Physiological and biochemical aspects of 17β-hydroxysteroid dehydrogenase type 2 and 3

Steroid hormones regulate a broad variety of physiological functions through the transcriptional regulation of target genes. The active steroid hormones discussed in this thesis largely elicit their physiological effects through the activation of nuclear receptors. Many of these receptors reside in their inactive form in the cytoplasm, translocate to the nucleus upon ligand binding and drive the transcription of target genes. The local interconversion of active and inactive steroid hormones is regulated by members of 17β-hydroxysteroid dehydrogenases/reductases (17β-HSDs). The enzyme 17β-HSD2 converts active estrogen estradiol and the potent androgen testosterone to its inactive keto-forms, whereas 17β-HSD3 mainly converts androstenedione into testosterone. The present thesis is split into three major projects that focus mainly on potential toxicological and therapeutic effects of inhibiting the enzyme 17β-HSD2 and biochemical aspects of the enzymes 17β-HSD3.
The first project was designed to develop a 17β-HSD2 pharmacophore model and subsequently use this model as a virtual screening tool to identify novel nonsteroidal 17β-HSD2 inhibitors. It has been hypothesized that pharmacological inhibition of the enzyme 17β-HSD2 expressed in osteoclasts could be a useful strategy to treat osteoporosis through increasing local concentrations of active sex hormones. Our approach was initiated through the development of a pharmacophore model that underwent several rounds of experimental validation and improvement. In silico screening of internal and external compound databases using the optimized pharmacophore model resulted in the identification of several novel nonsteroidal compounds that inhibit 17β-HSD2 at nanomolar concentrations in vitro. Furthermore, the virtual screening of a cosmetic ingredients database revealed several paraben compounds as potential 17β-HSD2 inhibitors. In vitro examinations revealed that all tested paraben compounds were found to significantly inhibit 17β-HSD2 at a concentration of 20 µM. However, parabens are rapidly metabolized to p-hydroxybenzoic acid, which does not influence 17β-HSD2 activity. We reported a novel potential estrogenic effect of paraben compounds by inhibiting 17β-HSD2 although their estrogenic potential is unlikely to be of toxicological concern due to their rapid metabolism by esterases.
In the second project we biochemically analyzed six mutations in the HSD17B3 gene that were associated to cause 17β-HSD3 deficiency in Egyptian and Tunisian patients with 46, XY disorder of sexual development (DSD). Patients with 17β-HSD3 deficiency are unable to synthesize sufficient amounts of testosterone during embryogenesis which leads to severe undervirilization of internal and external genitals. All tested mutations (G133R, C206X, T54A, M164T, and L194T) were confirmed to be nearly inactive in vitro and therefore unable to sufficiently convert androstenedione into testosterone. Further analyses showed that the G289S polymorphism exhibited a similar rate of testosterone formation as the wild type 17β-HSD3 enzyme and consequently cannot be causing the pathogenesis of 46, XY DSD. All HSD17B3 mutations associated with 46, XY DSD in this project were predicted by an in silico 17β-HSD3 homology model (based on the structure of 17β-HSD1) to interfere with either cofactor (NADPH) or substrate (androstenedione) binding sites. The G289S polymorphism however, was located on the surface of the enzyme without eliciting any effects on the cofactor or substrate active sites.
Besides its critical role in sexual differentiation, testosterone regulates a variety of physiological functions. The vast majority of testosterone is produced in testicular Leydig cells. The final project in the thesis focused on reviewing the in vitro models available for investigating androgen disruption by xenobiotics. Our in vitro investigations focused on validating the three most promising Leydig cell lines (MA-10, BLTK, TM3) for their potential to report androgen disruption by xenobiotic compounds through alteration of 17β-HSD3 activity and transcription. Our experiments revealed that these cell lines express minimal levels of 17β-HSD3 mRNA compared to freshly isolated mouse testes. Furthermore, the cell lines also exhibited a low rate of androstenedione to testosterone formation. In conclusion, all tested cell lines are not useful as screening tool to test androgen disruption by xenobiotic compounds due to the lack of endogenous 17β-HSD3.
Through the work described in this thesis, we have developed and optimized an in silico 17β-HSD2 pharmacophore model for the identification of inhibitors of the enzyme. We then confirmed and mechanistically evaluated mutations in the HSD17B3 gene associated with 46, XY DSD and finally investigated Leydig cell models to test testosterone disruption by xenobiotic compounds. Altogether, these findings significantly expand the knowledge about physiological and biochemical aspects of the enzymes 17β-HSD2 and 17β-HSD3.