Calcioic acid: In vivo detection and quantification of the terminal C24-oxidation product of 25-hydroxyvitamin D₃ and related intermediates in serum of mice treated with 24,25-dihydroxyvitamin D₃

Abstract

Calcitroic acid, the excretory form of vitamin D, is the terminal product of a 5-step pathway catalyzed by CYP24A1, commencing with C24-hydroxylation of 1,25-dihydroxyvitamin D₃ (1,25-(OH)₂D₃). Catabolism of 25-hydroxyvitamin D₃ (25-OH-D₃) proceeds via analogous steps culminating in calcioic acid; however this C23-truncated acid has not been reported in the circulation. It has recently been shown that 24,25-dihydroxyvitamin D₃ (24,25-(OH)₂D₃) is an important factor in optimal bone fracture healing acting via an effector molecule FAM57B2 to produce lactosylceramide. Administration of 24,25-(OH)₂D₃ was found to restore normal fracture repair in Cyp24a1^−/− mice devoid of 24,25-(OH)₂D₃. We set out to study the multi-step catabolism of D₃ metabolites in vivo using LC-MS/MS methods in vehicle or 24,25-(OH)₂D₃-treated mice. Vehicle-treated Cyp24a1^+/− mice possessed normal levels of serum 24,25-(OH)₂D₃ (7 ng/mL) and 25-OH-D₃-26,23-lactone (4 ng/mL). We also detected 24-oxo-25-OH-D₃ (3 ng/mL) and 24-oxo-23,25-(OH)₂D₃ (0.4 ng/mL); which were not detectable in vehicle-treated Cyp24a1^−/− mice. In 24,25-(OH)₂D₃-treated Cyp24a1^+/− mice, serum 24,25-(OH)₂D₃ rose to 200 ng/mL while 25-OH-D₃-26,23-lactone remained unchanged in comparison to vehicle-treated Cyp24a1^+/− mice Concentration of serum 24-oxo-25-OH-D₃ and 24-oxo-23,25-(OH)₂D₃ rose by 10-fold, when Cyp24a1^+/− mice were treated with 24,25-(OH)₂D₃ Calcioic acid was increased to 0.030 ng/mL for 24,25-(OH)₂D₃-treated Cyp24a1^+/− mice. In 24,25-(OH)₂D₃-treated Cyp24a1^−/− mice, serum 24,25-(OH)₂D₃ rose further to a striking 830 ng/mL due to lack of catabolism of the 24,25-(OH)₂D₃ dose. Serum 1,25-(OH)₂D₃ levels were suppressed in 24,25-(OH)₂D₃-treated Cyp24a1^+/− and Cyp24a1^−/− mice. Circulating 1,24,25-(OH)₃D₃ rose from 73 pg/mL to 106 pg/mL when Cyp24a1^+/− mice were treated with 24,25-(OH)₂D₃. While undetectable in vehicle-treated Cyp24a1^−/− mice, 1,24,25-(OH)₃D₃ rose unexpectedly to 153 pg/mL in 24,25-(OH)₂D₃-treated nulls suggesting conversion of 24,25-(OH)₂D₃ to 1,24,25-(OH)₃D₃ via 1-hydroxylation. Taken together, amplification of 24,25-(OH)₂D₃ catabolism by exogenous doses of this metabolite have enabled detection of downstream C24-oxidation pathway products in vivo, including calcioic acid; and provides a platform for studying alternative routes of vitamin D metabolism that may occur in pathological states including hypervitaminosis D and idiopathic infantile hypercalcemia caused by mutations of CYP24A1.

1. INTRODUCTION

Clearance of 25-hydroxyvitamin D (25-OH-D₃) and 1,25-dihydroxyvitamin D (1,25-(OH)₂D₃) is mediated by the cytochrome P450, CYP24A1 via pathways commencing with C24- or C23-hydroxylation and culminating in either a C-23 carboxylic acid which leaves the body via bile, or a 26,23-lactone product which retains significant binding affinity to the serum vitamin D-binding protein (DBP)¹. 25-OH-D₃ and 1,25-(OH)₂D₃ are catabolized by analogous steps of either the C24- or C23-hydroxylation pathway, and the degree of pathway utilization is species-dependent. Humans, mice and rats catabolize vitamin D metabolites mainly via C24-hydroxylation, but opossums and guinea pigs catabolize vitamin D to lactone products arising from an initial C23-hydroxylation step^2,3. While believed to have a primarily catabolic role, St-Arnaud and colleagues have identified an anabolic role for CYP24A1, revealing that 24,25-dihydroxyvitamin D₃ (24,25-(OH)₂D₃) is essential for optimal bone fracture healing in mice. A role for 24,25-(OH)₂D₃ in bone fracture repair was first reported in the 1990s^4,5,6, but the details of its vitamin D receptor (VDR)-independent mechanism of action remained elusive until recently. Using a combination of knockout mouse models and mRNA screening of the healing bone callus, St-Arnaud and colleagues revealed that 24,25-(OH)₂D₃ allosterically activates the membrane protein FAM57B2 which catalyzes the formation of lactosylceramide (LacCer) involved in downstream effects on bone healing⁷. Mice lacking either systemic CYP24A1 or FAM57B2 in chondrocytes exhibit impaired fracture healing. Administration of either 24,25-(OH)₂D₃ or LacCer restores optimal fracture healing in the Cyp24^−/− mouse, but Col2-Cre Fam57b^fl/fl mice respond only to LacCer and not exogenous 24,25-(OH)₂D₃ due to the absence of its downstream effector molecule⁷. Half of Cyp24a1^−/− mice die post weaning of hypercalcemia due to accumulation of 1,25-(OH)₂D; but the surviving animals retain a normal bone phenotype and are normocalcemic as they are able to circumvent lack of 1,25-(OH)₂D catabolism by downregulating CYP27B1 involved in 1,25-(OH)₂D formation⁸, thus permitting bone fracture healing studies on Cyp24a1-null mice devoid of 24,25-(OH)₂D₃. Using a sensitive liquid chromatography – tandem mass spectrometry (LC-MS/MS) assay, we set out to study serum vitamin D metabolites in Cyp24a1^+/− and Cyp24a1^−/− mice from the bone fracture healing study, given large doses of 24,25-(OH)₂D₃^7,8 to establish: What specific vitamin D metabolites can be detected in mice given 24,25-(OH)₂D₃? and How is vitamin D metabolism altered in the context of a large 24,25-(OH)₂D₃ dose?

2. MATERIALS AND METHODS

Development and treatment of Cyp24a1^−/− and Cyp24A1^+/− mice analyzed in this study have been described in previous work which identified the role of 24,25-(OH)₂D₃ in bone fracture repair.^7,8. The animal use protocol (AUP #4138) and all standard operating procedures were approved by the Institutional Animal Care and Use Committee and followed the guidelines of the Canadian Council on Animal Care. Analytical methods used to assay mouse serum for a range of vitamin D metabolites have been described previously^9,10,11. Briefly, 12-week old female Cyp24a1^−/− and Cyp24A1^+/− were subjected to osteotomy. Beginning 72-hours post surgery, mice were given rescue treatments of 6.7 μg/kg of 24,25-(OH)₂D₃ or 50 μL of propylene glycol daily for 15 days prior to sacrifice⁷. Vitamin D metabolites were measured in six mice from each of the four treatment groups. Cyp24A1^+/− are identical to Cyp24A1^+/+ in terms of body composition and metabolism. While kidney Cyp24a1 expression was not compared between Cyp24A1^+/− and Cyp24A1^+/+ mice, serum concentrations of 1,25-(OH)₂D₃, 1,24,25-(OH)₃D₃, 25-OH-D₃, 24,25-(OH)₂D₃, 25-OH-D₃-26,23-lactone were the same, suggesting similar CYP27B1 and CYP24A1 enzyme activities and that Cyp24A1^+/− littermates served as good controls for Cyp24a1^−/− mice⁷.

For analysis of 25-OH-D₃ and its metabolites, serum was extracted by liquid-liquid extraction with methyl-tert butyl ether and hexane, after protein precipitation with zinc sulfate and methanol. For 1,25-(OH)₂D₃ and 1,24,25-(OH)₃D₃, serum was extracted with 100 μL of anti-1,25-(OH)₂D antibody slurry (Immundiagnostik), and metabolites were eluted with ethanol. Both types of extract were evaporated to dryness and derivatized with 4-(2-(6,7-Dimethoxy-4-methyl-3-oxo-3,4-dihydroquinoxalinyl)ethyl)1,2,4-triazoline-3,5-dione) (DMEQ-TAD), and analyzed by LC-MS/MS on ACQUITY/Xevo TQ-S instruments (Waters) using a 20-minute methanol/water-based gradient on a C18+ column as previously described¹⁰. Mobile phase A and B comprised either water or methanol supplemented with 0.1% (by vol.) formic acid and 2 mM methylamine respectvely. Composition of mobile phase B was raised from 50 to 77 % (by vol.) over 18 minutes using a linear (curve 6) gradient. Samples prepared by liquid-liquid extraction were also analyzed using a more rapid LC separation⁹. Metabolites 25-OH-D₃, 24,25-(OH)₂D₃, 1,25-(OH)₂D₃ and 1,24,25-(OH)₃D₃ were quantified using specific deuterated internal standards and calibration lines, whereas concentration of the following metabolites was estimated on the basis of recovery of d₆-24,25-(OH)₂D₃ and the calibration line for 24,25-(OH)₂D₃: 25-OH-D₃-26,23-lactone, 24-oxo-25-OH-D₃, 24-oxo-23,25-(OH)₂D₃, and calcioic acid. We previously expressed human CYP24A1 in V79–4 Chinese hamster lung fibroblast cells using a targeted integration method (Flp-In). Incubation of these cells with 25-OH-D₃ using previously-described methods yielded a metabolite-rich extract which could be used as a LC-MS/MS standard¹². Identity of vitamin D metabolites detected in mouse serum was based on a combination of criteria where appropriate, including: 1) Presence of both 6S and 6R isomers of DMEQ-TAD adducts using an MRM transition based on the theoretical molecular mass of the metabolite 2) Co-chromatography with extracts derived from human recombinant CYP24A1-expressing cells incubated with 25-OH-D₃¹² 3) Co-chromatography with synthetic standards for 24,25-(OH)₂D₃ and calcioic acid. 4) Absence of a peak when CYP24A1 was ablated. Student’s T test was used to determine significance of changes in vitamin D metabolite concentrations and ratios.

3. RESULTS

Vehicle-treated Cyp24a1^+/− mice exhibited normal 25-OH-D₃, and 24,25-(OH)₂D₃ levels typical of that observed over a range of mouse studies using the same LC-MS/MS platform¹³, giving a 25-OH-D₃:24,25-(OH)₂D₃ ratio of 2.6 (Table 1, Figure 1A–C). With Cyp24a1 ablation, 24,25-(OH)₂D₃ levels decreased from 6.7 to 1.5 ng/mL, and 25-OH-D₃ rose 6-fold from 17.2 to 101 ng/mL, indicating a marked lack of catabolism of endogenous 25-OH-D₃ in the absence of CYP24A1, as such, the ratio of 25-OH-D₃:24,25-(OH)₂D₃ rose from 2.6 to 70 in Cyp24a1^−/− mice. The importance of CYP24A1 in the catabolism of 25-OH-D₃ was also manifested by the reduction in serum 25-OH-D₃-26,23-lactone in Cyp24a1^−/− mice (Table 1, Figure 1G). When given a 24,25-(OH)₂D₃ dose, Cyp24a1^+/− mice continued to exhibit normal 25-OH-D₃ levels, but possessed very high serum 24,25-(OH)₂D₃ as was expected, at almost 200 ng/mL. 25-OH-D₃ levels in 24,25-OH-D₃-treated Cyp24a1^−/− mice also appeared to accumulate in the absence of CYP24A1, and 24,25-(OH)₂D₃ was present at over 800 ng/mL, due to lack of catabolism of the exogenous 24,25-(OH)₂D₃ dose. Why the apparent accumulation of 25-OH-D₃ is not as extreme when Cyp24a1^−/− mice are treated with 24,25-(OH)₂D₃ as compared with vehicle remains unknown; but raises the possibility that extreme levels of 24,25-(OH)₂D₃ in Cyp24a1^−/− mice affects the expression of CYP2R1/CYP27A1 or enzymes involved in CYP24A1-independent catabolism of 25-OH-D₃. Because the metabolite 25-OH-D₃-26,23-lactone arises from further catabolism of 23,25-(OH)₂D₃, concentrations of serum 25-OH-D₃-26,23-lactone were identical between vehicle and 24,25-(OH)₂D₃-treated Cyp24a1^+/− mice, at 3.8 ng/mL. Furthermore, absence of 25-OH-D₃-26,23-lactone in the 24,25-(OH)₂D₃-treated Cyp24a1^−/− mice confirmed from the metabolic standpoint, that accumulation of the exogenous 24,25-(OH)₂D₃ dose in Cyp24a1^−/− mice was in fact due to a lack of CYP24A1.

Table 1:

Concentration of serum vitamin D metabolites in Cyp24a1^+/− and Cyp24a1^−/− mice.

Genotype	Treatment	Na	[25-OH-D3]	[24,25-(OH)2D3]	R b	[25-D3-26,23-lactone]	R c	[24-oxo-25–OH-D3]	[24-oxo-23,25-(OH)2D3]	[Calcioic acid]	[1,25-(OH)2D3]	[1,24,25-(OH)3D3]
			ng/mL	ng/mL		ng/mL		ng/mL	ng/mL	pg/mL	pg/mL	pg/mL
Cyp24a1 +/−	vehicle	6	17.2±2.4	6.7±1.1	2.6	3.8±0.7	4.5	2.8±0.4	0.4±0.1	<5	27.8±4.2	72.8±12.1
Cyp24a1 −/−	vehicle	6	101.3±20.6d	1.5±0.1d	70.6d	<0.1d	2000.2d	<0.1d	<0.1d	<5f	34.0±6.2f	<4d
Cyp24a1 +/−	24,25-D3	6	12.6±1.0e	196.9±68.5d	0.1d	3.8±0.3f	3.4d	31.4±9.2d	3.7±0.1d	31±11d	<4d	105.5±19.5d
Cyp24a1 −/−	24,25-D3	6	56.8±12.3d	827.9±327.8d	0.1d	<0.1d	2999.1d	2.7±1.2f	<0.1d	<5f	<4d	153.4±64.7g

^aNumber of mice in each treatment group.

^bDenotes the ratio 25-OH-D₃:24,25-(OH)₂D₃ or

^c25-OH-D₃:25-OH-D₃–26,23-lactone.

^dp<0.001 based on comparison to vehicle-treated Cyp24a1^+/− mice.

^ep<0.01 based on comparison to vehicle-treated Cyp24a1^+/− mice.

^fp>0.05 based on comparison to vehicle-treated Cyp24a1^+/− mice.

^gp<0.05 based on comparison to vehicle-treated Cyp24a1^+/− mice.

Figure 1: LC-MS/MS-based metabolomics of serum vitamin D metabolites:

Panels A-I depict mass chromatograms based on MRM transitions used to detect DMEQ-TAD adducts of specific serum vitamin D metabolites in Cyp24a1^−/− and Cyp24a1^+/− mice given vehicle (veh) or 24,25-(OH)₂D₃ (24,25-D₃), including: (A) 25-OH-D₃ (MRM, m/z 746→468), (B & C) 24,25-(OH)₂D₃ (plotted on two different intensity scales; MRM, m/z 762→468), (D) 24-oxo-25-OH-D₃ (MRM, m/z 760→468), (E) 24-oxo-23,25-(OH)₂D₃, (MRM, m/z 776→468), (F) calcioic acid (MRM, m/z 704→468), (G) 25-OH-D₃-26,23-lactone (MRM, m/z 774→468), (H) 1,25-(OH)₂D₃, (MRM, m/z 762→468+484), and (I) 1,24,25-(OH)₃D₃, (MRM, m/z 778→468+484).

The high concentration of serum 24,25-(OH)₂D₃ in treated mice, prompted us to screen for downstream metabolites of the C24-oxidation pathway. Somewhat surprisingly, the metabolites 24-oxo-25-OH-D₃ and 24-oxo-23,25-(OH)₂D₃ were easily detected at 2.8 and 0.4 ng/mL in vehicle-treated Cyp24a1^+/− mice, and present at 10-fold greater concentrations in Cyp24a1^+/− mice given 24,25-(OH)₂D₃ (Table, Figure 1D,E). While the absence of detectable peaks for these metabolites in most Cyp24a1^−/− mice confirms the CYP24A1-dependent catabolism of 24,25-(OH)₂D₃ in vivo, the residual 24-oxo-25-OH-D₃ in 24,25-(OH)₂D₃-treated Cyp24a1^−/− animals imply the presence of alternative enzymes that might act to clear high doses of 24,25-(OH)₂D₃ in the absence of CYP24A1. The terminal catabolite in the C24-oxidation of 25-OH-D₃, calcioic acid, could be detected at 30 pg/mL in 24,25-(OH)₂D₃-treated Cyp24a1^+/− animals only (Figure 1F). In order to confirm the identity of calcioic acid observed in vivo, we generated more concentrated extracts of serum from 24,25-(OH)₂D₃-treated Cyp24a1^+/− mice (Figure 2A) and a patient with hypervitaminosis D (Figure 2 B) and compared the LC-MS/MS profiles with extracts generated from human CYP24A1-expressing V79–4 cells incubated with 25-OH-D₃¹², as well as a synthetic standard for calcioic acid (Figure 2C,D). While we are currently unable to detect calcioic acid in normal human subjects, we were able to identify calcioic acid in a patient with hypervitaminosis D, with serum 25-OH-D of 200 ng/mL and 24,25-(OH)₂D₃ of 10 ng/mL, and 19 pg/mL of 1,25-(OH)₂D₃ who presented with severe hypercalcemia and suppressed parathyroid hormone. As this patient possessed elevated 24,25-(OH)₂D₃ and a normal 25-OH-D₃:24,25-(OH)₂D₃ ratio of 20, CYP24A1 mutation was ruled out, but patient history revealed prolonged and excessive cholecalciferol prophylaxis as the cause of hypercalcemia. These results reveal that amplified vitamin D catabolism through to calcioic acid can be observed both in mice given 24,25-(OH)₂D₃, as well as in a clinical case of hypercalcemia caused by over-supplementation with vitamin D₃.

Figure 2: Detection of calcioic acid by LC-MS/MS.

MRM transition m/z 704→468 was used to screen for the presence of calcioic acid in (A) Cyp24a1^+/− mice treated with 24,25-OH)₂D₃, (B) a patient with hypervitaminosis D, and (C) an extract generated from human CYP24A1-expressing V79–4 cells incubated with 25-OH-D₃. Retention times were compared against a synthetic calcioic acid standard (D).

Since the mice used in our study comprised normocalcemic Cyp24a1^−/− animals, 1,25-(OH)₂D₃ levels were normal, and did not differ from their heterozygotic littermates (28–34 pg/mL) However, super-physiological 24,25-(OH)₂D₃ concentrations appeared to suppress 1,25-(OH)₂D₃ levels to below the limit of detection of the assay, even in Cyp24a1^−/− mice incapable of catabolizing 1,25-(OH)₂D₃ (Table 1, Figure 1H). While 1,24,25-trihydroxyvitamin D₃ (1,24,25-(OH)₃D₃) was undetectable in vehicle-treated Cyp24a1^−/− mice, it rose to 153 pg/mL when treated with 24,25-(OH)₂D₃. Seventy three pg/mL of 1,24,25-(OH)₃D₃ were found in Cyp24A1^+/− mice when treated with vehicle, which increased to 106 pg/mL when treated with 24,25-(OH)₂D₃ (Table 1, Fig 1I). On the basis of apparent suppression of 1,25-(OH)₂D₃ formation in the 24,25-(OH)₂D₃-treated Cyp24a1^+/− and Cyp24a1^−/− mice, and absence of CYP24A1 in the 24,25-(OH)₂D₃-treated Cyp24a1^−/− mice, we conclude that the 1,24,25-(OH)₃D₃ observed in these animals arose from inter-conversion via 1-hydroxylation of 24,25-(OH)₂D₃.

4. DISCUSSION

Growing sensitivity of LC-MS/MS methods as well as sample preparation techniques, has circumvented the need to pool mouse samples enabling studies of biological variability of vitamin D metabolism in small volumes of serum taken from individual mice. In vitro and ex vivo studies of CYP24A1 enzyme activity have been successful at delineating individual steps in the C24- and C23-hydroxylation of 25-OH-D₃ and 1,25-(OH)₂D₃ using a combination of reconstituted enzyme systems and kidney perfusion models^14,15,16, but most in vivo studies have been limited to metabolites of high-abundance, or have necessitated pooling of sera from multiple animals, or required use of a tritiated tracer. We have shown that detection of downstream metabolites of C24-oxidation of 25-OH-D₃ is possible in individual mice, and represents the first report of detection of calcioic acid in both human and murine serum. Prior to calcioic acid formation, 24-oxo-23,25-(OH)₂D₃ undergoes cleavage at C23-C24 bond to form tetranor-23-OH-D₃ alcohol as well as an aldehyde in a process called disproportionation, in a similar manner as previously described for side chain cleavage of 24-oxo-1,23,25-(OH)₃D₃¹⁷. We were unable to detect these penultimate intermediates in mouse serum, presumably due to low stability and/or lack of sufficient binding affinity for the vitamin D-binding protein. While identification of the C24- and C23-oxidation products of 25-OH-D₃, paralleled that of 1,25-(OH)₂D₃ in most cases, it was only in 2006 that the water-soluble products of 25-OH-D₃ were specifically characterized using perfusates from vitamin D-intoxicated rats¹⁸. Reddy and co-workers concluded that calcioic acid was the major water-soluble product formed from 25-OH-D₃, but also found that a C24-truncated carboxylic acid termed cholacalcioic acid was also produced as a minor product (approximately 20%)¹⁸, analogous to the C24-acid produced as the terminal catabolite of 20-epi-1,25-(OH)₂D₃ identified by Sakaki’s group¹⁹. Cholacalcioic acid has been proposed to be formed via a peroxy adduct of 24-oxo-25-OH-D₃ leading to cleavage of the C24-C25 bond¹⁹. In vitro work conducted using the human CYP24A1 isoform where calcioic acid was identified as the major water-soluble product support our in vivo observations in the current study that detect calcioic acid in a patient with hypervitaminosis D (Figure 2B), as well as mice given 24,25-(OH)₂D₃ (Figures 1F, 2A). While chromatographic peaks potentially corresponding to cholacalcioic acid were identified in sera and cell extracts (MRM m/z 718→468), retention times and number of peaks were inconsistent between sample types, and we lacked a synthetic standard for this metabolite. On this basis we were unable to confirm the presence of cholacalcioic acid in serum, but we acknowledge that any cholacalcioic acid present in the relative amounts observed by Reddy and colleagues¹⁸ may be below the limit of detection of our serum assay³. The regioselectivity of CYP24A1 has been well documented based on in vitro studies. Different species of CYP24A1 have been noted for various degrees of C24- or C23-hydroxylation. It is the initial hydroxylation step on the side-chain that commits the substrate to a fate of either a C23-acid or a lactone product after multiple steps^3,17 each catalyzed by CYP24A1. Our results confirm from an in vivo perspective that an exogenous dose of 24,25-(OH)₂D₃ continues to proceed through the C24-oxidation pathway to form calcioic acid and not 25-OH-D₃-26,23-lactone which arises from 23,25-(OH)₂D₃, as lactone levels remained un-changed with administration of a 24,25-(OH)₂D₃ dose whereas metabolism via all other C24-pathway products was amplified (Figure 3).

Figure 3: Pathways of bioactivation, catabolism, and interconversion of detected vitamin D metabolites.

Vitamin D metabolites detected in mouse models using the current analytical platforms are indicated. Differential metabolism occurring as a result of Cyp24a1 ablation and/or 24,25-(OH)₂D₃ treatment are summarized in the legend by indicating the pathways that are either normal or unchanged (→), amplified (↑), or decreased (↓) in the specified test groups.

Treatment with 24,25-(OH)₂D₃ rescued Cyp24a1^−/− animals from impaired bone fracture healing. These animals possessed serum 24,25-(OH)₂D₃ of over 800 ng/mL, raising the possibility that other metabolites formed from 24,25-(OH)₂D₃ may have contributed to biological function at this supraphysiological dose, especially as we observed marked suppression of 1,25-(OH)₂D₃ in 24,25-(OH)₂D₃-treated mice. CYP27B1 has been noted for its ability to catalyze 1-hydroxylation of metabolites of 25-OH-D₃ in vitro²⁰, and this appears consistent with our observation of increased 1,24,25-(OH)₃D₃ production with 24,25-(OH)₂D₃ treatment (Figure 1H,I, Table 1, Figure 3). While we cannot specifically distinguish 1,24,25-(OH)₃D₃ formed by CYP24A1 or CYP27B1, we observed continued 1,24,25-(OH)₃D₃ formation in the absence of CYP24A1. The metabolite 1,24,25-(OH)₃D₃ retains approximately one tenth of the VDR-binding affinity of 1,25-(OH)₂D₃, but it is unlikely that this metabolite contributed to the bone fracture healing properties attributed to 24,25-(OH)₂D₃ for several reasons, including (1) 1,24,25-(OH)₃D₃ and 1,25-(OH)₂D₃ were shown not to induce production of LacCer via activation of FAM57B2 in vitro⁷ (2) 24,25-(OH)₂D₃ treatment of Col2-Cre Fam57b^fl/fl mice did not recover from bone fracture healing impairment, even with the presence of CYP24A1⁷. We anticipate that measurement of an expanding range of vitamin D metabolites made possible by improvements to analytical methodology, used in combination with pre-clinical models of vitamin D metabolism will help to advance the understanding of the role of vitamin D metabolism in certain disease states.

5. CONCLUSIONS

Using a sensitive LC-MS/MS assay, we have demonstrated the importance of CYP24A1 in the catabolism of both 25-OH-D₃, as well as exogenous doses of 24,25-(OH)₂D₃ using Cyp24a1^+/− and Cyp24a1^−/− mice used to establish a role for 24,25-(OH)₂D₃ in optimal bone fracture healing. The high sensitivity of our assay permitted assessment of CYP24A1 function on the basis of multi-step catabolism of 25-OH-D₃ and 24,25-(OH)₂D₃ through to terminal pathway products in the C24- and C23-hydroxylation pathways including 24-oxo-25-OH-D₃, 24-oxo-23,25-(OH)₂D₃, 25-OH-D₃-26,23-lactone and calcioic acid. High doses of 24,25-(OH)₂D₃ resulted in suppression of 1,25-(OH)₂D₃ formation and interconversion to 1,24,25-(OH)₂D₃ via 1-hydroxylation (Figure 3). This combination of mouse models and LC-MS/MS tools can be used to identify alternative routes of vitamin D metabolism that might occur in clinical cases of hypervitaminosis D and in idiopathic infantile hypercalcemia (IIH) where CYP24A1 is mutated.

HIGHLIGHTS.

• Serum calcioic acid can be detected in mice given 24,25-(OH)₂D₃.

• CYP24A1 is integral for the clearance of 24,25-(OH)₂D₃ to water-soluble products.

• •1,24,25-(OH)₃D₃ can be formed from 24,25-(OH)₂D₃ in vivo.

• 1,25-(OH)₂D₃ is suppressed when mice are given 24,25-(OH)₂D₃.

Abbreviations:

25-OH-D₃

25-hydroxyvitamin D₃

1,25-dihydroxyvitamin D₃

1,25-(OH)₂D₃

24,25-(OH)₂D₃

24,25-dihydroxyvitamin D₃

VDR

Vitamin D receptor

LacCer

Lactosylceramide

LC-MS/MS

liquid chromatography - tandem mass spectrometry

DMEQ-TAD

4-(2-(6,7-Dimethoxy-4-methyl-3-oxo-3,4-dihydroquinoxalinyl)ethyl)1,2,4-triazoline-3,5-dione)