Pregnanolone

Postpartum depression (PPD) affects ~10-15% of childbearing individuals, with deleterious consequences for two generations. Recent research has explored the biological mechanisms of PPD, particularly neuroactive steroids (NAS). We sought here to investigate associations between NAS levels and ratios during pregnancy and the subsequent development of depressive symptoms with postpartum onset. NAS levels and psychological scales were measured in individuals with and without mood disorders at up to eight visits across pregnancy and postpartum. Generalized linear mixed-effects regression models were used to assess relationships in euthymic pregnant individuals between each of the NAS biomarkers and ratios and subsequent PPD. Participants with a one-unit increase in the log isoallopregnanolone/pregnanolone ratio at the third trimester (T3) had higher odds (OR = 1.64, 95% CI: 1.13-2.37, FDR adjusted p = 0.038, C-index = 0.82), and those with a one-unit increase in the log pregnanolone/progesterone ratio at T3 had lower odds (OR = 0.64, 95% CI: 0.47-0.88, FDR adjusted p = 0.036, C-index = 0.82) of developing PPD; those with a one-unit increase in the log progesterone level at T3 had higher odds of developing PPD (OR = 4.00, 95% CI: 1.54-10.37, FDR adjusted p = 0.035, C-index = 0.80). We found key differences in the progesterone metabolic pathway at the third trimester, indicating likely decreased activity/expression of the 3α-HSD enzyme and/or increased activity/expression of 3β-HSD.

Neuroactive steroids are metabolites of progesterone, synthesised during pregnancy by the placenta. Here, we describe development of a novel liquid chromatography tandem mass spectrometry (LC-MS/MS) assay for quantitation of allopregnanolone, pregnanolone, isopregnanolone, epipregnanolone and allopregnan-20α-ol-3-one in maternal serum. Following addition of deuterated internal standards, 200 μL of serum was subjected to solid phase extraction. Chromatography was performed using a pentafluorophenyl column, and LC-MS/MS on a Sciex 6500+. Sample injection volume was 20 μL, and injection-to-injection time 10.0 min. The assay was validated according to published guidelines; assay linearity and lower limit of quantification were suitable for analysis of each steroid in maternal serum, for all analytes mean recoveries were 100 % ± 15 %, intra- and inter-assay imprecision <15 %, and matrix effects negligible, and specificity experiments confirmed nil interference from a wide range of endogenous metabolites of progesterone. The method was applied to human serum samples obtained from a large cohort of third trimester pregnancies which were subsequently characterised by normal fetal and maternal outcomes, and relationships between maternal neuroactive steroid concentrations and fetal gestational age assessed. Positive correlations between maternal serum concentration and fetal gestational age were observed for isopregnanolone, allopregnanolone and allopregnan-20α-ol-3-one. The LC-MS/MS method offers significant advantages over previously published approaches for quantitation of neuroactive steroids in human maternal serum, notably obviating the need for derivatisation, whilst achieving exceptional specificity. Characterisation of normal maternal neuroactive steroid concentrations will aid future research as dysregulated placental progesterone metabolism is observed in pregnancies with poor outcomes.