Metabolism and biomarkers of Vitamin D

Abstract

The last decade has witnessed a renaissance in the interest in the metabolism and biological actions of vitamin D₃. Part of this new found interest stems from the discovery that its active form, 1,25-dihydroxyvitamin D3 [1,25(OH)₂D3], through its nuclear vitamin D receptor [VDR], regulates hundreds of genes around the body including those coding for proteins involved in cell differentiation and cell proliferation as well as calcium and phosphate homeostasis. Furthermore, epidemiological association studies have suggested that levels of the main circulating form, 25-hydroxyvitamin D3 [25(OH)D3] correlate positively with various health outcomes connected to major diseases: cancer, immune function and infections and cardiovascular disease. Consequently, the biochemistry around the metabolism of vitamin D, its mechanism of its action in target cells and the clinical chemistry questions regarding its specific and sensitive assay remain relevant. This short article will review the current state of knowledge of the cytochrome P450-enzymes involved in activation and inactivation of vitamin D, as well as provide a synopsis of the biochemistry and physiology surrounding its roles in the body. The review will end by discussing the appropriate biomarkers to assess vitamin D metabolism and vitamin D status in various clinical disease settings.

Key Words::

• Vitamin D
• 25-hydroxyvitamin D
• 1,25-dihydroxyvitamin D3
• CYP2R1
• CYP27B1
• CYP24A1
• VDR
• DBP
• Calcium homeostasis
• phosphate homeostasis

Metabolism of Vitamin D

Vitamin D3 is formed by UVB (290–315 nm) irradiation of 7-dehydrocholesterol in the skin (Figure 1) [1]. Transport of vitamin D3 from skin to storage tissues or to liver for the first step of activation is carried out by a specific vitamin D binding globulin [DBP] [2]. Vitamins D can also be derived from dietary sources, in the form of vitamin D3 or vitamin D2. Vitamin D2 and D3 are thought to be activated and function in the same way [3], so for simplicity, this review will focus initially on defining the metabolism and mechanism of action of vitamin D3 .

Figure 1. Vitamin D Metabolism and Some Physiological Actions. Current knowledge of the cytochrome 450P-containing enzymes in the activation and inactivation of vitamin D together with some information about the physiological actions and regulation of levels of 1α,25(OH)₂D3. Reproduced with permission from NEJM [24].

Vitamin D3 is first activated by 25-hydroxylation, a step probably catalyzed by the liver cytochrome P450, CYP2R1. CYP2R1 is the best candidate to be the 25-hydroxylase involved since only CYP2R1 is able to 25-hydroxylate vitamin D2 or vitamin D3 equally [4,5,6]; a human mutation at Leu99Pro in CYP2R1 results in rickets [7] and the enzyme has been successfully crystallized with its vitamin D substrate bound in the active site [6]. Two recent genome-wide association studies [8,9] implicated CYP2R1 as one of the 4 major genetic determinants of serum 25-hydroxyvitamin D [25 (OH)D3], the others being: DBP, CYP24A1 and 7-dehydrocholesterol reductase (DHR7). It is possible that other 25-hydroxylases such as CYP27A1 or CYP3A4 may act upon vitamin D substrates if their concentration is raised into the high nanomolar or micromolar range [10].

25(OH)D3 is converted by 1α-hydroxylation to the active form of vitamin D, 1α,25-dihydroxyvitamin D3 [1α,25(OH)₂D3] [11], primarily in the kidney, by the action of the mitochondrial cytochrome P450, CYP27B1 [12]. Synthesis of circulating 1α,25(OH)₂D3 in the normal, non-pregnant mammal appears to be the exclusive domain of kidney, since patients with chronic kidney disease exhibit low renal-1α-hydroxylase enzyme activity and as a consequence, greatly reduced serum 1α,25(OH)₂D3 concentrations [13]. Renal CYP27B1 is up-regulated by PTH as part of the calcium homeostatic loop and down-regulated by FGF-23 as part of phosphate homeostatic loop [14](see Figure 1). Mutations in CYP27B1 cause vitamin D dependency rickets type 1 in humans [15], a disease mirrored in the CYP27B1-null mouse [16,17]. The CYP27B1 protein can be detected in extra-renal tissues and this has led to the concept that extra-renal CYP27B1 exists to catalyze “local” production of 1,25(OH)₂D3 from 25(OH)D3 in tissues such as skin, prostate, intestine, breast and possibly bone [8]. Granulomatous conditions, such as sarcoidosis involving activated macrophages, are often accompanied by over-expression of extra-renal CYP27B1 which results in excess local production of 1,25(OH)₂D3 which can leak out into the general circulation causing hypercalciuria and hypercalcemia [19].

The inactivation of 25(OH)D3 and 1,25(OH)₂D3 involves another cytochrome P450 known as CYP24A1, formerly known as the 24-hydroxylase. CYP24A1 is a multi-catalytic enzyme responsible for a 5-step, vitamin D-inducible, C-24-oxidation pathway which inactivates 1,25(OH)₂D3 to a water-soluble biliary form, calcitroic acid [20,21]. CYP24A1 also converts the circulating precursor, 25(OH)D3 into the inactive products: 24,25(OH)₂D3 and 25(OH)D3–26,23-lactone [22] (Figure 2). Over the past three decades, there have been claims that 24-hydroxylated vitamin D metabolites may play some role in bone mineralization or bone fracture healing [23] but currently this role remains unproven. Very recently [24], mutations of CYP24A1 have been implicated as a cause of idiopathic infantile hypercalcemia [IIH], the symptoms of which (hypercalciuria, hypercalcemia and nephrocalcinosis) resemble those also seen in the CYP24A1-knockout mouse [25] confirming that CYP24A1 plays a key role in vitamin D degradation.

Figure 2. Hydroxylation steps and structures of vitamin D metabolites. Sites of hydroxylation steps carried out by cytochrome 450P-containing enzymes in the activation and inactivation of vitamin D including the two potential catabolic products of the hormone 1α,25(OH)₂D3 produced by CYP24A1. Reproduced with permission from ABB [22].

Putting all of these cytochromes P450 into perspective, it seems that they should not be viewed as separate entities but as part of a well-integrated signaling system designed to operate as a classical hormonally-regulated endocrine unit in conjunction with a series of locally-operating paracrine units. In various vitamin D target cells, there appears to be a well-balanced equilibrium between CYP27B1-mediated synthesis and CYP24A1-mediated catabolism providing just the appropriate amount of 1,25(OH)₂D3 for normal function. In some hyperproliferative disease states (eg breast, prostate and colon cancer; cardiovascular hypertrophy), there is evidence that an imbalance of these two enzymes exists leading to a relative deficiency of 1,25(OH)₂D3 and inadequate gene expression [26–30].

At this point, it is worth reminding all readers that the biochemical process defined above for the metabolism and action of vitamin D3 is essentially the same for the other dietary form of vitamin D: vitamin D2 which is still widely used as a food supplement in North America. Evidence suggest that the metabolites 25(OH)D2, 1,25(OH)₂D2 are probably made by the same cytochrome P450 enzymes described above, and that the active form of vitamin D2, 1,25(OH)₂D2 interacts with the VDR and regulates vitamin D-dependent genes through a similar process to that described for 1,25(OH)₂D3 [3, 31]. The nutritionist and clinical chemist must consider monitoring vitamin D2 intake and metabolites, especially if there is drug treatment, supplement use or food fortification with vitamin D2. Thus, methods used by the clinical chemist must be designed to monitor serum 25(OH)D2 and 1,25(OH)₂D2 not just the vitamin D3 metabolite alone. In recent years, vitamin D2 supplementation has been criticized because it seems to result in an inferior increase in serum 25(OH)D as compared to vitamin D3 supplementation, especially when used in high monthly doses. In all likelihood, this is due to the more rapid clearance of 25(OH)D2 from the body reflecting differences in cytochromes P450 involved in catabolism, a mechanism that also explains the observed lower toxicity of high dose vitamin D2 therapy [32].

Mechanism of Action of 1,25(OH)₂D3

The hormonal form 1,25(OH)₂D3 , whether formed in the kidney or extra-renally, has calcemic roles which include the regulation of blood calcium and phosphate concentrations by actions at intestine, bone, parathyroid and kidney; as well as non-calcemic roles which include cell-differentiation and anti-proliferative actions in various cell types: bone marrow (osteoclast precursors & lymphocytes), immune system, skin, breast & prostate epithelial cells, muscle & intestine [33]. 1,25(OH)₂D3 achieves these functions through a vitamin D receptor (VDR)-mediated transcriptional mechanism, in which the hormone directly modulates gene expression of a wide variety of vitamin D-dependent genes in vitamin D-target cells. Liganded VDR recruits a partner known as the retinoid X receptor (RXR) and a plethora of other transactivators, termed a DRIP complex, in order to first expose DNA for transcription and then to efficiently transcribe genes to make mRNA transcripts [34]. Gene arrays performed on a number of tissues or cultured cells show that as many as 300–800 genes are regulated by 1,25(OH)₂D3 within a typical 40,000 gene chip array [35–37]. While some of these gene products are regulated by other modulators suggesting that some of the effects of 1,25(OH)₂D3 can be compensated for and making it possible for the vitamin D-deficient animal or VDR-null mouse to survive, 1,25(OH)₂D3 plays an essential role in the regulation of other vitamin D-dependent genes causing system malfunction (e.g. intestinal calcium absorption) [38,39]. Again the concept of the well-integrated vitamin D signaling system is reinforced by the up-regulation of CYP27B1 and 1,25(OH)₂D3 synthesis and down-regulation of CYP24A1 and 1,25(OH)₂D3 catabolism in the vitamin D-deficient animal or VDR-null mouse when vitamin D-signaling is inadequate.

Biomarkers of Vitamin D Metabolism and Status

Which biomarker to measure?

Serum 1,25(OH)₂D would seem at first glance to be the most appropriate biomarker to measure to indicate vitamin D action given the scientific dogma presented above. It is widely accepted by most researchers in the field that 1,25(OH)₂D is the sole active form derived from vitamin D under normal physiological circumstances. The only exception would be hypervitaminosis D, where excessive amounts of other vitamin D metabolites accumulate, the concentrations of 1,25(OH)₂D are actually suppressed and 25(OH)D may flood into the target cell and elicit a VDR-mediated response [40,41], but this process requires vitamin D intakes at least one order of magnitude higher than normal. Serum 1,25(OH)₂D [1,25(OH)₂D2 + 1,25(OH)₂D3] can be a useful clinical test [42] especially in hypercalcemic states such as sarcoidosis, IIH and some metabolic bone diseases (e.g. renal osteodystrophy) [19,24,13]. But except in these conditions, serum 1,25(OH)₂D has proven to be a rather poor biomarker of vitamin D action, probably because it is not an accurate predictor of the cellular 1,25(OH)₂D concentration in all target cells. The most likely explanation for serum 1,25(OH)₂D not equating with the cellular 1,25(OH)₂D concentration is the presence in some target cells of extra-renal CYP27B1 [43,44].

Serum 25(OH)D concentration, which is the sum of [25(OH)D2 + 25(OH)D3 concentrations], has emerged as the best measure of the vitamin D status of the animal in vivo largely because this parameter correlates well with non-calcemic clinical outcomes. It has been rationalized that this is due to the fact that serum 25(OH)D is a measure of the cellular 25(OH)D concentration which is turn is converted to cellular 1,25(OH)₂D3 by the loosely-regulated extra-renal CYP27B1. Using these arguments, increases in vitamin D intake cause increases in serum 25(OH)D which result in increased cellular 1,25(OH)₂D3 and increased action at vitamin D-dependent genes. It must be emphasized that this remains a theory to explain why serum 25(OH)D is a good biomarker of vitamin D status [33,44]. Clearly, the plausibility of the extra-renal CYP27B1 theory to explain the value of serum 25(OH)D should be tested and in fact this could be achieved if we could directly measure cellular 1,25(OH)₂D. While the emergence of ever more sensitive LC-MS/MS techniques [42] is making this idea potentially feasible, this would still require tissue biopsies/samples. However, in the immediate future, it is likely that we will still assume that serum 25(OH)D concentration is the best surrogate measure of vitamin D status to predict health outcomes.

How to measure serum 25(OH)D concentration?

There are now a wide range of antibody-based and LC-MS/MS-based methods available to measure serum 25(OH)C concentration [32,45,46] and their validity and performance are discussed in the adjoining papers in this symposium.

When to measure serum 25(OH)D concentration?

This parameter undergoes a seasonal fluctuation with a minimum in Feb/Mar and a maximum in Sept/Oct in northern latitudes. Vitamin D supplementation can usually be followed by monitoring on a quarterly basis since 25(OH)D has a half-life of 20 days and serum concentrations take 3–4 months to plateau.

Who is at risk of vitamin D deficiency?

The 2011 IOM Report [47] recently defined D deficiency as serum 25(OH)D < 50 nmol/L and identified several high risk groups including individuals who avoid sun exposure, those at high latitude, those with darkly-pigmented skin, those who are obese and those who suffer from chronic kidney disease. In these times of rising health-care costs, some health bodies have proposed that monitoring of serum 25(OH)D should be focused on these high-risk groups.

Conclusions

Vitamin D remains a hot topic because its functions around the body are so general and its disease implications are still hotly debated. There is still a large concern that some segments of our population are at high risk of vitamin D deficiency and thus monitoring of serum 25(OH)D concentration remains the most useful biomarker and the focus for clinical chemists. Vitamin D assays are improving and their use is being directed at those suspected of being vitamin D deficient rather than the assay being used as a routine screening tool. Time will tell if this is the correct decision.

Questions and Answers

R Vieth, Canada

You indicated 24,25 (OH)₂D3 as being interesting in metabolism but did not mention it at all when it came to laboratory assays. In my experience, 24,25(OH)₂D3 represents one-sixth of 25(OH)D concentration. Does this explain some of the differences between assays, when 24,25 (OH)₂D3 is being measured as part of 25(OH)D? Is 24,25(OH)2D biologically active?

G Jones

I think the measurement of 24,25(OH)₂D3 concentration or, as you have measured, the ratio between 24,25(OH)₂D3 and 25(OH)D concentrations, is important. In idiopathic infantile hypercalcaemia, its measurement is critical to the diagnosis. The assay therefore has a great value in a limited area. It may be, with all the polymorphisms being discovered, that we want to measure it more widely, in a normal population. It is an important clinical parameter to measure.

In my opinion, it is not biologically active.

E Delvin, Canada

The story regarding the 3-epimers came about because LC-MSMS methods can distinguish between the two epimers; is it really important in clinical use? Should we or should we not distinguish between them?

G Jones

Absolutely. I believe the 3-epimer story has profound implications, not only for clinical testing but to answer basic questions:

• Where does 3-epi-25(OH)D3 come from?

• Does it have any biological activity in vivo? It certainly works in biological systems in vitro

• Is it converted to 3-epi 1,25(OH)₂D3?

E Delvin

The 3-epimer concentration is higher in the very young and in cord blood; maybe 3-epimerases are responsible?

G Jones

Or it could be generated from some aspect of the neonatal diet.

J Welsh, USA

Could you give more information on the disease-causing mutations in CYP2R1 and CYP24A1? Is there complete loss of function?

G Jones

They are all ‘loss of activity’ mutations. L99Pis one which gives a truncated protein. More mutations of CYP2R1 are being discovered. We studied European families with CYP24A1 mutations which have more or less complete loss of activity and more mutations are being discovered as we get access to more patient material.

H Morris, Australia

What further information do you have on variations in enzymic activity related to identified CYP24A1 polymorphisms, other than a full loss of function? Can you speculate as to whether these polymorphisms may be responsible for the highly varied response amongst individuals given a single dose of 25(OH)D and could we have higher catabolisers and lower catabolisers?

G Jones

Yes, I believe the latter may well explain the genome-wide studies which have reported variance in CYP24A1 causing changes in 25(OH)D concentrations. I have been studying ‘man-made’ mutations for some time and we can make mutations of CYP24A1 which haven't lost full activity but have a change in qualitative activity. So far we have only found those with a complete loss or a near complete loss in humans. It maybe that our techniques for identifying the intermediate CYP24A1 or high CYP24A1 have not been pursued well enough.

M Hewison, USA

Coming back to 24,25(OH)₂D3 assays, I think one group in which there is increasing interest is early stage renal disease patients, who have very elevated FGF23. There is still some uncertainty as to whether this is impacting on CYP27B1 or whether it is boosting 24-hydroxylase activity. Either could compromise vitamin D function. It could well be that 24,25(OH)₂D3 measurements in relation to renal disease could become popular and be of great interest.

G Jones

I believe that CYP24A1 plays a major role in kidney disease and this raises the question as to whether patients with renal failure should be given both 25(OH)D3 and an active analogue. I have proposed the correction of 25(OH)D and 1,25(OH)₂D3 in renal patients, particularly in those on dialysis [43,44].

Declaration of interest: The author is a member of the Scientific Advisory Board of Cytochroma Inc, Markham, Ontario, Canada. The author alone is responsible for the content and writing of the paper.