Comparing the androgenic and estrogenic properties of progestins used in contraception and hormone therapy

Abstract

Progestins used in endocrine therapies bind to multiple steroid receptors and are associated with several side-effects. It is thus important to understand the relationship between steroid receptor cross-reactivity and the side-effect profile of progestins. In cell lines that express negligible levels of steroid receptors, we report for the first time the binding affinities, potencies and efficacies of selected progestins from different generations determined in parallel. We show that the progestins bind to the androgen receptor (AR) with similar affinities to each other and progesterone, while none bind estrogen receptor (ER)-β, and only norethisterone acetate, levonorgestrel and gestodene bind ERα. Comparative dose-response analysis revealed that progestins from the first three generations display similar androgenic activity to the natural androgen dihydrotestosterone for transactivation, while norethisterone acetate, levonorgestrel and gestodene are ERα agonists. We show for the first time that the anti-androgenic properties of progesterone and drospirenone are similar to the well-known AR antagonist hydroxyflutamide, while nomegestrol acetate is more potent and nestorone less potent than both hydroxyflutamide and progesterone. Moreover, we are the first to report that the older progestins, unlike progesterone and the fourth generation progestins, are efficacious ERα agonists for transrepression, while the selected progestins from the second and third generation are efficacious AR agonists for transrepression. Considering the progestin potencies and their reported free serum concentrations relative to dihydrotestosterone and estradiol, our results suggest that the progestins are likely to exert AR-, but not ERα- or ERβ-mediated effects in vivo.

Keywords: Androgen receptor; Contraception; Estrogen receptor; Hormone therapy; Progestins.

Fig. 1

(A) The selected progestogens all compete with [³H]-MIB for binding to the human AR, (C) while only NET-A, LNG and GES bind to human ERα. COS-1 cells expressing the human (A) AR or (C) ERα expression vector, were incubated for (A) 16 hours with 0.2 nM [³H]-MIB in the absence or presence of increasing concentrations of either unlabeled MIB (●), DHT (◆), LNG (▼), GES (▲), NES (■), NoMAC (△), or DRSP (*) or (C) 4 hours with 10 nM [³H]-E₂ in the absence or presence of increasing concentrations of either unlabeled E₂ (●), NET-A (○), LNG (▼) and GES (▲). Counts per minute (cpm) were measured and normalized to protein concentration determined by the Bradford method [40]. Total specific binding of (A) [³H]-MIB or (C) [³H]-E₂ only was set as 100% and the binding of unlabeled competitors plotted relative to this. Log K_d/K_i values of the ligands for the (B) AR and (D) ERα were plotted.

Fig. 2

(A) Second- and third-generation progestins display similar androgenic properties to DHT, while, like P₄, the fourth-generation progestins are AR antagonists. COS-1 cells, expressing the (A and D) human AR and the pTAT-2xPRE-E1b-luciferase reporter plasmid, were treated with varying concentrations of MIB (●), DHT (◆), P₄ (□), LNG (▼), GES (▲), NES (■), NoMAC (△) or DRSP (*) in the (A) absence or (D) presence of 0.1 nM MIB (set as 100%) for 24 hours. (G) LNG and GES are full agonists for transactivation via ERα, while NET-A is a partial agonist. The HEK293 cell line, expressing the (G) human ERα and the pGL3-2xERE-pS2-luciferase promoter-reporter plasmid, were treated with increasing concentrations of E₂ (●), NET-A (○), LNG (▼) and GES (▲) for 24 hours. Luciferase activity was measured in relative light units and normalized to protein concentration determined by the Bradford method [40]. (B, E and H) Maximal responses and (C, F and I) log EC₅₀ values were plotted. As P₄, NES, NoMAC and DRSP displayed very weak partial AR agonist activity, these log EC₅₀ values were not depicted in (C) and should be interpreted with caution.

Fig. 3

(A) Progestins from the first three generations are full agonist for transrepression via the AR, while the fourth-generation progestins are partial agonists. COS-1 cells expressing human AR and the 5xNFκB-luciferase reporter plasmid, were treated with vehicle (EtOH) and 10 ng/ml PMA in the absence (set as 100%) or presence of increasing concentrations of MIB (●), DHT (◆), P₄ (□), MPA (◊), NET-A (○), LNG (▼), GES (▲), NES (■), NoMAC (△) or DRSP (*) for 24 hours. (D) NET-A and GES are full agonists for transrepression via ERα, while LNG is a partial agonist. HEK293 cells expressing human ERα and the p(IL6κB)₃50hu.IL6P-luciferase reporter plasmid, were treated with vehicle (EtOH) and 20 ng/ml TNFα in the absence (set as 100%) or presence of increasing concentrations of E₂ (●), NET-A (○), LNG (▼) and GES (▲) for 24 hours. Luciferase activity was measured and normalized as in Fig. 2. Treatment with PMA or TNFα resulted in a ~12-fold and ~33-fold induction, respectively (Fig. 3A and 3D inserts). (B and E) Maximal repression and (C and F) log EC₅₀ values were plotted.