Circulating adrenal and gonadal steroid hormones heterogeneity in active young males and the contribution of 11-oxy androgens

Demographic characteristics of participants

There were 71 participants with a mean age of 25.1 years. Participants were mainly of Arab (54%) or Black (27%) ethnicity. Descriptive statistics of all participants are presented in Table 1. Participant demographics showed participants did not differ significantly on any demographic characteristics (p > 0.05).

UHPSFC-MS/MS analysis of free unconjugated adrenal and gonadal steroid hormones

Circulating mineralocorticoids, glucocorticoids and androgens (Supple. file, Supple. Table 1) and many steroids, being below the lower limit of detection (LOD) (Supple. Table 2) were therefore not detected. The classical androgens that were not detected were 5αDIONE and 3αDIOL and the C11-oxy androgens not detected were 11OH-5αDIONE, 11K-5αDIONE, 11OHDHT, 11KDHT, 11OHAST, 11KAST and 11OH3αADIOL. The C11-oxy progesterones that were not detected included 21dF, 21dE, Pdione, 11αOHP4, 11αOH-DHP4, 11βOHP4, 11KP4, 11βOHDHP4, 11KDHP4, 11OHPdiol, alfaxalone, 11OH-Pdione, 11K-Pdione, and 11K-Pdiol. DOC was the only adrenal steroid not detected at all while others were detected but at low frequencies: 11-deoxycortisol (0.37—2.15 nmol/L, 6/70), 18OHCORT (1.64–32.96 nmol/L, 4/70), ALDO (1.0—15.15 nmol/L, 3/70). Two C11-oxy steroids were only detected in three participants, 11K3αADIOL (1.28, 2.76 nmol/L, 2/70) and 3,11diOH-DHP4 (3.4 nmol/L, 1/70), and therefore not included in further analysis. Progesterones detected and not included in analysis were DHP4 (6.58 nmol/L, 1/70). Pdiol (7.36 nmol/L, 1/70) and pregnanetriol (22.26 nmol/L, 1/70).

Steroids subsequently included in analyses are listed in Table 2, showing the percentage at which each steroids was detected above lower limit of quantification (LLOQ) enabling accurate quantification and below the LLOQ but above the LOD. Steroids detected below the LOD in more than 75% of participant were not included in analyses.

Adrenal and gonadal steroids present at frequencies higher than 25% were there analysed and included mineralocorticoids, glucocorticoids and androgens. The steroid profile of one participant differed markedly from other profiles in the cohort. All the steroids analysed were below the LOD except for T, possibly indicative of an underlying clinical condition and was subsequently was excluded from analysis.

Analysis of steroids present at frequencies higher than 25% in participants

Steroid analysis of the participant group (n = 70) showed that while T, 24.5 nmol/L, was detected in all participants falling within the normal range, the hormone spanned a wide range 7.76–47.3 nmol/L. T precursors, A5 and A4, which were detected in 72% and 83% of participants, respectively, showed A4 concentrations 2.7-fold lower than A5. DHT, the potent SRD5A metabolite of T, was detected in far fewer participants, only 27% participants had detectable concentrations. No classical inactive androgens were detected (Fig. 1). Although the classical androgen levels were higher than the C11-oxy androgen levels, the latter were detected in a more participants with 11OHA4, 11OHT, 11KA4 and 11KT detected in 98.6, 95.7, 77 and 63% of the participants, respectively. 11OHA4 was detected at highest concentration followed by 11KA4 with 11OHT and 11KT which were all lower than 1 nmol/L. Taken together the C11-oxy androgens were present at far lower levels than the classical androgens.

Although the C11-oxy androgens were detected, no C11-oxy progesterones were detected. All of the progesterones were present at far lower concentrations than androgens with pregnenolone (P5) and progesterone (P4) detected in ~ 25% of participants. Of interest was 16α-hydroxyprogesterone (16OHP4), the product of P4, which was detected in 98.6% of participants, the most abundant progesterone detected.

Also detected in all participants were the glucocorticoids, cortisol and corticosterone (CORT) and their respective inactive forms, cortisone and 11-dehydrocorticosterone (11-DHCORT). Cortisol and cortisone were the most abundant steroids at 338 and 55.8 nmol/L, respectively with means all within the normal range. Deoxycortisol, the precursor steroid of cortisol was detected at very low levels in only 8% of participants. The inactive metabolites of cortisol and CORT, catalysed by 11βHSD2, cortisone and 11-DHCORT, were detected at lower levels than their active precursors, respectively. In addition, CORT levels as well as cortisol:cortisone were all within normal ranges. The downstream metabolite in the mineralocorticoid pathway, 18-hydroxycorticosterone (18OH-CORT) was detected in only four participants and aldosterone (ALDO) in three participants.

Upon further inspection it was evident that some participants had very low cortisol and cortisone concentrations together with low cortisol:cortisone. Cluster analysis showed that DHEA in these participants were within the lower normal range of circulating DHEA. We therefore analysed adrenal and gonadal steroids dividing the participants into two groups: those in the lower tertile (6.9 nmol/L), comprising 34% of the participants, and those with DHEA concentrations above (66%). Participant demographics of the participants with DHEA below and DHEA above 7 nmol/L showed participants in the two groups did not differ significantly on any demographic characteristics (p > 0.05) (Supple. Table 3.)

Comparing participants with low circulating DHEA (DHEA < 7 nmol/L) to those with DHEA > 7 nmol/L (Table 3), showed that although steroid profiles differed considerably, the classical androgen (A5, A4, T, DHT) levels remained comparable between the two groups. In contrast, the C11-oxyandrogens with the exception of 11KT were significantly lower in the group with low DHEA. In addition, together with significantly lower DHEA, the progesterones detected, P5, 17OHP5 and 16OHP4 were also significantly lower. The active glucocorticoids, cortisol and CORT, their inactive forms, cortisone and 11-DHCORT as well as cortisol:cortisone were also significantly lower in the group with low DHEA. Further analysis also showed that the presence of DHEA was significantly associated with 16OHP4 (p < 0.001) (Suppl. Table 4). The presence of DHEA was not significantly associated with any other steroid (p > 0.05).

Considering the mean steroid values of subjects with a DHEA lower than 7 nmol/L, the data clearly showed that the steroid profiles of these males differed markedly when compared to males with higher DHEA levels. Mean steroid concentrations of all steroids measured in this study were lower in males with DHEA lower than 7 nmol/L (Fig. 2), with data showing overall lower adrenal steroids in designated pathways.