The vitamin D system

The two main forms of vitamin D are vitamin D₂ (ergocalciferol) and vitamin D₃ (cholecalciferol)⁽Reference Windaus, Linsert and Lüttringhaus^24, Reference Windaus, Schenck and von Weder²⁵⁾; however, vitamin D₃ is the only form that is found naturally in human subjects and other animals. Vitamin D₃ is synthesised in the skin from 7-dehydrocholesterol through exposure to UVB irradiation⁽Reference Windaus, Schenck and von Weder²⁵⁾. The resulting pre-vitamin D (precholecalciferol) is converted to vitamin D₃ (cholecalciferol) via thermal isomerisation. Although the main source of vitamin D₃ is through endogenous synthesis in the skin, the vitamin can also be obtained from the diet and this is important for those who have limited exposure to the sun. The main dietary sources of vitamin D include oily fish, egg yolks and fortified milk. Vitamin D₃, whether derived from sunlight or the diet, enters the circulation bound to vitamin D-binding protein and is transported to the liver. Vitamin D₃ is hydroxylated in the liver to form 25-hydroxycholecalciferol (25(OH)D₃), the major circulating vitamin D metabolite. 25(OH)D₃ is then further hydroxylated by a 1α-hydroxylase enzyme (gene: CYP27B1), and this occurs primarily in the kidney to produce 1α,25-dihydroxycholecalciferol (1,25(OH)₂D₃), the biologically active form of vitamin D. In vivo studies have shown that the catabolism of vitamin D and its metabolites occurs mostly in the liver through a variety of cytochrome P450 enzymes that produce a number of catabolites⁽Reference Masuda, Byford and Arabian^26, Reference Xu, Hashizume and Shuhart²⁷⁾. The 24-hydroxylase (gene: CYP24), a mitochondrial cytochrome P450-containing enzyme, catalyses several steps in 1,25(OH)₂D₃ degradation through 24-hydroxylation and the formation of calcitrioic acid⁽Reference Jones, Strugnell and DeLuca²⁸⁾.

Vitamin D receptor

The action of 1,25(OH)₂D₃ is mediated through the vitamin D receptor (VDR), a member of the nuclear receptor superfamily, which regulates the transcription of many target genes⁽Reference Demay^29, Reference Mangelsdorf, Thummel and Beato³⁰⁾. VDR binds to 1,25(OH)₂D₃ with high affinity and specificity, which then heterodimerises with a retinoid X receptor⁽Reference Issa, Leong and Eisman³¹⁾. Once the heterodimer binds with vitamin D response elements in target genes, a genomic response is generated⁽Reference Carlberg^32, Reference Sutton and MacDonald³³⁾. In addition, there is also a plasma membrane VDR which mediates the acute, rapid actions of 1,25(OH)₂D₃⁽Reference Huhtakangas, Olivera and Bishop³⁴⁾. VDR has been identified in most human tissues, including in osteoblasts, skin keratinocytes, macrophages, smooth muscle, pancreatic β-cells and epithelial cells⁽Reference Norman^35, Reference Plum and DeLuca³⁶⁾. The ubiquitous expression of VDR may underlie the diverse effects of vitamin D and provide a mechanistic basis for the link between vitamin D deficiency and a number of disorders, including certain types of cancer, inflammatory bowel disease, CVD, diabetes (type 1 and type 2) and the metabolic syndrome⁽Reference Osei^11, Reference Plum and DeLuca^36–Reference Yiu, Chan and Yiu³⁹⁾.

Vitamin D deficiency and obesity

Clinically, vitamin D status is normally assessed by measurement of the serum level of 25(OH)D₃, the major form of vitamin D in the circulation, with a half life of between 15 and 50 d^(40–Reference Holick⁴²⁾. Vitamin D deficiency has been defined as a 25(OH)D₃ level of < 20 ng/ml or < 50 nmol/l⁽Reference Holick^43, Reference Dawson-Hughes, Heaney and Holick⁴⁴⁾. Studies on vitamin D status have suggested that there is a link between vitamin D deficiency and obesity as obese individuals tend to have low serum levels of 25(OH)D₃⁽Reference Goldner, Stoner and Thompson^17, Reference Khandalavala, Hibma and Fang^45, Reference Fish, Beverstein and Olson⁴⁶⁾. Serum 25(OH)D₃ levels are found to be inversely correlated with measures of obesity, including BMI, fat mass and waist circumference⁽Reference Ford, Ajani and McGuire^47–Reference Jorde, Sneve and Emaus⁵¹⁾. Furthermore, a negative association between 25(OH)D₃ and visceral fat has been demonstrated in African-Americans with diabetes⁽Reference Freedman, Wagenknecht and Hairston⁵²⁾. It is suggested that lower levels of circulating 25(OH)D₃ in obese individuals could be due to greater sequestration by adipose tissue, reducing its availability in the circulation⁽Reference Wortsman, Matsuoka and Chen^13, Reference Blum, Dolnikowski and Seyoum¹⁸⁾. There is evidence that increased dietary vitamin D intake and elevated serum 25(OH)D₃ are related to lower visceral adiposity and omental adipocyte size in women undergoing gynaecological surgery⁽Reference Caron-Jobin, Morisset and Tremblay⁵³⁾. In a recent double-blind, placebo-controlled trial, dietary supplementation with Ca and vitamin D for 16 weeks was associated with a beneficial reduction of visceral fat in overweight and obese adults⁽Reference Rosenblum, Castro and Moore⁵⁴⁾.

The relationship between obesity and the active form of vitamin D, 1,25(OH)₂D₃ (which has a short half life of approximately 15 h residence in the circulation⁽Reference Jones⁴¹⁾), is less clear, and this is probably due to the dynamic nature of the production and regulation of the active hormone. However, in healthy adults, low serum 1,25(OH)₂D₃ levels are associated with higher BMI and body fat mass⁽Reference Parikh, Edelman and Uwaifo^14, Reference Frost, Abrahamsen and Nielsen^55, Reference Kayaniyil, Vieth and Harris⁵⁶⁾. Similarly, data from a study of 2000 obese subjects have shown a negative correlation between BMI and serum 1,25(OH)₂D₃. Taken together, the present evidence strongly suggests that the vitamin D system is altered in obese subjects and this may have implications both for the development of obesity itself and of its co-morbidities.

Vitamin D system in adipose tissue

Recent studies suggest that the key components of the vitamin D system are evident in adipose tissue, and vitamin D could be involved in the function of the tissue⁽Reference Li, Byrne and Chang^21, Reference Narvaez, Matthews and Broun^57, Reference Ochs-Balcom, Chennamaneni and Millen⁵⁸⁾. Expression of the CYP27B1 gene, which encodes the enzyme converting 25(OH)D₃ to 1,25(OH)₂D₃, has been reported in mouse 3T3-L1 preadipocytes and in adipose tissue of Wistar rats⁽Reference Li, Byrne and Chang²¹⁾. Recent studies from our group and others have shown that CYP27B1 is expressed by Simpson–Golabii–Beymel Syndrome human adipocytes and preadipocytes⁽Reference O'Hara, Lim and Mazzatti^59, Reference Trayhurn, O'Hara and Bing⁶⁰⁾ and by human mammary adipocytes⁽Reference Ching, Kashinkunti and Niehaus¹⁹⁾. The CYP24 gene, which encodes the enzyme catalysing 1,25(OH)₂D₃, is also found to be expressed by murine 3T3-L1 adipocytes and by human adipocytes and preadipocytes⁽Reference Ching, Kashinkunti and Niehaus^19, Reference Li, Byrne and Chang^21, Reference O'Hara, Lim and Mazzatti^59, Reference Trayhurn, O'Hara and Bing⁶⁰⁾. Thus, adipocytes could be involved in the local synthesis as well as degradation of biologically active vitamin D.

Although expression of the VDR gene has been reported in mouse white and brown fat and in 3T3-L1 adipocytes⁽Reference Kamei, Kawada and Kazuki^23, Reference Querfeld, Hoffmann and Klaus⁶¹⁾, little is known on whether human adipose tissue expresses VDR. We have, however, recently observed that the VDR gene is expressed by human adipose tissue (subcutaneous and visceral) (unpublished results) as well as by human fat cells in culture (preadipocytes and differentiated adipocytes)⁽Reference O'Hara, Lim and Mazzatti^59, Reference Trayhurn, O'Hara and Bing⁶⁰⁾. Importantly, the bioactivation of 25(OH)D₃ to 1,25(OH)₂D₃ has been demonstrated very recently in human mammary adipocytes, together with the release of 1,25(OH)₂D₃⁽Reference Ching, Kashinkunti and Niehaus¹⁹⁾. The basal and substrate-induced 1,25(OH)₂D₃ secretion by mature adipocytes is reported to be higher than with human mammary epithelial cells; these cells are known to be highly efficient at bioactivating 25(OH)D₃. This finding suggests that mature adipocytes are capable of taking up 25(OH)D₃, converting it to 1,25(OH)₂D₃ and then releasing the biologically active hormone to the adjacent microenvironment – and possibly to the circulating pool (nevertheless, the kidney is the major source of circulating 1,25(OH)₂D₃).

Macrophages are known to play a role in vitamin D metabolism, with the ability to convert circulating 25(OH)D₃ to 1,25(OH)₂D₃⁽Reference Bikle⁶²⁾. Adipose tissue expansion in obesity is associated with an increase in macrophage accumulation in the tissue⁽Reference Xu, Barnes and Yang^63, Reference Weisberg, McCann and Desai⁶⁴⁾, which may facilitate the local hydroxylation of 25(OH)D₃. However, the specific characteristics of adipose tissue macrophages and their potential contribution to the local production of 1,25(OH)₂D₃ need to be clarified.

Data on the actual 1,25(OH)₂D₃ content of human adipose tissue are very limited. In a small study of seventeen morbidly obese subjects undergoing gastric bypass surgery, the 1,25(OH)₂D₃ concentrations measured by liquid chromatography–MS (LC/MS) were much higher (>10-fold) in abdominal subcutaneous fat than in serum⁽Reference Blum, Dolnikowski and Seyoum¹⁸⁾. Whether there is a rise in 1,25(OH)₂D₃ levels in adipose tissue, either by increased uptake or through local conversion in the obese state, is not known. Collectively, these findings suggest that human adipose tissue could well be a target for vitamin D, through endocrine as well as autocrine/paracrine actions of the hormone (Table 1).

Table 1 Vitamin D-related gene expression identified in adipose tissue and adipocytes

VDR, vitamin D receptor; SGBS, Simpson–Golabii–Beymel syndrome.

Role of vitamin D in adipogenesis

Vitamin D and its receptor VDR have been implicated in the modulation of preadipocyte differentiation into adipocytes (adipogenesis)⁽Reference Blumberg, Tzameli and Astapova⁶⁵⁾. The differentiation of 3T3-L1 (and other) preadipocytes is a highly controlled process through sequential induction of transcription factors that regulate the expression of adipocyte-specific markers. During adipogenesis, a series of cellular events begins with the rapid expression of CCAAT/enhancer-binding protein β (C/EBPβ), followed by the expression of C/EBPα, PPARγ and sterol-regulatory element-binding protein 1 (SREBP1)⁽Reference Mandrup and Lane⁶⁶⁾. As a result, there is increased expression of genes that produce the adipocyte phenotype, such as lipoprotein lipase, and adipocyte lipid-binding protein 2, which serves as a late marker of adipogenesis⁽Reference Christy, Yang and Ntambi^67, Reference Dani, Amri and Bertrand⁶⁸⁾. During differentiation, the expression of genes encoding lipogenic enzymes such as fatty acid synthase is highly induced and de novo fatty acid synthesis increases enormously⁽Reference Madsen, Petersen and Kristiansen⁶⁹⁾.

There is some evidence that 1,25(OH)₂D₃ inhibits 3T3-L1 preadipocyte differentiation in a dose-dependent manner, and this is in line with its inhibitory effect on the expression of adipogenic transcription factor (C/EBPβ, PPARγ and SREBP1) genes and of the downstream adipocyte markers (lipoprotein lipase, adipocyte lipid-binding protein 2 and fatty acid synthase), although 1,25(OH)₂D₃ does not block the induction of C/EBP⁽Reference Kong and Li²²⁾. The linkage between 1,25(OH)₂D₃ and adipocyte lipogenesis has also been supported by a study which demonstrated that the hormone strongly increased mRNA levels of insulin-induced gene-2 (Insig-2), a factor which blocks fatty acid synthesis in mature 3T3-L1 adipocytes and inhibits preadipocyte differentiation⁽Reference Lee, Lee and Choi⁷⁰⁾. During the differentiation of human mammary preadipocytes, exposure to 25(OH)D₃ or 1,25(OH)₂D₃ led to a significant reduction in lipid accumulation at day 7 but not at day 14, suggesting that vitamin D metabolites may inhibit the initiation of human preadipocyte differentiation⁽Reference Ching, Kashinkunti and Niehaus¹⁹⁾. Furthermore, in addition to reducing protein expression of C/EBPα, PPARγ and adipocyte lipid-binding protein 2 by 1,25(OH)₂D₃ alone, the combination of 1,25(OH)₂D₃ with genistein enhanced suppression of adipocyte lipid-binding protein 2 expression and lipid accumulation in 3T3-L1 adipocytes⁽Reference Rayalam, Della-Fera and Ambati⁷¹⁾. The effects of 1,25(OH)₂D₃ metabolites on adipogenesis may involve VDR, as 1,25(OH)₂D₃ combined with genistein significantly increased VDR protein expression⁽Reference Rayalam, Della-Fera and Ambati⁷¹⁾.

VDR has been shown to be expressed at the early stage of adipogenesis in 3T3-L1 cells and its expression levels are maintained by 1,25(OH)₂D₃ during the course of adipocyte differentiation⁽Reference Blumberg, Tzameli and Astapova⁶⁵⁾. In the presence of 1,25(OH)₂D₃, VDR inhibits adipogenesis by reducing C/EBPβ mRNA and C/EBPβ nuclear protein levels at a critical stage of differentiation⁽Reference Blumberg, Tzameli and Astapova⁶⁵⁾. In addition, 1,25(OH)₂D₃ induces the up-regulation of C/EBPβ core-repressor, eight twenty-one (ETO), which would further restrain the activity of remaining C/EBPβ⁽Reference Blumberg, Tzameli and Astapova⁶⁵⁾. A recent study has shown a positive association between VDR polymorphisms and the parameters of adiposity⁽Reference Ochs-Balcom, Chennamaneni and Millen⁵⁸⁾. VDR gene variants with polymorphisms on the 3′ UTR site, which affect the expression of VDR, are postulated to suppress the anti-adipogenic effect of vitamin D⁽Reference Ochs-Balcom, Chennamaneni and Millen⁵⁸⁾. Interestingly, a role for unliganded VDR in adipogenesis has been proposed, as VDR overexpression suppresses 3T3-L1 preadipocyte differentiation in the absence of 1,25(OH)₂D₃⁽Reference Kong and Li²²⁾. In contrast, the data from another study suggest that the unliganded VDR is required for lipid accumulation, as VDR knockdown with siRNA delays and prevents this process⁽Reference Blumberg, Tzameli and Astapova⁶⁵⁾. However, in vivo studies on VDR function suggest that VDR could promote adipogenesis. Mice with a global VDR knockout had little fat mass and higher rates of β-oxidation in adipose tissue in comparison with wild-type controls⁽Reference Wong, Szeto and Zhang⁷²⁾. Thus, additional studies, including adipose tissue-specific knockout models, are required to clarify the function of VDR in adipogenesis.

Vitamin D and lipid metabolism

There is some evidence that vitamin D could be involved in lipid mobilisation and utilisation in adipose tissue. An early study observed that 1,25(OH)₂D₃ induced a significant increase in lipoprotein lipase activity and of its mRNA level in 3T3-L1 adipocytes⁽Reference Querfeld, Hoffmann and Klaus⁶¹⁾. Concurrently, fatty acid synthase, which catalyses adipocyte lipogenesis, is down-regulated by 1,25(OH)₂D₃ in 3T3-L1 cells⁽Reference Kong and Li²²⁾. However, in vivo functional studies of VDR suggest that the receptor could inhibit lipid mobilisation and utilisation. VDR-null mice were reported to be resistant to high-fat diet-induced obesity, probably due to increases in fatty acid β-oxidation in white adipose tissue and the expression of uncoupling proteins in brown fat and of overall energy expenditure⁽Reference Wong, Szeto and Zhang⁷²⁾. On the other hand, targeted expression of VDR in adipocytes induces obesity in mice without changes in food intake, which is mainly caused by a marked decrease in energy expenditure together with reduced lipolysis and β-oxidation in adipose tissue⁽Reference Wong, Kong and Zhang²⁰⁾. In addition, the expression of genes involved in lipid metabolism, including hormone-sensitive lipase, adipose TAG lipase and uncoupling proteins 1, 2 and 3, is suppressed in VDR transgenic mice⁽Reference Wong, Kong and Zhang²⁰⁾.

Data on the effects of vitamin D in lipid metabolism in human subjects are scarce. A study of a small number of non-obese healthy subjects (n 10) has shown that vitamin D supplementation (2000 IU cholecalciferol/d), together with a low dietary Ca intake for 7 d, had no effect on energy expenditure, substrate metabolism or the expression of genes related to fat metabolism, such as hormone-sensitive lipase, fatty acid synthase and uncoupling protein 2 in adipose tissue, despite a significant increase in serum 1,25-OH₂D₃ levels⁽Reference Boon, Hul and Sicard⁷³⁾. These negative results could be due to the relatively short period (7 d) of treatment or because supplementation with cholecalciferol has little effect on healthy subjects who have adequate levels of vitamin D. Taken together, the present data are inconclusive and additional models and human studies are required to clarify the role of vitamin D metabolites in lipid metabolism.

Vitamin D and adipokine production

In addition to fuel storage, adipose tissue as an endocrine organ secretes a variety of bioactive proteins, and importantly a number of these adipokines, including adiponectin, TNF-α, IL-6 and monocyte chemoattractant protein-1 (MCP-1), are directly involved in inflammation⁽Reference Trayhurn, Bing and Wood^74, Reference Trayhurn and Wood⁷⁵⁾. There are as yet few studies that have examined the possible role of vitamin D in the modulation of adipokine production. It has been shown that mice lacking the VDR or CYP27B1 genes have reduced levels of serum leptin⁽Reference Narvaez, Matthews and Broun⁵⁷⁾. In a Middle-Eastern population of non-obese young subjects, serum 25(OH)D₃ was found to be positively correlated with adiponectin levels, though inversely associated with several metabolic risk factors, suggesting a possible link between vitamin D status and circulating adiponectin concentrations⁽Reference Gannagé-Yared, Chedid and Khalife⁷⁶⁾.

There is some evidence that vitamin D₃ may directly regulate adipokine expression and secretion by adipocytes. An in vitro study has shown that treatment with 1,25(OH)₂D₃ (10 nm) up-regulated the expression of the macrophage inhibitory factor (MIF), IL-6 and MCP-1 genes in 3T3-L1 adipocytes, and increased MIF mRNA levels in human adipocytes⁽Reference Sun and Zemel⁷⁷⁾. However, a recent in vivo study has reported that vitamin D supplementation reduced IL-6 protein content in adipose tissue of mice fed a high-fat diet, and 1,25(OH)₂D₃ (100 nm) inhibited LPS-induced IL-6 production by 3T3-L1 adipocytes⁽Reference Lira, Rosa and Cunha⁷⁸⁾. More recently, 1,25(OH)₂D₃ (100 nm) has been shown to inhibit the release of MCP-1 and adiponectin by human adipocytes⁽Reference Lorente-Cebrian, Eriksson and Dunlop⁷⁹⁾. The discrepancy between results might be due to the different doses used in these studies. Therefore, further work is needed to elucidate whether 1,25(OH)₂D₃ has a role in modulating the production of adipokines involved in inflammation.

Vitamin D and inflammation in adipose tissue

The potential role of vitamin D in modulating inflammation in obesity and other chronic diseases has received increasing attention. Evidence has accummulated that 1,25(OH)₂D₃ has potent immunoregulatory effects, such as inhibiting the production of IL-6, IL-8 and interferon-γ by peripheral blood mononuclear cells from psoriatic patients⁽Reference Inoue, Matsui and Nishibu⁸⁰⁾. It has also been shown that 1,25(OH)₂D₃ down-regulates the gene and protein expression of toll-like receptor (TLR) 2 and TLR-4 in human monocytes⁽Reference Do, Kwon and Park⁸¹⁾. 1,25(OH)₂D₃ also suppresses peripheral blood mononuclear cells’ proliferation and induces apoptosis in peripheral blood mononuclear cells of healthy subjects and inflammatory bowel disease patients⁽Reference Martinesi, Treves and d'Albasio⁸²⁾. Both 1,25(OH)₂D₃ and 25(OH)D₃ have been shown to reduce lipopolysaccharide-induced TNF-α and IL-6 production, probably by inhibiting p38 MAPK activation in human monocytes/macrophages⁽Reference Zhang, Leung and Richers⁸³⁾. Conversely, 1,25(OH)₂D₃-deficient T-cells isolated from CYP27B1 knockout mice are predisposed to overexpress IL-17⁽Reference Bruce, Yu and Ooi⁸⁴⁾, while VDR-null mice display a failure of T-cell homing to the gut with low levels of IL-10 in inflammatory bowel disease⁽Reference Yu, Bruce and Froicu⁸⁵⁾. Furthermore, in peripheral blood mononuclear cells from type-2 diabetic patients having a proinflammatory profile, 1,25(OH)₂D₃ is reported to act in an anti-inflammatory manner to decrease the expression of TNF-α, IL-1, IL-6 and IL-8⁽Reference Giulietti, van Etten and Overbergh⁸⁶⁾. In vivo, aged mice treated with vitamin D₃ showed a significant improvement in visual function by reducing retinal inflammation and amyloid-β accumulation⁽Reference Lee, Rekhi and Kam⁸⁷⁾.

With adipocyte hypertrophy in obesity, there is a marked increase in the synthesis and release of proinflammatory mediators (e.g. TNF-α, IL-6, IL-8 and MCP-1), and this contributes to the raised circulating levels as well as to the tissue inflammation⁽Reference Fontana, Eagon and Trujillo^88, Reference Skurk, Alberti-Huber and Herder⁸⁹⁾. Adipose tissue inflammation, characterised by increased infiltration of macrophages and other immune cells, is a central pathological process in adipose tissue dysfunction⁽Reference Cinti, Mitchell and Barbatelli^90, Reference Lolmede, Duffaut and Zakaroff-Girard⁹¹⁾. Work from our group and others has demonstrated that macrophage-conditioned medium potently stimulates the release of proinflammatory factors (e.g. MCP-1, IL-8, chemokine (C–C motif) ligand 5 (CCL-5) and IL-6) and a number of proteins involved in extracellular matrix remodelling from human preadipocytes and adipocytes; these factors may induce inflammation, fibrosis and insulin resistance in adipose tissue⁽Reference O'Hara, Lim and Mazzatti^59, Reference Lacasa, Taleb and Keophiphath^92–Reference Gao and Bing⁹⁵⁾. The immunoregulatory effects of 1,25(OH)₂D₃ suggest that the hormone may also modulate the inflammatory response in adipose tissue. Very recently, we have shown that treatment with 1,25(OH)₂D₃ (10 and 100 nm) led to a reduction in the protein release of MCP-1 and IL-6 by human preadipocytes, as well as preadipocyte-induced macrophage migration⁽Reference Gao, Trayhurn and Bing⁹⁶⁾. Concurrently, an inhibitory effect of 1,25(OH)₂D₃ on TNF-α-stimulated MCP-1 release by human adipocytes has been reported⁽Reference Lorente-Cebrian, Eriksson and Dunlop⁷⁹⁾.

Activation of the NF-κB signalling pathway is essential in the signal transduction of proinflammatory cytokines in many cell types, including adipocytes⁽Reference Gao and Bing^95, Reference Fujihara, Antunes and Mattar^97, Reference Chen, Kong and Sun⁹⁸⁾. NF-κB activation involves the degradation of IκBα protein and translocation of p65 into the nucleus⁽Reference Bonizzi and Karin⁹⁹⁾. Our recent work has shown that 1,25(OH)₂D₃ can increase IκBα protein abundance in human preadipocytes⁽Reference Gao, Trayhurn and Bing⁹⁶⁾. Blocking of NF-κB activation by 1,25(OH)₂D₃ has also been reported in mesangial cells, as 1,25(OH)₂D₃ can stabilise IκBα, leading to an inhibition of p65 NF-κB nuclear translocation⁽Reference Zhang, Yuan and Sun¹⁰⁰⁾. Very recently, we have observed that 1,25(OH)₂D₃ is able to reverse macrophage-elicited inhibition of IκBα and up-regulation of p65 NF-κB in differentiated human adipocytes (unpublished results). Overall, 1,25(OH)₂D₃ appears to be anti-inflammatory and it may ameliorate macrophage-induced inflammation in adipose tissue (Fig. 1).

Fig. 1 Proposed mechanism of vitamin D₃ signalling in adipocytes. 1,25-Dihydroxyvitamin D₃ (1,25(OH)₂D₃) may be transferred into the adipocyte (from the circulation or adjacent cells) or 25(OH)D₃ taken up from the circulation and hydroxylated in the fat cell by 1α-hydroxylase. 1,25(OH)₂D₃ acts to inhibit the phosphorylation of p38 MAPK and the activation of NF-κB signalling, with enhanced IκBα expression while reduced phosphorylation of p65, in human adipocytes. As a result, gene expression and protein release of proinflammatory mediators (i.e. monocyte chemoattractant protein-1 (MCP-1) and IL-6) by fat cells are reduced, leading to decreased recruitment of monocytes/macrophages and overall inflammation within adipose tissue.

Vitamin D and insulin resistance

Several clinical studies have associated low vitamin D status with the development of insulin resistance in adults⁽Reference Chiu, Chu and Go^8, Reference Gannagé-Yared, Chedid and Khalife^76, Reference Scragg, Holdaway and Singh¹⁰¹⁾ and children⁽Reference Parikh, Guo and Pollock^102, Reference Olson, Maalouf and Oden¹⁰³⁾. Higher basal levels of 25(OH)D₃ have been found to predict better β-cell function and lower glycaemia in subjects at risk for type 2 diabetes⁽Reference Kayaniyil, Retnakaran and Harris¹⁰⁴⁾. In rats, vitamin D deficiency impairs insulin release from the pancreas and reduces glucose tolerance, which is partially reversed following treatment with 1,25(OH)₂D₃⁽Reference Billaudel, Bourlon and Sutter^105, Reference Cade and Norman¹⁰⁶⁾. A recent clinical trial has shown that cholecalciferol (2000 IU daily) supplementation for 16 weeks improved β-cell function in adults at high risk of diabetes⁽Reference Mitri, Dawson-Hughes and Hu¹⁰⁷⁾. In type 2 diabetic patients with vitamin D deficiency, daily intake of a vitamin D-fortified yogurt drink increased serum 25(OH)D₃ levels and improved glycaemic status⁽Reference Nikooyeh, Neyestani and Farvid¹⁰⁸⁾. Therefore, vitamin D may have an effect on insulin secretion from pancreatic β-cells and further work in this area is warranted.

A potentially beneficial effect of vitamin D on insulin sensitivity has been proposed, as 1,25(OH)₂D₃ treatment increased insulin receptor mRNA levels and insulin-stimulated glucose transport in U-937 promonocytic cells, possibly via the up-regulation of phosphatidylinositol 3-kinase activity⁽Reference Maestro, Molero and Bajo¹⁰⁹⁾. Adipose tissue, in addition to skeletal muscle and liver, is a key organ exhibiting insulin resistance in obesity⁽Reference Hotamisligil¹¹⁰⁾. Although adipose tissue only accounts for approximately 10 % of insulin-stimulated whole-body glucose uptake, the reduction in insulin sensitivity of adipocytes increases NEFA release into the circulation, which may induce hepatic and muscle insulin resistance⁽Reference Smith¹¹¹⁾. In streptozotocin-induced diabetic rats, 1,25(OH)₂D₃ treatment has been reported to normalise the number of insulin receptors and improve the insulin response to glucose transport in epididymal adipocytes⁽Reference Calle, Maestro and Garcia-Arencibia¹¹²⁾. However, whether vitamin D metabolites have beneficial effects on glucose transport and insulin action in human adipose tissue remains to be investigated.