H.L.J. Makin

Measurement of 25-hydroxyvitamin D2, 25-hydroxyvitamin D3, 24,25-dihydroxyvitamin D2 and 25,26-dihydroxyvitamin D2 in a single plasma sample by mass fragmentography.

A specific and sensitive assay for the measurement of the concentration of 25-hydroxyvitamin D2, 25-hydroxyvitamin D3, 24,25-dihydroxyvitamin D2 and 25,26-dihydroxyvitamin D2 in a single plasma sample is described, using stable isotope dilution mass fragmentography. After addition of appropriate deuterium-labelled internal standards, plasma samples were treated with acetonitrile to precipitate protein, and vitamin D metabolites were extracted on prepacked microparticulate reverse-phase cartridges. Further purification was achieved using straight-phase cartridges and high-performance liquid chromatography. Gas chromatography-mass spectrometry was carried out after appropriate derivatisation of samples and standards. The method has been evaluated in terms of specificity, recovery of added standards, and reproducibility.

Gas chromatography-mass spectrometry and the measurement of vitamin D metabolites in human serum or plasma.

Although methods for the measurement of vitamin D metabolites continue to be developed, few have been properly validated by comparison with methods based on gas chromatography-mass spectrometry, widely accepted as being the definitive methodology. To the best of our knowledge, only three such comparisons have been carried out (14, 42, 83), all three examining HPLC assays for 25-OH-D. This lack of proper validation leads to lack of certainty as to the specificity of many assays widely used for clinical investigation. In our view there is an obvious need for the continuing development of mass fragmentographic assays for vitamin D and its metabolites, primarily for use as reference procedures for the evaluation of less rigorous methodologies. Provided standards, both labeled and unlabeled, become more widely available, development of specific mass fragmentographic assays for any metabolite of vitamin D should be possible. For metabolites where no specific binding protein or antiserum is available, mass fragmentography may be the only alternative.

Stable isotope-labeled vitamin D, metabolites and chemical analogs: Synthesis and use in mass spectrometric studies

Methods for the measurement of vitamin D and its metabolites using stable isotope-labeled internal standards and mass spectrometry are reviewed. The synthesis of both labeled and unlabeled standards is illustrated, and details of the synthesis of (26,26,27,27,27(-2)H5)-25,26-dihydroxyvitamin D3 and (28,28,28(-2)H3)-24,25-dihydroxyvitamin D2 are given. The use of in vitro biologic systems for the production of further metabolites of deuterated 25-hydroxyvitamin D3 is discussed. Use of deuterated 25-hydroxydihydrotachysterol3 as a substrate in the isolated perfused rat kidney has provided valuable data for the assignment of structure to a number of metabolites of 25-hydroxydihydrotachysterol3 formed in this system.

Specific mass fragmentographic assay for 25,26-dihydroxyvitamin D in human plasma using a deuterated internal standard

A specific mass fragmentographic assay for the measurement of 25,26-dihydroxyvitamin D3 [25,26(OH)2D3] in human plasma, using a stable isotope labelled internal standard ([26,27-2H5]25,26(OH)2D3), is described. Plasma samples (5 ml) were extracted with acetonitrile and applied to a C18 Sep-Pak cartridge, from which the vitamin D metabolites were eluted with methanol. The metabolites were then applied to a Sep-Pak SIL cartridge and three fractions were collected. The most polar fraction, containing the polyhydroxylated metabolites, was further purified by high-performance liquid chromatography on Zorbax SIL. The eluent containing 25,26(OH)2D3 was collected, and the 25,26-n-butylboronate cyclic ester 3-trimethylsilyl ether derivative was formed. Gas chromatography-mass spectrometry was carried out, monitoring the intensities of the ions at m/z 449 and m/z 454 (for the internal standard). These ions represent the loss of a methyl group and the 3-silanol group, (M-90-15)+. The minimum limit of detection of the assay was estimated to be approximately 0.05 microgram/l. Inter-assay (3.7%) and intra-assay (8.0%) precision was acceptable and added 25,26(OH)2D3, over the concentration range 0.5-1.5 microgram/l, was recovered quantitatively. The plasma 25,26(OH)2D3 level was estimated in 26 healthy volunteers and ranged from 0.05 to 1.30 microgram/l, with a mean value of 0.54 microgram/l.

Metabolism of a 20-methyl substituted series of vitamin D analogs by cultured human cells: apparent reduction of 23-hydroxylation of the side chain by the 20-methyl group

We describe here for the first time the effect of introducing a 20-methyl group on the side-chain metabolism of the vitamin D molecule. Using a series of 20-methyl-derivatives of 1alpha,25-(OH)2D3 incubated with two different cultured human cell lines, HPK1A-ras and HepG2, previously shown to metabolize vitamin D compounds, we obtained a series of metabolic products that were identified by comparison to chemically synthesized standards on HPLC and GC-MS. 24-Hydroxylated-, 24-oxo-hydroxylated-, and 24-oxo-23-hydroxylated products of 20-methyl-1alpha,25-(OH)2D3 were observed, but the efficiency of 23-hydroxylation was low as compared with that of the natural hormone and, in contrast to 1alpha,25-(OH)2D3, no truncated 23-alcohol was formed from the 20-methyl analog. These data, taken together with results from other analogs with changes in the vicinity of the C17-C20 positions, lead us to speculate that such changes must alter the accessibility of the C-23 position to the cytochrome P450 involved. Using the HepG2 cell line, we found evidence that the 24S-hydroxylated product of 20-methyl-1alpha,25-(OH)2D3 predominates, implying that the liver cytochrome involved in metabolism is a different isoform. Studies with a more metabolically resistant analog of the series, 20-methyl-Delta(23)-1alpha,25-(OH)2D3, gave the expected block in 23- and 24-hydroxylation, and evidence of an alternative pathway, namely 26-hydroxylation. 20-Methyl-Delta(23)-1alpha,25-(OH)2D3 was also more potent in biological assays, and the metabolic studies reported here help us to suggest explanations for this increased potency. We conclude that the 20-methyl series of vitamin D analogs offers new perspectives into vitamin D analog action, as well as insights into the substrate preferences of the cytochrome(s) P450 involved in vitamin D catabolism.

Contribution of several metabolites of the vitamin D analog 20-epi-22-oxa-24a,26a,27a-tri-homo-1,25-(OH)2 vitamin D3 (KH 1060) to the overall biological activity of KH1060 by a shared mechanism of action

The synthetic 1,25-dihydroxyvitamin D(3) (1,25-(OH)(2)D(3)) analog 20-epi-22-oxa-24a,26a,27a-tri-homo-1,25-(OH)(2)vitamin D(3) (KH1060) is considerably more potent than its cognate hormone. The mechanism of action of KH1060 includes interaction with the vitamin D receptor (VDR). We previously showed that KH1060 increases VDR stability in ROS 17/2.8 osteoblastic cells by inducing a specific conformational change in the VDR. KH1060 is metabolized, both in vivo and in vitro, into several stable products. In the present study, we investigated whether these metabolites might contribute to the increased biological activity of KH1060. We found that the potencies of two of these metabolites, 24a-OH-KH1060 and 26-OH-KH1060, were similar to that of 1,25-(OH)(2)D(3) in inducing osteocalcin production by the osteoblast cell line ROS 17/2.8. This report further showed that these metabolites had the same effects as KH1060 on VDR: they increased VDR stability in ROS 17/2.8 cells, while limited proteolytic analysis revealed that they caused a conformational change in the VDR, resulting in an increased resistance against proteolytic cleavage. Furthermore, as shown in gel mobility shift assays, both compounds clearly induced VDR binding to vitamin D response elements. Together, these results show that the potent in vitro activity of KH1060 is not only directed by the effects on the VDR conformation/stabilization of the analog itself, but also by certain of its long-lived metabolites, and emphasizes the importance of detailed knowledge of the metabolism of synthetic hormonal analogs.

Measurement of Vitamin D and its Metabolites by Gas Chromatography-Mass Spectrometry

Gas-liquid chromatography (GC) has been very successfully applied to the analysis of a number of different groups of biologically important compounds. In particular, the analysis of the multiplicity of metabolites of secreted steroid hormones found in human urine has been greatly simplified by the introduction of GC (1). Using conventional packed columns (2) and wall-coated capillary columns (3), ‘profiles’ of varying degrees of complexity have been produced. Plasma steroids have also been analysed by GC using electron capture detection (4) and more recently using gas chromatography-mass spectrometry (GC-MS, 5).

Use of vitamin D4 analogs to investigate differences in hepatic and target cell metabolism of vitamins D2 and D3

In this study, we used molecules with either of the structural differences in the side chains of vitamin D(2) and vitamin D(3) to investigate which feature is responsible for the significant differences in their respective metabolism, pharmacokinetics and toxicity. We used two cell model systems-HepG2 and HPK1A-ras-to study hepatic and target cell metabolism, respectively. Studies with HepG2 revealed that the pattern of 24- and 26-hydroxylation of the side chain reported for 1alpha-hydroxyvitamin D(2) (1alpha-OH-D(2)) but not for 1alpha-OH-D(3) is also observed in both 1alpha-OH-D(4) and Delta(22)-1alpha-OH-D(3) metabolism. This suggests that the structural feature responsible for targeting the enzyme to the C24 or C26 site could be either the C24 methyl group or the 22-23 double bond. In HPK1A-ras cells, the pattern of metabolism observed for the 24-methylated derivative, 1alpha,25-(OH)(2)D(4), was the same pattern of multiple hydroxylations at C24, C26 and C28 seen for vitamin D(2) compounds without evidence of side chain cleavage observed for vitamin D(3) derivatives, suggesting that the C24 methyl group plays a major role in this difference in target cell metabolism of D(2) and D(3) compounds. Novel vitamin D(4) compounds were tested and found to be active in a variety of in vitro biological assays. We conclude that vitamin D(4) analogs and their metabolites offer valuable insights into vitamin D analog design, metabolic enzymes and maybe useful clinically.

Gas chromatography—mass spectrometry in the investigation of on-column dehydration of steroid hormones during gas—liquid chromatography

Some underivatized steroids when injected onto conventional packed columns for gas-liquid chromatography underwent varying degrees of dehydration. This problem was traced to the presence of small pieces of broken glass on the top of the column at the point of injection. This observation provoked an examination of the effect of pre-column dehydration on a number of different types of steroids. Powdered aluminium was placed in the injection liner of a Hewlett-Packard gas chromatograph fitted with an HP1 capillary column connected to a mass selective detector, and injections were made using a new high temperature septumless injection system at temperatures between 200 and 400 degrees C. 5 alpha-androstan-3 alpha-ol, a simple monofunctional C19 steroid chosen as a model to establish optimum conditions, underwent dehydration at injection temperatures greater than 250 degrees C and the product reached a maximum at 400 degrees C when no unchanged steroid was present. Monohydroxylated androgens and oestrogens underwent dehydration at 400 degrees C producing products whose mass spectra indicated they were monenes, although the position of the double bond could not be assigned. Polyfunctional androgens and oestrogens and corticosteroids underwent complex changes producing a number of products some of whose structures could not be determined. The dehydration products had the advantage that they had relatively intense high mass ions and for suitable steroids this might provide enhanced sensitivity of detection during mass fragmentography. In such cases dehydration was reproducible and straight line standard curves were obtained. C27 and C28 secosteroids (vitamins D2 and D3) and some of their metabolites (e.g. 25-hydroxyvitamin D) underwent efficient dehydration, again producing products with intense molecular ions. In the case of 24,25-dihydroxyvitamin D3 and 25,26-dihydroxyvitamin D3, dehydration produced different products which were easily resolved in the chromatographic system used. Dehydration of vitamin D metabolites eliminates the need for derivatization and gives enhanced sensitivity of measurement by gas chromatography-mass spectrometry.

Effects of ‘Non-Calcaemic’ Vitamin D Analogues on 24-Hydroxylase Expression in MG-63 Osteoblast-Like Cells

Background: New ‘non-calcaemic’ analogues of 1,25-dihydroxyvitamin D₃ (1,25(OH)₂D₃) are entering the clinical arena and some of them have been shown to have differential effects in bone. This may have a bearing on the evolution of bone lesions in uraemic patients receiving vitamin D therapies. A potential mechanism for differential effects of analogues lies in their target cell inactivation. Methods: Using a human osteoblastic cell line, MG-63, three analogues, 22-oxacalcitriol (OCT), 19-nor-1,25-dihydroxyvitamin D₂ (paricalcitol) and 1α,25-dihydroxydihydrotachysterol₂ (1,25(OH)₂DHT₂), were compared with 1,25(OH)₂D₃ for (1) their affinity for the vitamin D receptor (VDR) by competitive displacement of tritiated 1,25(OH)₂D₃ from calf thymus VDR; (2) effects on 24-hydroxylase mRNA expression using comparative RT-PCR, and (3) rates of metabolism, using high performance liquid chromatography, over a 24-hour time course. Results: Relative VDR-binding affinities (IC₅₀) were 1,25(OH)₂D₃ (100%), OCT (25%), paricalcitol (14%) and 1,25(OH)₂DHT₂ (0.3%). A ≧3-fold increase in 24-hydroxylase mRNA expression was observed for all compounds at 2 h peaking at 7- to 8-fold above control levels by 12 h, with no significant difference between the analogues and 1,25(OH)₂D₃. Differences in their rates of metabolism were observed [calculated t½ values = OCT (1.2 h) > paricalcitol (2.3 h) > 1,25(OH)₂D₃ (2.6 h) > 1,25(OH)₂DHT₂ (3.4 h)], with OCT having a significantly shorter half-life. Conclusion: In MG-63 cells these analogues up-regulate 24-hydroxylase mRNA expression with similar potency, in each case accelerating ligand inactivation, despite significant differences in VDR affinity. VDR affinity did not correspond to either 24-hydroxylase mRNA expression or the rates of ligand disappearance, suggesting cellular metabolism is one of several factors that determine the analogue specificity of these agents in bone.