Calcium as a cellular regulator

Ca performs its regulatory functions at the cellular level in the form of Ca²⁺ ions. Ca²⁺ is considered as the most versatile, ubiquitous intracellular messenger⁽Reference Carafoli, Santella and Branca^20–Reference Sergeev²³⁾. It reversibly binds to specific proteins that act as Ca²⁺ sensors to decode its information before passing it on to targets. The membrane Ca²⁺ transport systems control the cellular homeostasis of Ca²⁺. Remarkably, Ca²⁺ is a universal, ambivalent signalling agent. It carries information to virtually all processes important to cell life, but also transmits signals that promote cell death. Spatial and temporal characteristics of the Ca²⁺ signal determine the type and magnitude of biological responses; for example, oscillations of cytosolic Ca²⁺ in pancreatic β-cells underlie the oscillatory pattern of insulin release⁽Reference Sergeev, Norman, Norman, Bouillon and Thomasset^24, Reference Sergeev and Rhoten²⁵⁾.

Calcium intake

The dietary reference intake for Ca was updated by the United States Food and Nutrition Board of the Institute of Medicine in 2010, with the biggest change made being the conversion from adequate intake to estimated average requirement and RDA^{(26, 27)}. The RDA varies from 700 mg/d for younger children to 1300 mg/d for adolescents; 1200 mg/d should be considered sufficient for most groups. The nutritional guidelines provided by the European Union Commission are similar⁽²⁸⁾.

The best source of Ca is dairy foods⁽Reference Heaney²⁹⁾. Although some individuals are intolerant to dairy products or do not consume dairy products for ethical reasons, other good sources of Ca exist. However, Ca deficiency is still an important issue. Adolescents and older individuals are the most likely groups to be deficient in Ca⁽Reference Wright, Wang and Kennedy-Stephenson³⁰⁾, and Ca deficiency is considered in the USA to warrant a national effort to increase average intake levels⁽³¹⁾.

Calcium and body mass

A growing body of literature suggests that a low Ca intake is associated with a greater fat mass. The potential effect of Ca on weight loss has been shown in animal models⁽Reference Pilvi, Harala and Korpela^32–Reference de Wit, Bosch-Vermeulen and Oosterink³⁵⁾. Importantly, the results obtained with murine models of obesity appear to translate to humans. Several observational studies (mainly cross-sectional and retrospective cohort studies)⁽Reference Leite and Sampaio^36–Reference Goldberg, da Silva and Peres⁴²⁾ have demonstrated a negative relationship between Ca intake and body weight. Furthermore, this association has been observed across multiple ages and ethnicities and in both sexes. However, it is necessary to consider that the overall energy density of diets characterised by a low Ca content can be higher than comparable diets characterised by a high Ca content⁽Reference Tidwell and Valliant⁴¹⁾. Moreover, daily food consumption in experimental animals fed a high-Ca diet ad libitum can be lower than that of animals fed diets with normal or low Ca levels. These observations may provide an additional explanation of the possible effect of Ca on body weight.

The intervention studies examining the effects of Ca on body weight are summarised in Table 1. The studies conducted during an energy-restriction programme demonstrated that Ca intake has no significant effect on body weight⁽Reference Torres, da Silva Ferreira and Carvalho^43–Reference Thomas, Wideman and Lovelady⁴⁸⁾. Importantly, the studies conducted without requiring energy restriction failed to demonstrate the effect of Ca intake on body-weight loss as well⁽Reference Yanovski, Parikh and Yanoff^49, Reference Chailurkit, Saetung and Thakkinstian⁵⁰⁾. Several studies showed no effect of a high Ca intake on body-weight loss⁽Reference Wagner, Kindrick and Hertzler^44, Reference Teegarden, White and Lyle^46, Reference Riedt, Schlussel and von Thun^47, Reference Yanovski, Parikh and Yanoff^49, Reference Chailurkit, Saetung and Thakkinstian⁵⁰⁾, although one of these studies demonstrated a significant increase in fat oxidation⁽Reference Teegarden, White and Lyle⁴⁶⁾. The meta-analysis review that included randomised controlled trials also failed to find evidence of a beneficial effect of Ca on body weight⁽Reference Trowman, Dumville and Hahn⁵¹⁾. A study conducted by Torres et al. ⁽Reference Torres, Francischetti and Genelhu⁵²⁾ found that subjects consuming a diet containing 1200–1300 mg Ca per d exhibited a greater reduction in waist circumference and waist:hip ratio as compared with subjects who consumed a similar diet with a low Ca content ( < 500 mg/d). However, there were no statistically significant differences between groups with respect to body weight. Faghih et al. ⁽Reference Faghih, Abadi and Hedayati⁴⁵⁾ found an inverse relationship between dairy Ca consumption and body weight, but there were no significant differences between body weight and BMI in the soya milk, Ca supplement and control groups.

Table 1 Intervention studies examining the effects of calcium on body weight

Generally, results of the reviewed studies failed to demonstrate the direct effect of dietary Ca on body weight; however, several studies suggest the role of Ca in increasing fat oxidation and decreasing waist circumference. The inconsistency between observational studies and randomised controlled trials might arise from missing data on energy intake and physical activity in most observational studies (we found only one study demonstrating that, in adolescents, a lower Ca intake is accompanied by a higher intake of fat and energy⁽Reference Tidwell and Valliant⁴¹⁾).

Vitamin D as a hormonal and cellular regulator

The biological effects of vitamin D₃ result from its sequential metabolism in the liver into 25-hydroxyvitamin D₃ (25(OH)D₃), and then in the kidney into the steroid hormone 1,25-dihydroxyvitamin D₃ (1,25(OH)₂D₃)⁽Reference Norman⁵³⁾. 1,25(OH)₂D₃ is considered as the principal Ca²⁺-regulatory hormone, which exerts its regulatory effects by increasing intestinal Ca²⁺ absorption, bone Ca²⁺ mobilisation and renal Ca²⁺ reabsorption⁽Reference Sergeev, Rhoten and Spirichev^21, Reference Norman^53, Reference Spirichev and Sergeev⁵⁴⁾. 1,25(OH)₂D₃ regulates not only Ca²⁺ metabolism, but also a wide spectrum of cellular processes, including proliferation, differentiation, development, apoptosis and secretion⁽Reference Sergeev, Rhoten and Spirichev^21, Reference Sergeev and Rhoten^25, Reference Sergeev, Rhoten, Norman, Norman, Bouillon and Thomasset^55, Reference Sergeev and Rhoten⁵⁶⁾. 1,25(OH)₂D₃ produces its biological effects via both receptor-mediated regulation of nuclear events and rapid actions independent of genomic pathways⁽Reference Sergeev, Rhoten and Spirichev^21, Reference Sergeev, Rhoten, Norman, Norman, Bouillon and Thomasset^55, Reference Norman, Nemere and Zhou⁵⁷⁾. The genomic responses utilise signal-transduction pathways linked to the nuclear/cytosolic VDR, while the rapid responses utilise signal-transduction pathways linked to the membrane VDR localised in the plasma and, possibly, endoplasmic reticulum membranes⁽Reference Sergeev⁵⁸⁾ as well as to the plasma membrane-associated rapid response steroid hormone-binding protein⁽Reference Nemere, Garbi and Hammerling^59, Reference Nemere and Hintze⁶⁰⁾.

Nuclear and membrane VDR have been demonstrated in many (over forty) tissues, including adipose tissue⁽Reference Sergeev, Rhoten and Spirichev^21, Reference Norman^53, Reference Sergeev and Spirichev⁶¹⁾. In these target tissues, the VDR functions both as a transcriptional factor to influence more than 3% of the human genome and a modulator of a number of cellular signal-transduction pathways, including Ca²⁺ signalling. Analogues of vitamin D that can act as agonists of genomic and non-genomic pathways have been identified⁽Reference Norman, Sergeev and Bishop^62–Reference Norman and Silva⁶⁴⁾. It appears that Ca²⁺ signals (transient and prolonged) triggered by 1,25(OH)₂D₃ can be linked to both membrane and nuclear VDR⁽Reference Sergeev^22, Reference Sergeev⁶⁵⁾.

Significant evidence indicates that biological actions of 1,25(OH)₂D₃ are mediated or influenced by intracellular Ca²⁺-signalling events. Rapid (within seconds to minutes) effects of 1,25(OH)₂D₃ on intracellular Ca²⁺ have been demonstrated in mammary and intestinal epithelial cells, pancreatic β-cells, osteoblasts and adipocytes⁽Reference Sergeev, Rhoten and Spirichev^21, Reference Sergeev and Rhoten^25, Reference Sergeev, Rhoten, Norman, Norman, Bouillon and Thomasset^55, Reference Sergeev and Rhoten^56, Reference Sergeev and Rhoten^66–Reference Rhoten and Sergeev⁶⁹⁾. It has been also shown that 1,25(OH)₂D₃ rapidly induces Ca²⁺ uptake, activates Ca²⁺ channels and stimulates phosphoinositide turnover in different cell types⁽Reference Sergeev, Rhoten, Norman, Norman, Bouillon and Thomasset^55, Reference Norman, Nemere and Zhou^57, Reference Farach-Carson, Sergeev and Norman⁶³⁾. Thus, activation of the intracellular Ca²⁺-signalling pathways is essential for the actions of 1,25(OH)₂D₃. It is interesting to note that, in turn, Ca²⁺ can interact with the VDR via Ca²⁺-binding γ-carboxyglutamate residues of the receptor protein, which results in down-regulation of the vitamin D-mediated biological responses⁽Reference Sergeev and Spirichev^61, Reference Sergeev and Spirichev^70, Reference Sergeev and Norman⁷¹⁾.

Determination of vitamin D status

Vitamin D status is determined by measuring the circulating concentration of its transport form, 25-hydroxyvitamin D (25(OH)D). The serum concentration of 25(OH)D is approximately 1000 times higher than that of 1,25(OH)₂D, and 25(OH)D has a half-life of 2–3 weeks compared with that of 4–6 h for the hormone⁽Reference Holick, Feldman, Glorieux and Pike⁷²⁾. The circulating concentration of 25(OH)D is considered to be the most reliable indicator of vitamin D production, intake and stores⁽Reference Verreault, Semba and Volpato⁷³⁾.

Currently, there is no consensus on how the circulating levels of 25(OH)D should be classified to describe vitamin D nutritional status. The concentration of 25(OH)D in severe vitamin D deficiency is ‘undetectable’ (i.e. <  1–5 ng/ml); normal range is traditionally defined as 20–50 ng/ml; levels below 10–20 ng/ml may indicate vitamin D insufficiency, while levels above 100 ng/ml represent a risk of toxicity^{(26, 27)}. A higher range of normal or ‘optimal’ 25(OH)D concentrations have been also suggested (for example, 30–70 ng/ml)⁽Reference Norman and Bouillon⁷⁴⁾.

Vitamin D status depends on sunlight exposure and dietary intake. The skin has a large vitamin D₃ production capacity and is capable of supplying body requirements in vitamin D. Latitude, season, skin pigmentation, sunscreen use and ozone air pollution influence the cutaneous production of vitamin D₃. The levels of UVB irradiation to produce a significant increase in the serum 25(OH)D concentration are not possible to achieve in winter at latitudes above and below 40°, thus limiting endogenous production of vitamin D₃ for several months of the year⁽Reference Foss⁷⁵⁾. Consumption of dietary or supplemental vitamin D or artificial UV irradiation should be considered under these circumstances⁽²⁶⁾.

The dietary reference intake allowance of vitamin D – the value sufficient to meet the needs of virtually all individuals – recommended in 2010 by the United States Food and Nutrition Board of the Institute of Medicine is 15 μg/d (600 IU/d)⁽²⁶⁾. For infants, adequate intake is 10 μg/d (400 IU/d). Estimated average requirement of vitamin D is 400 IU/d for all life-stage groups (it is not established for infants). Possible benefits of vitamin D intake at the level of 2000–4000 IU/d (25–100 μg/d) are actively debated⁽Reference Norman and Bouillon⁷⁴⁾. The Institute of Medicine does not recommend such an increase⁽²⁶⁾. The recommended upper intake level of vitamin D for adults is 4000 IU/d (100 μg/d)⁽²⁶⁾; long-term intake at the level of 5000–10 000 IU/d may represent a risk of toxicity. The nutritional guidelines for vitamin D set forward by the European Union Commission are similar⁽²⁸⁾. The US and European Union recommendations are based on the established actions of vitamin D on Ca and bone metabolism.

Vitamin D status and obesity

The potential link between vitamin D and obesity was first observed in 1971 by Rosenstreich et al. ⁽Reference Rosenstreich, Rich and Volwiler⁷⁶⁾. They demonstrated the association between increased body fat and low serum 25(OH)D concentrations and attributed this to sequestration of the fat-soluble vitamin D in the adipose tissue. Zemel et al. ⁽Reference Zemel, Shi and Greer⁷⁷⁾ showed that dietary Ca can reduce adiposity and suggested that the regulation by Ca of the production of 1,25(OH)₂D plays a role. These two studies laid the groundwork for examining the connection between vitamin D and body composition as well as for elucidating the underlying mechanisms.

Recently, a growing body of epidemiological evidence has emerged suggesting the role of vitamin D in obesity, including observational studies that demonstrated the association between vitamin D intake and body composition⁽Reference Goldberg, da Silva and Peres^42, Reference Foss^75, Reference Aasheim, Hofso and Hjelmesaeth^78–Reference Lagunova, Porojnicu and Lindberg¹⁰⁰⁾. In these studies, a statistically significant inverse correlation suggesting that a low vitamin D status is associated with a larger fat mass as well as a greater risk of weight gain over time has been reported. However, it is important to mention that obese individuals have limited mobility and avoid outdoor activity⁽¹⁰¹⁾. In those individuals, exposure of skin to UVB irradiation may become inadequate to maintain optimal vitamin D status⁽Reference Sharma, Barr and Macdonald¹⁰²⁾. It is also important to mention that vitamin D₃ is sequestrated in the obese adipose tissue⁽Reference Mawer, Backhouse and Holman¹⁰³⁾. These should be taken into consideration in explaining the inverse correlation between vitamin D status and body weight demonstrated in observational studies.

The effect of vitamin D supplementation (often together with an increased dietary Ca intake) on body fat in human subjects has been evaluated in several randomised controlled trials (Table 2) ⁽Reference Zittermann, Frisch and Berthold^104–Reference Caan, Neuhouser and Aragaki¹⁰⁹⁾. The 4-year study by Caan et al. ⁽Reference Caan, Neuhouser and Aragaki¹⁰⁹⁾ demonstrated that the consumption of 1000 mg Ca plus 400 IU vitamin D per d has a small effect on the prevention of weight gain, which was observed primarily in women who reported inadequate Ca intakes. In contrast, in five studies no effect was demonstrated⁽Reference Zittermann, Frisch and Berthold^104–Reference Zhou, Zhao and Watson¹⁰⁸⁾. For example, the 15-week weight-loss study by Major et al. ⁽Reference Major, Alarie and Dore¹⁰⁶⁾ showed that the consumption of 600 mg Ca plus 1100 IU vitamin D per d enhanced the beneficial changes in HDL-cholesterol, TAG and total cholesterol in overweight or obese women with a low daily Ca intake, but had no effect on body weight. The study by Zhou et al. ⁽Reference Zhou, Zhao and Watson¹⁰⁸⁾ demonstrated that percentage of trunk fat was significantly different between the Ca intervention groups and the placebo group, but vitamin D supplementation had no additional effect. Furthermore, the 1-year study conducted by Zittermann et al. ⁽Reference Zittermann, Frisch and Berthold¹⁰⁴⁾ in Norway revealed that a significant weight reduction in overweight and obese subjects is unlikely to occur with vitamin D supplementation.

Table 2 Intervention studies examining the effects of vitamin D on body weight*

25(OH)D, 25-hydroxyvitamin D; N/A, not applicable.

* 1 IU vitamin D₃ = 0·025 μg vitamin D₃.

The results of observational studies and randomised controlled trials do not provide conclusive causal or mechanistic insights into whether or how vitamin D (and Ca) might regulate body weight. It is necessary to note that confounding effects could explain the association between vitamin D levels and fat stores in a non-casual or reverse-casual way; for example, low 25(OH)D levels in obesity could be observed because of its increased sequestration in the obese adipose tissue or because higher levels of physical outdoor activities could increase sun exposure and 25(OH)D levels or decrease adiposity directly.

Calcium intake and faecal fat excretion

Ca can impair the absorption of fat in the intestine via formation of insoluble Ca fatty acid soaps or by binding of bile acids. This mechanism was demonstrated in animal and human studies, indicating that supplemental Ca and dairy foods increase faecal excretion of fat. deWit et al. ⁽Reference de Wit, Bosch-Vermeulen and Oosterink³⁵⁾ observed a significant increase in the faecal excretion of fatty acids and bile acids in mice fed a high-Ca diet. Buchowski et al. ⁽Reference Buchowski, Aslam and Dossett¹¹⁰⁾ also found that a high-Ca diet causes an increase in faecal fat excretion independent of the Ca source. Specifically, they showed that, compared with a low-Ca diet, a high-Ca diet increased the faecal fat excretion in human subjects by 1·8 g/d, or about 3% of daily fat intake. This increase in fat excretion could cause 0·4–0·7 kg of body-fat loss over a 1-year period. Christensen et al. ⁽Reference Christensen, Lorenzen and Svith¹¹¹⁾ examined the effect of Ca on fat excretion by performing a systematic meta-analysis review and estimated that increasing daily Ca intake by 1241 mg/d resulted in an increase in faecal fat of 5·2 (1·6–8·8) g/d. This effect was most pronounced in subjects with a low habitual dietary Ca intake. However, the effect of Ca on faecal fat excretion has not been investigated in long-term studies.

Intracellular calcium, 1,25-dihydroxyvitamin D₃ and apoptosis

Apoptosis, a highly regulated form of cell death, plays an important role during development and adult life via intimate involvement in cellular and tissue homeostasis. Apoptosis is the main mechanism for controlling cell number in most tissues⁽Reference Reed¹¹²⁾. Obesity is the result of an increase in the adipose tissue mass, which can result from both hypertrophy, an increase in adipocyte size, and hyperplasia, an increase in adipocyte number⁽Reference Lane and Tang^9, Reference Prins and O'Rahilly¹¹³⁾. Weight loss can be caused not only by a decrease in adipocyte size (i.e. increasing lipolysis with a potential for lipotoxic effects), but also in adipocyte number (for example, by stimulating apoptosis). An increase in the rate of adipocyte apoptosis will prevent excessive accumulation of adipose tissue and may result in a significant loss of adipose tissue mass over time, in contrast to that which occurs after energy restriction. However, studies on the role of apoptosis in fat tissue have been limited by the fact that mature, differentiated adipocytes are extremely stable and not thought to be capable of undergoing apoptosis. Induction of death of adipocytes through apoptosis may emerge as a promising strategy for the prevention and treatment of obesity because removal of adipocytes via this mechanism will result in reducing body fat and a long-lasting maintenance of weight loss.

Cellular Ca²⁺ has been implicated in the induction of apoptosis and regulation of the apoptotic signalling pathways⁽Reference Carafoli, Santella and Branca^20–Reference Sergeev^22, Reference Sergeev^65, Reference Berridge, Bootman and Lipp^114, Reference Orrenius, Zhivotovsky and Nicotera¹¹⁵⁾; however, mechanisms of Ca²⁺ signalling in apoptosis remain obscure. We⁽Reference Sergeev^22, Reference Sergeev, Rhoten, Norman, Norman, Bouillon and Thomasset^55, Reference Sergeev and Rhoten^56, Reference Sergeev^65, Reference Sergeev and Rhoten^68, Reference Sergeev, Colby, Norman, Norman, Bouillon and Thomasset^116–Reference Sergeev and Norman¹¹⁸⁾ and others⁽Reference Carafoli, Santella and Branca^20, Reference Berridge, Bootman and Lipp^114, Reference Orrenius, Zhivotovsky and Nicotera^115, Reference McConkey and Orrenius¹¹⁹⁾ have shown that increases in the concentration of intracellular Ca²⁺ occur in the early and late stages of apoptosis. It appears that the critical characteristic of the apoptotic Ca²⁺ signal is a sustained increase in intracellular Ca²⁺ concentration, reaching elevated, but not cytotoxic levels. Although there is little doubt that such an increase in intracellular Ca²⁺ concentration triggers cell death via apoptosis, the mechanisms of action of intracellular Ca²⁺ in apoptotic pathways are not known and, particularly, interactions of the cellular Ca²⁺ signal with molecular Ca²⁺ targets in cells undergoing apoptosis have not been identified. A family of intracellular cysteine proteases, the caspases, is responsible for most biochemical and morphological alterations during apoptosis, although additional or alternative apoptosis initiation and execution pathways have been demonstrated⁽Reference Carafoli, Santella and Branca^20, Reference Sergeev^22, Reference Welsh, VanWeelden and Flanagan¹²⁰⁾. Ca²⁺-dependent caspases and Ca²⁺-dependent neutral proteases, the calpains, are considered as the primary Ca²⁺-activated apoptotic targets⁽Reference Mathiasen, Sergeev and Bastholm^117, Reference Carafoli and Molinari^121–Reference Welsh and Byrne¹²⁵⁾.

Interaction of the Ca²⁺ signal with intracellular Ca²⁺ buffers plays a particularly important role in the apoptotic process. A key element of the cytosolic Ca²⁺-buffering system is the vitamin D-dependent Ca²⁺-binding proteins, calbindins. Elevated levels of calbindins dramatically increase the cytosolic Ca²⁺-buffering capacity and an increase in Ca²⁺ buffering via forced expression of calbindin-D_28k protects cells against Ca²⁺-mediated apoptosis⁽Reference Sergeev^65, Reference Sergeev, Rhoten and Carney^67, Reference Rhoten and Sergeev^69, Reference Mathiasen, Sergeev and Bastholm¹¹⁷⁾.

We have shown⁽Reference Sergeev, Rhoten and Spirichev^21, Reference Sergeev^22, Reference Sergeev⁶⁵⁾ that a sustained (not reaching cytotoxic levels) increase in intracellular Ca²⁺ concentration signals the cell to enter the apoptotic pathway via activation of the Ca²⁺-dependent protease μ-calpain followed by activation of the Ca²⁺/calpain-dependent caspase-12 and other executor caspases (for example, caspase-3). A lack of expression or low levels of the cytosolic Ca²⁺-binding proteins (for example, calbindin-D) diminish the ability of the cell to buffer intracellular Ca²⁺ concentration increases and, thus, facilitate induction of apoptosis. On the other hand, agents that induce expression of the intracellular Ca²⁺ buffers or suppress pathways for the generation of the apoptotic Ca²⁺ signal may protect against Ca²⁺-mediated apoptosis.

It is well established that 1,25(OH)₂D₃ can induce Ca²⁺ signals in different cell types, including adipocytes. 1,25(OH)₂D₃ activates the voltage-dependent and voltage-insensitive Ca²⁺ entry pathways and triggers Ca²⁺ release from the endoplasmic reticulum stores through the inositol 1,4,5-trisphosphate and ryanodine receptors⁽Reference Sergeev, Rhoten and Spirichev^21, Reference Sergeev^22, Reference Sergeev and Rhoten^56, Reference Sergeev^65, Reference Mathiasen, Sergeev and Bastholm¹¹⁷⁾. Importantly, we have also shown that 1,25(OH)₂D₃ induces apoptosis in adipocytes⁽Reference Sergeev^23, Reference Sergeev⁵⁸⁾ and that apoptosis induced by 1,25(OH)₂D₃ in these cells depends on Ca²⁺ signalling⁽Reference Sergeev^22, Reference Sergeev, Norman, Norman, Bouillon and Thomasset^24, Reference Mathiasen, Sergeev and Bastholm¹¹⁷⁾.

Below we summarise the present results regarding the role of 1,25(OH)₂D₃ in generating Ca²⁺ signals in adipocytes and provide evidence that 1,25(OH)₂D₃-induced Ca²⁺ signals can determine adipocyte fate by apoptosis (Fig. 1). These findings may help in the rational search for therapeutic and preventive agents for obesity that act via Ca²⁺-dependent molecular targets in apoptotic pathways.

Fig. 1 Role of cellular Ca²⁺ and 1,25-dihydroxyvitamin D₃ (1,25(OH)₂D₃; 1,25-D) in the regulation of adipocyte functions. 1,25(OH)₂D₃ generates the cellular Ca²⁺ signal via the regulation of Ca²⁺ entry from the extracellular space, Ca²⁺ mobilisation from the intracellular stores and intracellular Ca²⁺ buffering. The mechanism includes interactions of 1,25(OH)₂D₃ with the membrane vitamin D receptor (VDR) and Ca²⁺ channels/receptors in the plasma and endoplasmic reticulum (ER) membrane. The 1,25(OH)₂D₃-induced cellular Ca²⁺ signal (a sustained increase in concentration of cytosolic Ca²⁺) inhibits lipogenesis, stimulates lipolysis and induces apoptosis (via activation of the Ca²⁺-dependent apoptotic pathway) in mature adipocytes. 1,25(OH)₂D₃ also inhibits adipogenesis by suppressing adipocyte differentiation. The process involves down-regulation of a transcriptional factor CCAAT enhancer-binding protein-α (C/EBPα) and up-regulation of the nuclear PPARγ. 1,25(OH)₂D₃ appears to inhibit expression of the mitochondrial uncoupling proteins (UCP) in adipocytes, which may counteract body-fat loss (a high UCP expression may promote the shift toward thermogenesis and, thus, a loss of body fat). [Ca²⁺]_i, intracellular Ca²⁺ concentration; nVDR, nuclear VDR; RXR, retinoid X receptor; Mit, mitochondria.

1,25-Dihydroxyvitamin D₃ induces Ca²⁺-mediated apoptosis in adipocytes

The mechanism controlling adipocyte apoptosis is unknown and even the ability of adipocytes to undergo apoptosis has not been conclusively demonstrated. We have recently shown⁽Reference Sergeev²³⁾ that 1,25(OH)₂D₃ induces apoptosis in mature mouse 3T3-L1 adipocytes via activation of the Ca²⁺-dependent calpain and Ca²⁺/calpain-dependent caspase-12. Treatment of adipocytes with 1,25(OH)₂D₃ induced, in a concentration- and time-dependent fashion, a sustained increase in the basal level of intracellular Ca²⁺. The increase in intracellular Ca²⁺ concentration was associated with induction of apoptosis and activation of μ-calpain and caspase-12. Importantly, susceptibility of mature, differentiated adipocytes to the Ca²⁺-elevating effect of 1,25(OH)₂D₃ appears to be dependent on the reduced Ca²⁺-buffering capacity of these cells. The results demonstrated that Ca²⁺-mediated apoptosis can be induced in mature, differentiated adipocytes and that the apoptotic molecular targets activated by 1,25(OH)₂D₃ in these cells are Ca²⁺-dependent calpains and caspases. It is interesting to note that a different class of compounds, flavonoids, which are present in the common human diet, also induces Ca²⁺-mediated apoptosis in adipocytes⁽Reference Sergeev, Li and Ho^126–Reference Sergeev, Li and Colby¹²⁸⁾. These findings provide a strong rationale for evaluating the role of vitamin D in the prevention and treatment of obesity.

1,25(OH)₂D₃ is not only an important determinant of adipocyte apoptosis, but also appears to regulate adipocyte differentiation. It has been shown that 1,25(OH)₂D₃ inhibits adipogenesis in 3T3-L1 cells by blocking their differentiation to mature adipocytes⁽Reference Kong and Li¹²⁹⁾. The process involves suppression of a transcriptional regulator CCAAT-enhancer-binding protein (C/EBPα) and up-regulation of PPARγ. PPARγ is a primary regulator of fatty acid storage and glucose metabolism, and the genes activated by PPARγ are linked to stimulation of lipid uptake by adipocytes and adipogenesis⁽Reference Jones, Barrick and Kim¹³⁰⁾. C/EBPα promotes adipogenesis by inducing the expression of PPARγ⁽Reference Clarke, Robinson and Gimble¹³¹⁾. The nuclear VDR appears to play an essential role in adipogenesis via C/EBPα and PPARγ (for example, in the absence of 1,25(OH)₂D, ‘knock-down’ of the VDR using siRNA delays the formation of adipocytes⁽Reference Blumberg, Tzameli and Astapova¹³²⁾).

It is worth mentioning that ‘pharmacological’ concentrations of 1,25(OH)₂D (10–100 nmol/l) are used in in vitro studies. Under normal physiological conditions, the serum circulating concentration of 1,25(OH)₂D is tightly regulated at the level of 100–125 pmol/l, and normal levels of 1,25(OH)₂D in blood can be maintained within a wide range of 25(OH)D concentrations (10–250 nmol/l)⁽Reference Sergeev, Rhoten and Spirichev²¹⁾. On the other hand, concentrations of 1,25(OH)₂D in different tissues can approach a nanomolar range due to in situ biosynthesis of the hormone⁽Reference Sergeev, Rhoten and Spirichev^21, Reference Norman and Bouillon⁷⁴⁾. Therefore, it is difficult to make a direct comparison of in vitro studies using 1,25(OH)₂D and clinical studies using either supplementation with vitamin D or measuring the circulating concentration of 25(OH)D.

Calcium and lipid metabolism

Ca²⁺ may play a role in the regulation of lipid metabolism and TAG storage. The hypothesis advocated by Zemel & Sun⁽Reference Zemel and Sun¹³³⁾ suggests that depressed levels of 1,25(OH)₂D (due to a high dietary Ca intake) cause a decrease in intracellular Ca²⁺ concentration, and, thus, stimulate lipolysis, and inhibit fatty acid synthase and de novo lipogenesis. However, a basal, steady-state intracellular Ca²⁺ concentration is maintained by mechanisms (pumps and buffers) independent of Ca intake, and a decrease in circulating 1,25(OH)₂D will not directly affect basal intracellular Ca²⁺ concentration. Interestingly, Sampath et al. ⁽Reference Sampath, Havel and King¹³⁴⁾ administered 1500 mg supplemental Ca per d for 3 months to overweight or obese subjects and measured the rates of lipolysis in adipose tissue, whole-body lipid oxidation and circulating concentrations of several hormones. They failed to demonstrate the effect of Ca supplementation on the rate of lipid oxidation or lipolysis. This was confirmed in the study by Bortolotti et al. ⁽Reference Bortolotti, Rudelle and Schneiter¹³⁵⁾, where subjects receiving 800 mg dairy Ca per d for 5 weeks demonstrated no effects of Ca supplementation on markers of lipid metabolism.

Vitamin D and mitochondrial uncoupling proteins

An increase in the expression of mitochondrial uncoupling proteins (UCP) promotes a shift toward thermogenesis and away from ATP synthesis. As ATP production diminishes, the dynamics of the catabolic breakdown and biosynthesis of stored nutrients shifts toward catabolism in an effort to replenish the ATP stores⁽Reference Nedergaard, Ricquier and Kozak¹³⁶⁾. Theoretically, therefore, a higher expression of UCP could be beneficial for body-fat loss. The study conducted by Sun & Zemel⁽Reference Sun and Zemel¹³⁷⁾ found that treatment of human adipocytes with 1,25(OH)₂D₃ inhibits UCP-2 mRNA and the protein level via a mechanism linked to the nuclear VDR. The potential role of UCP was also demonstrated in the normocalcemic, VDR knockout mouse⁽Reference Narvaez, Matthews and Broun¹⁷⁾. The lean phenotype of VDR knockout mice was characterised by a reduced serum leptin concentration, a compensatory increase of food intake and was associated with elevated levels of UCP-1 in adipose tissue. These studies show that 1,25(OH)₂D₃ may suppress UCP expression in adipocytes. However, such an effect would not necessarily contribute to fat accumulation in normal adipose tissue because ATP synthesis by mitochondria is tightly regulated, and physiological thermogenesis occurs mainly in brown adipose tissue.