rel="nofollow" href="#ulink_b59e98ad-dd24-5680-8591-c77963d2a839">69,] given its relatively high concentrations in the blood (0.7–24 nM) [70], modern assays for this metabolite use LC-MS exclusively. Frequently this is done as part of a multimetabolite profile. Like that for 25(OH)D and 1,25(OH)2D the C3-beta epimer of 24,25(OH)2D has been identified [71]. Moreover, 24,25(OH)2D exists as both the 24R,25(OH)2D and 24S,25(OH)2D epimer, but only the R epimer is biologically active [72]. Therefore, careful separation of these epimers as well as other dihydroxylated vitamin D metabolites prior to MS is required to obtain accurate results. As noted, 24,25(OH)2D measurement is often part of a multimetabolite profile. The approaches to such assays have recently been reviewed [40]. Most of these assays used ESI for ionization, triplequadrupole instruments for MS, and nonspecific water loss transitions for monitoring. Derivitazation is often employed to increase sensitivity for the less abundant metabolites in the profile, but PTAD derivatization of 25(OH)D was found to interfere with the separation of C3-beta epi-25(OH)D from 25(OH)D [73]. Other methods of derivitazation have been developed that may circumvent this problem [74], and derivitazation may not be required if larger samples are used [75].
Free Vitamin D Metabolite Measurements
Up to this point, I have focused on measurements of total vitamin D metabolite levels. However, the free levels may be a better marker of vitamin D status. The vitamin D metabolites circulate in blood extensively bound to 2 liver-produced proteins, vitamin DBP and albumin. Approximately 85% of this binding is to DBP, 15% to albumin. Given the high affinity of these metabolites for DBP, and the abundance of DBP in normal individuals, free vitamin D metabolite levels are very low (approximately 0.03% of total for free 25(OH)D and 0.4% of total for free 1,25(OH)2D) [76, 77]. However, in conditions such as liver disease (and likely nephrotic syndrome and protein losing enteropathies), DBP and albumin levels are low. The reverse is true during the 3rd trimester of pregnancy when DBP levels are increased. In these situations, the free level of 25(OH)D and 1,25(OH)2D is likely to provide a better assessment of vitamin D status than total level [76–78]. Prior to the development of a commercially available direct measure of free 25(OH)D, investigators have calculated the free level using the published affinity constants of 25(OH)D (and 1,25[OH]2D) for DBP and albumin. This method has several drawbacks in that it assumes that these affinity constants do not change with different physiologic states and are identical from one individual to the next. Neither assumption is valid [79–81]. Moreover, these calculations rely on accurate measurement of the levels of DBP and albumin. Although the measurement of albumin is standard from laboratory to laboratory, the measurement of DBP has been more problematic. DBP has a number of polymorphisms, which affect not only its affinity for the vitamin D metabolites [81], but also its measurement. A commercially available monoclonal antibody used in an ELISA from R&D Systems was found to underestimate the levels of the 1F allele of DBP, an allele common in Africans (Gambians) and African Americans, but seldom found in whites of European descent [82]. Using this ELISA resulted in substantially lower measurements of DBP levels in African Americans compared to Caucasians [83], results that were not substantiated when polyclonal antibodies were used either as ELISAs or with radial immunodiffusion [82]. When mass spectrometry was used to measure DBP, the results confirmed that DBP levels differed little on the basic of allelic differences, but measurements were substantially below that obtained with the polyclonal assay [84]. This disparity between the polyclonal immunoassays and mass spectrometry has not been resolved, but this disparity further points to the difficulty in calculating free vitamin D metabolite levels rather than measuring these levels directly. Centrifugal ultrafiltration was the first of the direct assays for the free metabolites, and provided much of the initial data on affinity constants and levels in different groups varying in DBP levels, but because of its labor intensive requirements it is no longer used. Instead, an immunoassay has been developed by Future Diagnostics that is now commercially available, and measurement of free 25(OH)D by LC-MS may soon be available.
Centrifugal Ultrafiltration. The centrifugal ultrafiltration assay consists of an inner vial capped on one end with dialysis membrane resting on filter pads at the bottom of an outer vial. The serum sample following incubation with freshly purified 3H-labeled vitamin D metabolite and 14C-labeled glucose as a marker of free water was placed in the inner vial and centrifuged at 37°C for 45 min. The ratio of 3H/14C in the ultrafiltrate to that in the sample determined the % free. The free concentration was then calculated by multiplying the % free times the total metabolite concentration [77]. This method is dependent on the purity of the labeled vitamin D metabolite requiring purification by HPLC before each assay, as degradation products could increase the fraction filtered. Also, the assembly of the ultrafiltration apparatus is labor intensive, as no suitable commercial equipment is available that does not bind to the filtered vitamin D metabolite. This assay is not commercially available.
Immunoassay. In this assay, developed by Future Diagnostics B.V., Wijchen, The Netherlands, antibodies reactive to 25(OH)D are immobilized in a microtiter well (solid phase). Standards, controls, and patient samples are added to the wells, binding to the solid-phase antibodies. The solid phase is then washed and a biotin-labeled analog of 25(OH)D is added to react with the remaining antibody in a second incubation. After washing, the wells are incubated with a streptavidin-peroxidase conjugate and bound enzyme is quantitated using a colorimetric reaction. Intensity of the signal is inversely proportional to the level of free 25(OH)D in the sample. This assay like all immunoassays is dependent on the specificity of the antibody. It is reported to modestly (70–90%) underestimate 25(OH)D2 relative to 25(OH)D3 (http://www.future-diagnostics.nl/).
The measurements of free 25(OH)D using this immunoassay are in agreement with the measurements obtained with centrifugal ultrafiltration in subjects with a large range of DBP and albumin concentrations [76, 85].
LC-MS. As noted earlier, LC-MS has been used to detect 25(OH)D in saliva, which is expected to be free of DBP and albumin and so represents free 25(OH)D [54]. In this method, 1 mL of saliva was deproteinized with acetonitrile, purified using a Strata-X cartridge, derivatized with PTAD, ionized by ESI and subjected to LC-MS. The limits of detection were reported as 2 pg/mL. The range of values obtained in normal controls was between 3 and 15 pg/mL, correlating well with total serum 25(OH)D (10–30 ng/mL). The intercept was positive, but the free fraction in the mid range of the assay was approximately 0.04%, in line with the results from centrifugal ultrafiltration and the Future Diagnostics immunoassay.
