Vitamin D signaling maintains intestinal innate immunity and gut microbiota: potential intervention for metabolic syndrome and NAFLD

Abstract

A lack of sunlight exposure, residence in the northern latitudes, and dietary vitamin D insufficiency are coprevalent with metabolic syndrome (MetS), Type 2 diabetes (T2D), and nonalcoholic fatty liver diseases (NAFLD), implying a potential causality and underlying mechanism. Whether vitamin D supplementation or treatment can improve these disorders is controversial, in part, because of the absence of large-scale trials. Experimental investigations, on the other hand, have uncovered novel biological functions of vitamin D in development, tumor suppression, and immune regulation, far beyond its original role as a vitamin that maintained calcium homeostasis. While the large intestine harbors massive numbers of microbes, the small intestine has a minimal quantity of bacteria, indicating the existence of a gating system located in the distal region of the small intestine that may restrain bacterial translocation to the small intestine. Vitamin D receptor (VDR) was found to be highly expressed at the distal region of small intestine, where the vitamin D signaling promotes innate immunity, including the expression of α-defensins by Paneth cells, and maintains the intestinal tight junctions. Thus, a new hypothesis is emerging, indicating that vitamin D deficiency may impair the intestinal innate immunity, including downregulation of Paneth cell defensins, leading to bacterial translocation, endotoxemia, systemic inflammation, insulin resistance, and hepatic steatosis. Here, we review the studies for vitamin D for innate immunity and metabolic homeostasis, and we outline the clinical trials of vitamin D for mitigating MetS, T2D, and NAFLD.

VITAMIN D, MORE THAN A VITAMIN

Having been discovered almost 100 years ago, and named as the fourth vitamin (54), vitamin D in many ways does not fit the classic definition for vitamins, since humans are able to synthesize it, given sufficient sunlight exposure. Differing biochemically and differentiating itself from other vitamins, which often serve as coenzymes for metabolic reactions, vitamin D is more like a steroid hormone that regulates gene expression through nuclear receptor (vitamin D receptor, VDR). As a consequence of evolution and migration from Africa to the northern latitudes with clothing covering their bodies, humans gradually developed mechanisms for de novo synthesis of vitamin D (18). To cope with insufficient sunshine exposure in current societies worldwide, dairy products are often fortified with vitamin D. Thus, vitamin D has become a micronutrient, and 10–20 µg/daily (400–800 IU) is recommended in many countries. Determination of the vitamin D status in the body is often assessed by measuring the plasma 25(OH)D₃. The 25(OH)D₃ has a half-life of ~3 wk, while the biological active component, 1,25(OH)₂VD is unstable, and its half-life is measured in hours. The definition of vitamin D deficiency or insufficiency is a matter of debate; but the plasma 25(OH)VD₃ levels at 50–75 nM is generally regarded as a healthy condition, as defined by the Institute of Medicine (IOM) and Endocrine Society (ES) (1, 37). Vitamin D insufficiency (plasma 25OHVD₃ at 25–50 nM) and deficiency (25OHVD₃<25 nM) are overwhelmingly prevalent worldwide. Vitamin D deficiency or insufficiency is a global health issue that afflicts more than 1 billion people worldwide (37). Recent studies showed that in Europe, vitamin D deficiency (<30 nmol/L) and insufficiency (<50 nmol/L) account for 13% and 40% of the general population, respectively (67). Public health agencies in many countries, including National Academy of Medicine in the United States, increased the recommendation of daily value of vitamin D (RDA) to 600 IU for people 1–70 yr-old, and 800 IU for 71+ yr-old (5).

In response to UVB exposure and through light-catalyzed reactions, the ring of 7-dehyroxyl-cholesterol is broken and converted to secosteroid, as cholecalciferol or vitamin D3 (Fig. 1). It is estimated that 20 min of direct sunlight exposure may synthesize 400–800 IU VD₃. Otherwise, fortified milk or vitamin D-containing foods, such as fish and yolk, are the major source of vitamin D. In sunlight, an equivalent secosteroid can be made in plants, for example, mushrooms, called vitamin D₂ (ergocalciferol), which has undistinguishable biological functions as VD₃. Vitamin D supplements (VD₂ or VD₃) are widely accepted in developed countries. Calcitriol, 1,25(OH)₂VD₃, the biologically active component of vitamin D, is synthesized by two steps of hydroxylation that occurs sequentially in the liver (CYP2R1/CYP27A1 for 25-hydroxylation) and kidney (CYP27B1 for 1α-hydroxylation). Turnover of calcitriol takes ~5 to 8 h in humans through hydroxylation at position-24 of the secosteroid (CYP24A1) followed by oxidation to vitamin D acid for excretion in the urine. Calcitriol, similar to the steroid hormones, assumes its biological functions mainly through the vitamin D receptor (VDR), a typical nuclear receptor and transcriptional factor, which binds at particular DNA response elements called vitamin D response elements (VDRE) for gene expression in genome-wide scales and epigenetic determination (42).

Fig. 1.

Vitamin D biosynthesis and major biological functions. With exposure to sunlight, animals, including humans and certain plants, can synthesize vitamin D through light reactions in the epithelial cells. As the consequence of urbanization and a lack of sunlight exposure, vitamin D has been regarded as a micronutrient in supplements and dairy products. The biologically active vitamin D, 1,25(OH)₂VD, also known as calcitriol, is synthesized sequentially in the liver and kidney. In the course of evolution with regular exposure to sunlight, humans and other terrestrial animals have developed the vitamin D-dependent metabolisms, mediated through vitamin D receptor (VDR) and signaling for transcriptional control at a genome-wide scale. In addition to promoting calcium/phosphorus adsorption in the small intestine, vitamin D exerts pleiotropic functions. It is known that vitamin D signaling maintains immune homeostasis through suppression of adaptive immunity and promoting innate immunity.

The prevalence of vitamin D deficiency overlaps with the incidence of Type 2 diabetes (T2D), metabolic syndrome (MetS), and nonalcoholic fatty liver disease (NAFLD). Metabolic diseases, including obesity, T2D, MetS, and NAFLD, often share common etiology basis, such as systemic inflammation, insulin resistance, cardiovascular disorders, and hepatic steatosis. These metabolic diseases also have common causes, such as excessive calorie intake and insufficient physical activity. On the other hand, a lack of sunlight exposure is often linked to obesity and MetS, suggesting the contribution of vitamin D deficiency. Many epidemiological surveys clearly demonstrate an association of vitamin D deficiency with T2D, MetS, and NAFLD. For example, a cohort study in the European Prospective Investigation with 9,671 incident T2D cases found that the plasma total 25(OH)D levels were inversely associated with the incidence of T2D (97). A cross-sectional study from the United States found a significant inverse relationship between 1,25(OH)₂D (n = 1,048) and 25(OH)D (n = 2,096) concentrations, with high triglyceride and low high-density lipoprotein levels (11). A report from the National Health and Nutrition Examination Surveys 2003–2006 concluded that higher 25(OH)D and dietary vitamin D intake are associated with reduced prevalence of metabolic syndrome (51). The association of vitamin D deficiency and metabolic risk was also reported in Chinese adults with prediabetes (81). One study of 4,364 postmenopausal Korean women demonstrated that an increased serum 25(OH)D level was associated with low incidence of elevated blood pressure, elevated TGs, and reduced HDL-C (22). Vitamin D levels are inversely associated with the adolescent obesity and T2D (30). One study of people living in Northern Finland (latitude 65°N) noted that nearly 80% of the subjects had low vitamin D levels, which were associated with a high prevalence of MetS, while the subjects with vitamin D supplementation had a significantly lower incidence of MetS (59). A recent meta-analysis has suggested a significant reduction in alkaline phosphatase after vitamin D supplementation in NAFLD patients (52). Low 25(OH)vitamin D levels are associated with the presence of NAFLD, and VDR expression in the liver is negatively associated with the severity of liver histology in nonalcoholic steatohepatitis (NASH) (8, 9). A study of biopsy-proven NAFLD found that serum 25OHD₃ levels and the VDR gene single nucleotide polymorphism (SNP) were significantly and independently associated with the severity of liver fibrosis (3).

Despite the public enthusiasm and apparently strong association between vitamin D deficiency and these metabolic diseases, it is still unclear whether vitamin D deficiency contributes or enhances the metabolic syndrome and NAFLD. Although clinical surveys and trials have sometimes given contradictory results, the experimental models based on animal studies have clearly demonstrated the critical roles of vitamin D signaling in physiological functions beyond its roles in calcium and phosphorus homeostasis (18). More well-designed human studies and trials are needed. In this review, we focus on vitamin D functions in innate immunity for antagonizing the MetS, T2D, and NAFLD. Additional studies for the association of vitamin D (VD) status with MetS, T2D, and NAFLD are summarized in Table 1.

Table 1.

Clinical investigation for the association of vitamin D status with metabolic syndrome, T2D, and NAFLD

First Author	Study Design	n	Participants	Vitamin D Measure	Efficacy Measurement	Conclusion
Ceglia et al. (20)	Cross-sectional	2,161	Adults with prediabetes	Weight and 25OHD at baseline	Body weight, BMI, waist circumference, and adipose tissue distribution	Inversely associated
Krul-Poel et al. (45)	Cross-sectional	241	Dutch patients with Type 2 diabetes	50,000 IU vitamin D once a month for 6 mo	Glycemic control	No significant association with health-related quality of life
Naderpoor et al. (60)	Cross-sectional	120	Healthy volunteers with no history of liver disease	Serum 25OHD concentration and liver enzymes	Liver enzymes (GGT, ALT, and ALP)	No association
Park et al. (63)	Cross-sectional	7,514	Healthy Korean adults	Serum 25OHD levels and visceral fat mass	NAFLD was diagnosed by using abdominal ultrasonography	Vitamin D status was associated with NAFLD in men but not in women.
Zhai et al. (95)	Cross-sectional	5,066	Adults in 16 areas of East China	25OHD levels, plasma glucose and lipid profile and insulin resistance	NAFLD evaluation by liver ultrasonography	A significantly higher prevalence of NAFLD in men with lower vitamin D
Wang et al. (87)	Cross-sectional	4,161	2,700 men with a mean age of 53 yr and 1,461 postmenopausal women over 55	25OHD and sex hormone binding globulin SHBG levels	The degree of fatty liver based on ultrasonography following guidelines of the American Diabetes Association	The combined association of low SHBG and low vitamin D was much higher in moderate-severe NAFLD

ALP, alkaline phosphatase; ALT, amino transferase; GGT, glutamyl transpeptidase; NAFLD, nonalcoholic fatty liver disease; SHBG, sex hormone-binding globulin; T2D, Type 2 diabetes.

Vitamin D suppresses inflammation and Th1 response. Decades of work have shown that vitamin D plays critical roles in modulation of adaptive and innate immunity (56). For example, the vitamin D metabolite 1,25(OH)₂VD₃ can inhibit T-helper 1 (T_H1)- and enhance T_H2-cell responses. It also decreases T_H17 cell differentiation, with reciprocal upregulation of forkhead box protein 3 (FOXP3)⁺ regulatory T (T_Reg) cells and T regulatory type 1 (T_R1) cells (99). 1,25(OH)₂VD₃ also inhibits the proliferation of B cells and their differentiation into antibody-secreting cells. 1,25(OH)₂VD₃ modulates dendritic cell functions by impairing their maturation to suppress Th1 cells (64). Thus, VD signaling contributes to immune homeostasis by suppression of adaptive immunity. It has been known for centuries that sunlight exposure and fish liver oil could mitigate tuberculosis. Vitamin D as an immune adjuvant can suppress Mycobacterium tuberculosis grown in macrophages through Toll-like receptor (TLR) activation (48). The gastrointestinal tract is the major tissue in humans that expresses VDR in high quantity, suggesting physiological roles of vitamin D signaling in the innate immunity of the gut. Intestinal tissues of vitamin D₃-deficient mice displayed increased inflammatory cell infiltrates, as well as significantly higher gene transcript levels of inflammatory mediators, such as TNF-α, IL-1β, IL-6, TGF-β, IL-17A, and IL-17F, as well as the antimicrobial peptide REG3γ (16). One study identified the inhibitory effects of vitamin D on LPS-stimulated inflammatory response in human blood monocytes and noticed that both 1,25OH₂D₃ and 25OHD₃ dose dependently inhibited LPS-induced p38 phosphorylation through induction of MAPK phosphatase-1 (MKP-1) in monocytes/macrophages (96).

Oral immune tolerance for food antigens and microbes from the gut is mediated by intestinal immunity, which is mediated, in part, by the peripheral mucosal iT_reg cells (induced T-regulatory cells) and tolerogenic antigen-presenting cells in the gut-associated lymphoid tissue. Vitamin D signaling contributes significantly to the immune tolerance in many ways. For example, in a colitis animal model, mice carrying VDR deletion in gut epithelial cells (Villin-Cre-VDR) developed more severe colitis than VDR control mice, characterized by more robust T1 and T17 responses, with greater increases in mucosal IFN-γ, IL-17, and IFN-γ IL-17 T cells (34). The functions of VD in immune tolerance were further demonstrated by clinical studies in which providing human subjects a high dose of VD supplement promoted peripheral T_Reg (13, 69).

Vitamin D promotes intestinal innate immunity and barrier functions. The intestinal mucosal barrier through its physical, biochemical, and immune machineries constitutes the first barrier in the gut. First, mucus separates the hydrolytic luminal contents from the intestinal epithelium (83). The mucus consists of a highly glycosylated hydrated gel (mucin) that is secreted by Goblet cells (G cells). The mucus also works as a molecular lubricator, preventing large food particles from injuring the epithelial cell layer, while allowing small molecules to pass through. Mucin 2 coding gene, muc2, has VDRE cis-elements in its proximal promoter, and dietary vitamin D depletion downregulated muc2 expression and disruption of the mucus lining (76). In animal models, VD signaling was found to regulate mucin gene expression, and dietary vitamin D deficiency led to impaired mucus and increased permeability. Tight junctions in the lateral sides of intestinal epithelial cells are critical for innate immunity. Work on animal models has demonstrated that vitamin D deficiency, either dietary depletion of VD or VDR genetic knockout (KO), can significantly downregulate mucin genes, as well as tight junction genes, such as ZO1, occludin, and claudin (76). In dextran sulfate sodium (DSS)-induced colitis model, VDR KO mice were much more susceptible for intestinal injury than the wild-type mice (43). From in vitro cell culture, it was known that calcitriol could promote the tight junctions and improve the impermeability for bacterial endotoxin infiltration (4). Intestinal bacterial translocation to the liver is often found in chronic liver disease patients, in part, due to loss of the intestinal barrier. On the basis of the rat model of liver disease, one found that 1,25(OH)₂D₃ could improve the cirrhosis through increased intestinal tight junction protein expression and barrier exclusion (88).

The gut is an active site of vitamin D-targeting tissue, while Cyp27B1, a critical enzyme for 1α-hydroxylation of 25OHVD was found expressed in the gut, suggesting potential paracrine of the VD signaling for resident immune and intestinal epithelial cells (2). Animal work showed that vitamin D deficiency altered intestinal mucosal defense and increases susceptibility to Citrobacter rodentium-induced colitis (71). The vitamin D-deficient mice suffered greater bacterial translocation, including the mesenteric lymph nodes, spleen, and liver.

One study showed that pigs subjected to antigen exposure together with the addition of 1α,25(OH)₂D₃ could enhance the antigen-specific IgA-response and activate the lymphocytes in the GALT tissues (84). Innate lymphoid cells (ILC) are critical for protective immunity and mucosal barrier in the gut, and a study showed that vitamin D receptor signaling regulates gut ILC3. VDR deletion led to reduction in colonic ILC3 populations at steady state and impaired ILC3 responses following Citrobacter rodentium infection, resulting in substantial increases in intestinal bacterial growth and mouse mortality (35)

The innate response is often initiated when foreign microbes are recognized by pattern recognition receptors (PRR), a group of sensor proteins on the cellular membrane or inside cells, which recognize components that are common for the microorganisms. The PRR of the innate immune cells, including dendritic cells (DCs), macrophages, monocytes, neutrophils, and epithelial cells in the small intestine, can recognize two categories of targeting components: the pathogen-associated molecular patterns (PAMPs), which are derived from the foreign microbes or antigens; and damage-associated molecular patterns (DAMPs), which are derived from the host’s damaged or dead components. Among the PRR superfamily are 1) the TLR, located either in the plasma membrane or in the cytoplasm; 2) the cytosolic receptors, such as NOD-like receptors and RIG (retinoic acid-inducible genes)-like receptors; and 3) inflammasomes that make matured IL-1β and IL-18. Vitamin D signaling was found to suppress the expression of PRR to prevent the host from excessive immune reactions for tissue homeostasis (41). It is well known how VD signaling upregulates the gene expression through VDR/VDRE-promoted transcription initiation, while it is less known how VD signaling suppresses the genes for inflammation.

Vitamin D promotes the Paneth cell defensins to restrain microbial growth in the small intestine (Fig. 2). A remarkable function of innate immunity in the gut is the production antimicrobe peptides (AMPs) that can directly suppress or kill the microbes in the lumen of the small intestine. In the small intestine, Paneth cells (PCs), located at the bottom of crypts of Lieberkühn, are the major source of lumen-secreted antimicrobial components, including α‐defensins along with lysozyme and phospholipase A2. Paneth cells are exclusively found throughout the intestinal epithelium, and there are a greater number of PCs in the distal small intestine compared with the proximal portions. Most of defensin peptides made by innate immune cells, such as neutrophils and NK cells, are secreted into the bloodstream in an endocrine fashion, while α-defensin 5 (DEFA5) and α-defensin 6 (DEFA6) (cryptdins for mice), produced by Paneth cells, are secreted exclusively into the lumen of the small intestine. Defensins are 3–4 kDa, small conserved amphipathic peptides rich in cationic and hydrophobic amino acid residues. α-Defensins have two peptide chains with six conserved cysteine residues to form three pairs of disulfide bonds. In Paneth cells, prodefensins are proteolytically processed and activated by matrix metalloproteinase-7 (MMP7) (6). In sensing of bacterial antigens, Paneth cells can quickly release the defensins (7). Lacking VD signaling, either dietary vitamin D depletion or VDR-KO can downregulate the expression of MMP7 and α-defensin 5 in Paneth cells for gut dysbiosis (77). Autophagy was found to be essential for Paneth cells to sense bacterial invasion and secrete antibacterial peptides (12, 15). One study found that vitamin D through induction of autophagy controlled expression of cathelicidin, a protein that has direct antimicrobial activity (93). Using transgenic mice to express human DEFA5, authors of one study showed reduction of gut Firmicutes, and concomitantly promoted the relative abundance of Bacteroidetes. However, MMP7 KO mice, presumably lacking of α-defensin maturation and activation, had opposite consequences, which collectively demonstrated the critical roles of α-defensins from Paneth cells in maintaining intestinal eubiosis (73). Likewise, the impaired expression of the Paneth cell-specific defensins has been found in obese human subjects (36).

Fig. 2.

Vitamin D in modulating intestinal innate immunity to determine gut microbiota. While the large intestine harbors massive number of microbes, the small intestine has a minimal level of the microbiome, indicating the existence of a gating system at the distal region of the ileum, which restrains the translocalization of gut microbes into the small intestine. The vitamin D receptor is highly expressed at the end of small intestine, and vitamin D signaling can upregulate the Paneth cell defensins that restrain bacterial growth in the small intestine. Moreover, vitamin D signaling also upregulates the tight junctions of intestinal epithelial cells and mucin expression in the Goblet cells for innate immune homeostasis. Conversely, vitamin D deficiency, which is often associated with Type 2 diabetes and nonalcoholic fatty liver disease, may lead to gut dysbiosis and endotoxemia for systemic inflammation, insulin resistance, and metabolic disorders.

Vitamin D regulates gut innate immunity in the pattern recognition receptor (PRR) pathways. PRRs are essential for the innate immune system. PRRs are cellular sensors, which detect molecules typical for the pathogen types. PRRs in the intestinal epithelial cells can identify two classes of molecules: pathogen-associated molecular patterns (PAMPs) from microbial pathogens and damage-associated molecular patterns (DAMPs) from damaged tissues. PRRs also initiate antigen-specific adaptive immune response and release of inflammatory mediators. There are membrane-bound PRRs, including Toll-like receptors (TLRs) and C-type lectin receptors (CLRs), as well as cytoplasmic PRRs, such as NOD-like receptors (NLRs) and RIG-I-like receptors (RLRs). An in vitro study showed that calcitriol decreased TLR-induced cytokine production and enhanced cytokine levels induced by muramyl dipeptide (MDP), the NOD2 ligand. However, calcitriol inhibited Mo-D-TLR-induced cytokines, but costimulation of NOD2 resulted in increased IL-10 and IL-23 (25). In another in vitro experiment, calcitriol was found to stimulate expression of NOD2/CARD15/IBD1 gene and protein in primary human monocytic and epithelial cells, whereas NOD2 was often downregulated in inflammatory bowel diseases (89). It was shown that calcitriol can suppress TLR4-mediated inflammation through induction of IκB, a cellular inhibitor IKK pathway (28). Likewise, vitamin D downregulates proinflammatory cytokine response to Mycobacterium tuberculosis through PRR, while inducing protective cathelicidin production (41).

Vitamin D, through epigenetic remodeling, regulates intestinal physiology. Approximately 3% of the mouse and human genomes are regulated directly or indirectly by vitamin D signaling (19). VDR, functioning as heterodimer with retinoid X receptor (RXR) as VDR/RXR, can directly bind to specific genomic sequences on the promoter regions of target genes, called vitamin D response elements (or VDREs), which can further recruit transcription factors, cotranscriptional factors, or cosuppressors to either activate or suppress specific gene transcription. The nature of VDR binding to specific genes relies on cell types and their existing epigenetic code, namely, the chromatin modifications, such as methylation and acetylation. Furthermore, VD-VDR binding may also impact the local epigenetic modeling that may lead to various consequences, such as cell cycle, immune function, cell proliferation and differentiation, growth regulation, and tumor suppression. For example, VDR along with histone modification enzymes may be involved in acetylation, deacetylation, methylation, and demethylation, of different epigenetically regulated target genes (94). Vitamin D changes the status of DNA methylation and histone modifications, resulting in the activation of tumor suppressors and inhibition of oncogenes. In addition, vitamin D activates the expression of tumor-suppressing miRNAs, which contribute to the tumor-suppressive activity (78). Epigenetic studies showed that 1,25(OH)₂D₃ can regulate thousands of genes at genome-wide scale, and ~2,000 to 8,000 VDR binding sites can be detected following activation by 1,25(OH)₂D₃ stimulation in a cell-type-dependent manner. On the other hand, large amounts of VDR are persistently bound in chromatin independent of 1,25(OH)₂D₃ binding (65). Moreover, there are reports showing how 1,25(OH)₂D₃ regulates transcription via multiple enhancers at distal regions of the targeting genes.

Nutritional abnormalities are common in patients with gastroparesis, a disorder that may affect gastric motility and may delay emptying. One report showed that low 25(OH)VD levels is associated with slow gastric emptying (40). Clinical trials are under way to examine the influence of vitamin D on the effects of epigenetic aging (21). A recent report found that neuroprotective effects of vitamin D for high-fat diet and palmitic acid-induced enteric neuronal loss in the mice (47).

SUNLIGHT EXPOSURE AND VITAMIN D IN MODULATING GUT MICROBIOTA

Whether sunlight exposure or vitamin D can modulate gut microbiota is an intriguing topic. In zoology and at an ecogeographical scale, Bergmann’s Rule states that the body size of the warm-blooded animals is related to the local temperature and sunlight at various latitudes (55). One study found a positive correlation between Firmicutes and northern latitudes, and a negative correlation between Bacteroidetes, showing that variation of gut microbial composition is consistent with the pattern expected by Bergmann’s rule (79). Importantly, this finding also implies that sunlight exposure may impact on gut microbiota and the body’s metabolic capacity and homeostasis. It is well known that a high abundance of Firmicutes and a low abundance of Bacteroidetes in gut microbiome are often related to obese and metabolic syndromes. It is known that Firmicutes promote energy harvesting (82). Using conditional knockout mice, one group found that a lack of vitamin D signaling in intestinal epithelial cells resulted in gut dysbiosis and exacerbated drug-induced inflammatory bowel diseases (92). In the VDR KO mice, Lactobacillus was depleted in the fecal stool, whereas Clostridium and Bacteroides were enriched (38).

Using animal models, we demonstrated the importance of vitamin D signaling in maintaining gut microbiota in an eubiosis state through functioning Paneth cells (76). In particular, we found VDR is highly expressed in the distal region of small intestine where the VDR levels are ~1,000-fold greater than that in the liver. VDR is abundantly expressed in the cells of crypts of Lieberkühn, the location of the Paneth cells. We found that vitamin D signaling is critical for maintaining the high-level expression of α-defensin 5, and MMP7, a converting enzyme for pro-α-defensin. Conversely, dietary depletion of vitamin D or VDR knockout resulted in downregulation of α-defensins and MMP7, as well as tight junction proteins in the ileum. Likewise, in an animal model of high-fat feeding, we found that sufficient vitamin D in the standard chow (AIN93), by which the VD levels are equivalent to the daily value of diet for humans, could significantly suppress the metabolic disorders featured by partial restoration of insulin sensitivity, reduction of plasma triglycerides, improved glucose tolerance, and attenuation of hepatic steatosis. Importantly, vitamin D deficiency or VDR knockout in the animal models exacerbated gut dysbiosis, showing increased abundance of Firmicutes and decreased Bacteroidetes. In particular, in mice in C57/BL6 background, Helicobacter hepaticus, a pathogenic species was substantially increased in the small intestine, while Akkermansia muciniphila, a symbiotic, was downregulated in the mice fed the vitamin D-depleted chow or in VDR knockout mice. Conversely, oral administration of synthetic human DEFA5 was able to suppress Helicobacter hepaticus and restore Akkermansia muciniphila in the gut. A major signature of gut dysbiosis is the increased Firmicutes and downregulation of Bacteroidetes, while how that contributes to metabolic diseases is less understood. Most of Bacteriodetes are Gram-negative bacteria, and their endotoxin, lipopolysaccharide (LPS), is a prominent ligand as PAMP for PRR activation and inflammation. Chronic infection and systemic inflammation are the key driving force for insulin resistance. One study showed that subcutaneous infusion of LPS could produce hyperglycemia and insulinemia and fatty liver in the mice (17). Decreased Bacteroidetes, presumably due to their death, may release LPS to drive systemic inflammation, which is a key driving force for insulin resistance and metabolic disorders. We found that oral administration of cholestyramine, polycationic resins, could sufficiently deplete gut endotoxin and improve insulin resistance and metabolic disorders and fatty liver in the high-fat feeding animal models (98). Another study showed that probiotics Lactobacillus rhamnosus and Lactobacillus plantarum were able to increase VDR protein expression in both mouse and human intestinal epithelial cells, while probiotic treatment also enhanced numbers of Paneth cells (91). Cyp27b1 (Cyp) knockout mice have impaired synthesis of 1,25-hydroxylation of VD, and in these mice, dysbiosis appeared. Giving 1,25(OH)₂D₃ to the Cyp KO mice decreased the colitis severity and suppressed Helicobacteraceae in the feces (62). A genome-wide association analysis (GWAS) in 1,812 individuals identified the variation in levels of vitamin D receptor among other host factors influencing the gut microbiota, and nongenetic and genetic factors, each accounting for ~10% of the variation in gut microbiota (86). A meta-analysis showed that 12 of 14 human studies found association of vitamin D deficiency and altered gut microbiota (90).

VITAMIN D INTERVENTION FOR METABOLIC SYNDROME AND NAFLD

While compelling data from epidemiological surveys show that vitamin D deficiency is associated with the incidence of MetS, T2D, and NAFLD, controlled clinical trials have emerged testing causality and efficacy. One controlled and double-blind intervention was conducted to test the impact of vitamin D on insulin resistance by administering 4,000 IU vitamin D₃ (n = 42) or placebo (n = 39) daily for 6 mo to the T2D patients with vitamin D deficiency. With the treatment, the serum 25OHD₃ increased significantly from 21 to 75 nmol/l with supplementation and significant improvements in insulin sensitivity were achieved (85). Another trial showed that daily supplementation of vitamin D for 12 wk, along with moderate endurance physical activity, significantly increased 25OHD₃ concentration and induced a significant reduction in lipid profile in metabolic syndrome patients (26). Similarly, for postmenopausal women with VD deficiency (n = 80), supplementation with 1,000 IU vitamin D₃ for 9 mo was associated with a reduction in the MetS risk profile, showing a reduction of hypertriglyceridemia and hyperglycemia (27). Elevation of plasma von Willebrand factor is associated with increased risk of metabolic syndrome and stroke. Data from 14 clinical trials that contained a total of 1,253 participants, and one meta-analysis found that vitamin D supplementation (≤4,000 IU/day) could significantly decrease von Willebrand factor (vWF) (80). Polycystic ovary syndrome (PCOS) is often related to metabolic syndrome, such as obesity and T2D. One trial showed that 50,000 IU vitamin D supplementation every other week for 8 wk had beneficial effects on insulin metabolism and lipid profile in PCOS patients.

The large-scale clinical trials using vitamin D on T2D are emerging. For example, a total of 2,423 participants with pre-T2D-diabetic conditions underwent randomization (1,211 participants to the vitamin D group, 4,000 IU/daily, and 1,212 to the placebo group). By month 24, while the serum 25-hydroxyvitamin D level in the vitamin D group was significantly improved to the placebo group, the incidence of progression to T2D stage did not change significantly (68). In a small scale, randomized, placebo-controlled study, 47 patients of T2D were given vitamin D at a dose of 1,000 U/day for 12 mo. Metabolic parameters, including fasting glucose, lipid profile, HbA1C, insulin, hs-CRP, 25 OH vitamin D, adiponectin and leptin, HOMA-IR, and central aortic augmentation index (AI) were evaluated (14). The results showed improved AI, while other parameters did not change statistically. Additional information regarding the clinical interventions using VD supplementation for MetS and NAFLD is summarized in Table 2.

Table 2.

Clinical intervention with vitamin D for metabolic syndrome, T2D, and NAFLD

First Author	Study Design	n	Participants	Intervention	Goals and Primary Outcomes
Soric et al. (75)	Randomized, interventional study (NCT00985361)	37	Type 2 diabetes patients with hemoglobin A1C >7%, aged 21 to 75 yr	Vitamin D3 2,000 IU tablets daily for 3 mo	Hemoglobin A1c N.A.
Breslavsky et al. (14)	Randomized, interventional study (NCT01854463)	47	Type 2 diabetes patients	25-hydroxy vitamin D 1,000 IU daily for 12 mo	Glycemic control status Significant decrease in AI during 1 yr of treatment
Kampmann et al. (39)	Randomized, interventional study (NCT00812578)	16	Diabetes patients with 25(OH)D <50 nmol/L	8 tablets cholecalciferol equivalent to 280 µg for 2 wk and 4 tablets equivalent to 140 µg for 10 wk	Insulin secretion and insulin resistance Not improve insulin resistance, blood pressure, inflammation or HbA1c, but might increase insulin secretion in patients with established Type 2 diabetes
Krul-Poel et al. (44)	Randomized double-blind placebo-controlled trial (NTR3154)	300	T2D patients receiving lifestyle advice, metformin, or sulfonylurea derivatives	placebo or 50,000 IU vitamin D3 at monthly intervals	change in glycated hemoglobin level between baseline and six months N.A.
Sadiya et al. (72)	Randomized controlled trial (NCT02101151)	87	T2DM patients with obesity	In Phase 1 the vitamin D arm received 6,000 IU vitamin D3/day (3 mo) followed by Phase 2 with 3,000 IU vitamin D3/day. During follow up (phase 3), both of the arms were unblinded and supplemented with 2,200 IU vitamin D3/day for another 6 mo.	Anthropometric measures High daily dose of vitamin D3 we did not achieve target levels of serum-25(OH)D above 75 nmol/L
Gulseth et al. (32)	Randomized controlled trial (NCT00992797)	62	T2DM patients with VD deficiency	400,000 IU with an additional 200,000 IU D3 if serum 25(OH)D was <100 nmol/L after 4 wk	Euglycemic hyperinsulinemic clamp and intravenous glucose tolerance test IVGTT No improvement of change insulin sensitivity or insulin secretion
Rezagholizadeh et al. (70)	Randomized controlled trial	40	T2DM patients	4000 IU vitamin D3 for 2 mo	Fasting blood glucose FBG, HbA1c, and insulin Increase serum anti-inflammatory adipokine SFRP5 but decrease serum pro-inflammatory Wnt5a and TNF-α
Harris et al. (33)	Randomized controlled trial	22	Overweight subjects	60,000 IU monthly vitamin D3 for 16 wk	Endothelial function by low-mediated dilation Testing Improving vascular endothelial function in African-American adults
Larsen et al. (46)	Randomized controlled trial (NCT01166165)	112	Hypertension patients	3,000 IU daily cholecalciferol for 20 wk	Central blood pressure measurement No improvement of blood pressure
Martins et al. (53)	Randomized controlled trial	130	Overweight and obese subjects	100,000 IU vitamin D3 monthly for 3 mo	Adipocyte cytokine expression and arterial stiffness Increased serum 25(OH)D levels, decreased intact PTH level and the levels of select inflammatory and oxidative stress mediators of arterial stiffness. Longer term prospective studies are warranted to evaluate the effect of high dose vitamin D supplementation on arterial stiffness.
Pilz et al. (66)	Randomized controlled trial (NCT02136771)	188	Patients with hypertension	2,800 IU vitamin D3 daily for 8 wk	24-h systolic blood pressure Not significant
de Courten et al. (24)	Randomized, interventional study (NCT02112721)	50	Nondiabetic patients with 25(OH)D <50 nmol/L, aged 18 to 60 yr	Each participant was given an initial dose of 100,000 IU of vitamin D3. Thereafter, participants took 4,000 IU daily for a period of 16 wk.	Insulin sensitivity measure using euglycemic glucose clamp N.A.
Forouhi et al. (29)	Randomized controlled trial	340	Patients with prediabetes	100,000 IU vitamin D2 or D3 monthly for 4 mo	HbA1c Short-term supplementation with vitamin D2 or D3 had no effect on HbA1c.
Moreira-Lucas et al. (57)	Randomized controlled trial	71	Patients with prediabetes with VD deficiency	28,000 IU weekly vitamin D3 for 24 wk	2-h 75-g oral glucose tolerance test No improvement in oral glucose tolerance
Mousa et al. (58)	Randomized controlled trial	65	Overweight or obese subjects with VD deficiency	A bolus oral dose of 100,000 IU cholecalciferol followed by 4,000 IU cholecalciferol/day or a matching placebo for 16 wk	Change in insulin resistance by IVGTT No effects
Niroomand et al. (61)	Randomized controlled trial	162	Prediabetes patients with VD deficiency	50,000 IU weekly for 3 mo	Fasting plasma glucose FPG/ 2-h oral glucose tolerance test OGTT In patients with prediabetes and hypovitaminosis D, high-dose vitamin D improves insulin sensitivity and decreases risk of progression toward diabetes.
Farag et al. (26)	Randomized controlled trial	70	Metabolic syndrome patients	Vitamin D 2,000 IU daily for 12 wk with or without endurance physical activity	Lipid profiles (cholesterol, LDL-C, HDL-C) Daily supplementation of vitamin D for 12 wk, along with moderate-endurance physical activity, significantly increases vitamin D concentration and induces a significant reduction in lipid profile in metabolic syndrome patients.
Sharifi et al. (74)	Randomized controlled trial	53	NAFLD patients	50,000 IU every other week for 4 mo	Serum ALT levels, hs-CRP, HOMA-IR, TNF-α, malondialdehyde (MDA), total antioxidant capacity, and TGF-β improved vitamin D status and led to amelioration in serum hs-CRP and MDA in patients with NAFLD.
Barchetta et al. (10)	Randomized controlled trial	65	NAFLD patients with T2DM	2000 IU cholecalciferol daily over 24 wk	The reduction of MRI-measured hepatic fat fraction, change in transaminase/CK18-M30/P3NP levels and fatty liver index did not improve hepatic steatosis or metabolic/cardiovascular parameters.
Lorvand et al. (49)	Randomized controlled trial	73	NAFLD patients with vitamin D defficiency	25 μg calcitriol with hypocaloric diet	The grade of fatty liver disease by liver fat accumulation may have positive effects on lipid profile, liver enzyme tests, and insulin sensitivity during a weight-loss program.
Lorvand et al. (50)	Randomized controlled trial	120	NAFLD, overweight patients	25 μg calcitriol with or without 500 mg calcium for 12 wk	Lipid profile and glucose measures, Liver enzyme tests, and ultrosonography Calcium plus calcitriol supplementation for 12 wk may be potentially effective for biochemical parameters in NAFLD patients.
Dabbaghmanesh et al. (23)	Randomized controlled trial	91	NAFLD patients	50,000 IU vitamin D3 (cholecalciferol) pearl or 0.25 mg calcitriol weekly for 3 mo	Serum ALT, AST, and GGT No beneficial effects
Geier et al. (31)	Randomized controlled trial	22	NASH patients	2,100 IU vitamin D over 48 wk	Change in ALT level from baseline to the end of treatment Significant improvement of serum ALT levels

AI, augmentation index; ALT, amino transferase; AST, aspartate aminotransferase; BMI, body mass index; FPG, fasting plasma glucose; GGT, glutamyl transpeptidase; HDL-C, high-density lipoprotein-cholesterol; hs-CRP, high-sensitivity C-reactive protein; IU, international unit; IVGTT, intravenous glucose tolerance test; LDL-C, low-density lipoprotein-cholesterol; NAFLD, nonalcoholic fatty liver disease; NASH, nonalcoholic steatohepatitis; OGTT, oral glucose tolerance test; P3NP, procollagen type III NH₂-terminal peptide; T2D, Type 2 diabetes; T2DM, Type 2 diabetes mellitus.

CONCLUSIONS AND PERSPECTIVE

As terrestrial animals, humans have evolved relying on the sunlight, which not only provides the warmth, but initiates the synthesis of vitamin D to regulate metabolic homeostasis in many ways. With human migration to northern latitudes, we have to depend on vitamin D as a micronutrient through supplementation. Vitamin D regulates the genome on a wide scale. As an immune adjuvant, vitamin D suppresses adaptive immunity but upregulates innate immunity. In particular, VDR is highly expressed in the gut epithelial cells, including Paneth cells to regulate α-defensins to restrain microbial overgrowth in the small intestine. Epidemiological surveys show an association of the prevalence of vitamin D deficiency and obesity, metabolic syndrome, T2D, and NAFLD, indicating potential linkages. Evidence from animal studies demonstrate that lacking VD or VD signaling pathways impairs gut innate immunity, leading to gut dysbiosis and low-grade systemic inflammation, a key driving force for insulin resistance and metabolic disorders. Evidence is emerging that vitamin D supplementation improves metabolic syndrome, although the impact on NAFLD is not known. Large-scale and controlled trials are needed to address the role of VD in these metabolic disorders (76).

GRANTS

The work was supported by the Natural Science Foundation of China Grants 31571165 and 31771288 to Y.-P. Han, the National Institutes of Health (Bethesda, MD) Grants P01CA163200, P50 AA011999, and P01DK-098108 to S. J. Pandol, and the Sichuan Science and Technology Program (No. 2017TJPT0013) and Public Health and Clinical Center of Chengdu Scientific Research Project (Grant 2018K01) to Y. L. Zeng.

DISCLOSURES

No conflicts of interest, financial or otherwise, are declared by the authors.

AUTHOR CONTRIBUTIONS

Y.Z. and M.L. conceived the review; Liwei Pan, S.G., D.L., L. Z., Y. L, Lisha Pan, S. X., R. Z., C. Z., P. W., and L.G helped literature search and analysis; Y.C. prepared figures; Y.-P.H. drafted manuscript; and M.N., S.J.P., and Y.-P.H. edited and revised manuscript.