Review Article: Vitamin D and Inflammatory Bowel Disease

Established Concepts and Future Directions

M. Garg; J. S. Lubel; M. P. Sparrow; S. G. Holt; P. R. Gibson


Aliment Pharmacol Ther. 2012;36(4):324-344. 

In This Article

Vitamin D Synthesis, Metabolism and Cellular Effects

Synthesis & Absorption

Vitamin D exists in two main forms, vitamin D3 (VD3, cholecalciferol) and vitamin D2 (VD2, ergocalciferol), differing in their side chain structure. In humans, the majority of VD3 is produced in the skin from exposure to sunlight (Figure 1), with a small proportion obtained from animal sources such as oily fish and egg yolk.[4] VD2 is predominantly obtained from plant sources. Supplements of both VD2 (produced from irradiation of the yeast sterol ergosterol) and VD3 are commercially available.[5] Commonly, vitamin D refers collectively to VD2 and VD3.

Figure 1.

Synthesis, metabolism and cellular effect of vitamin D. The classical pathway involves renal conversion of 25(OH) vitamin D to 1,25(OH)2 vitamin D, which circulates with subsequent 'endocrine' action on bone metabolism. A novel pathway involves uptake of 25(OH) vitamin D, either free or bound to DBP, followed by intracellular 1α hydroxylation in nonrenal cells including monocytes and colonic epithelial cells to 1,25(OH)2 vitamin D leading to 'intracrine' actions localised to those cells. The latter may be important in many of the extra-skeletal effects of vitamin D, particularly in cells involved in immune responses.

Human skin-derived VD3 is produced from 7-dehydroxycholesterol upon exposure to ultraviolet B radiation (UVB, wavelength 290–315 nm).[4] As a fat-soluble vitamin, dietary vitamin D is incorporated into chylomicrons and transported via lymphatics into the venous circulation. Some of the dietary vitamin D is extracted by adipose tissue and muscle, but the remainder and most of the endogenously synthesised vitamin D is transported to the liver. Here, it is metabolised by the cytochrome P450 enzymes vitamin D 25-hydroxylases (microsomal CYP2R1 and mitochondrial CYP27A1) to 25-hydroxy vitamin D (25(OH)D).[6] In classical calcium-related responses, another cytochrome P450 enzyme, 1α-hydroxylase (CYP27B1), converts 25(OH)D to the biologically active form of vitamin D, 1,25-hydroxy vitamin D (1,25(OH)2D) in the proximal tubule of the kidneys.[7,8]

Storage, Circulation and Excretion

Previously believed to be biologically inert at physiological levels, 25(OH)D is the major storage and circulating form of vitamin D and frequently measured as an index of vitamin D status. In the human body, the highest concentration of 25(OH)D is noted in the plasma (usually measured in the serum as 20–150 nmol/L or 8–60 ng/mL), but the largest pool of 25(OH)D is in adipose tissue and muscle.[9] Hence, although a circulating half-life of 25(OH)D is approximately 10–15 days,[6,10] release from tissue stores effectively results in a half-life in vivo of 2–3 months.[11]

Renally produced 1,25(OH)2D circulates in the blood at levels in the picomolar range, about one thousandth those of 25(OH)D. Contrasting with the relative lack of regulation of 25-hydroxylation, 1α-hydroxylation is under control by serum parathyroid hormone (PTH) and fibroblast growth factor 23 (FGF23) in response to serum calcium and phosphate and represents the rate-limiting step in the synthetic pathway.

Metabolites in the vitamin D pathway are transported in the circulation predominantly (about 85–90%) bound to vitamin D-binding protein (DBP, also known as group-specific component globulin, Gc-globulin) and albumin (about 10–15%), with <1% in the free form.[12] DBP is a liver-derived, 58 kDa glycosylated α-globulin structurally similar to albumin,[13] which circulates at concentrations of 0.6–11 μmol/L.[12] The affinity of DBP for 25(OH)D is 5 × 108 mol/L, about an order of magnitude greater than that for vitamin D (1 × 105 to 1 × 107 mol/L) or 1,25(OH)2D (2 × 107 mol/L).[14,15] This difference in affinity partly accounts for the shorter plasma half-life of vitamin D (~4–6 h) and 1,25(OH)2D (~4–20 h).[4,6]

Tight regulation of 1α-hydroxylation and the short half-life mean than the serum concentration of 1,25(OH)2D is not an accurate measurement of total body vitamin D status, and measurement is of most use in altered states of 1α-hydroxylation such as chronic kidney disease (reduced) or granulomatous disease (increased).

The catabolic enzyme 24-hydroxylase (CYP24A1) is responsible for the conversion of both 25(OH)D and 1,25(OH)2D to the inactive metabolites, 24,25(OH)2D and 25(OH)D-26,23-lactone,[6] and via a multistep pathway to the water soluble calcitroic acid (1α-hydroxy-23 carboxy-24,25,26,27-tetranorvitamin D3), which undergoes urinary and biliary excretion.[16,17]

The Vitamin D-binding Protein

Vitamin D-binding protein has been shown to regulate the effect of vitamin D metabolites in target organs. As a reservoir with vastly greater circulating levels than that of vitamin D metabolites, DBP may control availability in tissues, allowing only the small free fraction of vitamin D metabolites to passively enter cells through diffusion across cell membranes.[18] In a small series of 49 young adults, bone mineral density (BMD) was positively correlated with only free and bioavailable 25(OH)D, not total 25(OH)D.[12] Furthermore, BMD was negatively correlated with DBP concentrations (r = −0.296).[12] Second, DBP itself actively facilitates the uptake of bound 25(OH)D into renal tubular cells via the membrane glycoproteins, megalin and cubilin. The megalin-DBP-25(OH)D complex is internalised via endocytosis, in co-operation with cubulin, and presented to mitochondrial 1α-hydroxylase.[19] A similar process has been described in mammary cells[20] and osteoblasts.[21,22] Third, DBP may modulate intracellular actions of 25(OH)D and 1,25(OH)2D. Interestingly, DBP has been shown to attenuate monocyte response in vitro to 25(OH)D and 1,25(OH)2D, seemingly independent of megalin.[23] Loss of DBP in the urine in proteinuric diseases has been suggested to be a cause for vitamin D deficiency associated with diabetes.[24]

The gene for DBP has three dominant single nucleotide polymorphisms (SNPs), which correlate with protein products (GC-1s, GC-1f and GC-2) separated by single amino acid substitutions.[25] Functionally, GC-1s has twice and GC-1f has four times the binding affinity for 25(OH)D than GC-2.[26] In individuals, these SNPs result in six diplotypes (1f-1f, 1f-1s, 1s-1s (collectively GC-1–1), 1f-2, 1s-2 (collectively GC-1–2) and 2–2 (GC-2–2)). Subsequently, total 25(OH)D levels are lowest in subjects with GC-2–2, intermediate in those with GC-1–2 and highest in those with GC-1–1.[27–30] The functional significance of this polymorphism is not yet completely understood, but may cause very different circulating free and hence intracellular levels of 25(OH)D3.[31] This means that it is difficult to ascertain deficiency or repletion without knowledge of free levels and DBP genotype.

Apart from binding vitamin D metabolites, DBP also binds with high affinity globular (G)-actin via a domain near its C-terminus, preventing the formation of filamentous (F)-actin, which plays a role in vascular occlusion following cellular damage.[18,25] Subsequently, any condition which results in cell death, and hence the release of large amounts of G-actin, may reduce free plasma DBP.[18,32] This has been noted in systemic sepsis.[32] However, given DBP is produced in the liver with close homology to albumin, its synthesis may also be reduced in any condition of physiological stress.[33] Other described functions of DBP include enhancement of the neutrophil chemotactic effect of complement 5a,[34] and deglycosylated DBP serves as a macrophage-activating factor.[35]

Several studies have investigated the association of risk of specific diseases with DBP genotype, but no consistent relationship has yet been established.[36–38]

The Vitamin D Receptor (VDR)

The VDR was first recognised by Haussler and colleagues in 1969,[39,40] and its structure subsequently described in 1988.[41] Over the past three decades, increasing characterisation of the role of the VDR and its presence in multiple tissues has resulted in a vast expanse in our knowledge of the pleiotropic nature of the vitamin D axis in human physiology.

Most currently described cellular actions of vitamin D occur via genomic regulation. 1,25(OH)2D, on entry into cells or via intracellular conversion from 25(OH)D, binds to nuclear VDR. The VDR-ligand complex subsequently forms a heterodimer with retinoid-X receptor and binds vitamin D responsive elements located predominantly in the promoter regions of target genes.[16] In combination with transcription factors and co-regulatory proteins, this complex has been shown to promote or suppress the transcription of a wide range of genes. Through recent advances using chromatin precipitation with massively parallel sequencing (ChIP-seq) in lymphoblastoid cell lines, 623 genomic regions were shown to be occupied by VDR in the basal state, increasing to 2776 regions on stimulation with calcitriol, with significant changes in expression in 229 identifiable genes.[42]

In addition to regulation of gene transcription, VDR also participates in nontranscriptional rapid cellular responses by translocating to the plasma membranes within seconds or minutes (compared with hours to days required for genomic regulation).[43] Examples of these actions include voltage-gated calcium and chloride channel regulation in osteoblasts, skeletal muscle cell calcium entry, contractility and myogenesis, calcium uptake by intestinal epithelial cells and insulin secretion by pancreatic islet beta cells.[43,44]

Inactivating mutations in the VDR gene result in hereditary vitamin D-resistant rickets, a condition manifested by rickets, hypocalcaemia, hypophosphatemia and alopecia.[45] More subtle mutations or polymorphisms in the VDR gene have been described. The more widely studied polymorphisms include TaqI, BsmI, ApaI, Tru9I and EcoRV between exons 8 and 9, FokI in exon 2 and Cdx2 near exon 2.[46] These changes may affect gene promotion, RNA transcription efficiency or protein structure.[46] For example, a FokI polymorphism resulting from a T to C change shifts the start codon, and hence results in a protein that is truncated by three amino acids. This shorter protein has greater transcriptional activity due to increased binding affinity to transcription factor IIb.[47] Intense research is actively underway to find associations of various polymorphisms with diseases, but much data to date have been conflicting.[46]


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.
Post as: