Analysis of vitamin D and its metabolites in biological samples - Part II: Optimization of a sample preparation method for liver tissue

Extraction of vitamin D, including its hydroxylated and esterified metabolites, from soft tissues such as the liver is challenging due to the lipophilic character of matrix and analytes that are expected in very low concentration levels. In this study, we aimed at the optimization of two-step extraction using solid-liquid extraction as the first step, followed by solid -phase extraction. Various solvents, including ethanol, acetonitrile, methanol, acetone, heptane, and heptane with isopropanol, were investigated to isolate vitamin D compounds from liver tissue in the first step. Acetone was finally selected as the most suitable solvent for the solid-liquid extraction, with the highest recovery in the range of 67 - 98% for polar hydroxylated forms and 3 - 28% for lipophilic vitamin D and esters. Two solid phase extraction (SPE) based on the (i) "bind and elute strategy" and (ii) "removal strategy" using hydrophilic-lipophilic balanced SPE sorbent were optimized as a proceeding step for acetone extracts to increase the method selectivity. Finally, two optimized methods, combining solid-liquid extraction and individual SPE strategy, were examined in terms of sensitivity, recovery, matrix effect, accuracy, and precision. The limits of quantification were in the range of 1 - 10 ng/mL and 3 - 20 ng/mL analyzed by ultra -high performance supercritical fluid chromatography and ultra -high performance liquid chromatography hyphenated a with tandem mass spectrometer, respectively. The absolute recovery determined for the "bind and elute strategy" protocol was in the range of 3 - 24 %. Nevertheless, this method was free of matrix effects, which were determined to be in the 73 - 120 % range. On the contrary, the "removal strategy" approach provided higher recovery values for all compounds (47 - 123 %), but the results for nonpolar vitamin D and esters were strongly affected by signal suppression (matrix effects 3 - 51 %). Both methods fulfilled the criteria for accuracy and precision requested by the European Medicine Agency Guideline on Bioanalysis. "Removal strategy" SPE with decreased manual intervention and lower solvent consumption was finally applied to mouse liver tissue to determine vitamin D and its hydroxylated and esterified metabolites for the first time. The results, i.e., vitamin D esters detected in liver tissue, supported the notion that esters of vitamin D can be stored in lipophilic tissues to release vitamin D.