In the last decade, the importance of vitamin D beyond its effects on calcium homeostasis has become more evident and its functions in immune modulation, cell differentiation and proliferation, and the inflammatory response have all been well described. However, an interesting paradox has developed: while our understanding of the physiological functions of vitamin D have become better known, vitamin D deficiency (VDD) has silently become increasingly more common in Western populations. Concomitantly, the last decade has seen nonalcoholic fatty liver disease (NAFLD) rise to become the most common cause of chronic liver disease in Western nations. Given that VDD and NAFLD have both indirect and direct associations with obesity and sedentary lifestyle, it is not unexpected that VDD would coexist with NAFLD. A growing body of evidence points to a linked and potentially causative relationship between VDD and NAFLD. This review will assess the role of VDD in the pathogenesis of NAFLD, identify trends in the epidemiology of VDD and NAFLD, and evaluate available evidence on the clinical utility of vitamin D replacement in NAFLD populations.
Vitamin D is a secosteroid with known effects on calcium homeostasis that has recently been shown to have other significant functions regarding immune modulation, cell differentiation and proliferation, and the inflammatory response. As our understanding of the many functions of vitamin D has grown, the presence of vitamin D deficiency (VDD) has become more evident in Western populations. Concomitantly, nonalcoholic fatty liver disease (NAFLD) has become the most common cause of chronic liver disease. NAFLD and VDD are often found together, and while this is not unexpected, given their similar associations with obesity and sedentary lifestyle, a growing body of evidence points to a closely linked and potentially causative relationship between VDD and NAFLD. The epidemiologic association between VDD and NAFLD as well as the role of VDD in the pathogenesis of NAFLD and the available evidence on the clinical utility of vitamin D replacement in NAFLD populations are discussed. (Hepatology 2013;53:1166–1174)
vitamin D-binding protein
farnesoid X receptor
- insulin resistance NAFLD
nonalcoholic fatty liver disease
randomized controlled trial
retinoid X receptor
single nucleotide polymorphism
vitamin D deficiency.
Vitamin D Metabolism
Vitamin D is a secosteroid that is obtained both exogenously (Vitamin D2) and endogenously (Vitamin D3). Since dietary vitamin D (D2) is naturally contained in very few foods, dietary supplementation and dermal synthesis (D3) are the primary sources of vitamin D. Previtamin D3 is synthesized in skin by UV sunlight from 7-dehydrocholesterol which is then converted to vitamin D3 or cholecalciferol (Fig. 1). Dietary vitamin D2 is fat soluble and is absorbed by the small intestine and incorporated into chylomicrons where it is transported to the liver bound to vitamin D-binding protein (DBP). In the liver, vitamin D undergoes hydroxylation by 25-hydroxylase (CYP2R1) leading to formation of 25-hydroxyvitamin D [25(OH)D3] or calcidiol. 25(OH)D3 is transported to the kidney where it undergoes hydroxylation by 1a-hydroxylase (CYP27B1) to its active form 1a,25-dihydroxyvitamin D [1,25(OH)2D3]. Secondary to its long half-life and relatively stable levels in the blood, 25(OH)D3 is what is typically measured to assess an individual's vitamin D status, although 1,25(OH)2D3 is the hormonally active form of vitamin D.
Vitamin D3 Homeostasis
Vitamin D homeostasis is maintained by the synthetic activity of 1a-hydroxylase and catabolic activity of 24-hydroxylase (CYP24A1). 1,25(OH)2D3 regulates 1a-hydroxylase activity both directly through negative feedback but also by way of inhibition of parathyroid hormone (PTH) activity. Conversely, in response to hypocalcemia, PTH increases 1a-hydroxylase transcription and, therefore, 1,25(OH)2D3 synthesis through a cyclic adenosine monophosphate (cAMP)-dependent pathway.
Another mediator of vitamin D homeostasis is fibroblast growth factor 23 (FGF23) which is produced primarily by osteoblasts and osteocytes and influences vitamin D metabolism through down-regulation of 1a-hydroxylase activity and promotion of 24-hydroxylase activity. Sex hormones, calcitonin, and prolactin can also affect vitamin D homeostasis, though 1a-hydroxylase activity remains the primary factor in vitamin D homeostasis.
In addition to sun exposure and diet, vitamin D levels may also be affected by genetic factors and high heritability of VDD has been shown in several epidemiologic studies.[5, 6] The exact genes involved have only recently been investigated, with the most substantial study to date showing single nucleotide polymorphisms (SNPs) in the genes encoding CYPR21 and DBP were associated with vitamin D status in an initial cohort of 156 unrelated healthy Caucasians and a similar replication cohort of 340 patients. Given the essential role of CYPR21 and DBP in vitamin D homeostasis, these findings are not surprising and have been replicated in other studies. Interestingly, the study by Ramos-Lopez et al. associated the CYP2R1 gene with both vitamin D levels and type 1 diabetes, although no data exist evaluating the SNPs associated with vitamin D levels in NAFLD patients. Conversely, the genes associated with a high incidence of NAFLD have not been evaluated for a putative role in vitamin D metabolism.
Vitamin D3 Mechanism of Action
The primary mediator of vitamin D is the vitamin D nuclear receptor (VDR), which is a member of the superfamily of nuclear hormone receptors. VDR has four major domains that interact to confer ligand-activated transcription factor activity: a ligand-binding domain, a retinoid X receptor (RXR) heterodimerization domain, a DNA binding domain to vitamin D response elements, and a recruitment domain of VDR coregulators. VDR bound to RXR forms a heterodimer that interacts with vitamin D response elements (VDRE) located within promoter regions of target genes and leads to activation or repression of transcription. Target genes of the VDR are broad and include functions of hormone secretion, immune regulation, cellular proliferation, and differentiation.
Vitamin D Targets
The nonclassic actions of vitamin D can be grouped into three primary categories to include modulation of immunologic function, hormone secretion, and cellular proliferation and differentiation (Fig. 1).
Immunologic Functions of Vitamin D
The expression of VDR in immunologic cells including antigen-presenting cells (APCs) and lymphocytes as well as evidence of 1a-hydroxylase expression by activated macrophages suggests a role for vitamin D in the immune system.
Vitamin D's effects specific to the innate immune system are mediated by transmembrane pathogen receptors that recognize cell membrane patterns from pathogenic organisms called Toll-like receptors (TLRs) located in lymphopoietic cells, including Kupffer cells and epithelial cells. Activation of these TLRs through cellular production of 1a-hydroxylase and VDR leading to 1,25(OH)2D3 synthesis results in synthesis of reactive oxygen species and antimicrobial peptides including cathelicidin in both macrophages and epithelial cells. VDD may predispose individuals to endotoxin exposure secondary to decreased activation of this pathway.
The clinical application of VDD in the antimicrobial response was shown by Liu et al., who demonstrated human macrophage TLR activation led to expression of VDR and 1a-hydroxylase and thus cathelicidin, leading to death of intracellular Mycobacterium tuberculosis. Furthermore, African Americans, who have significantly decreased 1,25(OH)2D3 levels because of skin melanin content compared to Caucasian counterparts, were shown to have decreased production of cathelicidin. When vitamin D was repleted to physiologic levels, TLR-induced cathelicidin production was restored.
Vitamin D also influences the adaptive immune system through modulation of both T and B lymphocytes as well as production of cytokines and immunoglobulins. Chen et al. examined the role of vitamin D in regulation of autoantibody production and found that vitamin D inhibited proliferation of activated B cells, induced their apoptosis, and inhibited immunoglobulin secretion, suggesting that vitamin D-dependent B-cell regulation may be important in maintaining B-cell homeostasis.
VDD may also contribute to B-cell hyperactivity. Vitamin D acts on dendritic cells to reduce APC to CD4 cells, inhibit proliferation and differentiation of CD4 cells into T-helper1 (Th1) and Th17 cells, and promote differentiation into Th2 and Treg cells.[15, 16] The decrease in Th1 cells leads to decreased production of interferon-gamma (IFN-γ) and interleukin-2 (IL-2) as well as decreased macrophage activation, while the increase in TH2 cells leads to the production of IL-4, IL-5, and IL-10. This association suggests that vitamin D tempers the adaptive immune response. Specific effects of vitamin D on liver-related adaptive immunity remains to be determined but early evidence suggests that human T cells may be inactive against hepatitis C virus (HCV) infection in the setting of VDD.
Hormonal Function of Vitamin D
While the role of vitamin D in regulating bone homeostasis is well characterized, its role in the regulation of other hormones that are important in NAFLD, such as insulin and adiponectin, is less well defined. The potential association between vitamin D and diabetes was first described by Campbell et al., who noted glycemic control was worse in the winter months in 12 patients living in the Antarctic when the prevalence of VDD was higher. A subsequent systematic review of vitamin D and type 2 diabetes mellitus (DM) identified several longitudinal, observational studies reporting an inverse association between vitamin D status and risk of developing DM. Analysis of randomized controlled trials (RCTs) revealed no benefit from vitamin D supplementation in patients with normal glucose tolerance, but did show an improvement in glycemic indices in patients with baseline glucose intolerance or insulin resistance (IR).
Mechanistically, vitamin D is thought to act on pancreatic β cells, which have been shown to contain the both VDR and 1a-hydroxylase. Furthermore, the human insulin gene has been shown to contain a VDRE in its promoter region as well as transcriptional activation through vitamin D ligand-dependent binding.
Data suggest an association between vitamin D and adiponectin expression. A recent study demonstrated vitamin D supplementation with or without calcium was associated with an increase in serum adiponectin. Similarly, another study demonstrated an association between VDD and low adiponectin in type 2 diabetics. A potential explanation pertains to the renin-angiotensin system (RAS), where vitamin D decreases the expression of renin leading to decreased activation of the RAS. Adipocytes are known to stimulate a “local” RAS, which leads to inhibition of adiponectin secretion. Increased adipose-tissue RAS activation can therefore explain the low adiponectin levels seen with obesity, and conversely, vitamin D's inhibitory effects on the RAS can increase adiponectin levels.
Cellular Proliferation and Differentiation
Vitamin D also has effects on cellular proliferation and differentiation, predominantly in epidermal tissues and in the setting of malignancy. Vitamin D has been shown to promote differentiation of keratinocytes and inhibit their proliferation. Similarly, vitamin D has been shown to be involved in several malignancies where multiple neoplasms express the VDR. In keratinocytes with DNA damage, vitamin D promotes the repair of DNA damage, reduces apoptosis, and increases cell survival. A 4-year prospective trial suggested a clinical benefit of vitamin D therapy where treatment with 1,100 IU vitamin D and 1,400-1,500 mg calcium daily showed a 77% reduction in certain malignancies, including breast and colon cancer. Unfortunately, the benefit of vitamin D does not appear to extend to treating cancer, although study has been limited to small case series. Further studies are needed to determine if the antineoplastic effects of vitamin D are clinically relevant.
NAFLD and Vitamin D
NAFLD is by far the most common chronic liver disease in Western nations and carries an increased all-cause mortality, particularly in those patients who meet criteria for nonalcoholic steatohepatitis (NASH).[33, 34] The association between vitamin D levels and NAFLD has been increasingly recognized. Using the NHANES III database, Liangpunsakul and Chalasani reviewed over 6,800 patients and found 308 with unexplained elevation in alanine aminotransferase (ALT) and compared their serum vitamin D concentrations with those of 979 matched controls. NHANES III patients with elevated ALT were found to have lower vitamin D levels than the control group, even when controlling for metabolic syndrome, IR, and serum triglyceride level. This was confirmed in a study of 262 patients referred to an endocrinology clinic where the relationship between NAFLD and reduced vitamin D levels persisted regardless of age, sex, triglycerides, and IR.
Targher et al. confirmed the association between NAFLD and VDD and importantly evaluated the relationship of liver histology to vitamin D levels. Vitamin D concentrations were lower in NASH patients when compared to those with isolated fatty liver and inversely correlated with liver histology.
NAFLD/NASH Pathophysiologic Mechanisms
The understanding of NASH pathogenesis has evolved from the relatively simplistic “two-hit” hypothesis and includes a number of metabolic pathways resulting in hepatic steatosis, steatohepatitis, and hepatic fibrosis. A number of these pathways can be affected by vitamin D and relate to the hormonal, immunologic, and cellular differentiation “nonclassical” effects of vitamin D.
Hepatic steatosis is generally thought to arise from lipolysis derived flux of free fatty acids (FFA) from adipocytes, as well as dietary lipids, de novo lipogenesis, and impaired lipid disposal. The buildup of FFA results in insulin signaling defects and impairment of cellular glucose metabolism, with the resulting hyperglycemia leading to increased lipogenesis through increased activation of sterol regulatory element binding proteins (SREBP) as well as activation of carbohydrate response element binding proteins (CHREBP).
Visceral adipose tissue also plays an important role in a variety of inflammatory and immune reactions pertinent to NASH by way of secretion of adipocytokines such as adiponectin, resistin, and omentin. Adiponectin has been described as the prototypic adipocytokine by way of its function as an antiinflammatory agent. Low adiponectin levels are independently associated with obesity and NASH and adiponectin levels increase after weight loss. In murine models, high levels of adiponectin have been experimentally shown to decrease necroinflammation and steatosis in alcoholic and nonalcoholic fatty liver disease, as well as improved insulin resistance, suggesting that, in humans, adiponectin may improve hepatic inflammation and hepatic insulin sensitivity. Indeed, data suggest that when pioglitazone is given to NASH patients, adiponectin levels increase 2-fold to 3-fold with an associated improvement in IR as well as improved steatosis, necroinflammation, and fibrosis.
The role of vitamin D in adipokine activity is an active area of research. Vaidya et al. showed a positive association between vitamin D concentrations and levels of adiponectin in a large cohort of 1,645 patients. Interestingly, this relationship was not modified by body mass index (BMI) and has been duplicated in other smaller studies, although those populations were notably leaner and younger.[48, 49] This could potentially be explained by the inhibitory effects of vitamin D on the RAS as previously discussed, although further study is required. A recent study in Iranian type 2 diabetic patients showed that vitamin D therapy in the form of a fortified yogurt drink significantly improved adiponectin levels.
Another key adipokine is leptin, which is secreted by adipose tissue in response to a triglyceride-mediated expansion in adipocytes. Leptin oxidizes hepatic fatty acids (FA) by way of decreasing SREBP-1 expression and prevents FA accumulation in nonadipose tissues. In addition to promoting hepatic steatosis, leptin is thought to have proinflammatory and profibrotic effects, which are important in NASH pathogenesis. Resistin is similarly produced by adipose tissue and is thought to promote the development of NASH by way of activation of c-Jun-terminal kinase (JNK) and nuclear factor kappa B (NF-κB), which leads to increased IR.
Tumor necrosis factor alpha (TNF-α) and IL-6 are proinflammatory cytokines secreted by adipocytes from obese and insulin-resistant patients and weight loss has been shown to lead to a decrease in serum levels. Continuous exposure to TNF-α and IL-6 is associated with hepatic IR, suggesting that the liver may be an important target for these adipocytokines and inhibition of TNF-α activity through anti-TNF antibodies has been shown to prevent inflammation and improve NAFLD.
The effect of VDD on adiponectin, leptin, resistin, TNF-α, and IL-6 was recently investigated by Roth et al. in a rat model where Sprague-Dawley rats were fed either a low-fat diet (LFD) or a high-fat Western diet (WD). WD/VDD mice showed increased hepatic steatosis compared to both VDD and vitamin D replete LFD groups. Hepatic histology also correlated to VDD with increased lobular inflammation and NAFLD activity score (NAS) seen in the WD/VDD mice versus WD/vitamin D replete. Resistin and IL-6 levels were also significantly higher in the WD/VDD group compared to WD/vitamin D replete. In total, these findings suggest VDD worsens NAFLD related to up-regulation of hepatic inflammatory and oxidative stress genes.
Intestinal Microbiome, VDD, and NAFLD
The role of the intestinal tract, nutrients, and their relationship to gut microbiota in immune response and pathogenesis of NAFLD is also intriguing and may relate to VDD. Bacterial lipopolysaccharides (LPS) play an important role in activation of the immune system and are involved in the development of both systemic inflammation and obesity. Increased intake of fat and carbohydrate has been shown to increase circulating LPS, which in turn leads to increased hepatic IR. NAFLD patients have increased gut permeability, suggesting a role for gut-derived endotoxins in the development of this disease.
The interaction of the intestinal microbiome with gut epithelium is also mediated in part through TLRs expressed on gut epithelium. As previously mentioned, TLRs have a key role in mediating immune function. TLR-4 binding of fatty acids leads to production of proinflammatory cytokines in macrophages and epithelial cells, suggesting a role for fatty acid-bound gut-derived TLR-4 in the pathogenesis of obesity-associated inflammation and IR (Fig. 2).
Similarly, TLR-5 is implicated in the development of metabolic syndrome and alterations in gut microbiota. Vijay-Kumar et al. showed that TLR-5-deficient mice develop hyperphagia, obesity, IR, and hepatic steatosis. Subsequent transfer of microbiota from TLR-5-deficient mice to healthy mice led to development of de novo disease, confirming the relationship between TLR-5 and intestinal microbiota.
Finally, TLR-9 has been implicated in the development of murine hepatic steatohepatitis, as evidenced by TLR-9-deficient mice failing to develop inflammation versus controls when exposed to IL-1β. The importance of intestinal microbiota, particularly by way of the TLRs in the pathogenesis of NAFLD, is clear; the role of vitamin D in this process is likely, as demonstrated in the aforementioned study by Roth et al. In addition to the effects on the adipocytokines previously discussed, VDD in WD rats led to increased levels of messenger RNA of TLR-2, TLR-4, and TLR-9. These authors speculated that VDD contributed to NAFLD by increased endotoxin exposure to the liver mediated by these TLRs. Further study is needed to address whether vitamin D replacement is beneficial in suppressing the effects of TLR-2, TLR-4, and TLR-9.
Bile acids are important in the pathogenesis of NAFLD, as they affect the absorption of dietary lipids and regulate glucose and lipid homeostasis. Once absorbed from the distal ileum, bile acids act as ligands for a variety of nuclear hormone receptors. The farnesoid X receptor (FXR) affects multiple pathways of lipid biosynthesis decreasing de novo lipogenesis as well as acting locally in the intestinal defense against inflammation controlling bacterial growth and maintain mucosal integrity. As previously discussed, VDRs are present in epithelial tissues throughout the gastrointestinal tract and vitamin D is known to increase bile acid absorption. Preliminary data suggest that vitamin D treatment in rats increased hepatic portal bile acid concentration and elevated expression of FXR. Further study addressing the role of vitamin D with this and other nuclear hormone receptors is required.
Vitamin D Deficiency in the Progression From NAFLD to NASH
The development of fibrosis in NASH is associated with disease progression. Hepatic stellate cell (HSC) activation is responsible for collagen deposition and fibrosis. HSCs are activated in part by platelet-derived growth factor (PDGF) through proinflammatory cytokines such as TNF-α and IL-6 as well as by way of autophagy-mediated pathways. This activation leads to increased cellular proliferation and transformation into a myofibroblast-like cell resulting in increased synthesis and deposition of extracellular matrix proteins, particularly type I collagen.
The role of vitamin D in HSC proliferation appears to be one of inhibition. Abramovitch et al. demonstrated that inhibition of HSC proliferation by vitamin D was associated with antifibrotic effects in an in vivo murine model. Further study in vitro has suggested a benefit of vitamin D supplementation to suppress activity of HSCs even in the presence of FFAs. Appropriately powered clinical trials are required to determine if vitamin D supplementation produces a clinically significant improvement in fibrosis in NASH patients.
The pathogenesis of NAFLD encompasses pathways that lead to hepatic steatosis and resulting in an excess of FFAs. The steps and factors that promote steatohepatitis and hepatic fibrosis are more complicated and have not been completely elucidated. As discussed, vitamin D appears to interact at multiple steps in both the development of hepatic steatosis as well as steatohepatitis and even fibrosis. VDD is known to be associated with NAFLD and even has been correlated with disease severity. Cumulatively, this would suggest that vitamin D replacement may be effective in the treatment of NAFLD and potentially those with NASH.
Vitamin D as a Treatment for NAFLD/NASH and Other Liver Disease
There is extremely limited evidence that vitamin D replacement provides clinical benefit in NAFLD and NASH patients, with most of the available evidence derived from other chronic liver diseases. The most compelling evidence to date to suggest that vitamin D replacement may be efficacious in NAFLD comes from a recent study in rats by Nakano et al. Rats with steatohepatitis (induced by special diet) received either phototherapy or no treatment. Those that underwent phototherapy had higher vitamin D levels, as expected, but also demonstrated statistically significant elevations in adiponectin levels and decreased markers of hepatic fibrosis including TGF-β and alpha-smooth muscle actin (α-SMA).
Clinical work done with vitamin D repletion is limited by dose-limiting hypercalcemia that results from the pharmacological doses of vitamin D required to obtain similar immunologic, hormonal, and cellular effects seen in bench research. The use of other medications that affect the VDR in the liver such as ursodiol (UDCA) do not cause hypercalcemia but have shown limited benefit in NASH populations.
Other liver diseases where preliminary evidence suggests a possible benefit of vitamin D supplementation include hepatocellular carcinoma (HCC) and chronic hepatitis C. A preliminary study of 33 patients with inoperable HCC treated with the VDR agonist orseocalcitol showed stable disease in 12, improvement in 2, and nonresponse in 19 patients. The benefits of vitamin D therapy have also been shown in chronic hepatitis C (CHC) patients where adding vitamin D (1,000-4,000 IU daily) to standard of care pegylated interferon and ribavirin resulted in 13/27 versus 5/31 patients obtaining a sustained virologic response (SVR). These authors speculated the increase in SVR was related to improvements in IR, which would also be relevant to NAFLD populations.
An interesting potential confounder that has not been addressed in the few studies to date is the potential association between VDD and inactivity, perhaps from leading a sedentary indoor lifestyle. Further appropriately powered RCTs are required to better evaluate the efficacy of vitamin D replacement and parameters of therapy in NAFLD and other chronic liver diseases.
Recommendations and Conclusion
VDD is increasingly diagnosed in Western patients and is commonly found in NAFLD populations. Given the pleiotropic effects of vitamin D ranging from hormonal to immunologic to cellular differentiation, it is quite possible vitamin D replacement in VDD may produce significant biochemical and histologic benefit, although more data from appropriately powered prospective randomized placebo-controlled trials are needed. The levels of 25(OH)D that constitute deficiency versus sufficiency are debatable, although 20 ng/mL (50 nmol/L) has been suggested to be a minimal acceptable level.
Optimal replacement regimens have not been established. Some studies suggest that cumulative dose is more important than dosing frequency. Our typical practice is to replace VDD patients with 50,000 IU vitamin D3 weekly for 12 weeks. A daily supplement of 800-2,000 IU is then recommended, typically in conjunction with calcium. Vitamin D levels are then checked in 3-6 months to confirm adequate replacement and rule out toxicity.
In conclusion, the relationship of vitamin D and NAFLD requires further study but evidence to date confirms an intimate and potentially therapeutic association.