Fourth-Generation Progestins Inhibit 3β-Hydroxysteroid Dehydrogenase Type 2 and Modulate the Biosynthesis of Endogenous Steroids

Abstract

Progestins used in contraception and hormone replacement therapy are synthetic compounds designed to mimic the actions of the natural hormone progesterone and are classed into four consecutive generations. The biological actions of progestins are primarily determined by their interactions with steroid receptors, and factors such as metabolism, pharmacokinetics, bioavailability and the regulation of endogenous steroid hormone biosynthesis are often overlooked. Although some studies have investigated the effects of select progestins on a few steroidogenic enzymes, studies comparing the effects of progestins from different generations are lacking. This study therefore explored the putative modulatory effects of progestins on de novo steroid synthesis in the adrenal by comparing the effects of select progestins from the respective generations, on endogenous steroid hormone production by the H295R human adrenocortical carcinoma cell line. Ultra-performance liquid chromatography/tandem mass spectrometry analysis showed that the fourth-generation progestins, nestorone (NES), nomegestrol acetate (NoMAC) and drospirenone (DRSP), unlike the progestins selected from the first three generations, modulate the biosynthesis of several endogenous steroids. Subsequent assays performed in COS-1 cells expressing human 3βHSD2, suggest that these progestins modulate the biosynthesis of steroid hormones by inhibiting the activity of 3βHSD2. The Ki values determined for the inhibition of human 3βHSD2 by NES (9.5 ± 0.96 nM), NoMAC (29 ± 7.1 nM) and DRSP (232 ± 38 nM) were within the reported concentration ranges for the contraceptive use of these progestins in vivo. Taken together, our results suggest that newer, fourth-generation progestins may exert both positive and negative physiological effects via the modulation of endogenous steroid hormone biosynthesis.

Fig 1. The biosynthesis of human steroid hormones consists of multiple reactions which are catalysed by specific steroidogenic enzymes (grey boxes).

The conversion of 17α-hydroxyprogesterone (17OH-Prog) to androstenedione (A4) by CYP17A1 is shown as a dashed box as 17OH-Prog is a poor substrate for the 17,20-lyase activity of human CYP17A1 [22, 31].

Fig 2. Chemical structures of the endogenous steroids progesterone (Prog) and testosterone, the synthetic MR antagonist spironolactone, and the progestins used in this study

: Medroxyprogesterone acetate (MPA), nestorone (NES), nomegestrol acetate (NoMAC), norethisterone/norethindrone acetate (NET-A), levonorgestrel (LNG), gestodene (GES) and drospirenone (DRSP). The inserts (1^st, 2^nd, 3^rd and 4^th) denote the four consecutive generations of progestins.

Fig 3. Effect of selected progestins on total steroid production by the human H295R adrenocortical carcinoma cell line under basal and FSK-stimulated conditions.

Cells were incubated with DMSO (vehicle control) or 1 μM MPA, NET-A, LNG, GES, NES, NoMAC or DRSP, in the absence and presence of 10 μM FSK for 48 hours. Steroid metabolites were extracted from the cell culture medium and analyzed by UPLC–MS/MS. The concentrations of total steroid produced (μM) were normalized to protein concentration (mg/ml). The insert graph shows the total steroid production (μM/mg protein) in the absence of progestin treatment (DMSO) under basal and FSK-stimulated conditions. This total steroid production for both conditions was set as 100%, and the percentage change upon treatment with progestin relative to the vehicle control (DMSO) of each condition was plotted. Results shown are the average of three independent experiments with each condition performed in triplicate (± SEM).

Fig 4. NES is metabolized by the H295R cells under both basal and FSK-stimulated conditions.

H295R cells were treated with DMSO or 1 μM NES, NoMAC or DRSP in the absence and presence of 10 μM FSK for 48 hours. Medium containing the test compounds (no cells) was added to the wells of a 12-well plate as a negative control for metabolism. Steroids were extracted and analyzed by UPLC–MS/MS. The amount of progestin present in the medium after incubation with the cells was expressed as a % relative to the amount of progestin in the negative control for metabolism, which was set as 100%. Result shown is the average of at least two independent experiments with each condition performed in triplicate (± SEM).

Fig 5. NES, NoMAC and DRSP inhibit the activity of human 3βHSD2, while only DRSP inhibits CYP17A1 activity.

COS-1 cells were transiently transfected with plasmids expressing human (A) 3βHSD2 (pCDNA6-hHSD3β2-V5), (B) CYP17A1 (pIRES-hCYP17A1-V5-X-hCYPB5-6HIS) or (C and D) CYP21A2 (pCDNA6-hCYP21A2-V5), respectively. Cells were subsequently treated with 1 μM Preg (A) or Prog (B and C) or 17OH-Prog (D), in the absence (DMSO) and presence of 1 μM MPA, LNG, GES, NES, NoMAC or DRSP for 20 minutes (A), 4 hours (B) or 90 minutes (C and D), respectively. The steroid metabolites produced by the cells in the medium were extracted and analyzed by UPLC–MS/MS. The concentration of the steroids produced by the cells was normalized to the total protein concentration using the Bradford protein assay method. The % conversion of substrate to product was plotted, with the substrate only response (DMSO) set as 100% and everything else relative to that. Results shown are the average of at least two independent experiments with each condition performed in triplicate (± SEM).

Fig 6. Inhibition of 3βHSD2 activity by NES, NoMAC, DRSP and trilostane.

COS-1 cells were transiently transfected with a plasmid expressing human 3βHSD2 (pCDNA6-hHSD3β2-V5), and subsequently treated with increasing concentrations (0.5, 1, 2, 4 and 8 μM) of Preg (substrate) in the presence of 0.0, 0.2 or 0.5 μM (A) NES, (B) NoMAC, (C) DRSP or (D) trilostane. The conversion of Preg to Prog was analyzed using UPLC-MS/MS. Michaelis-Menten plots are shown in the absence (black symbols and lines) and presence of 0.2 μM (red symbols and lines) and 0.5 μM (blue symbols and lines) of NES, NoMAC, DRSP and trilostane. A K_m of 0.85 ± 0.05 μM and V_max of 31.1 ± 0.7 nmol/min/mg were obtained in the absence of inhibitor. Three inhihitory mechanisms were fitted (S2 Fig) and the best fit mechanism is shown. Each data point represents the mean ± SE of one experiment performed in duplicate. These results were validated by a model predicting 3βHSD2 activity in an independent experiment (S3 Fig, dashed green line).

Fig 7. NES, NoMAC and DRSP do not inhibit the mRNA expression of 3βHSD2 and CYP17A1 in the human H295R adrenocortical carcinoma cell line.

The H295R cell line was incubated with DMSO (vehicle control) or 1 μM NES, NoMAC or DRSP for 6 hours. Total RNA was isolated, reversed transcribed to cDNA and real-time qPCR performed to determine the relative mRNA expression levels of (A) 3βHSD2 and (B) CYP17A1. GAPDH was used as the reference gene. Results shown are the average of four independent experiments with each condition performed in duplicate (± SEM).