Oral contraceptives containing ethinyl estradiol and drospirenone increase hydroxylation and methylation of endogenous estrogen but not genotoxic estrogen DNA-adduct formation

Abstract

Combined oral contraceptives (COCs) are commonly prescribed for the prevention of pregnancy, as well as numerous other non-contraceptive health reasons. COCs act by suppressing the natural hormonal fluctuations of the menstrual cycle that result in ovulation. No studies have investigated the effects of COC use on endogenous estrogen biotransformation and the production of estrogen metabolites. This is important since imbalances in hormone biotransformation (e.g., inefficient methylation by catechol O-methyltransferases; COMT) are implicated in the initiation of breast cancer through the generation of genotoxic metabolites (i.e., estrogen quinones) and reactive oxygen species (ROS), and the depletion of vital antioxidants and metabolic cofactors. Here, we quantified the urinary levels of various estrogen precursors and metabolites in healthy young women who were using COCs containing drospirenone (DRSP) and ethinyl estradiol (EE) (n = 24) and controls (n = 25) via liquid chromatography-tandem mass spectrometry (LC‒MS/MS). In addition, we analysed several circulatory intermediates of the methylation cycle that are linked to the methylation of catechol estrogens via LC‒MS/MS. We found that free urinary estradiol (E2) and estrone (E1) were significantly lower, while 2-methoxyestrone (2-MeOE1) levels were significantly higher in COC users. Excretion of some metabolites including 16-hydroxylation pathway metabolites, glutathione conjugates, and DNA-adducts were also lower in COC users, although total hormone and metabolite excretion levels were not significantly different. Estrone metabolite ratios were higher in COC users, including 2&4-MeOE1:E1, 2&4-OHE1:E1, E1-3-sulphate: E1, and E1-3-glucuronide: E1. There was a positive correlation between 2-hydroxyestrogen and 2-methoxyestrogen levels in controls but not in COC users. In addition, the serum betaine and dimethylglycine (DMG) levels, as well as the betaine: choline ratio, were reduced in COC users, whereas the levels of choline and serine and the DMG: betaine ratio were significantly increased. DMG levels positively correlated with methoxyestrogens and methoxyestrogen: hydroxyestrogen ratios in COC users, while S-adenosylmethionine (SAM) negatively correlated with 2-MeOE2. Our data suggests that the use of EE/DRSP increases the flux of endogenous hormones into the hormone biotransformation pathway, resulting in increased conversion of estrogens (especially E1) into conjugated, catechol, and methylated estrogens but that the latter is limited by methyl-group donor availability. Interestingly, the increased oxidation of estrogens in COC users does not result in increased DNA-adduct formation.

Keywords: Catechol estrogens; Combined oral contraceptives; Drospirenone; Estrogen DNA-adducts; Estrogen biotransformation; Ethinyl estradiol; Methylation.

Fig. 1

Estrogen metabolite abundances. Estrogen metabolites were grouped into different branches of the estrogen biotransformation pathway, and the abundances were calculated as a percentage of the total metabolite load (fraction of the whole). These percentages are indicated on the right-hand side of the figure. The total metabolite load was calculated as the sum of all parent estrogens and all estrogen metabolites and precursors. The percentages in bold indicate values that are > 1.5-fold greater than those of the other group.

Fig. 2

Spearman correlation between estrogen metabolite groups. Correlation analyses were performed for each experimental group separately using the grouped metabolite abundances calculated for Fig. 1. Spearman r values for each correlation are indicated in the individual squares of the heatmap. P values are given in Supplementary Tables S3 and S4.

Fig. 3

Spearman correlations between metabolites and intermediates of the methylation cycle and catechol and methoxyestrogens. Correlation analyses were performed for each experimental group separately. Spearman r values for each correlation and the associated p values are given in Supplementary Tables S5, S6, S7, and S8.