CSIRO PUBLISHING

www.publish.csiro.au/journals/rfd

Reproduction, Fertility and Development, 2008, 20, 861870

Oestradiol-induced spermatogenesis requires a functional androgen receptor

Patrick LimA , Charles M. AllanA , Amanda J. NotiniB , Anna-Maree AxellB , Jennifer SpalivieroA , Mark JimenezA , Rachel DaveyB , Julie McManusB , Helen E. MacLeanB , Jeffrey D. ZajacB and David J. HandelsmanA,C
A Andrology

Laboratory, ANZAC Research Institute, Concord Hospital and University of Sydney, Sydney, NSW 2139, Australia. B Department of Medicine, Austin Health, University of Melbourne, Heidelberg, Vic. 3084, Australia. C Corresponding author. Email: djh@anzac.edu.au

Abstract. Spermatogenesis requires androgen but, paradoxically, oestradiol (E2) treatment stimulates spermatogenic development in gonadotrophin- and androgen-deficient hypogonadal (hpg) mice. The mechanisms of E2-induced spermatogenesis were investigated by determining intratesticular E2 levels and testis cell populations in E2-treated hpg male mice, and E2 spermatogenic actions were determined in androgen receptor-knockout (ARKO) mice. Despite increased serum E2 concentrations (150300 pmol L1 ), intratesticular E2 concentrations declined fivefold (P < 0.001) in E2-treated v. untreated hpg male mice. Serum FSH reached 40% of normal and total testicular numbers of known FSHresponsive Sertoli, spermatogonia and meiotic spermatocyte populations were significantly (P < 0.001) elevated 1.7-, 4and 13-fold, respectively. However, E2 administration also increased androgen-dependent pachytene spermatocytes and post-meiotic spermatids to levels comparable with testosterone-treated hpg testes. Selective investigation of androgen receptor involvement used E2-treated ARKO mice, which were found to exhibit increased (1.6-fold; P < 0.05) intratesticular E2 concentrations and suppression of the elevated serum gonadotrophins, although FSH remained twofold higher than normal. However, testis size and total Sertoli, spermatogonia and spermatocyte numbers were not increased in E2-treated ARKO male mice. Therefore, E2-stimulated murine spermatogenic development occurs with markedly suppressed and not elevated intratesticular E2 levels and displays an absolute requirement for functional androgen receptors. We propose that this paradoxical E2 spermatogenic response is explained by predominantly extratesticular E2 actions, increasing FSH to combine with residual androgen activity in hpg testes to stimulate pre- to post-meiotic development. Additional keywords: FSH, hypogonadal, mouse, testis.

Introduction Mammalian spermatogenesis is regulated by the two pituitary hormones FSH and LH, the latter exerting its effects by stimulating Leydig cell production of testosterone (T; Spaliviero et al. 2004). Androgen is necessary (Vanha-Perttula et al. 1970; Yeh et al. 2002) and sufficient (Singh et al. 1995) to induce qualitative spermatogenic development in mice, with the androgen receptor (AR) in Sertoli cells a key requirement for completion of male germ cell meiosis and post-meiotic development (Chang et al. 2004; De Gendt et al. 2004; Holdcraft and Braun 2004), noting that the AR is not expressed (Bremner et al. 1994; Zhou et al. 2002) or required (Lyon et al. 1975; Johnston et al. 2001) within male germ cells themselves. FSH is essential for quantitative completion of spermatogenic development, acting via Sertoli cells uniquely expressing FSH receptors (Simoni et al. 1997). Although FSH alone cannot initiate full spermatogenesis in androgen-deficient hypogonadal (hpg) mice (Allan et al. 2001, 2004; Haywood et al. 2003) or men (Schaison et al. 1993), it promotes Sertoli cell proliferation
CSIRO 2008

and increases spermatogonia and meiotic spermatocyte numbers in hpg mice (Haywood et al. 2003; Allan et al. 2004) and prepubertal administration may enhance subsequent FSH + human chorionic gonadotrophin (hCG)-induced pubertal onset and testis function in hypogonadotrophic hypogonadal boys (Raivio et al. 2007). Furthermore, when combined with adequate T to ensure meiotic competence, FSH completes the quantitative restoration of murine spermatogenic development, with FSH and androgen activity having additive or synergistic meiotic and post-meiotic effects (Haywood et al. 2003). The role of oestrogens in spermatogenesis has been of considerable interest (ODonnell et al. 2001), partly prompted by alleged toxicological effects of xeno-oestrogens (Safe 2005; Delbes et al. 2006) and pharmacological studies (Atanassova et al. 1999). Yet, any direct physiological role of E2 in spermatogenic development remains to be established. Full spermatogenesis develops in mice with targeted disruption of the aromatase gene (Robertson et al. 1999) or the oestrogen receptor (ER) (Eddy et al. 1996) or ER (Krege et al. 1998) gene
10.1071/RD08144 1031-3613/08/080861

862

Reproduction, Fertility and Development

P. Lim et al.

(or both; Dupont et al. 2000), noting spermatogenesis is subsequently impaired by secondary extratesticular effects in mice lacking either aromatase and ER, but not ER (ODonnell et al. 2001). Using the hpg mouse, which has functional FSH and LH/T deficiency from birth (Cattanach et al. 1977), we established that T or non-aromatisable 5-dihydrotestosterone (DHT) alone induce qualitatively complete spermatogenesis, although Sertoli cell stock and ultimate testis size remained subnormal (Singh et al. 1995; Haywood et al. 2003). DHT-induced spermatogenesis led to the conclusion that aromatisation (i.e. E2) was not required for spermatogenesis (Singh et al. 1995). Surprisingly, it was subsequently shown that E2 could also induce spermatogenic development in hpg mice (Ebling et al. 2000), although the mechanisms were not well defined. It is possible that E2 acts directly upon intratesticular ER, such as ER expressed in Sertoli and germ cells (Saunders et al. 1998; Zhou et al. 2002). However, direct actions of E2 in the testis are difficult to evaluate because intratesticular E2 concentrations were not determined (Ebling et al. 2000; Baines et al. 2005) and E2 administration also stimulated pituitary secretion of FSH, raising the possibility for an extratesticular pathway (Ebling et al. 2000). In addition, anti-androgen (bicalutamide) reduced the E2-stimulated spermatogenic response in hpg testes without decreasing pituitary FSH content, suggesting a role for androgen signalling in this paradoxical E2-stimulated spermatogenic response (Baines et al. 2005). In that study, direct effects of the anti-androgen on the testis, whether acting as a partial agonist or antagonist, could not be excluded, nor were the effects of androgen blockade on serum FSH levels reported. To further define the mechanism of E2-induced spermatogenesis and the possible involvement of androgen, FSH or intratesticular E2, we have carefully investigated the effects of E2 in androgen-deficient hpg, as well as AR-knockout (ARKO; Notini et al. 2005), mouse models. AR-deficient mice exhibit low serum T levels, unlike the high T concentrations in androgen-insensitive humans (Quigley et al. 1995), due to androgen sensitivity of the murine 17hydroxylase and 17-ketosteroid reductase enzymes involved in T biosynthesis (Murphy and OShaughnessy 1991). Hence, the androgen-deficient and -insensitive background of ARKO male mice provides a definitive approach to determine whether E2 may induce spermatogenesis by pathways independent of the classical AR-mediated mechanism. Materials and methods Experimental animals All experimental procedures were approved by either the Animal Welfare Committee of the Central Sydney Area Health Service or the Austin Health Animal Ethics Committee within National Health and Medical Research Council of Australia guidelines for animal experimentation. Mice were housed (19 22 C, 12 h lightdark cycle) in cages with free access to water and standard chow. Hypogonadal (hpg) mice from the ANZAC Institute colony were genotyped by polymerase chain reaction (PCR), as described previously (Singh et al. 1995). ARKO male mice were produced using the Cre/Lox system, crossing mice carrying floxed Ar gene exon-3 with transgenic (Tg)

CMV-Cre mice, as described elsewhere (Notini et al. 2005). CMV-Cre mice (originally from Dr U Lichtenberg, University of Cologne, Cologne, Germany) express X-linked Tg Cre via the ubiquitous CMV promoter (Schwenk et al. 1995), so generation of ARKO male mice first required breeding floxed Ar male and Tg Cre female mice to generate females heterozygous for the mutant Ar exon-3 deletion, which were then bred with wild-type males to generate ARKO male progeny. Cremediated disruption of the mouse Ar gene was determined using genomic DNA from toe or tail biopsy and PCR, using three forward PCR primers for the floxed mouse Ar gene (ARin2-F, CAGAAATCCACCTGCCTCTACC in intron 2 upstream of the first LoxP site; ARex3-F, CTTCTCTCAGGGAAACAGAAGT in exon-3; and ARneo-F, TAGATCTCTCGTGGGATCATTG in the inserted neo) with a common reverse primer (ARin3-R, GGGAGACACAGGATAGGAAATT in intron 3). To confirm ARKO XY males, theY chromosome was detected using primers and PCR for mouse Sry as described by Notini et al. (2005). Steroid treatments Weanling (21-day-old) male hpg, ARKO or phenotypically normal littermate control mice were administered E2 doses by insertion of Silastic implants under the flank skin with mice under ketamine + xylazine anaesthesia. To achieve low-dose E2 delivery, crystalline E2 (catalogue no. E950; Steraloids, Newport, RI, USA; purity 98% by thin-layer chromatography) was diluted with crystalline cholesterol (Sigma-Aldrich, St Louis, MO, USA) at a ratio of 1 : 1000, 1 : 2000 or 1 : 4000 of E2 : cholesterol by weight, solubilised in 100% ethanol by gentle heating and shaking for at least 48 h, recrystallised overnight by evaporation at 37 C and ground (mortar and pestle) to a homogeneous powder. The powder was packed manually into 1 cm of a 1.4-cm long Silastic tube (i.d. 1.47 mm, o.d. 1.95 mm; Dow-Corning, Sydney, NSW, Australia) and then sealed at both ends with Silastic adhesive, as described for 1 cm T and DHT implants (Singh et al. 1995). Implants were rinsed in ethanol and presoaked in saline for 2 h before implantation to reduce initial burst release. ARKO mice were treated with two or three 1 cm E2 implants at a 1 : 1000 dilution. After 6 weeks of E2 treatment (at 9 weeks of age), experiments were terminated by cardiac exsanguination of mice under ketamine + xylazine anaesthesia (blood collection), followed by whole-body perfusion via the left ventricle with warm (37 C) heparinsaline solution and then 30 mL Bouins fixative. Dissected fixed tissues were weighed and serum was collected and stored at 20 C until assayed. Stereology Fixed testes were embedded in methacrylate resin using the Technovit 7100 embedding kit (Heraeus, Wehrheim, Germany) before sections were cut using a Reichert-Jung microtome (Polycut S; Reichert-Jung, Heidelberg, Germany). Thin (7 m) sections stained with 0.5%Toluidine blue were used for photomicrographs and macroscopic histology. Thick (25 m) sections were prepared for stereology from upper, lower and mid sections of the testis and were stained with periodic acid-Schiff (PAS). Sections were evaluated for numerical cell density of Sertoli and germ cells (classified according to Russell et al. 1990)

E2-initiated spermatogenesis via AR and FSH

Reproduction, Fertility and Development

863

using the optical disector technique and CAST software (Olympus, Copenhagen, Denmark) as described previously (Haywood et al. 2003; Spaliviero et al. 2004). Volume fraction (%) of testicular seminiferous tubules and interstitium was estimated in testis sections (representing three different areas) by counting the number of hits in a 144-point grid frame at 100 magnification, with frames (70 frames per mouse) assigned by uniform random sampling using CAST software. Mass (or volume, assuming a specific gravity of 1.0) of each compartment was calculated by multiplying volume fraction by testis mass. Mean tubular diameter per testis was calculated by measuring the diameter of 100 tubules in symmetrical cross-section. Mean tubular length was calculated by dividing total tubular mass by the average cross-sectional area defined by the mean tubular diameter. Hormone assays Mouse serum FSH and LH levels were measured by specific two-site immunofluorometric assays, each with <0.01% crossreactivity to mouse LH (Jimenez et al. 2005) or FSH (van Casteren et al. 2000). Serum and intratesticular T levels were determined after extraction by radioimmunoassay (cross-reactivity: 5-androstane-3-17-diol = 0.96%; 5-androstane-3-17-diol = 0.51%; androstanedione = 0.29%; androstenediol = 0.20%; dehydroisoandrostenne = 0.011%; cortisol, E2, androsterone, dihydroepiandrosterone (DHEA) sulphate <0.009%), as described previously (Allan et al. 2004). Serum and intratesticular E2 concentrations were determined by Delfia time-resolved fluoroimmunoassay (Perkin-Elmer, Foster City, CA, USA; cross-reactivity: oestrone, oestrone sulfate, oestriol (E3), 2-OH-E3, 16-OxoE2, 16OH-oestrone, E2-SO4 , E-glucuronide <1%; testosterone, progesterone, DHEA, cortisol <0.01%) according to the manufacturers instructions, and as described previously (Bianco et al. 2002), after sample extraction with 2.5 volumes of diethyl ether-ethyl acetate (1 : 1 v/v). All samples were determined in duplicate. Statistics Data are expressed as the mean s.e.m. and were evaluated by analysis of variance followed by suitable post hoc tests and linear contrasts, as required, using NCSS software (Statistical Analysis and Data Analysis Software, Kaysville, UT, USA). P < 0.05 was considered significant. Results Effects of E2 on hpg and ARKO testis weight The weight of the testes of untreated 9-week-old hpg males (3.7 0.3 mg; n = 11) was significantly less (P < 0.005) than that of age-matched ARKO testes (5.8 0.1 mg; n = 8) and both were significantly less than that of age-matched control testes (77 3 mg; n = 7). In hpg mice, E2 treatment had no significant effect on bodyweight (P = 0.20), but produced a progressive dose-dependent increase (r2 = 0.98; P < 0.001) in testis weight (n = 610 per group), reaching 35% of normal at the highest doses of E2 examined (two 1-cm, 1 : 1000 E2 implants), which was similar to the increased testis weight stimulated by a maximal T response (i.e. 1 cm T implant; Singh et al. 1995), as shown in Fig. 1. In contrast, E2 treatment (two or three 1-cm,

30 Testis weight (mg)

(a) hpg

(b) ARKO

30

20

20

10

10

00 T 0 0 0 00 00 00 /10 1/4 1/2 1/1 2 1

E2

0 0 00 00 T 1/1 1/1 3 2 E2

Fig. 1. Testis weights in (a) untreated (white; n = 11), oestradiol (E2)treated (black; n = 610 per group) or testosterone (T)-treated (grey; n = 15) hpg male mice and (b) untreated (white; n = 8), E2-treated (black; n = 68 per group) or T-treated (grey; n = 6) androgen receptor-knockout (ARKO) male mice, all at 9 weeks of age. E2-treated animals received between one and three 1-cm implants with E2 : cholesterol dilutions as indicated on the x-axis; T-treated animals received one 1-cm T implant, as described in the Materials and methods.

1 : 1000 E2 implants) had no significant effect on the testis size of ARKO male mice (Fig. 1; n = 68 per group). ARKO males were completely androgen insensitive, as indicated by no testicular response to T treatment (Fig. 1) and no effect of T or DHT on bodyweight or the weight of sex accessory glands compared with controls (data not shown). Effects of E2 on serum and intratesticular hormone levels In hpg male mice (n = 611 per group), E2 treatment produced a progressive dose-dependent increase in serum E2 levels (r2 = 0.96; Fig. 2), reaching 284 40 pmol L1 at the highest dose of E2 used. Serum E2 levels were positively correlated (r2 = 0.80; P < 0.001) with increased testis size. Serum FSH levels were also increased (r2 = 0.78) by E2 administration, reaching 40% of normal FSH levels in wild-type mice (13.7 0.7 ng mL1 ; Jimenez et al. 2005), and were also positively correlated with the increased hpg testicular weight (r2 = 0.81, P < 0.001). Serum LH levels were undetectable in hpg male mice, with or without E2 treatment (data not shown). Intratesticular E2 concentrations in treated hpg males was markedly reduced (approximately fivefold; P < 0.001) at all doses of E2 examined, despite the progressive increase in serum E2 levels (Fig. 2). Similarly, intratesticular T levels in treated hpg males were reduced (approximately fourfold; P < 0.001) at all doses of E2 (Fig. 2). In ARKO male mice (n = 8 per group), E2 treatment significantly decreased (P < 0.001) the elevated circulating gonadotrophin levels found in androgen-insensitive mice (Scott et al. 1992), with serum FSH levels reduced by 69%, but remaining nearly double (23.9 3.2 ng mL1 ) the level in normal male mice, and serum LH reduced by 95% (Fig. 3). In contrast with decreased intratesticular E2 levels in treated hpg males, the intratesticular E2 concentration in treated ARKO male mice was significantly elevated (1.6-fold; P < 0.05), as shown in Fig. 3.

864

Reproduction, Fertility and Development

P. Lim et al.

(a) 300 200 100 0 (c)

Serum E2 (pmol L 1)

(b) Serum FSH (ng mL

1)

Serum FSH (ng mL

4 2 0
Intratesticular E2 (pmol mg 1 testis)

60

0.4

*
40

0.3 0.2 0.1 0.0

*
20 0 (c) (d )

(d )

0.08

Intratesticular T (pmol mg 1 testis)

0.8

1)

Serum LH (ng mL

0.04

3 2 1

0.03 0.02 0.01

0.00

0 1/ 0 20 0 1/ 0 10 2 0 1/ 0 10 00

0 1/ 0 20 0 1/ 0 10 2 0 1/ 0 10 00

0.0

Fig. 2. Serum oestradiol (E2) and FSH levels (a, b) and intratesticular E2 and testosterone (T) concentrations (c, d) in untreated (white) and E2treated (black) hpg adult male mice (n = 611 per group). E2-treated animals received one or two 1-cm implants with E2 : cholesterol dilutions as indicated on the x-axis. The reference range for FSH in adult male mice is indicated (Jimenez et al. 2005).

*
0 _ E2 _ E2 0.00

Fig. 3. Serum FSH and LH levels (a, c) and intratesticular testosterone (T) and oestradiol (E2) concentrations (b, d) in untreated (white; n = 8) and E2-treated (black; n = 8; three 1 : 1000 E2 : cholesterol 1-cm implants) androgen receptor-knockout (ARKO) adult male mice. P < 0.05 compared with untreated mice.

In contrast, intratesticular T concentrations were reduced (1.8fold; P < 0.001) in E2-treated ARKO male mice (Fig. 3). Effects of E2 on spermatogenesis in hpg and ARKO males Histology The seminiferous tubules of 9-week-old untreated ARKO or hpg testes displayed no lumen or post-meiotic spermatids (Fig. 4). The sparse appearance of pachytene spermatocytes was the maximal level of germ cell development observed in untreated ARKO or hpg testes. Within the interstitial space, the Leydig cells exhibited a hypoplastic or hypertrophic/hyperplastic appearance in the hpg or ARKO testis, respectively. The larger ARKO testis (Fig. 1) was attributable to a greater interstitial space volume compared with hpg testis (267 68 v. 24 3 mm3 , respectively; P < 0.02), as well as a larger tubule volume (330 42 v. 181 10 mm3 , respectively; P < 0.05).After 6 weeks E2 treatment, testes from hpg male mice exhibited enlarged tubules with a lumen, basally located Sertoli cells and the presence of post-meiotic spermatids (Fig. 4). In contrast, the histological appearance of testes from E2-treated ARKO males remained equivalent to that of untreated ARKO testes (Fig. 4). Stereology Total Sertoli cell numbers were similar in 9-week-old untreated hpg (n = 13) and ARKO testes (n = 5), each approximately 40% of numbers in wild-type testes (n = 7; Fig. 5). Administration of E2 significantly increased the total number of Sertoli cells in hpg testes compared with untreated hpg testes (by 65%; n = 6; P < 0.001), reaching 68% of that in wild-type

controls. The effect of E2 on the total number of Sertoli cells in hpg testes was significantly greater than that seen after T treatment alone (29% increase; P < 0.001; Fig. 5). In contrast, Sertoli cell numbers in ARKO testes (n = 5) were not altered by E2 treatment (Fig. 5). Total spermatogonia and early spermatocyte (preleptotene zygotene) numbers were equivalent in untreated hpg and ARKO testes (approximately 12% and 7%, respectively, of wild-type testes). In hpg testes, E2 significantly increased (P < 0.001) the spermatogonia and early spermatocyte numbers by four- and 13-fold, respectively, reaching half or equivalent wild-type levels, respectively. In contrast, T treatment of hpg males had no significant effect on total spermatogonia numbers, consistent with previous findings (Haywood et al. 2003), and stimulated a smaller increase in early spermatocyte numbers (fivefold; P < 0.01). However, stereological analysis further revealed that E2 and T treatment each increased total pachytene spermatocyte, round and elongated spermatid numbers to equivalent levels in hpg testes (to approximately 30%, 29% and 20% of wild-type respectively), although the effects of E2 were more variable than those of T (Fig. 5). In E2-treated ARKO testes, there was no significant change in any germ cell population examined (Fig. 5) and no post-meiotic germ cells were observed (Fig. 4). Discussion The requirement of androgen and FSH for complete spermatogenic development is well established, whereas any direct physiological role of oestrogen remains unclear. Ebling et al. showed that E2 administration to hpg mice induced spermatogenesis and

Intratesticular E2 (pmol mg 1 testis)

0.4

0.04

1/ 40

1/ 40

Intratesticular T (pmol mg 1 testis)

1)

80

(a)

(b) 0.5

E2-initiated spermatogenesis via AR and FSH

Reproduction, Fertility and Development

865

(a)

(b)

IS

IS IS (c) (d ) IS

IS

IS

(e)

IS

Fig. 4. Histology of (a) untreated hpg testis, showing small tubules with no lumen, containing scattered Sertoli cells (arrows) and premeiotic germ cells, adjacent to areas of interstitial space (IS); (b) untreated androgen receptor-knockout (ARKO) testis, showing vastly increased areas of IS due to hypertrophy of Leydig cells compared with hpg and larger tubules with no lumen but epithelium containing more basal Sertoli cells (arrows) and premeiotic germ cells; (c) oestradiol (E2)-treated (two 1 : 1000 E2 : cholesterol 1-cm implants) hpg testis showing enlarged tubules with a lumen (L), basally located Sertoli cells and complete spermatogenetic development; (d) E2-treated (three 1 : 1000 E2 : cholesterol 1-cm implants) ARKO testis exhibiting no change relative to untreated ARKO testis; and (e) wild-type control testis showing larger tubules with distinct lumen containing basally located Sertoli cells and complete spermatogenesis. Sections (3 m) were stained with Toluidine blue. (Original magnification 40.)

testis development without any increase in serum or testicular T (Ebling et al. 2000; Baines et al. 2005). A key issue is whether this effect of E2 was exerted directly on the testis via mechanisms independent of androgen and/or indirectly via the unexpected increase in circulating FSH levels reported in E2-treated hpg mice (Ebling et al. 2000). We have determined the effects of E2 in two complementary mouse models, namely the hpg male mouse with complete postnatal functional androgen deficiency and the ARKO male mouse with complete functional AR deficiency combined with low serum T concentrations. Our detailed analysis of testicular hormone levels and cell populations in E2treated androgen-deficient hpg and ARKO male mice shows that E2-induced spermatogenic development resembles the typical

actions of FSH and androgen, with an absolute requirement for the presence of functional AR. Stimulation of spermatogenesis occurred with marked suppression of intratesticular E2 concentrations, which we predict supports a predominantly extratesticular mode of E2 action for this unusual spermatogenic response. In E2-treated hpg mice, increased testicular size and serum FSH levels (reaching 40% of normal) were both positively correlated with serum E2 levels. The E2-stimulated increase in FSH supports findings first described by Ebling et al. (2000). In contrast, intratesticular T levels were significantly reduced in E2-treated hpg mice. Previously, no change was reported for total testicular T content in E2-treated hpg compared with

866

Reproduction, Fertility and Development

P. Lim et al.

SC 4 3 4 2 1 0 14 12 4 2 0 PS 40 30 20 2 10 0 0 E2 Wt Total cells ( 106) testis

1

Sg 6 6 4 2 0 RS 30 20 10 0 E2 hpg T _ E2 Wt

Sc

ES

E2 hpg

E2 hpg

E2 Wt

ARKO

ARKO

ARKO

Fig. 5. Histograms showing absolute numbers of Sertoli cells (SC), spermatogonia (Sg), preleptotenezygotene spermatocytes (Sc), pachytene spermatocytes (PS), round (RS) and elongated spermatids (ES) determined by stereology in untreated (n = 13), testosterone (T)-treated (1-cm implant; n = 15) and oestradiol (E2)-treated (two 1 : 1000 E2 : cholesterol 1-cm implants; n = 6) hpg testes (grey) or untreated (n = 5) and E2-treated (three 1 : 1000 E2 : cholesterol 1-cm implants; n = 5) androgen receptor-knockout (ARKO) testes (black), and age-matched wild-type (Wt, n = 7) testes, as described in the Materials and methods.

untreated hpg mice (Ebling et al. 2000); however, the larger E2-treated hpg testes indicate that intratesticular T concentration declined at least fourfold (Ebling et al. 2000; Baines et al. 2005), consistent with the findings of the present study. Pharmacological oestrogen treatment of neonatal (Atanassova et al. 2005) or adult (DSouza et al. 2005) rats, or transgenic overexpression of aromatase in mice (Li et al. 2004), can decrease testicular androgen production and concentration due to pharmacological suppression via negative feedback of pituitary LH and testicular T production (ODonnell et al. 2001). Other evidence suggests that E2 may also directly inhibit steroidogenic enzymes in Leydig cells (Nozu et al. 1981; Abney 1999) and immature testes independently of the effects of LH (Delbes et al. 2006). Reduced circulating levels of FSH in these models of oestrogen excess contrast with the paradoxical rise of serum FSH in E2-treated hpg males were exhibiting physiological range serum E2 levels. However, elevated serum FSH has been reported in adult male mice (Dalterio et al. 1985) and rats (Atanassova et al. 1999) after neonatal diethylstilboestrol (DES) treatment, suggesting that regulation of circulating FSH is dependent on age and the dose of oestrogen (Atanassova et al. 2000; Tena-Sempere et al. 2000; Mikkila et al. 2006) and, in mice, is likely to be mediated via ER expressed in the hypothalamus and/or pituitary (ODonnell et al. 2001). It was reported recently that male hpg mice exhibit abundant ER expression in the pituitary gland (Baines et al. 2008) and E2 stimulation of pituitary FSH secretion may reflect the neonatalimmature state of the pituitary gonadotrophs in mature hpg mice.

In the present study, two observations support the possibility that E2 acts via an extratesticular pathway. First, despite E2 treatment increasing serum E2 levels, the intratesticular E2 concentration was markedly reduced in treated hpg males exhibiting qualitatively complete spermatogenic development. Conversely, in E2-treated ARKO males, intratesticular E2 levels were increased, yet there was no detectable germ cell response. This apparent inverse relationship between intratesticular E2 levels and the induction of spermatogenesis suggests that either E2 has no direct action in the testis or, alternatively, high testicular E2 levels may be inhibitory to germ cell development, in particular meiotic and post-meiotic development. Deleterious effects of supraphysiological doses of oestrogens (E2, DES or other weak xeno-oestrogenic chemicals) on rodent spermatogenesis involve primarily toxicological or pharmacological suppression via negative feedback of pituitary gonadotrophin and testicular T production (Abney 1999; ODonnell et al. 2001; Akingbemi et al. 2004; Delbes et al. 2006). In mice, expression of transgenic aromatase, which elevates serum E2, results in spermatogenic arrest at meiosis (Li et al. 2001), which can be explained by decreased serum FSH and T. During perinatal development, elevated gonocyte numbers in ER-null relative to normal mouse testes was proposed to show that oestrogens directly inhibit early germ cell growth (Delbes et al. 2004), but these effects appear limited because adult ER-null males exhibit normal spermatogenesis and fertility (Krege et al. 1998). Our findings do not rule out an inhibitory effect of E2 in the testis; however, the inverse association between intratesticular E2 concentrations and the

E2-initiated spermatogenesis via AR and FSH

Reproduction, Fertility and Development

867

induction of spermatogenesis does not support a direct role for E2 testicular activity stimulating germ cell development. The second observation supporting an extratesticular E2 pathway is that elevated serum FSH levels in E2-treated hpg males, in addition to being correlated with testis growth, were associated with increased total numbers of the known FSH-responsive Sertoli cell, spermatogonia and early spermatocyte populations in hpg testes (Allan et al. 2004). In another study, postnatal treatment of normal rats with oestradiol benzoate (EB) reduced Sertoli cell, spermatogonia and spermatocyte numbers, whereas combined postnatal EB + exogenous FSH treatment provided the expected FSH-induced Sertoli and spermatogonia proliferation and further elevated spermatocyte numbers (Kula et al. 2001). Reduction of T levels by EB treatment was proposed as one possible E2-driven mechanism to enhance the induction of spermatogenesis, although intratesticular T or E2 levels and longer-term consequences on adult rat spermatogenesis were not reported (Kula et al. 2001). Taken together, these findings indicate that E2 via, or combined with, increased FSH may stimulate early spermatogenesis in immature testes. However, the effects of FSH alone do not fully explain the meioticpost-meiotic spermatogenic development in E2-treated hpg males, which we reveal was comparable to the levels of post-meiotic development in T-treated hpg males. We have established previously that FSH without LH activity has limited effects on post-meiotic development in mice (Allan et al. 2004), whereas FSH has strong synergistic effects with T on meiotic post-meiotic development (Haywood et al. 2003). Although intratesticular T levels were reduced in E2-treated hpg males, sufficient T may remain for synergistic androgenFSH activity, because intratesticular T levels 1015% of normal can provide strong synergy with FSH (Haywood et al. 2003). Furthermore, the failure of E2 to induce spermatogenesis in ARKO testes, despite increasing E2 levels, demonstrates an absolute requirement for the presence of functional AR, definitively extending the finding that E2-induced spermatogenesis in hpg mice could be pharmacologically reduced by an anti-androgen (Baines et al. 2005), in which anti-androgens may act as partial agonists rather than pure antagonists (Nguyen et al. 2007), especially in the context of low endogenous androgen secretion in the hpg mouse. The presence of the AR is also required for any significant FSH response in the testis, because total Sertoli cell numbers in either untreated or E2-treated ARKO mice, despite highly elevated serum FSH levels, remained equivalent to those in untreated FSH/T-deficient hpg mice. Earlier work has shown that the E2-induced increase in hpg pituitary FSH content is prevented by pharmacological blockade with an anti-oestrogen (fulvestrant) but not an anti-androgen (bicalutamide; Baines et al. 2005); however, serum FSH levels were not reported in those studies, so that any direct effects of FSH on the testis could not be determined. We propose that the effects of E2 on hpg testes involve stimulation of pituitary FSH secretion, combined with sufficient intratesticular T levels to induce the spectrum of spermatogenic development associated with combined FSHT actions (Haywood et al. 2003). Whether other potential E2regulated mechanisms, such as the growth hormoneinsulin-like growth factor-I pathway (Leung et al. 2004), also contribute the E2-induced spermatogenic response remains to be determined.

Considering the requirement for AR, direct activation by E2 seems an unlikely explanation due to the markedly reduced E2 concentration in hpg testes, as well as serum E2 levels well below the concentrations that activate AR (Yeh et al. 1998). Using immunohistochemistry, we have found that the AR is expressed in Sertoli, Leydig and peritubular cells of untreated hpg testes (K. McTavish, D. J. Handelsman, C. M. Allan, unpubl. obs.), ruling out a requirement for initiation of AR expression. There is in vitro evidence for cross-talk between AR and ER, such as heterodimer formation between AR and ER (but not ER, which is present in Sertoli cells; Zhou et al. 2002) in yeast mammalian two-hybrid models (Panet-Raymond et al. 2000) and the potential for ternary ERARSrc complexes (Migliaccio et al. 2000) or competition for common co-activators (Heinlein and Chang 2002; Holter et al. 2002). In vitro studies also describe rapid membrane-initiated E2 actions (Vasudevan and Pfaff 2007) or other putative ER (Hammes and Levin 2007). However, the physiological relevance of such molecular interactions has yet to be established in the testis. Furthermore, there is currently little evidence for testicular expression of ER and ER to play any major role in stimulating postnatal spermatogenesis. In mouse testis, ER is predominantly restricted to interstitial Leydig cells (Jefferson et al. 2000; Zhou et al. 2002) and ER-null germ cells develop normally into fertile spermatozoa when transplanted into a germ cell-depleted but oestrogen-sensitive host testis (Mahato et al. 2001), indicating that germ cell ER is not required in adult testes. Similarly, although ER is consistently found in mouse Leydig, Sertoli and some early germ cell populations (Rosenfeld et al. 1998; Jefferson et al. 2000; Zhou et al. 2002), loss of ER has no effect on the completion of spermatogenetic development (Krege et al. 1998). Importantly, genetic loss of aromatase (or ER, but not ER) activity ultimately leads to late developing spermatogenic damage, attributable to tubular fluid back-pressure from defective epididymal fluid resorption (Hess et al. 2000). Therefore, any physiological role for E2 in postnatal spermatogenesis appears likely to involve extratesticular actions or the somatic, but not testicular, germ cells. In summary, E2 treatment of the hpg mouse increased serum FSH and FSH-responsive Sertoli and germ cell populations, and stimulated meiotic and post-meiotic development to levels equivalent to androgen-treated hpg testes. These actions of E2 occurred despite reduced intratesticular E2 and T concentrations. Conversely, E2 was not able to induce spermatogenesis in androgen-resistant ARKO mice, despite increased intratesticular E2 and high serum FSH levels. Thus, the findings from two distinct genetic models indicate that E2 induction of spermatogenesis in the mouse depends on an intact, functional AR. We propose that E2-stimulated spermatogenesis in hpg testes involves elevated FSH activity acting in synergy with low intratesticular T levels, and that the suppressed intratesticular E2 levels in hpg testes do not support a direct testicular oestrogen pathway stimulating spermatogenic development. Acknowledgements
The authors are grateful to Dr W. Alexander (Walter and Eliza Hall Institute, Melbourne, Vic., Australia) for kindly providing the CMV-Cre mice. This research was supported by funding from the National Health and Medical

868

Reproduction, Fertility and Development

P. Lim et al.

Research Council of Australia. The authors declare that there is no conflict of interest that would prejudice the impartiality of this scientific work.

References
Abney, T. O. (1999). The potential roles of estrogens in regulating Leydig cell development and function: a review. Steroids 64, 610617. doi:10.1016/S0039-128X(99)00041-0 Akingbemi, B. T., Sottas, C. M., Koulova, A. I., Klinefelter, G. R., and Hardy, M. P. (2004). Inhibition of testicular steroidogenesis by the xenoestrogen bisphenol A is associated with reduced pituitary luteinizing hormone secretion and decreased steroidogenic enzyme gene expression in rat Leydig cells. Endocrinology 145, 592603. doi:10.1210/ EN.2003-1174 Allan, C. M., Haywood, M., Swaraj, S., Spaliviero, J., Koch, A., et al. (2001). A novel transgenic model to characterize the specific effects of folliclestimulating hormone on gonadal physiology in the absence of luteinizing hormone actions. Endocrinology 142, 22132220. doi:10.1210/ EN.142.6.2213 Allan, C. M., Garcia, A., Spaliviero, J., Zhang, F. P., Jimenez, M., Huhtaniemi, I., and Handelsman, D. J. (2004). Complete Sertoli cell proliferation induced by follicle-stimulating hormone (FSH) independently of luteinizing hormone activity: evidence from genetic models of isolated FSH action. Endocrinology 145, 15871593. doi:10.1210/ EN.2003-1164 Atanassova, N., McKinnell, C., Walker, M., Turner, K. J., Fisher, J. S., Morley, M., Millar, M. R., Groome, N. P., and Sharpe, R. M. (1999). Permanent effects of neonatal estrogen exposure in rats on reproductive hormone levels, Sertoli cell number, and the efficiency of spermatogenesis in adulthood. Endocrinology 140, 53645373. doi:10.1210/ EN.140.11.5364 Atanassova, N., McKinnell, C., Turner, K. J., Walker, M., Fisher, J. S., Morley, M., Millar, M. R., Groome, N. P., and Sharpe, R. M. (2000). Comparative effects of neonatal exposure of male rats to potent and weak (environmental) estrogens on spermatogenesis at puberty and the relationship to adult testis size and fertility: evidence for stimulatory effects of low estrogen levels. Endocrinology 141, 38983907. doi:10.1210/EN.141.10.3898 Atanassova, N. N., Walker, M., McKinnell, C., Fisher, J. S., and Sharpe, R. M. (2005). Evidence that androgens and oestrogens, as well as folliclestimulating hormone, can alter Sertoli cell number in the neonatal rat. J. Endocrinol. 184, 107117. doi:10.1677/JOE.1.05884 Baines, H., Nwagwu, M. O., Furneaux, E. C., Stewart, J., Kerr, J. B., Mayhew, T. M., and Ebling, F. J. (2005). Estrogenic induction of spermatogenesis in the hypogonadal (hpg) mouse: role of androgens. Reproduction 130, 643654. doi:10.1530/REP.1.00693 Baines, H., Nwagwu, M. O., Hastie, G. R., Wiles, R. A., Mayhew, T. M., and Ebling, F. J. (2008). Effects of estradiol and FSH on maturation of the testis in the hypogonadal (hpg) mouse. Reprod. Biol. Endocrinol. 6, 4. doi:10.1186/1477-7827-6-4 Bianco, J. J., Handelsman, D. J., Pedersen, J. S., and Risbridger, G. P. (2002). Direct response of the murine prostate gland and seminal vesicles to estradiol. Endocrinology 143, 49224933. doi:10.1210/EN.2002220493 Bremner, W. J., Millar, M. R., Sharpe, R. M., and Saunders, P. T. (1994). Immunohistochemical localization of androgen receptors in the rat testis: evidence for stage-dependent expression and regulation by androgens. Endocrinology 135, 12271234. doi:10.1210/EN.135.3.1227 Cattanach, B. M., Iddon, C. A., Charlton, H. M., Chiappa, S. A., and Fink, G. (1977). Gonadotrophin-releasing hormone deficiency in a mutant mouse with hypogonadism. Nature 269, 338340. doi:10.1038/269338A0 Chang, C., Chen, Y. T., Yeh, S. D., Xu, Q., Wang, R. S., Guillou, F., Lardy, H., and Yeh, S. (2004). Infertility with defective spermatogenesis and hypotestosteronemia in male mice lacking the androgen

receptor in Sertoli cells. Proc. Natl Acad. Sci. USA 101, 68766881. doi:10.1073/PNAS.0307306101 Dalterio, S., Bartke, A., Steger, R., and Mayfield, D. (1985). Neonatal exposure to DES in BALB/c male mice: effects on pituitarygonadal function. Pharmacol. Biochem. Behav. 22, 10191024. doi:10.1016/ 0091-3057(85)90312-0 De Gendt, K., Swinnen, J. V Saunders, P. T., Schoonjans, L., Dewerchin, M., ., et al. (2004). A Sertoli cell-selective knockout of the androgen receptor causes spermatogenic arrest in meiosis. Proc. Natl Acad. Sci. USA 101, 13271332. doi:10.1073/PNAS.0308114100 Delbes, G., Levacher, C., Pairault, C., Racine, C., Duquenne, C., Krust, A., and Habert, R. (2004). Estrogen receptor beta-mediated inhibition of male germ cell line development in mice by endogenous estrogens during perinatal life. Endocrinology 145, 33953403. doi:10.1210/ EN.2003-1479 Delbes, G., Levacher, C., and Habert, R. (2006). Estrogen effects on fetal and neonatal testicular development. Reproduction 132, 527538. doi:10.1530/REP.1.01231 DSouza, R., Gill-Sharma, M. K., Pathak, S., Kedia, N., Kumar, R., and Balasinor, N. (2005). Effect of high intratesticular estrogen on the seminiferous epithelium in adult male rats. Mol. Cell. Endocrinol. 241, 4148. doi:10.1016/J.MCE.2005.04.011 Dupont, S., Krust, A., Gansmuller, A., Dierich, A., Chambon, P., and Mark, M. (2000). Effect of single and compound knockouts of estrogen receptors alpha (ERalpha) and beta (ERbeta) on mouse reproductive phenotypes. Development 127, 42774291. Ebling, F. J., Brooks, A. N., Cronin, A. S., Ford, H., and Kerr, J. B. (2000). Estrogenic induction of spermatogenesis in the hypogonadal mouse. Endocrinology 141, 28612869. doi:10.1210/EN.141.8.2861 Eddy, E. M., Washburn, T. F., Bunch, D. O., Goulding, E. H., Gladen, B. C., Lubahn, D. B., and Korach, K. S. (1996). Targeted disruption of the estrogen receptor gene in male mice causes alteration of spermatogenesis and infertility. Endocrinology 137, 47964805. doi:10.1210/EN.137.11.4796 Hammes, S. R., and Levin, E. R. (2007). Extranuclear steroid receptors: nature and actions. Endocr. Rev. 28, 726741. doi:10.1210/ ER.2007-0022 Haywood, M., Spaliviero, J., Jimenez, M., King, N. J. C., Handelsman, D. J., and Allan, C. M. (2003). Sertoli and germ cell development in hypogonadal (hpg) mice expressing transgenic follicle-stimulating hormone alone or in combination with testosterone. Endocrinology 144, 509517. doi:10.1210/EN.2002-220710 Heinlein, C. A., and Chang, C. (2002). Androgen receptor (AR) coregulators: an overview. Endocr. Rev. 23, 175200. doi:10.1210/ER.23.2.175 Hess, R. A., Bunick, D., Lubahn, D. B., Zhou, Q., and Bouma, J. (2000). Morphologic changes in efferent ductules and epididymis in estrogen receptor-alpha knockout mice. J. Androl. 21, 107121. Holdcraft, R. W., and Braun, R. E. (2004). Androgen receptor function is required in Sertoli cells for the terminal differentiation of haploid spermatids. Development 131, 459467. doi:10.1242/DEV .00957 Holter, E., Kotaja, N., Makela, S., Strauss, L., Kietz, S., Janne, O. A., Gustafsson, J. A., Palvimo, J. J., and Treuter, E. (2002). Inhibition of androgen receptor (AR) function by the reproductive orphan nuclear receptor DAX-1. Mol. Endocrinol. 16, 515528. doi:10.1210/ ME.16.3.515 Jefferson, W. N., Couse, J. F., Banks, E. P., Korach, K. S., and Newbold, R. R. (2000). Expression of estrogen receptor beta is developmentally regulated in reproductive tissues of male and female mice. Biol. Reprod. 62, 310317. doi:10.1095/BIOLREPROD62.2.310 Jimenez, M., Spaliviero, J. A., Grootenhuis, A. J., Verhagen, J., Allan, C. M., and Handelsman, D. J. (2005). Validation of an ultrasensitive and specific immunofluorometric assay for mouse follicle-stimulating hormone. Biol. Reprod. 72, 7885. doi:10.1095/BIOLREPROD.104. 033654

E2-initiated spermatogenesis via AR and FSH

Reproduction, Fertility and Development

869

Johnston, D. S., Russell, L. D., Friel, P. J., and Griswold, M. D. (2001). Murine germ cells do not require functional androgen receptors to complete spermatogenesis following spermatogonial stem cell transplantation. Endocrinology 142, 24052408. doi:10.1210/EN.142. 6.2405 Krege, J. H., Hodgin, J. B., Couse, J. F., Enmark, E., Warner, M., Mahler, J. F., Sar, M., Korach, K. S., Gustafsson, J. A., and Smithies, O. (1998). Generation and reproductive phenotypes of mice lacking estrogen receptor beta. Proc. Natl Acad. Sci. USA 95, 15 67715 682. doi:10.1073/PNAS.95.26.15677 Kula, K., Walczak-Jedrzejowska, R., Slowikowska-Hilczer, J., and Oszukowska, E. (2001). Estradiol enhances the stimulatory effect of FSH on testicular maturation and contributes to precocious initiation of spermatogenesis. Mol. Cell. Endocrinol. 178, 8997. doi:10.1016/S03037207(01)00415-4 Leung, K. C., Johannsson, G., Leong, G. M., and Ho, K. K. (2004). Estrogen regulation of growth hormone action. Endocr. Rev. 25, 693721. doi:10.1210/ER.2003-0035 Li, X., Nokkala, E., Yan, W., Streng, T., Saarinen, N., Warri, A., Huhtaniemi, I., Santti, R., Makela, S., and Poutanen, M. (2001). Altered structure and function of reproductive organs in transgenic male mice overexpressing human aromatase. Endocrinology 142, 24352442. doi:10.1210/EN.142.6.2435 Li, X., Strauss, L., Makela, S., Streng, T., Huhtaniemi, I., Santti, R., and Poutanen, M. (2004). Multiple structural and functional abnormalities in the p450 aromatase expressing transgenic male mice are ameliorated by a p450 aromatase inhibitor. Am. J. Pathol. 164, 10391048. Lyon, M. F., Glenister, P. H., and Lamoreux, M. L. (1975). Normal spermatozoa from androgen-resistant germ cells of chimaeric mice and the role of androgen in spermatogenesis. Nature 258, 620622. doi:10.1038/258620A0 Mahato, D., Goulding, E. H., Korach, K. S., and Eddy, E. M. (2001). Estrogen receptor-alpha is required by the supporting somatic cells for spermatogenesis. Mol. Cell. Endocrinol. 178, 5763. doi:10.1016/S03037207(01)00410-5 Migliaccio, A., Castoria, G., Di Domenico, M., de Falco, A., Bilancio, A., et al. (2000). Steroid-induced androgen receptoroestradiol receptor betaSrc complex triggers prostate cancer cell proliferation. EMBO J. 19, 54065417. doi:10.1093/EMBOJ/19.20.5406 Mikkila,T. F.,Toppari, J., and Paranko, J. (2006). Effects of neonatal exposure to 4-tert-octylphenol, diethylstilbestrol, and flutamide on steroidogenesis in infantile rat testis. Toxicol. Sci. 91, 456466. doi:10.1093/ TOXSCI/KFJ156 Murphy, L., and OShaughnessy, P. J. (1991).Testicular steroidogenesis in the testicular feminized (Tfm) mouse: loss of 17 alpha-hydroxylase activity. J. Endocrinol. 131, 443449. Nguyen, T. V Yao, M., and Pike, C. J. (2007). Flutamide and ., cyproterone acetate exert agonist effects: induction of androgen receptor-dependent neuroprotection. Endocrinology 148, 29362943. doi:10.1210/EN.2006-1469 Notini,A. J., Davey, R.A., McManus, J. F., Bate, K. L., and Zajac, J. D. (2005). Genomic actions of the androgen receptor are required for normal male sexual differentiation in a mouse model. J. Mol. Endocrinol. 35, 547555. doi:10.1677/JME.1.01884 Nozu, K., Dufau, M. L., and Catt, K. J. (1981). Estradiol receptor-mediated regulation of steroidogenesis in gonadotropin-desensitized Leydig cells. J. Biol. Chem. 256, 19151922. ODonnell, L., Robertson, K. M., Jones, M. E., and Simpson, E. R. (2001). Estrogen and spermatogenesis. Endocr. Rev. 22, 289318. doi:10.1210/ER.22.3.289 Panet-Raymond, V Gottlieb, B., Beitel, L. K., Pinsky, L., and Trifiro, M. A. ., (2000). Interactions between androgen and estrogen receptors and the effects on their transactivational properties. Mol. Cell. Endocrinol. 167, 139150. doi:10.1016/S0303-7207(00)00279-3

Quigley, C. A., De Bellis, A., Marschke, K. B., el-Awady, M. K., Wilson, E. M., and French, F. S. (1995). Androgen receptor defects: historical, clinical, and molecular perspectives. Endocr. Rev. 16, 271321. doi:10.1210/ER.16.3.271 Raivio, T., Wikstrom, A. M., and Dunkel, L. (2007). Treatment of gonadotropin-deficient boys with recombinant human FSH: longterm observation and outcome. Eur. J. Endocrinol. 156, 105111. doi:10.1530/EJE.1.02315 Robertson, K. M., ODonnell, L., Jones, M. E., Meachem, S. J., Boon, W. C., Fisher, C. R., Graves, K. H., McLachlan, R. I., and Simpson, E. R. (1999). Impairment of spermatogenesis in mice lacking a functional aromatase (cyp 19) gene. Proc. Natl Acad. Sci. USA 96, 79867991. doi:10.1073/PNAS.96.14.7986 Rosenfeld, C. S., Ganjam, V K., Taylor, J. A., Yuan, X., Stiehr, J. R., . Hardy, M. P., and Lubahn, D. B. (1998). Transcription and translation of estrogen receptor-beta in the male reproductive tract of estrogen receptoralpha knock-out and wild-type mice. Endocrinology 139, 29822987. doi:10.1210/EN.139.6.2982 Russell, L. D., Ettlin, R. A., Sinha Hikim, A. P., and Clegg, E. D. (1990). In Histological and Histopathological Evaluation of the Testis. pp. 119 161. (Cache River Press: Clearwater.) Safe, S. (2005). Clinical correlates of environmental endocrine disruptors. Trends Endocrinol. Metab. 16, 139144. doi:10.1016/J.TEM.2005. 03.004 Saunders, P. T., Fisher, J. S., Sharpe, R. M., and Millar, M. R. (1998). Expression of oestrogen receptor beta (ER beta) occurs in multiple cell types, including some germ cells, in the rat testis. J. Endocrinol. 156, R13R17. doi:10.1677/JOE.0.156R013 Schaison, G., Young, J., Pholsena, M., Nahoul, K., and Couzinet, B. (1993). Failure of combined follicle-stimulating hormonetestosterone administration to initiate and/or maintain spermatogenesis in men with hypogonadotropic hypogonadism. J. Clin. Endocrinol. Metab. 77, 15451549. doi:10.1210/JC.77.6.1545 Schwenk, F., Baron, U., and Rajewsky, K. (1995). A cre-transgenic mouse strain for the ubiquitous deletion of loxP-flanked gene segments including deletion in germ cells. Nucleic Acids Res. 23, 50805081. doi:10.1093/NAR/23.24.5080 Scott, I. S., Bennett, M. K., Porter-Goff, A. E., Harrison, C. J., Cox, B. S., et al. (1992). Effects of the gonadotrophin-releasing hormone agonist Zoladex upon pituitary and gonadal function in hypogonadal (hpg) male mice: a comparison with normal male and testicular feminized (tfm) mice. J. Mol. Endocrinol. 8, 249258. Simoni, M., Gromoll, J., and Nieschlag, E. (1997). The folliclestimulating hormone receptor: biochemistry, molecular biology, physiology, and pathophysiology. Endocr. Rev. 18, 739773. doi:10.1210/ ER.18.6.739 Singh, J., ONeill, C., and Handelsman, D. J. (1995). Induction of spermatogenesis by androgens in gonadotropin-deficient (hpg) mice. Endocrinology 136, 53115321. doi:10.1210/EN.136.12.5311 Spaliviero, J. A., Jimenez, M., Allan, C. M., and Handelsman, D. J. (2004). Luteinizing hormone receptor mediated effects on initiation of spermatogenesis in gonadotropin-deficient mice are replicated by testosterone. Biol. Reprod. 70, 3238. doi:10.1095/BIOLREPROD.103. 019398 Tena-Sempere, M., Navarro, J., Pinilla, L., Gonzalez, L. C., Huhtaniemi, I., andAguilar, E. (2000). Neonatal exposure to estrogen differentially alters estrogen receptor alpha and beta mRNA expression in rat testis during postnatal development. J. Endocrinol. 165, 345357. doi:10.1677/ JOE.0.1650345 van Casteren, J. I., Schoonen, W. G., and Kloosterboer, H. J. (2000). Development of time-resolved immunofluorometric assays for rat follicle-stimulating hormone and luteinizing hormone and application on sera of cycling rats. Biol. Reprod. 62, 886894. doi:10.1095/ BIOLREPROD62.4.886

870

Reproduction, Fertility and Development

P. Lim et al.

Vanha-Perttula, T., Bardin, C. W., Allison, J. E., Gunbreck, L. G., and Stanley, A. J. (1970). Testicular feminization in the rat: Morphology of the testis. Endocrinology 87, 611619. Vasudevan, N., and Pfaff, D. W. (2007). Membrane-initiated actions of estrogens in neuroendocrinology: emerging principles. Endocr. Rev. 28, 119. doi:10.1210/ER.2005-0021 Yeh, S., Miyamoto, H., Shima, H., and Chang, C. (1998). From estrogen to androgen receptor: a new pathway for sex hormones in prostate. Proc. Natl Acad. Sci. USA 95, 55275532. doi:10.1073/PNAS.95.10.5527 Yeh, S., Tsai, M. Y., Xu, Q., Mu, X. M., Lardy, H., et al. (2002). Generation and characterization of androgen receptor knockout (ARKO)

mice: an in vivo model for the study of androgen functions in selective tissues. Proc. Natl Acad. Sci. USA 99, 13 49813 503. doi:10.1073/PNAS. 212474399 Zhou, Q., Nie, R., Prins, G. S., Saunders, P. T., Katzenellenbogen, B. S., and Hess, R. A. (2002). Localization of androgen and estrogen receptors in adult male mouse reproductive tract. J. Androl. 23, 870881.

Manuscript received 24 June 2008, accepted 25 July 2008

http://www.publish.csiro.au/journals/rfd