Analytical considerations for accurately capturing the relevant species contributing to vitamin D status in liquid chromatography‐tandem mass spectrometry assays

Abstract This tutorial review focuses on analytical challenges encountered with the liquid chromatography‐tandem mass spectrometry determination of 25‐hydroxyvitamin D, which is currently still considered the metabolite that is most representative of vitamin D status. It describes how multiple binding states of circulating 25‐hydroxyvitamin D (phase II metabolites, epimers, free/bioavailable/protein‐bound species) can influence the accuracy of the analytical determination. It also summarizes important chemical species that can inadvertently contribute to vitamin D status and thus cause systematic errors. These interfering endogenous and exogenous compounds might be isomers of vitamin D, constitutional isomers or isobars and the article outlines techniques to eliminate or minimize these interferences, including chromatographic separations, ion mobility spectrometry, and high‐resolution mass spectrometry.


INTRODUCTION
This tutorial review was prompted by a recent article by Jenkinson et al, 1 who presented a novel method that permits simultaneous analysis of phase II metabolites of vitamin D 3 (sulfates and glucuronides) and unconjugated circulating 25-hydroxyvitamin D 3 (25(OH)D 3 ) using identical instrumental method conditions. The authors' aim was to answer the question whether combined serum measurement of unconjugated and conjugated forms is better suited for assessing vitamin D status than current methods. Importantly, they demonstrated that sulfated conjugates form a significant proportion of the circulating vitamin D metabolites, whereas glucuronide conjugates are less important. The authors suggest that a combination of both conjugated and unconjugated measurements may provide a more accurate assessment of vitamin D status. 1 This is an open access article under the terms of the Creative Commons Attribution License, which permits use, distribution and reproduction in any medium, provided the original work is properly cited. © 2021 The Authors. Analytical Science Advances published by Wiley-VCH GmbH Generally, there is an ongoing discussion over what exactly needs to be measured to properly assess vitamin D status. 2, 3 Premer and Schulman recently posed the question "Have we been measuring the wrong form of vitamin D?" 2 The question referred to the role of vitamin D in coronary artery disease, which has been under scrutiny, with unreliable results concerning its prognostic value and role in therapy. Moreover, recent studies on differences of vitamin D binding protein (DBP) and vitamin D levels in various populations have triggered the question of whether "technical errors" in using total rather than free/bioavailable 25(OH)D concentrations are the reasons for observed discrepancies. 3 This article focuses only on the 25(OH)D metabolite, which is currently still considered the metabolite that is most representative of vitamin D status, and its assessment by liquid chromatographytandem mass spectrometry (LC-MS/MS) assays (although many of the issues described here will equally apply to other analytical methods). 4 Tutorial doi.org/10.1002/ansa.202100057 Instead, it will summarize important chemical species that might inadvertently contribute to vitamin D status in its current definition and thus cause systematic errors. These interfering species might be isomers of vitamin D (without being biologically active), constitutional isomers, or isobars that exhibit similar or identical physicochemical properties during analysis. In addition, a fraction of the circulating target 25(OH)D marker compound in the serum/plasma sample may be 'hidden' or masked by other endogenous components and thus unavailable in the analysis under the conditions used.
In the following text, we will focus on phase II metabolites, epimers, free/bioavailable/protein-bound species, isomeric vitamin D metabolites as well as other endogenous and exogenous interferences.

PHASE II METABOLITES
We start the discussion with vitamin D conjugates (phase II metabolites: glucuronides and sulfates, Figure 1 In many applications, the sulfate conjugates are analyzed in a separate assay from 25(OH)D, however, utilizing, for example, the deprotonated molecules formed from the acidic sulfate group for sensitive detection in ESI negative ionization mode. 10

EPIMERS
The parallel epimerization pathway of 25(OH)D provides epimer- together. 13 In fact, the presence of the 3α epimer has been suggested to overestimate serum 25(OH)D 3 concentrations by up to 25%. 14 In adults, the 3β-25(OH)D 3 levels are usually much lower (<10%) 15,16 and it has sometimes been suggested that this contribution is negligible for adults, not requiring an LC-MS/MS method that differentiates the two epimers, only for samples of patients younger than 1 year. 17 For example, we previously reported a linear correlation between 3α-25(OH)D 3 levels and 25(OH)D 3 in a cohort of 91 patients with chronic liver diseases. 13   The unbound 25(OH)D is hypothesized to play an important role in the numerous non-classical actions of vitamin D. As reflected in the "free hormone hypothesis," it is the free 25(OH)D metabolite that is able to enter cells, thus exerting biological actions. 24,25 The clinical relevance of these metabolites remains to be unraveled; however, in order to do this, accurate quantification methods need to become widely available.
For further information in this area of research, the present authors refer the interested reader to, for example, refs. [24][25][26][27][28] . Finally, it remains to be seen whether ion mobility spectrometry separations may provide sufficient selectivity and resolving power in the future to separate hydroxylated vitamin D isomers prior to mass spectrometry.  Of course, liquid chromatography will remove many of the interfering isobars and isomers. Nevertheless, because of the large number of possible endogenous and exogenous metabolites, many of which have the potential to mimic the mass spectral behavior of 25(OH)D, 4 it is very likely that multiple isobaric interferences will still co-elute with 25(OH)D.

CONCLUSIONS
This tutorial has addressed two areas of concern with analytical assays for quantification of the 25 However, an often-overlooked general issue in quantitative bioanalytical LC-MS/MS assays is the interference that the stable isotope standard might experience. The stable isotope standard's mass is usually 3-6 Da higher than the precursor ion of the analyte and the level of interference may be different from the analyte, which would then can lead to systematic errors. HRMS assays with sufficient resolution are expected to reduce these effects but cannot address isomeric interferences.
In conclusion, since one never knows in advance, which exact Open access funding enabled and organized by Projekt DEAL.

AUTHOR CONTRIBUTION
The manuscript was written through contributions of all authors.