Exposure to ultraviolet radiation (UVR) from sun exposure stimulates intradermal synthesis of vitamin D and is the principal natural source of this vitamin. Higher lifetime UVR has been associated with a decreased risk of multiple sclerosis (MS) in whites [1
], and recently we reported similar findings in blacks and Hispanics. However, higher serum 25-hydroxyvitamin D (25OHD) levels [1
] are associated with a reduced risk of MS only in whites. We and others have been unable to detect an association between serum 25OHD levels and MS risk in blacks [5
] or Hispanics [7
]. Whether this is because serum 25OHD is not a good indicator of vitamin D status in non-whites or due to the many vitamin D-independent immunological effects of UVR is unknown.
Blacks and Hispanics have lower 25OHD levels than whites, yet there is substantial evidence that low 25OHD levels in blacks do not lead to any excess ill health consequences [8
]. Recent studies of the general population have argued that total serum 25OHD levels are a poor indicator of vitamin D status in blacks and that bioavailable or ‘free’ 25OHD and/or its metabolites may be superior indicators [8
]. Blacks have different dominant common polymorphisms in the vitamin D-binding protein gene than whites that directly affect binding avidity and bioavailability of vitamin D metabolites. Vitamin D-binding protein (DBP), among other functions, is the main transporter of 25OHD to target tissues [10
]. Thus, it is biologically plausible that differences in these DBP polymorphisms could explain why we were unable to detect an association between 25OHD and MS risk in blacks and Hispanics.
The primary purpose of the MS Sunshine study was to test the vitamin D-MS hypothesis in blacks and Hispanics including variations in the gene that encodes DBP. The results presented herein address whether high serum 25OHD, independent of UVR, reduces the risk of MS in those individuals that carry the common DBP polymorphisms that are dominant in whites, regardless of race/ethnicity.
That common polymorphisms in the vitamin D-binding protein gene could modify the relationship between vitamin D and MS risk in blacks and whites is biologically plausible. Vitamin D-binding protein is the main transporter of 25OHD and its metabolites. SNPs in the gene encoding DBP at rs7041 and to a lesser extent rs4588 influence bioavailable fractions because they result in DBP isoforms that have different avidity for 25OHD and its metabolites [9
]. The dominant allele at rs7041 in whites is the minor allele in blacks. The resulting DBP isoform most commonly found in blacks has the highest avidity and is the most efficient transporter of 25OHD and its metabolites to target tissues [9
]. It is also associated with higher ‘free’ or bioavailable 25OHD levels which may explain why, despite quite low total 25OHD levels, most blacks are not physiologically vitamin D deficient [7
]. Following this line of reasoning, we hypothesized that these known racial differences in the prevalence of common genetic polymorphisms in the vitamin D-binding protein gene could explain why we and others [5
] have been unable to detect an association between 25OHD and MS risk in blacks.
Yet, the analyses presented herein demonstrate that racial/ethnic variations in bioavailable vitamin D resulting from differences in dominant polymorphisms in the vitamin D-binding protein gene do not explain the lack of association between 25OHD and MS in blacks and Hispanics. Taken together with our previously reported findings that higher levels of UVR exposure from sunshine are independently and consistently associated with lower MS risk across all 3 racial/ethnic groups [7
], these findings further challenge the biological plausibility of vitamin D deficiency as causal for MS.
While it may be tempting to cling to the idea that somehow vitamin D is causally linked to MS risk in whites but just not in blacks and Hispanics, it is difficult to conceive how a factor could be causal in only one racial/ethnic group. One could contend that the consistently protective effect of UVR across all three groups is still ultimately mediated by vitamin D, but that serum 25OHD is the wrong measure of vitamin D status in blacks and Hispanics. However, there are several problems with this line of reasoning: (1) the common polymorphisms we measured are in coding regions and directly influence the bioavailable fraction of 25OHD and its metabolites; (2) the prevalence of these common polymorphisms in whites and Hispanics are almost identical yet we cannot establish a relationship between MS risk and 25OHD in Hispanics. Furthermore, other studies of Hispanics with MS have reported that 25OHD levels do not differ between prevalent cases or controls [20
], or decline with increasing disability as has been shown in whites [21
]; (3) There is increasing evidence that UVR causes a plethora of immunological changes including effects that could influence autoimmune diseases independent of vitamin D. These include generation of T regulatory cells, B suppressor (regulatory) cells and production of immunosuppressive lipid mediators and alarmins [22
] as well as suppression of the animal model of MS (experimental autoimmune encephalomyelitis) [23
]. Lastly (4), the amount of UVR exposure, while consistently associated with MS risk, is strongly related to latitude of residence. It is important to remember that the vitamin D-MS hypothesis originated from the observation that the prevalence of MS is higher at latitudes further away from the equator where the intensity of UVR exposure is lower. However, UVR may not be the only reason for the observed latitude gradient as there are other potential MS risk factors that traditionally varied with latitude, including endemic infectious agents, food sources and genetic make-up of native inhabitants.
Our data are consistent with previous studies that examined the relationship between vitamin D-binding protein genotype and serum 25OHD levels in blacks, Hispanics and whites [6
]. We found similar distributions of rs7041 and rs4588 polymorphisms and similar associations of these genes with lower 25OHD levels across all three groups [7
]. Our findings are also consistent with the results of randomized controlled trials that have failed to demonstrate convincing health benefits of vitamin D supplementation [26
] in other diseases.
Intriguingly, we also found that the strong association of higher sun exposure and lower MS risk in blacks, Hispanics and whites is most pronounced in those individuals who carry at least one copy of the C allele at rs7041. This difference is most noticeable in blacks even though only 30.4% carry at least one C allele at this location. We think this stronger effect in blacks may be because they are less likely to use sunscreen than whites; thus blacks effectively get more sun exposure for the same amount of time outdoors at a particular ambient UVR than whites. We cannot exclude the possibility that this is a chance finding, as we did not detect a significant multiplicative interaction between UVR and rs7041 in the pooled cohort where only genetic ancestry and not self-reported race/ethnicity was considered. Rather, this association was uncovered when we examined the results of the stratified models that were run after detecting an interaction between 25OHD and rs7041 in the pooled cohort.
Alternatively, we speculate that the different DBP isoforms may have novel interactions with the immunomodulatory effects of UVR independent of vitamin D. Aside from transporting vitamin D, DBP also directly influences macrophage and neutrophil function and plays a role in tissue injury and repair [10
]. Only 5% of the binding sites on DBP are occupied by vitamin D, implying that this protein’s major role may be through other actions [10
]. UVR exposure results in immunomodulation through multiple mechanisms, many of which are independent of vitamin D. Such a non-classical interaction between DBP and UVR would be a more satisfying unifying explanation of our findings. Yet the evidence supporting such an interaction is lacking as no studies directly address the role of DBP in photoimmunology and very few studies have examined whether specific DBP genotypes are associated with differences in other DBP functions such as macrophage activation [10
If this finding can be replicated, it would have wide-reaching implications for the role of vitamin D in MS and other disease states as it implies that variation in the DBP genotype may affect MS risk through a vitamin D-independent pathway. Such a finding would invalidate the current instruments for Mendelian randomization studies of vitamin D because it violates the assumption that the genotypes affect the risk of disease only through influencing vitamin D status. This has important implications for the MS-vitamin D hypothesis, as Mendelian randomization studies are the strongest evidence to date to support a causal relationship between 25OHD and MS risk in whites. The main limitation of this study is that we were unable to identify an existing cohort of blacks or Hispanics with incident MS and sun exposure, 25OHD measurements and genotype in which these findings could be replicated. We also cannot exclude the possibility that we are underpowered to detect a protective association with 25OHD and MS in the small subgroups of white or Hispanic AA carriers and black carriers of AC/CC. Other limitations of this study include the case-control design necessitating that most 25OHD measures were obtained after symptom onset (although very close to the time of diagnosis). MS often results in sun avoidant behavior due to heat sensitivity which can cause an exaggerated
association between low 25OHD and MS particularly in prevalent cases. This most likely explains the observed lower 25OHD levels in a previous study of prevalent black MS cases compared to controls [29
] and may explain the strong association we found in whites, but does not explain the lack
of association we found in Hispanics and blacks. Likewise, multiple statistical tests would be expected to lead to false positive results rather than consistently negative findings. Data on sunscreen use over the life course were limited. Sunscreen is more likely to have been used by whites than other racial groups (since they are more sun sensitive), and regular and correct use of a high SPF sunscreen would reduce the effective UVR exposure of the skin. Thus, our effect estimates presented herein likely underestimate the protective effect of the received UVR dose in whites.
This study highlights how multi-ethnic studies can lead to novel insights into disease. Important racial variations in the gene encoding DBP do not explain the lack of association between 25OHD and MS risk in blacks and Hispanics, yet higher levels of UVR may be particularly important in protecting against the risk of MS in those carrying the DBP dominant in whites. Replication of our key findings should be addressed in future studies.