A physiological role for androgen actions in the absence of androgen receptor DNA binding activity

https://doi.org/10.1016/j.mce.2011.08.017Get rights and content

Abstract

We tested the hypothesis that androgens have physiological actions via non-DNA binding-dependent androgen receptor (AR) signaling pathways in males, using our genetically modified mice that express a mutant AR with deletion of the 2nd zinc finger of the DNA binding domain (ARΔZF2) that cannot bind DNA. In cultured genital skin fibroblasts, the mutant ARΔZF2 has normal ligand binding ability, phosphorylates ERK-1/2 in response to 1 min DHT treatment (blocked by the AR antagonist bicalutamide), but has reduced androgen-dependent nuclear localization compared to wildtype (WT). ARΔZF2 males have normal baseline ERK-1/2 phosphorylation, with a 1.5-fold increase in Akt phosphorylation in ARΔZF2 muscle vs WT. To identify physiological actions of non-DNA binding-dependent AR signaling, ARΔZF2 males were treated for 6 weeks with dihydrotestosterone (DHT). Cortical bone growth was suppressed by DHT in ARΔZF2 mice (6% decrease in periosteal and 7% decrease in medullary circumference vs untreated ARΔZF2 males). In conclusion, these data suggest that non-DNA binding dependent AR actions suppress cortical bone growth, which may provide a mechanism to fine-tune the response to androgens in bone.

Highlights

► We investigate non-DNA binding dependent (DBD) androgen receptor (AR) actions. ► AR from male mice with deletion of the 2nd zinc finger of the DBD (ARΔZF2) has normal ligand binding ability in vitro. ► AR from ARΔZF2 mice phosphorylates ERK-1/2 in response to DHT in vitro. ► DHT treatment of ARΔZF2 male mice suppresses cortical bone growth in vivo. ► Non-DNA binding dependent AR actions oppose classical AR actions in bone.

Introduction

Androgens, including testosterone and 5α-dihydrotestosterone (DHT), are essential in males for fetal masculinization and induction of secondary sexual characteristics during puberty. Androgens also have anabolic actions in non-reproductive tissues, including increasing muscle strength and bone mineral density (Brill et al., 2002, Bhasin et al., 2005). Androgens act predominantly via the androgen receptor (AR) in target cells, to activate downstream signaling pathways. In the classical, or DNA binding-dependent model of AR action, the AR acts directly as a ligand-dependent transcription factor, with ligand-bound AR translocating to the nucleus, dimerizing and binding directly to androgen response elements in target genes (MacLean et al., 1997). The AR recruits co-activators or co-repressors and other transcription factors, resulting in activation or repression of gene transcription. Other AR signaling pathways, which do not involve binding of AR to DNA and are collectively termed non-classical or non-genomic signaling pathways, may also be present in cells (Estrada et al., 2003, Unni et al., 2004, Kousteni et al., 2001, Peterziel et al., 1999). While some non-DNA binding-dependent androgen actions may occur through receptors other than the AR, the focus of this study is AR-mediated non-DNA binding-dependent signaling. Non-genomic pathways can be divided into two broad categories, involving either rapid activation of 2nd messenger signaling pathways, or indirect gene repression in the absence of the AR binding directly to DNA.

A number of studies have demonstrated that androgens activate the MAPK/ERK cascade in cell lines, including osteoblasts and osteocytes, muscle and prostate cancer cells (Unni et al., 2004, Kousteni et al., 2001, Peterziel et al., 1999). Ligand-bound AR can induce phosphorylation of CREB via activation of ERK (Unni et al., 2004, Kousteni et al., 2001), and in human genital skin fibroblasts, DHT-stimulated ERK phosphorylation is AR-dependent, but does not require the DNA binding domain (Unni et al., 2004). Non-DNA binding-dependent AR signaling can also activate the PI3K/Akt cascade, which requires direct interaction between AR and PI3K (Sun et al., 2003), and other intracellular messengers (Estrada et al., 2000).

The other category of non-DNA binding-dependent AR signaling, involving indirect gene transrepression, has also been characterized in vitro. The AR binds and sequesters transcription factors such as AP-1 and Ets-related transcription factor to block their transactivation ability, resulting for example in the repression of nerve growth factor receptor (Ngfr) and matrix metalloproteinases (Mmp)-3, -7 and -13, respectively (Kallio et al., 1995, Schneikert et al., 1996). Although the AR does not bind directly to DNA in this transrepression, the AR DNA binding domain is still required for the interaction with some transcription factors (Kallio et al., 1995), but not others (Schneikert et al., 1996).

Although non-DNA binding-dependent AR signaling pathways have been identified in vitro, the physiological relevance of these actions remains unclear. Supraphysiological doses of testosterone induce rapid vasodilatation in humans and other mammals (Wu and von Eckardstein, 2003), but this occurs in response to exogenous androgen treatment. One of the few demonstrated physiological roles of non-DNA binding-dependent androgen actions via the AR is to induce meiosis during mouse oocyte maturation (Gill et al., 2004). Recent evidence from other steroid hormone receptors has also shown physiological actions of non-DNA binding-dependent receptor signaling. A knock-in mouse model in which estrogen receptor (ER)α is unable to bind DNA, shows that non-DNA binding-dependent ERα actions play a role in female reproductive tract development (Jakacka et al., 2002). Similarly, mouse models demonstrate that some glucocorticoid actions are independent of glucocorticoid receptor (GR) DNA binding (Reichardt et al., 1998). These findings highlight the importance of the requirement for in vivo models to elucidate non-DNA binding-dependent pathways of steroid hormone receptors.

AR knockout (ARKO) models are AR-null, with deletion of both DNA binding-dependent and non-DNA binding-dependent AR actions (Yeh et al., 2002, Kawano et al., 2003, De Gendt et al., 2004). Therefore, AR-null mice cannot be used to distinguish between DNA binding-dependent and non-DNA binding-dependent actions in vivo. We generated a mouse model with an in-frame deletion of exon 3 of the AR gene, which encodes the 2nd zinc finger of the DNA binding domain (ARΔZF2) (Quigley et al., 1992, Notini et al., 2005), in contrast to AR-null ARKO mice, which produce no AR protein (Yeh et al., 2002, Kawano et al., 2003, De Gendt et al., 2004). In humans, deletion of exon 3 causes complete androgen insensitivity syndrome (CAIS), and AR lacking exon 3 is unable to bind DNA from the classic hormone response element of the MMTV-LTR with high affinity, or transactivate (Quigley et al., 1992). Male ARΔZF2 mice have a female external phenotype, small intra-abdominal testes and no internal reproductive organs (Notini et al., 2005), similar to CAIS, demonstrating the mutant AR cannot activate the classical DNA binding-dependent signaling pathway. We have previously used this ARΔZF2 mouse model to demonstrate the importance of DNA binding-dependent actions of the AR in male and female physiology (MacLean et al., 2008, MacLean et al., 2010, Chiang et al., 2009, Notini et al., 2007). Unlike AR-null mouse models, our ARΔZF2 provides the unique opportunity to investigate the physiological role of the non-DNA binding-dependent actions of the AR in vivo. We hypothesize that AR signaling through non-DNA binding-dependent pathways plays a physiologically significant role in mediating androgen actions in males. The aim of the present study is to test this hypothesis using our ARΔZF2 model.

Section snippets

Mice

Global ARΔZF2 male mice were generated by crossing ARWT/ΔZF2 heterozygous females with WT males as described (Notini et al., 2005, MacLean et al., 2008), and maintained on a congenic C57BL/6J background. WT male littermates were used as controls. Mice were housed in a conventional facility, with a 12-h light–dark cycle and standard chow and water provided ad libitum. All studies were undertaken with approval from the Austin Health Animal Ethics Committee.

Androgen treatment of ARΔZF2 males

To determine the effects of DHT acting

Results

The goal of this study was to determine the physiological role of androgens acting through non-DNA binding-dependent AR signaling pathways using our ARΔZF2 male mice. Firstly, it was important to demonstrate that the mutant ARΔZF2 expressed in ARΔZF2 mice was capable of the functions required for non-DNA binding-dependent signaling.

Discussion

We have identified a physiological role of non-DNA binding-dependent AR signaling pathways in bone using our ARΔZF2 male mice, which have a mutant AR lacking the 2nd zinc finger of the DNA-binding domain (ARΔZF2). We previously showed that ARΔZF2 males have female external genitalia, small intra-abdominal testes, undeveloped internal reproductive tract and reduced serum testosterone levels (Notini et al., 2005). Male ARΔZF2 mice also have decreased muscle mass and decreased trabecular bone

Acknowledgments

We thank Rebecca Sawyer of the Hanson Institute for her excellent technical assistance. This work was supported by NHMRC (Australia) Project Grant Nos. 350334, 350346 and 509328 and Career Development Award No. 359226 (HEM); and an Endocrine Society of Australia Research Higher Degree Scholarship (TPSP). Funding sources were not involved in any aspects of the research study.

References (50)

  • K.M. Wiren et al.

    Targeting of androgen receptor in bone reveals a lack of androgen anabolic action and inhibition of osteogenesis: a model for compartment-specific androgen action in the skeleton

    Bone

    (2008)
  • C.M. Allan et al.

    Estradiol induction of spermatogenesis is mediated via an estrogen receptor-α mechanism involving neuroendocrine activation of follicle-stimulating hormone secretion

    Endocrinology

    (2010)
  • A.M. Axell et al.

    Continuous testosterone administration prevents skeletal muscle atrophy and enhances resistance to fatigue in orchidectomized male mice

    Am. J. Physiol. Endocrinol. Metab.

    (2006)
  • D.T. Baran et al.

    Effect of testosterone therapy on bone formation in an osteoporotic hypogonadal male

    Calcif. Tissue Res.

    (1978)
  • S. Bhasin et al.

    Older men are as responsive as young men to the anabolic effects of graded doses of testosterone on the skeletal muscle

    J. Clin. Endocrinol. Metab.

    (2005)
  • K.T. Brill et al.

    Single and combined effects of growth hormone and testosterone administration on measures of body composition, physical performance, mood, sexual function, bone turnover, and muscle gene expression in healthy older men

    J. Clin. Endocrinol. Metab.

    (2002)
  • C. Chiang et al.

    Mineralization and bone resorption are regulated by the androgen receptor in male mice

    J. Bone Miner. Res.

    (2009)
  • K.H. Chuang et al.

    Neutropenia with impaired host defense against microbial infection in mice lacking androgen receptor

    J. Exp. Med.

    (2009)
  • A. Crozat et al.

    Comparison of androgen regulation of ornithine decarboxylase and S-adenosylmethionine decarboxylase gene expression in rodent kidney and accessory sex organs

    Endocrinology

    (1992)
  • M.L. Cutress et al.

    Structural basis for the nuclear import of the human androgen receptor

    J. Cell Sci.

    (2008)
  • R.A. Davey et al.

    Effects of amylin deficiency on trabecular bone in young mice are sex-dependent

    Calcif. Tissue Int.

    (2006)
  • K. De Gendt et al.

    A Sertoli cell-selective knockout of the androgen receptor causes spermatogenic arrest in meiosis

    Proc. Natl. Acad. Sci. USA

    (2004)
  • M. Estrada et al.

    Aldosterone- and testosterone-mediated intracellular calcium response in skeletal muscle cell cultures

    Am. J. Physiol. Endocrinol. Metab.

    (2000)
  • M. Estrada et al.

    Testosterone stimulates intracellular calcium release and mitogen-activated protein kinases via a G protein-coupled receptor in skeletal muscle cells

    Endocrinology

    (2003)
  • P.A. Frischmeyer et al.

    Nonsense-mediated mRNA decay in health and disease

    Hum. Mol. Genet.

    (1999)
  • Cited by (0)

    1

    The work performed in this study arose equally from these authors’ laboratories.

    View full text