthermore, the study also found that moderate-to high grade (grades 4���6) actinic damage, a marker for lifetime sun exposure, was independently associated with a decreased risk of multiple sclerosis (OR 0.32, 95% CI 0.11 to 0.88, adjusted for the same variables and sun exposure post-MS diagnosis). Similar findings were reported in Norway where increases in outdoor activities in early life, particularly at 16���20 years of age, were associated with decreased MS risk (OR 0.55, 95% CI 0.39 to 0.78) [51]. A North American study of 79 pairs of identical twins discordant for MS found that the unaffected twin reported more sun exposure during childhood than did the twin with MS: Each one-unit rise in the sun exposure index score (range ���9 to +9 0 indicating no sun exposure difference, 9 indicating more relative sun exposure compared to twin in each variable) was associated with an OR 0.75 (95% CI 0.62 to 0.90) [52]. Finally, a case���control study consisting of participants from Cuba, Martinique and Sicily���regions of varying latitudes, ambient UVR, and MS prevalences���also observed a consistently reduced risk of MS related to measures of sun exposure before age 15, and increased risk of MS related to sun protection practices before age 15 years of age [53]. For instance, in multivariate analyses, weekday sun exposure of ���1 h per day was associated with decreased MS risk (OR 0.90, 95% CI 0.85 to 0.98) while wearing pants when exposed to sunlight was associated with increased risk (OR 1.90, 95% CI 1.10 to 3.20). These four studies provide evidence supporting the hypothesis that sun exposure in childhood conveys protection against MS. Further support for the importance of sun exposure in childhood in determining MS risk also comes from studies investigating place of childhood residence, migration patterns, and ethnicity of MS popula- tions. Migration between areas of disparate MS prevalence before or during adolescence results in the individual adopting the risk of the new region. Migration in adulthood, however, does not influence MS risk [54���58]. In a study comparing the ancestry of pediatric and adult MS patients living in the same city, the pediatric MS patients were far more likely to be first generation Canadians, and to have parents born in world regions of low MS prevalence [59]. 4.1.3. Vitamin D status prior to MS diagnosis In a case���control study nested within a prospective cohort of over 7 million US military personnel, a decreased risk of MS (OR 0.38, 95% CI 0.19 to 0.75) was observed among white participants (148 cases, 296 controls) with serum 25(OH)D concentrations in the highest quintile (99.1���152.9 nmol/l) compared with the lowest quintile (b63.3 nmol/l) [60]. This paper will be discussed further below in the section on dose���response. 4.1.4. Vitamin D status at the clinical onset of MS The first clinical manifestation of MS presents with acute neurological deficits in vision, strength, balance, or sensation, typically associated with evidence for CNS inflammation in cerebro- spinal fluid (oligoclonal bands) and on brain imaging [61]. This first attack of demyelination can also represent a monophasic illness 203 H.E.C. Hanwell, B. Banwell / Biochimica et Biophysica Acta 1812 (2011) 202���212