Sertoli Cell-Specific Modification of The Androgen Receptor DNA-Binding Domain Disrupts Sertoli Cell Function and Completion of Spermatogenesis.

Abstract

Androgen receptor (AR) actions are vital for spermatogenesis, but male germ cells do not express AR, highlighting its key role in testicular somatic cells. Recent genetic models with targeted Sertoli cell (SC) AR disruption confirmed the vital role for Sertoli AR-activated pathways in spermatogenesis. However, complete loss of functional AR in these SC AR knockout (SCARKO) models excludes selective analysis of classical (via direct AR DNA binding) versus potential indirect non-classical or non-genomic AR actions. We have used a Cre-loxP strategy to selectively study the in vivo requirement of AR DNA-binding in SC function. Transgenic (Tg) mice with SC-specific Cre were crossed with mice carrying a floxed-AR gene (Arflox) to target exon 3. Cre-mediated inframe excision of exon 3, encoding a zinc finger essential for DNA-binding, was predicted to generate mutant ARΔex3 to selectively disrupt AR-regulated gene transcription via direct DNA interaction. We established two unique SC-specific ARΔex3 (SCARΔex3) models: i) AMH.SCARΔex3 derived from [Tg antimullerian hormone (AMH) promoter-driven Cre mice] X [Arflox mice], and ii) Abp.SCARΔex3 derived from [Tg androgen-binding protein (Abp) promoter-driven Cre mice] X [Arflox mice]. ROSA-LacZ reporter mice showed that the Abp promoter directs SC-specific Tg expression consistent with our past work, with testis Abp.Cre expressed as early as day 17 postcoitum. RT-PCR detected testis ArΔex3 RNA in both SCARΔex3 models. Using hemizygous TgCre mice, Abp.SCARΔex3 (49.3 ± 1.8 mg) testes were significantly larger (p < 0.001) than AMH.SCARΔex3 testes (29.9 ± 0.9 mg), and both were smaller (p < 0.001) than age-matched 9 week old controls (105.4 ± 4.4 mg). Homozygous TgCre mice were used to increase Cre expression levels. Testis size was further reduced in homozygous Abp.SCARΔex3 (31.6 ± 1.9 mg, p < 0.001) but not homozygous AMH.SCARΔex3 (32.1 ± 1.5 mg) mice relative to corresponding hemizygous TgCre SCARΔex3 testes, suggesting a maximal effect for SC Ar exon 3 disruption was achieved using Tg homozygous Abp.Cre or hemizygous AMH.Cre mice. Immunohistochemistry revealed SC nuclear AR expression was retained in SCARΔex3 testes. Despite histological detection of Leydig cell hypertrophy in SCARΔex3 testes, serum testosterone levels were equivalent to normal (22.8 ± 5.0 nM) in Abp.SCARΔex3 (15.7 ± 8.0 nM) and AMH.SCARΔex3 (16.6 ± 7.0 nM) 9 week old males, unlike the elevated or reduced androgen levels reported in SCARKO models with complete SC AR loss. Stereology revealed total SC numbers were normal in SCARΔex3 testes, whereas spermatogenesis in both SCARΔex3 models showed predominant meiotic arrest, noting larger Abp.SCARΔex3 testes using hemizygous Tg Abp.Cre mice exhibited more post-meiotic development. The testicular phenotype of these unique SCARΔex3 models demonstrates that despite nuclear ARΔex3 localisation, a functional AR DNA-binding domain is vital for Sertoli cell function and full spermatogenic development. The SCARΔex3 testes resemble SCARKO testes with complete SC AR loss, suggesting that non-genomic or pathways independent of direct AR-DNA binding play no major role in SC function. In combination, our novel SCARΔex3 models provide unique opportunity to further determine dose-dependent effects of AR activity in Sertoli cells. This research was supported by NHMRC funding.