Vitamin D-tour : cognition and depression: the role of vitamin D and its interplay with glucose homeostasis

According to recent estimations approximately 35.6 million people have dementia worldwide. Globally, 350 million people experience one or more depressive episodes during their life. As the therapeutic options for dementia and depression are limited, these conditions form a major challenge for public health and society. More and more researchers have initiated research on potential preventive factors for dementia and depression, including the potential effects of nutritional factors. The aim of this PhD-thesis was to study the role of vitamin D and its potential interplay with glucose homeostasis, in the development of cognitive decline and depression, using epidemiological data as well experimental animal data. <strong>Chapter 2</strong> recapitulates a debate between vitamin D experts that was organized to make a step towards the harmonization on the formulation of optimal vitamin D intake levels and serum 25(OH)D concentrations across Europe. It was concluded that based on the current evidence-base 25(OH)D concentrations ≥50 nmol/L are sufficient with respect to optimal bone health. For health outcomes beyond bone health evidence was considered insufficient to formulate optimal levels. In order to achieve and maintain a 25(OH)D concentration ≥50 nmol/L, older adults aged ≥65 years were recommended to adhere to a vitamin D intake of 20 μg/day. <strong>Chapter 3</strong> shows that there is a high prevalence of 25(OH)D inadequacy in a population of Dutch older adults that participated in the B-PROOF study (n=2857), namely 45% had 25(OH)D concentrations <50 nmol/L. Mean vitamin D intake was 4.9±2.9 µg/day and only 20% of the participants reported to use vitamin D containing supplements. Exploration of the determinants of 25(OH)D status showed significant associations between vitamin D ‘raising’ SNPs (n=2530), higher sun exposure (n=1012), vitamin D intake (n=596) and higher 25(OH)D concentrations. Including all the potential relevant predictors in one model explained 35% of the variance in 25(OH)D status (R2=0.35). In <strong>chapter 4</strong> the associations between 25(OH)D status and global cognitive performance (n=116), depressive symptoms (n=118), and surrogate markers of glucose intolerance (n=593) were evaluated using data of European adults aged 70-75 years. None of the associations reached significance. Studying the potential role of vitamin D in domain-specific cognitive performance and depression in 127 Dutch pre-frail and frail older adults aged ≥65 years (<strong>chapter 5</strong>), showed an association between 25(OH)D concentration and executive functioning, and a tendency towards an association with information processing speed. Stratification for ‘low’ and ‘high’ fasting glucose concentrations did not suggest an interaction between vitamin D and glucose homeostasis in the association with domain-specific cognitive performance. Moreover, adding fasting glucose or insulin did not substantially influence the associations between 25(OH)D status and domain-specific cognitive performance, and hence a mediation effect of glucose homeostasis was considered unlikely. We furthermore observed associations of 25(OH)D status with attention and working memory (n=787) (<strong>chapter</strong> <strong>6</strong>), depression (n=2839) (<strong>chapter</strong> <strong>7</strong>) and grey matter volume of the brain (n=217) (<strong>chapter</strong> <strong>8</strong>) in a population community-dwelling Dutch older adults aged ≥65 years. Again, these studies did not provide evidence that the associations were modified or mediated by glucose intolerance. However, it should be emphasized that glucose intolerance in these three chapters was defined sub-optimally, specifically using blood samples that may have been collected in a non-fasting state, or by using self-reported diabetes data. Hence, the mediation and interaction effects should be interpreted cautiously. Finally, <strong>chapter 9</strong> shows the results of a proof of principle study on the effect of a long-term vitamin D deficiency on cognitive decline and emotional reactivity in old C57BL/6j mice. Modest tendencies were shown for a relation between vitamin D and spatial learning, but these tendencies did not reach significance. Vitamin D deficiency did not affect recognition memory, spatial memory or emotional reactivity. Mice that received a higher dietary fat load, which was given to induce an impaired glucose tolerance, did not respond differently to a vitamin D deficiency than mice that received a low fat diet did. Overall, it is concluded that the evidence for an effect of vitamin D on cognitive performance/decline, depression or brain volume is insufficient to formulate disease specific cut-off values for vitamin D intake or 25(OH)D status. However, given the high prevalence of 25(OH)D concentrations <50 nmol/L we do call for a more active promotion of the current vitamin D intake recommendations.