Endocrine Abstracts

Introduction: Lipidomics is an emerging approach to characterize circadian rhythm alterations due to diseases of the hypothalamus-pituitary-adrenal axis. Glycerophospholipids are membrane structural constituents and key players in cell signalling, homeostasis and inflammatory and immune responses. Among blood derivatives, serum may reflect the soluble fraction, whereas dried blood spot (DBS) may reflect the cell membrane fraction of these lipids. However, little information is available about individual glycerophospholipid distribution among different blood derivatives as well as about their circadian fluctuation.

Aim: To compare glycerophospholipid levels and their variation during the daytime as assessed in serum and DBS.

Methods: Six normal-weight healthy adults (3 males, 3 females) aged 26-50 y were enrolled. For 7 days, ending with the observation day, subjects had a Mediterranean diet standardized in composition and time. Subjects underwent six paired venepunctures and finger-pricks to obtain serum and DBS, respectively, at pre-established times: 30 min before and 2 hours after breakfast (7.30), lunch (13.30) and dinner (19.30). Ninety glycerophospholipids were quantified by flow injection - tandem mass spectrometry.

Results: Eighty-four metabolites were detected both in serum (mean minimum concentration: 0.117 µM, phosphatidylcholine-diacyl (PC-aa)-C24:0; mean maximum concentration: 413.7 µM, PC-aa-C34:2) and DBS (mean minimum concentration: 0.284 µM, PC-acyl-alkyl(ae)-C44:3; mean max concentration: 218.6 µM, PC-aa-C34:2). Among these, 41 analytes were higher and 35 lower in basal serum vs DBS (P: <0.001-0.046). Further 1 and 4 metabolites were only detected in serum and DBS (mean range concentration: 0.746-2.40 µM), respectively; 1 was not detectable. Levels of 75 glycerophospholipids significantly correlated between DBS and serum (P: <0.0001-0.040). Highest correlations were found for PC-aa-C34:4 (r=0.933), PC-aa-C32:2 (r=0.912) and PC-aa-C34:3 (r=0.909). Significant variations along the six time points were observed for serum PC-ae-C32:1, lysoPC-a-C16:0, lysoPC-a-C16:1, lysoPC-a-C17:0, lysoPC-a-C18:1, lysoPC-a-C18:2, lysoPC-a-C20:3 and lysoPC-a-C20:4 (P: <0.001-0.031), showing a positive linear or quadratic trend. The latter three compounds showed a parallel fluctuation in DBS (P: 0.001-0.014). Forty-six other glycerophospholipids showed significant variations in DBS along the six time points (P: <0.001-0.049), mostly with negative linear trend. Strongest variations were found for PC-aa-C24:0, PC-ae-C30:0, PC-ae-C36:0, PC-ae-C38:0, PC-ae-C38:1 and lysoPC-a-C26:0 (all P: <0.001).

Conclusions: In our healthy cohort, significant differences in serum and DBS glycerophospholipid levels were observed. DBS displayed daily variations of a much larger panel of glycerophospholipids than serum. This is consistent with the presence of the cellular fraction in DBS. DBS may therefore represent a convenient and more informative tool to study glycerophospholipid involvement in diseases related to circadian rhythm imbalance.