Assessment of the potential risk of oteseconazole and two other tetrazole antifungals to inhibit adrenal steroidogenesis and peripheral metabolism of corticosteroids

Abstract

The triazole antifungals posaconazole and itraconazole can cause pseudohyperaldosteronism with hypertension and hypokalemia, edema, and gynecomastia by inhibiting steroid synthesis and metabolism. Mechanisms underlying pseudohyperaldosteronism include inhibition of adrenal 11β-hydroxylase cytochrome-P450 (CYP) 11B1 and 17α-hydroxylase (CYP17A1) as well as peripherally expressed 11β-hydroxysteroid dehydrogenase type 2 (11β-HSD2). To enhance specificity for fungal CYP51, tetrazoles have been developed. This study employed H295R adrenocortical cells and enzyme activity assays to assess the potential risk of oteseconazole and two other tetrazoles, VT-1598 and quilseconazole, to inhibit adrenal steroidogenesis or 11β-HSD2. Steroidomic footprint analyses of H295R cell supernatants using untargeted liquid-chromatography-high-resolution mass-spectrometry (LC-HRMS) indicated overall patterns common to oteseconazole, quilseconazole and itraconazole, as well as similarities between VT-1598 and isavuconazole. Additionally, more specific features of the steroid signatures were observed. Targeted quantification of nine adrenal steroids in supernatants from treated H295R cells revealed an overall inhibition of adrenal steroidogenesis by the three tetrazoles, itraconazole and isavuconazole, providing an explanation for their similar steroidomic pattern. Applying recombinant enzymes indicated that this effect is not due to direct inhibition of steroidogenic enzymes because no or only weak inhibition could be observed. Moreover, oteseconazole and the two other tetrazoles did not inhibit 11β-HSD2, suggesting that they do not pose a risk of pseudohyperaldosteronism. Furthermore, oteseconazole did not alter steroid concentrations in a recent clinical study. Nevertheless, follow-up studies should assess the mechanism underlying the observed overall steroidogenesis inhibition by tetrazoles, itraconazole and isavuconazole, and whether concentrations achievable in a subgroup of susceptible patients might cause adrenal insufficiency and hyperplasia.

Keywords: H295R; adverse drug reaction; azole antifungal; cytochrome P450; enzyme; inhibition; steroid profile; steroidogenesis.

FIGURE 1

Structures of triazole and tetrazole antifungals with their characteristic aromatic heterocycle depicted in red and green, respectively.

FIGURE 2

Schematic presentation of adrenal steroidogenesis. In the adrenals cytochrome-P450 (CYP) and hydroxysteroid dehydrogenase (HSD) enzymes are responsible for the synthesis of glucocorticoids (red), mineralocorticoids (yellow) and androgen precursors (blue). Inhibition of CYP17A1 17α-hydroxylase or CYP11B1 results in feedback stimulation of adrenal steroidogenesis to maintain circulating cortisol concentrations, accompanied by an excessive production of mineralocorticoids. In peripheral tissues, 11β-HSDs interconvert cortisol and its inactive metabolite cortisone. Inhibition of 11β-HSD2 in kidney results in cortisol-mediated activation of mineralocorticoid receptors (MR). Both mechanisms result in an excessive MR activation with hypertension and hypokalemia.

FIGURE 3

Principal component analysis (PCA) of steroid patterns obtained from H295R cells. H295R cells were treated with different concentrations of triazole (A) and tetrazole (B) antifungals (rectangles displays 3 µM treatment, triangles 1 µM and circles 0.3 µM). Cell culture supernatants were analyzed using an untargeted LC-HRMS approach with steroid annotation and identification. Experiments were performed three times independently, each in duplicate. Results were normalized to the DMSO control. Individual experimental results appear as one point on the PCA scores plot.

FIGURE 4

Effects of selected tetrazole antifungals on the steroid profile of forskolin-stimulated H295R cells. Forskolin-stimulated H295R cells were incubated with the positive control prochloraz or different concentrations of oteseconazole, quilseconazole and VT-1598 for 48 h. Steroid concentrations in cell culture supernatants were determined by UHPLC-MS/MS. Experiments were performed three times independently, each in duplicate. Data were normalized to the forskolin control (FC) and fold changes are shown as mean ± SD. n.d. indicates steroids for which no peak could be observed. LLOD/2 was used for calculations when the signal to noise was below five (indicated with an asterisk, see Supplementary Table S1 for method sensitivity). Color-coding indicates percent reduction in steroid formation relative to forskolin with ≥67% remaining steroid formation depicted in green, 33%–66% in yellow, and ≤32% in red. Values ≥ 200% are highlighted in blue.

FIGURE 5

Inhibitory effects of selected azole antifungals on CYP11B1, CYP17A1, CYP21A2, 11βHSD1, 11βHSD2, and 3βHSD2. Initial screenings were performed at 10 µM of the tested substances. CYP17A1 and CYP21A2 activities were determined using microsomal fractions of transfected COS-1 cells, measuring the formation of 17α-hydroxyprogesterone from progesterone (CYP17A1 17α-hydroxylase reaction) (A, B), DHEA from 17α-hydroxypregnenolone (CYP17A1 17,20-lyase reaction) (C, D) and 11-DOC from progesterone (CYP21A2) (E, F). For the CYP11A1 activity assay, the formation of pregnenolone from 20α-hydroxycholesterol was measured using mitochondrial fractions of transfected V79-4 cells (G). CYP11B1 activity was assessed using mitochondrial preparations of transfected V79-4 cells and measuring the conversion of 11-DOC to corticosterone, with ketoconazole serving as positive control. Activities of 11β-HSDs and 3β-HSD2 were measured using lysates of transfected HEK293 cells. Cortisol and cortisone were quantified as substrate and product for 11βHSD2 and the reverse reaction for 11βHSD1 activity. Progesterone was quantified as product of the 3β-HSD2 catalyzed reaction with pregnenolone as substrate. Glycyrrhetinic-acid and bisphenol A were used as positive controls to inhibit 11β-HSDs and 3β-HSD2, respectively (H). Experiments were performed at least three times independently. Results were normalized to the DMSO control and represent mean ± SD.