Abstract
Vitamin D deficiency is a major public health concern in the UK. As the natural sources of vitamin D in the UK are limited, supplementation or food fortification are possible strategies for achieving the dietary recommendations of 10 μg/d that will be introduced in 2016 for the whole population. However, there is controversy as to whether vitamin D2 and vitamin D3 are equally effective at raising vitamin D status (25OHD concentration). The primary and secondary aims of this PhD project were: to investigate the effects of both these forms of vitamin D independently on vitamin D status, markers of bone and cardiovascular health, and gene expression; as well as to examine whether common genetic variants affect response to either form of vitamin D. A cohort of 90 South Asian and 245 Caucasian women were recruited onto a randomised-controlled trial; the D2-D3 Study. Participants were given either 15 µg/d of vitamin D2, 15 µg/d of vitamin D3 or placebo, in fortified foods, for 12 weeks. At baseline, serum total 25OHD concentrations were significantly lower in the South Asian women (27.6 nmol/L) than the Caucasian women (60.3 nmol/L). In both the South Asian and Caucasian women, 25OHD concentrations significantly decreased in the placebo intervention (-5% and -15% respectively, p<0.001), and significantly increased in both the vitamin D2 (112% and 39% respectively, p<0.001) and the vitamin D3 interventions (243% and 72% respectively, p<0.001), with significantly greater increases seen in the vitamin D3 intervention (p<0.001). In the vitamin D3 groups, parathyroid hormone (PTH) concentrations decreased in the South Asian women (p<0.001), who had higher baseline concentrations, and were maintained in the Caucasian women, who had healthy baseline PTH concentrations. This effect was not seen with vitamin D2 fortification. Over the 12 weeks, there were no clinically relevant changes in blood lipid concentrations in response to either vitamin D2 or D3, in the South Asian and Caucasian women. Interestingly, whole blood transcriptome analysis indicated that the vitamin D2 and D3 interventions triggered a difference in expression of entirely different genes, and predicted therefore a difference in the activity of the respective metabolic and cellular pathways. The associations between genetic polymorphisms and change in 25OHD concentration in response to vitamin D also appear to differ depending on the form of vitamin D taken, although baseline 25OHD concentration may be a confounder. The implications of this work, as the largest RCT conducted to date and showing conclusively that vitamin D3 is more effective than vitamin D2 at raising total 25OHD concentration and achieving or maintaining a healthy PTH concentration, are important: in the clinical setting vitamin D3 may be preferable in the treatment of vitamin D deficiency. The novel findings that vitamin D2 and vitamin D3 lead to different metabolic/cellular responses requires further research to determine whether the response to vitamin D2 is due to a decrease in 25OHD3 concentration (observed in this study following vitamin D2 treatment) or whether it is in response to the increase in 25OHD2 concentration.