Elevated Steroidogenesis, Defective Reproductive Organs, and ...

83 downloads 59 Views 889KB Size Report
sitization of steroidogenesis after prolonged exposure to high levels of ...... Ding YQ, Huhtaniemi I 1989 Human serum LH inhibitor(s): behaviour and contribution ...
0013-7227/03/$15.00/0 Printed in U.S.A.

Endocrinology 144(11):4980 – 4990 Copyright © 2003 by The Endocrine Society doi: 10.1210/en.2003-0403

Elevated Steroidogenesis, Defective Reproductive Organs, and Infertility in Transgenic Male Mice Overexpressing Human Chorionic Gonadotropin ¨ KELA ¨ , JORMA TOPPARI, MATTI POUTANEN, SUSANA B. RULLI, PETTERI AHTIAINEN, SARI MA AND ILPO HUHTANIEMI Departments of Physiology (S.B.R., P.A., J.T., M.P., I.H.) and Anatomy (S.M.), Institute of Biomedicine, University of Turku, FIN-20520 Turku, Finland We previously developed a transgenic (TG) mouse model that overexpresses the human chorionic gonadotropin (hCG) ␤-subunit under the universal human ubiquitin C promoter, displaying in males a modest 3-fold increase in circulating levels of LH/hCG bioactivity. The males were fertile and presented with a mild reproductive phenotype. To achieve higher levels of hCG, a double TG model was generated by crossbreeding the hCG␤-expressing mice with another TG line harboring a ubiquitin C/common ␣-subunit fusion gene. The double-TG mice expressed excessive levels of dimeric hCG, with 2000-fold elevated circulating LH/hCG bioactivity. These male mice were infertile, primarily due to inability to copulate, and they showed enhanced testicular androgen production despite clear down-regulation of LH/hCG receptors. Their intratesticular inhibin B was unaltered, but serum FSH was markedly reduced. Apparently the chronic hCG hyperstimulation led to focal Leydig cell proliferation/hypertrophy at 6

H

UMAN CHORIONIC gonadotropin (hCG) is a glycoprotein hormone normally secreted by the placenta. It is a heterodimeric protein formed through noncovalent association between the common ␣- and the hormonespecific ␤-subunit (1). Dimerization is obligatory for hormonal activity, as the individual subunits of hCG are devoid of bioactivity (2). Being structurally and biologically closely related to the pituitary-derived LH, hCG binds to the same receptor and acts as a potent LH agonist. The LH/hCG receptor belongs to the seven-transmembrane domain, G protein-coupled receptor family (3). Activation of the receptor by LH/hCG leads to activation of Gs, the G protein coupled to adenylyl cyclase, and to an increase in cAMP. In vitro studies demonstrate that high concentrations of LH/hCG can also activate the inositol phosphate and MAPK signaling cascades (4). In males, LH plays a role in both normal and abnormal reproductive function by modulating testicular Leydig cell differentiation and steroidogenesis. Testosterone secreted by Leydig cells, in turn, promotes male sexual differentiation, pubertal androgenization, and fertility. In the testis, functional LH receptors are expressed in Leydig cells during fetal development, transiently in early postnatal life, and from Abbreviations: hCG, Human chorionic gonadotropin; IFMA, immunofluorometric assay; PMSG, pregnant mare’s serum gonadotropin; TG, transgenic; wt, wild-type.

months of age, but failed to promote testicular tumors. Even though full spermatogenesis occurred in most of the seminiferous tubules, progressive tubule degeneration was apparent as the males grew older. The prostate and seminal vesicles were enlarged by distension of glandular lumina. Functional urethral obstruction was indicated by distension and sperm accumulation in distal vas deferens as well as by dilated urinary bladder and enlarged kidneys. The abnormal function of accessory sex glands and/or lower urinary tract as a consequence of the disturbed sex hormone balance or direct action of hCG may be the main cause of infertility in this model. The present study provides in vivo evidence that exposure of male mice to chronically elevated levels of hCG severely affects their urogenital tract function at multiple sites and causes infertility, but, unlike in LH/hCG overexpressing female mice, it is not tumorigenic. (Endocrinology 144: 4980 – 4990, 2003)

puberty to adult life (5). Although primarily expressed in gonads, LH receptors are also found in several extragonadal sex organs, including the prostate (6, 7), epididymis (8), and seminal vesicles (9), but the physiological significance of these extratesticular receptors remains unclear. In man, naturally occurring mutations of the LH/hCG receptor have been associated with disrupted reproductive function (10). Inactivating mutations of the receptor cause pseudohermaphroditism associated with different degrees of Leydig cell hypoplasia, whereas constitutively activating mutations induce gonadotropin-independent, male-limited precocious puberty with autonomous hypersecretion of testosterone (10). In addition, Leydig cell adenomas are associated with a specific activating mutation of the LH receptor (Asp578His) (11, 12). This finding emphasizes the potential role of gonadotropins as tumor promoters, as has also been proposed by recent clinical and experimental findings (10) and in particular in female transgenic (TG) mice overexpressing LH or hCG (13, 14). Genetically modified animal models are of great importance in corroborating the findings on humans related to disrupted or enhanced gonadotropin function. In fact, the abolition of LH action in mice by targeted disruption of the LH receptor did not affect normal masculinization at birth, but resulted in postnatal blockade of testicular growth, arrest of external genital and accessory sex organ maturation, and azoospermia (15, 16). Animal models harboring activating

4980

Rulli et al. • hCG-Overexpressing Transgenic Mice

LH receptor or inactivating LH␤ mutations have not yet been reported. Many attempts have been made to elucidate the effects of elevated chronic LH/hCG stimulation on testicular function in vivo. A TG mouse model overexpressing LH has been produced (13), but only female mice produced elevated levels of LH. Nothing is known at the moment about the phenotypic consequences of LH overproduction in males. Findings on long- and short-term hCG treatments in rats are inconsistent and seem to be age, dose, and time dependent (17–23), and the formation of antibodies against the foreign protein poses a confounding factor. In fetal and neonatal rat testes, the regulation of LH/hCG receptor expression and function appears to differ from that in the adult. Unlike those in adults, fetal and neonatal testes are refractory to gonadotropin-induced receptor down-regulation, and treatment with LH/hCG causes marked up-regulation of the cognate receptors and enhancement of steroidogenesis (20, 24, 25). This finding is intriguing because normally the male is effectively protected from high LH/hCG action through down-regulation of testicular LH/hCG receptors and desensitization of steroidogenesis after prolonged exposure to high levels of LH/hCG. To study the effects of chronically elevated levels of LH/ hCG on male reproductive functions, we generated and characterized two TG mouse models expressing either the hCG ␤-subunit alone or together with the common ␣-subunit, both under the same ubiquitin C promoter. TG expression of hCG␤ produced moderate elevation of circulating LH/hCG bioactivity, whereas the double-TG hCG␣␤ mice presented with pharmacological levels of the hormone. We present in this paper the phenotypic characterization of males of the two TG mouse models. Materials and Methods Transgene construct The 2.4-kb hCG␣ minigene was released from the original plasmid pIB␣ [provided by Dr. I. Boime (26)] and subcloned into the BamHI site of pGEM-4Z vector (Promega, Madison, WI). Thereafter, the 1.2-kb BglII-BamHI human ubiquitin C promoter fragment was inserted upstream of the hCG␣ minigene. The 3.6-kb ubiquitin C/hCG␣ fragment was released from the backbone vector by digestion with EcoRI and SphI. The fragments were resolved in a 1% agarose gel and were isolated by electroelution, followed by purification with Elutip-D columns (Schleicher & Schuell, Keene, NH). Finally, the fragments were diluted in TE buffer (10 mm Tris-HCl and 5 mm EDTA, pH 7.5) at a concentration of 2 ng/␮l for microinjection. The 2.8-kb ubiquitin C/hCG␤ transgene was constructed and used for TG mouse generation as described previously (14).

Generation of TG mice The hCG␣⫹ TG mice carrying the ubiquitin C/hCG␣ transgene were generated by microinjecting the transgene into pronuclei of fertilized oocytes from FVB/N mice, and the microinjected embryos were implanted into oviducts of pseudopregnant female mice of the NMRI strain. The generation of ubiquitin C/hCG␤ TG mice has been described previously (14). PCR analyses of genomic DNA from tail biopsies were used to identify the TG animals. One microgram of genomic DNA was added to the 50-␮l PCR containing 10 mm Tris-HCl (pH 8.8), 1.5 mm MgCl2, 50 mm KCl, 200 ␮m deoxynucleotide triphosphate mix, 0.2 ␮m primers, and 2.5 U DNA polymerase. A 600-bp DNA fragment of the ubiquitin/hCG␣ transgene was amplified using primers specific for the ubiquitin promoter (5⬘-CGCGCCCTCGTCGTGTC-3⬘) and the hCG␣

Endocrinology, November 2003, 144(11):4980 – 4990 4981

minigene (5⬘-CCGGCTGGGAGAAGAATGG-3⬘). An 830-bp DNA fragment of the ubiquitin/hCG␤ transgene was obtained by using specific primers for the ubiquitin C promoter (see above) and the hCG␤-cDNA (5⬘-AAGCGGGGGTCATCACAGGTC-3⬘). Genotyping of the double-TG mice (see below) was performed by separate PCRs for the two transgenes. The DNA was denatured at 94 C for 4 min, followed by PCR: 94 C for 0.5 min, 57 C for 1 min, and 72 C for 1.5 min for 32 cycles. The resulting PCR products were analyzed by electrophoresis on 2% agarose gel, and the fragments were UV-visualized with ethidium bromide. The animals were housed in a specific pathogen-free environment under controlled conditions of temperature and light, and they were provided with tap water and commercial mouse chow ad libitum. All mice were produced and handled in accordance with the institutional animal care policies of the University of Turku.

Fertility studies Wild-type (wt), hCG␣⫹, hCG␤⫹, or hCG␣␤⫹ male mice were housed individually at 6 – 8 wk of age with randomly cycling wt female mice for at least 2 months for fertility and fecundity tests (n ⫽ 8 –10/group). The number of offspring born and frequency of births were recorded. Shortterm mating studies were conducted to determine the relative reproductive performance of the males. Six-week-old wt females were superovulated by sc administration of 7.5 IU pregnant mare’s serum gonadotropin (PMSG), followed by 5 IU hCG 47 h later, and immediately housed individually with wt, hCG␤⫹, or hCG␣␤⫹ male mice. Vaginal plugs were monitored on the following morning to confirm mating.

Measurement of hormone levels Six-month-old male mice were killed by cervical dislocation in the morning, blood was collected from the heart, and serum samples were separated by centrifugation and stored at ⫺20 C until hormone measurements. Serum hCG␣ was measured by an immunofluorometric assay technique (IFMA; Delfia, Wallac Oy, Turku, Finland), adapted to detect specifically the hCG␣-subunit in mouse serum samples. Briefly, microtitration wells coated with streptavidin and a biotinylated monoclonal antibody against the ␣-subunit were used in combination with a europium-labeled mouse monoclonal antibody against the ␣-subunit from the human FSH Delfia kit. The reference preparation from the NIDDK, NIH (Bethesda, MD), ␣hCG CR-119, was used as the standard. The detection limit was 50 ng/liter. Serum hCG␤ levels were measured by IFMA, adapted to specifically detect the hCG␤-subunit in mouse serum samples, as previously described (14). The reference preparation from NIH, hCG␤ CR-121, was used as the standard. The detection limit was 40 ␮g/liter. Serum levels of dimeric hCG were measured by IFMA using the Delfia hCG kit according to the manufacturer’s instructions; the detection limit was 0.5 IU/liter. FSH levels were measured by IFMA as described previously (27); the sensitivity of the assay was 50 ng/liter. Intratesticular testosterone and progesterone were determined by homogenizing one testis in 200 ␮l PBS. The homogenates or sera were extracted twice with 2 ml diethyl ether and evaporated to dryness. After reconstitution into PBS, testosterone and progesterone were measured by conventional RIAs. The bioactivity of circulating hCG was determined by the mouse interstitial cell in vitro bioassay (28). Testosterone production, determined by RIA, was used as an index of the hCG response. Recombinant hCG (specific activity, 14,800 IU/mg; Organon, Oss, The Netherlands) was used as the standard. The sensitivity of the bioassay was 0.5 IU/liter, and the intra- and interassay coefficients of variation were less than 5% and 10%, respectively. Intratesticular inhibin B was measured with a serum inhibin B kit (Oxford Bio-Innovation Ltd., Oxford, UK) for the human according to the manufacturer’s protocol. Before assay, frozen testis samples were weighed and homogenized in saline phosphate buffer. Samples were centrifuged at 13,000 rpm for 10 min, and supernatants were used for protein and inhibin B measurements.

Histological analysis Testes and epididymides were fixed overnight in Bouin’s reagent or 4% paraformaldehyde. Seminal vesicles, ventral prostates, and kidneys were fixed overnight in 4% paraformaldehyde. Tissues were dehydrated and embedded in paraffin and 5-␮m-thick sections were stained with hematoxylin and eosin.

4982

Endocrinology, November 2003, 144(11):4980 – 4990

Rulli et al. • hCG-Overexpressing Transgenic Mice

LH/hCG receptor binding assay Testicular LH/hCG receptor binding was measured as previously described (29). Highly purified hCG (NIH CR-125; 13,000 IU/mg) was radioiodinated using a solid phase lactoperoxidase method (29). Briefly, one testis was homogenized in 700 ␮l Dulbecco’s PBS and 0.1% BSA. Thereafter, 100-␮l aliquots of testicular homogenate were incubated in triplicate in the presence of saturating concentrations of [125I]iodo-hCG (150,000 cpm/tube; ⬃3 ng). Nonspecific binding was assessed in the presence of 50 IU unlabeled hCG (Pregnyl, Organon, Oss, The Netherlands). After overnight incubation at room temperature, the homogenates were washed with 4 ml ice-cold Dulbecco’s PBS and 0.1% BSA and centrifuged at 2000 ⫻ g for 20 min. The supernatants were aspirated, and pellets were counted in a ␥-spectrometer. Protein concentrations were measured using the Bradford method (30), and specific hCG binding was expressed as a percentage of the control, based on counts per minute per milligram of protein.

Statistical analysis SigmaStat for Windows 2.03 was used for t test or Student-NewmanKeuls test. Significance was set at P ⬍ 0.05. The values are presented as the mean ⫾ sem.

edly elevated (hCG␣␤⫹, 9970 ⫾ 1770 IU/liter; hCG␤⫹, ⬍0.5 IU/liter; wt, ⬍0.5 IU/liter), indicating that the TG hCG␣- and ␤-subunits were properly synthesized, dimerized, and secreted. In the hCG␤⫹ mice, in which hCG/LH bioactivity depends upon the association of hCG␤-subunit with endogenously expressed ␣-subunit present in pituitary gonadotrope and thyrotrope cells, the circulating levels of bioactive hCG/LH were 3- to 4-fold higher than those in wt males. In hCG␣␤⫹ male mice, levels of the bioactive form of hCG was more than 2000-fold the respective levels in wt mice (Fig. 1A). Fertility

Fertility studies were performed by mating wt, hCG␤⫹, or hCG␣␤⫹ male mice with randomly cycling wt females (Table 2). During the 2-month mating period, wt and hCG␤⫹ males sired an average of eight offspring per litter every 21–22 d,

Results Establishment of hCG␤⫹ and hCG␣␤⫹ mouse lines

Independent lines of hCG␣⫹ and hCG␤⫹ mice were established by mating founders with wt FVB/N counterparts. Double-TG hCG␣␤⫹ mice were obtained by cross-breeding hCG␣⫹ and hCG␤⫹ mice. To succeed in producing double-TG mice, the combination of four different hCG␣⫹ lines with one established hCG␤⫹ TG line was attempted (Table 1). Pregnancy and delivery of double TG offspring were successful only when hCG␣⫹ females of line 59 were used, coincident with the lowest levels of transgene expression. As the double TG males and females were infertile, constant cross-breeding of hCG␤⫹ with the hCG␣⫹ line mentioned was necessary to obtain hCG␣␤⫹ mice. Transgene expression and hCG bioactivity

Transgene expression under the ubiquitin C promoter was observed in a variety of tissues, including the pituitary gland, as previously demonstrated by RT-PCR, immunohistochemistry, and circulating peptide levels (14). The circulating hCG␤-subunit concentration was elevated in hCG␤⫹ males (hCG␤⫹, 5.7 ⫾ 1.3 mg/liter; wt, ⬍40 ␮g/liter). In double-TG hCG␣␤⫹ mice, circulating levels of dimeric hCG were markTABLE 1. Efficiency of cross-breeding between hCG␣⫹ and hCG␤⫹ mice to obtain double-TG mice Breeding with hCG␤⫹ malesa

hCG␣-subunit (mg/liter)b

Fertility ratec

Litters/ male

Offspring/ litter

wt hCG␣/59 hCG␣/80 hCG␣/20 hCG␣/82

0.7 7 11 136

4/4 4/4 2/5 3/4 0/5

10 6.25 1.2 1.5 0

8.0 ⫾ 0.4 6.5 ⫾ 0.6 6.5 ⫾ 0.7 3.5 ⫾ 0.9 0

a Six- to 8-wk-old hCG␤⫹ males were housed individually with two hCG␣⫹ females from the different lines for 4 months. Litters and offspring per litter were recorded. b Serum levels of hCG␣-subunit from founders of each independent line, as a reference value. c Number of males able to produce at least one litter/number of males studied was recorded. Individual males were mated with two females for 4 months.

FIG. 1. Serum levels of bioactive hCG/LH (A), serum concentrations of FSH (B), and testicular concentrations of inhibin B (C) in 6-monthold wt, hCG␤⫹, and hCG␣␤⫹ male mice are shown (n ⫽ 5–10/group). Different letters above the bars denote a statistically significant difference between the groups (P ⬍ 0.05).

Rulli et al. • hCG-Overexpressing Transgenic Mice

Endocrinology, November 2003, 144(11):4980 – 4990 4983

whereas hCG␣␤⫹ males sired no offspring. In a short-term fertility test in which wt female mice were PMSG/hCG hormone-primed to induce ovulation, hCG␤⫹ male mice mated normally (visible vaginal plugs). However, eight of eight double-TG hCG␣␤⫹ males failed to mate with superovulated wt females, confirming that these male mice were infertile due to abnormal mating behavior. On the other hand, aggressive behavior against the female was evidenced with these males, resulting in biting and severe injuries, even death, of their female partners when caged for breeding. The fertility and fecundity of both male and female hCG␣⫹ mice were normal when bred with wt counterparts (data not shown). Body and reproductive organ weights

At the age of 6 months, a slight increase in the body weight of hCG␤⫹ mice was detected, but no difference was found in hCG␣␤⫹ mice compared with wt animals (Table 3). The testicular weights of hCG␤⫹ and hCG␣␤⫹ mice were 75% and 50% of those of wt mice, respectively. Seminal vesicle weights (with coagulating glands) were 3-fold increased in the double-TG males, whereas epididymal weight was slightly decreased in hCG␤⫹ mice and was unaltered in hCG␣␤⫹ mice compared with wt mice. Serum and intratesticular hormone levels and testicular LH receptor binding

In 6-month-old adult hCG␤⫹ and hCG␣␤⫹ mice, the serum FSH levels were reduced to 40% and 3% of control levels, respectively (Fig. 1B). The intratesticular inhibin B contents were decreased by about 50% (P ⬍ 0.05; data not shown) in both adult hCG␤⫹ and hCG␣␤⫹ males, but no significant changes were observed when the inhibin B concentration were adjusted to testis weight (Fig. 1C). Serum progesterone was significantly increased in adult hCG␣␤⫹ mice compared with wt or hCG␤⫹ males (Table 3). As serum testosterone levels are normally highly fluctuating in mice (31), the intratesticular concentration of testosterone TABLE 2. Male reproductive performance of wt, hCG␤⫹, and hCG␣␤⫹ mice

wt (n ⫽ 10) hCG␤⫹ (n ⫽ 10) hCG␣␤⫹ (n ⫽ 8)

Mating ratea

Days between birthsb

Pups/litterb

9/10 9/10 0/8

21.4 ⫾ 0.4 22.7 ⫾ 1.5

8.4 ⫾ 0.5 8.1 ⫾ 0.4 0

a Number of males able to plug/number of males studied. Eightweek-old males were housed overnight with superovulated wt females, and vaginal plugs were monitored the next morning. b Eight-week-old males were mated with random cycling wt females for 2 months, and the number of pups and days between each birth were scored.

was determined as a more stable parameter to evaluate testicular hormone production. Both intratesticular testosterone and progesterone levels were significantly increased in adult hCG␣␤⫹ compared with wt or hCG␤⫹ mice (Fig. 2, A and B). Radioreceptor assay of hCG binding was performed to determine the free receptor concentration in the testes of adult hCG␤⫹ and hCG␣␤⫹ mice (Fig. 2C). hCG binding was reduced to 42% in hCG␤⫹ testes with respect to wt samples, and a more pronounced reduction to 1.4% occurred in the hCG␣␤⫹ testes, apparently due to receptor occupancy and down-regulation by the very high circulating hCG levels. Histology of the testes

Histological examination of the adult testes revealed progressive changes in the structure of the seminiferous epithelium according to age (Fig. 3). At the age of 2– 4 months, the testes of both hCG␤⫹ and hCG␣␤⫹ mice showed normal tubular structure and full spermatogenesis. At the age of 6 months, most seminiferous tubules were still normal, and spermatogenesis was complete in both TG lines (Fig. 3, C–E). At this age, incipient tubular degeneration was observed in hCG␣␤⫹ males, and large vacuoles were also present in the basal compartment of some seminiferous tubules (Fig. 3F). In addition, clear Leydig cell hypertrophy was observed in hCG␣␤⫹ mice, in particular under the capsule. At the age of 8 –9 months, the degeneration of seminiferous tubules had progressed, and a more severe damage was observed (Fig. 3, G and H). At this age, Leydig cell hypertrophy was also evident. No sign of Leydig cell tumor was observed in any of the animals studied. Morphology of the urinary tract and accessory sex organs

In the urinary tract of adult hCG␣␤⫹ male mice, there were clear signs of infravesical (urethral) obstruction. The kidneys were enlarged (Fig. 4, upper panel), and the renal pelvis, ureters, and urinary bladder were markedly distended (Fig. 5). In renal cortex, tubular lumina were distended, and cortical cells were enlarged (Fig. 4, lower panel). LH receptor expression was undetectable in kidneys from wt and hCG␣␤⫹ mice. As further signs of infravesical obstruction in hCG␣␤⫹ mice, the distal parts of vasa deferentia were distended and filled with sperm (Fig. 6, A and B). The epithelium in the posterior wall of the urethra was hyperplastic, and the cells were filled with vacuoles (Fig. 6C). However, open urethral lumen was visible, indicating that the obstruction was functional, rather than anatomical/mechanical. Upon macroscopic examination, the seminal vesicles and ventral prostate of the adult hCG␣␤⫹ mice appeared greatly enlarged compared with those in age-matched wt controls (Fig. 5). The glandular lumina of seminal vesicles were distended, the epithelium was flattened, mucosal folding was

TABLE 3. Body and organ weights, and serum testosterone and progesterone in adult wt, hCG␤⫹, and hCG␣␤⫹ male mice

wt hCG␤⫹ hCG␣␤⫹ a,b,c

Body weight (g)

Testis (mg)

Seminal vesicle and coagulating gland (mg)

Epididymis (mg)

Serum testosterone (nmol/liter)

Serum progesterone (nmol/liter)

29.8 ⫾ 0.6a 37.0 ⫾ 2.0b 33.2 ⫾ 1.6a,b

91.3 ⫾ 5.6a 67.9 ⫾ 2.2b 46.0 ⫾ 4.1c

410 ⫾ 46a 398 ⫾ 31a 1098 ⫾ 165b

38.6 ⫾ 1.6a 34.2 ⫾ 1.0b 41.9 ⫾ 2.3a

45.0 ⫾ 11.8a,b 23.9 ⫾ 6.5b 68.6 ⫾ 13.1a

2.2 ⫾ 0.4a 1.7 ⫾ 0.2a 5.7 ⫾ 1.9b

Different superscripts denote statistical differences, P ⬍ 0.05 (n ⫽ 6 –12).

4984

Endocrinology, November 2003, 144(11):4980 – 4990

Rulli et al. • hCG-Overexpressing Transgenic Mice

FIG. 3. Testicular histology of 6-month-old wt (A and B), hCG␤⫹ (C and D), hCG␣␤⫹ (E–H), and 8-month-old hCG␣␤⫹ (G and H) mice. The arrow in F indicates tubular vacuolization. The asterisks in F and H indicate degenerative seminiferous tubules. Bars, 50 ␮m.

cauda epididymis at 2– 6 months of age in hCG␣␤⫹ males (Fig. 8, E and G), but abnormal sperm structure associated with bent tails was observed at 8 –9 months (Fig. 8H). FIG. 2. Intratesticular concentrations of testosterone (A), intratesticular concentrations of progesterone (B), and specific [125I]iodo-hCG binding to testicular homogenates (C) in 6-month-old wt, hCG␤⫹, and hCG␣␤⫹ male mice (n ⫽ 5–7/group). The mean hCG binding measured (counts per minute per milligram of protein) in wt testes was assigned a value of 100%. Different letters above the bars denote a statistically significant difference between the groups (P ⬍ 0.05).

less prominent, and the fibromuscular layer was thickened (Fig. 6, D and E). Enlarged glands with flat single-layer epithelium were frequently seen in the ventral prostates of hCG␣␤⫹ mice (Fig. 7). In addition, abnormal epithelial structures were present in multiple foci. These included hyperplastic, multilayered epithelium with vacuolated cytoplasm and the presence of structures resembling mucus-filled goblet cells (Fig. 7, E–K). The histology of epididymides of hCG␣␤⫹ mice at 6 months did not reveal any changes compared with wt mice (Fig. 8, A and B). At the age of 8 –9 months, increased connective tissue and smooth muscle cells were observed in cauda epididymis, and these changes were most prominent around the tubules (Fig. 8, C and D). The tubule structure was totally abnormal, resembling that of sperm granulomas. In some cases increased smooth muscle and connective tissue were also observed in the distal part of corpus epididymides. Morphologically normal, motile sperm were present in

Discussion

We have recently reported the characterization of female TG mice overexpressing hCG, in which aberrant ovarian function leads to multiple reproductive and endocrine disturbances, including pituitary lactotrope adenomas and mammary gland tumors (14). In the present study we demonstrated that chronic exposure of male mice to elevated levels of hCG caused pathologies in the testes and urogenital organs. The transgenic hCG␤ protein was highly produced and secreted in males (on the order of milligrams per liter) as previously demonstrated in females. However, circulating bioactive hCG, which depends on the rate-limiting dimerization of hCG␤ with endogenous ␣-subunit, was only moderately elevated in males due to a strong down-regulation of ␣-subunit gene expression at the pituitary level (1). In accordance, hCG␤⫹ males presented with only a mild reproductive phenotype and were fertile. To achieve higher levels of hCG, double-TG hCG␣␤⫹ mice were generated by crossbreeding TG mice harboring the hCG␣-subunit with those overexpressing hCG␤. In this way excessive pharmacological levels of circulating bioactive hCG were achieved in the double-TG hCG␣␤⫹ mice, and consequently, these males were infertile and presented significant reproductive disturbances. Fertility tests demonstrated that the failure of hCG␣␤⫹

Rulli et al. • hCG-Overexpressing Transgenic Mice

Endocrinology, November 2003, 144(11):4980 – 4990 4985

FIG. 4. Histology of kidneys of 6-monthold wt (left panels) and hCG␣␤⫹ mice (middle and right panels). The upper panels show longitudinal sections through kidney (bar, 0.5 mm). The lower panels are from sections showing renal cortex (bar, 50 ␮m). In hCG␣␤ double-TG mice, the kidneys are enlarged and show signs of hydronephrosis. The renal pelvis is distended, and the cortical cells are enlarged.

FIG. 5. The abdominal cavity (upper panels), ventral view of the urinary bladder and urethroprostatic block (middle and lower panels), and ventral prostate (arrow), urinary bladder (asterisk), and seminal vesicles (arrowhead) of wt (left panels), hCG␤⫹ (middle panels), and hCG␣␤⫹ (right panels) mice at 6 months of age. The urinary bladder is distended, and the accessory sex glands (prostate and seminal vesicles) are enlarged.

males to produce progeny was primarily due to their inability to copulate. Motile and morphologically normal sperm were observed in cauda epididymis at the ages when fertility was tested, indicating that the problem is at the level of copulatory or ejaculatory function and not in sperm production or maturation at that age. No vaginal plugs were observed when hCG␣␤⫹ males were mated with PMSG-treated females. This could be due to either altered reproductive behavior or abnormal function or anatomy of the accessory sex glands or urethra. Locally produced estrogens from circulating testosterone mediate the activating effects of testosterone on male copulatory behavior and are crucial for fetal brain masculinization (32, 33). This could explain the behavioral defect, when altered sex steroid profile during specific periods of embryonic and neonatal life may have influenced

the sexual development of the brain and imprinted sexual behavior in adulthood. Additional studies should be carried out to investigate whether the reproductive deficit observed in our mice reflects primarily motivational or consummatory aspects of behavior. On the other hand, testosterone has been shown to increase aggressive behavior in both male and female mice (34), which could explain the aggressive behavior of the hCG␣␤⫹ males toward females. Finally, the possibility that an anatomical impediment resulted in painful coituses cannot be ruled out. FSH production was clearly reduced in adulthood in both TG models, apparently due to elevated androgen feedback regulation to the hypothalamic-pituitary level (1). The reason why FSH levels were also reduced in hCG␤⫹ mice despite the minimal changes found in testosterone secretion is not clear.

4986

Endocrinology, November 2003, 144(11):4980 – 4990

Rulli et al. • hCG-Overexpressing Transgenic Mice

FIG. 6. Histology of the accessory sex glands of 6-month-old hCG␣␤⫹ mice. A, Transverse section through the prostatic urethra (U), showing the posterior urethral wall and collecting ducts and distal parts of vas deferens (asterisk). The latter structures are distended and filled with sperm. The epithelium in the posterior wall of urethra is hyperplastic and filled with vacuoles (C). Bar, 0.5 mm. B, A higher power magnification of vas deferens (area marked with rectangle in A). Bar, 100 ␮m. C, A higher power magnification of the epithelium in the posterior wall of urethra (area marked with rectangle in A). Bar, 100 ␮m. D and E, Seminal vesicles of wt (D) and hCG␣␤⫹ (E) mice. D, Typical, highly folded, single layer columnar epithelium of a wt mouse. E, Epithelium is flattened, folding is less prominent, and stromal layer is thick in a hCG␣␤⫹ mouse. Bar, 100 ␮m.

The possibility that other androgen metabolites not measured in this study were influencing gonadotropin feedback regulation cannot be ruled out. FSH is known to play a major role in Sertoli cell proliferation in the maturing testis, which, in turn, influences Leydig cell function and vice versa (35). The reduced levels of FSH found in both hCG␤⫹ and hCG␣␤⫹ mice could be a contributing factor to the abnormal testis growth by reducing the finite number of Sertoli cells, a determinant of total tubular length. Inhibin B, in addition to being a negative feedback regulator of FSH secretion, is a

marker of spermatogenesis and Sertoli cell function (36). FSH is recognized to stimulate testicular inhibin secretion, but the possibility that Leydig cells contribute to the synthesis and secretion of inhibin B through LH/hCG action is less clear. mRNAs encoding inhibin ␣ and ␤B and the subunits themselves have been localized in Sertoli as well as Leydig cells (36, 37). Even though it was suggested that in male Gottingen miniature pigs, Leydig cells are the predominant source of inhibin B (38), studies carried out in normal or hypogonadal men indicate that LH or hCG failed to stimulate inhibin B

Rulli et al. • hCG-Overexpressing Transgenic Mice

FIG. 7. Histology of ventral prostate of 6-month-old wt and hCG␣␤⫹ mice. A and B, Ventral prostate of a wt mouse, showing glands lined with single layer cuboidal or columnar epithelium. Bars, 100 ␮m (A) and 50 ␮m (B). C–K, Ventral prostates of hCG␣␤⫹ mice. In most parts of the ventral lobe, the glandular lumina are enlarged (C), and the epithelium is very flat (D). In addition, highly altered glandular/epithelial structures are present in multiple foci (E–K). These include hyperplastic, multilayered epithelium with vacuolized cytoplasm (E and F) and the presence of structures resembling mucus-filled goblet cells (arrows in H, J, and K). Bars, 100 ␮m (C and E) and 50 ␮m (D and F–K).

Endocrinology, November 2003, 144(11):4980 – 4990 4987

4988

Endocrinology, November 2003, 144(11):4980 – 4990

Rulli et al. • hCG-Overexpressing Transgenic Mice

FIG. 8. The histology of cauda epididymis of wt (A) and hCG␣␤⫹ (B) at 6 months of age shows no significant changes. In contrast, at 8 months, granuloma-like structures, increased connective tissue, and lymphocyte and macrophage infiltration are seen around the dysmorphic epididymal tubules (C and D). Bar, 200 ␮m. Morphologies of spermatozoa taken from cauda epididymis of a 4-month-old wt mouse (E), a 4-month-old hCG␣␤⫹ mouse (F), a 6-month-old hCG␣␤⫹ mouse (G), and an 8-month-old hCG␣␤⫹ mouse (H) are shown. Spermatozoa in G and H were taken from the contralateral cauda epididymis of those shown in B and C, respectively. Arrows indicate hairpin flagella.

secretion (39, 40). No evidence was obtained for changes in testicular inhibin B concentration in our model. Whether this is the result of a compensatory effect between reduced FSHmediated inhibin B production from Sertoli cells and a putative increased hCG-mediated effect remains to be elucidated. Long-term treatment of prepubertal or adult rats with LH/hCG was shown to produce hyperplasia, hypertrophy, and an increase in the steroidogenic capacity of adult Leydig cells (17, 18, 21, 23). In agreement with these reports, one of the most conspicuous findings in the double-TG mice was Leydig cell hyperplasia/hypertrophy. These results support the concept that LH/hCG is critical for adult Leydig cells to induce proliferation, hypertrophy, and enhanced steroidogenesis. Although LH/hCG signaling plays a major role in Leydig cell proliferation, the prolonged exposure to hCG hyperstimulation failed to induce Leydig cell tumors in hCG␣␤⫹ mice. In humans, one specific somatic activating mutation of the LH receptor gene (Asp578His) is associated with Leydig cell adenomas (11, 12). Besides the classical cAMP-mediated signaling, this particular mutation activates

the phospholipase C pathway. Activation of alternative intracellular signaling pathways of LH receptor action was therefore proposed to be responsible for the neoplastic transformation (11). Whether LH/hCG signaling is altered upon chronic stimulation remains to be studied. The unaltered intratesticular inhibin concentrations in our model are consistent with the concept that missing inhibin action is related to tumor formation in gonads (41). Exposure to LH/hCG has a biphasic effect on the control of Leydig cell function. Low doses of the hormone maintain the receptors and steroidogenic enzymes, whereas higher doses cause homologous and heterologous (e.g. PRL receptor) receptor down-regulation and uncoupling of signal transduction (25). In the present study chronic hCG overexpression of hCG␣␤⫹ mice resulted in enhanced testicular steroidogenesis, as demonstrated by elevated intratesticular testosterone and progesterone. Significant LH receptor down-regulation in this system in the face of elevated testicular testosterone synthesis suggests that the small fraction of free LH/hCG receptors is sufficient for the steroidogenic capacity of the Leydig cells, in keeping with the spare re-

Rulli et al. • hCG-Overexpressing Transgenic Mice

ceptor concept (42). Activation of alternative signal transduction pathways apart from cAMP, such as phospholipase C and MAPKs, may occur with high doses of LH/hCG. Seminiferous tubules of the hCG-overexpressing testes first presented with full spermatogenesis and normal sperm quality, but progressive tubular degeneration occurred in hCG␣␤⫹ mice as the animals grew older. This event correlated with the appearance of epididymal structure abnormalities and altered sperm structure with bent tails. Foci of degenerative seminiferous tubules were characteristic of the hCG␣␤⫹ phenotype, but were absent in hCG␤⫹ mice. hCGinduced high testicular blood flow (43, 44), precapillary vasoconstriction (45), and inflammatory-type response in the intertubular tissue with localized degeneration of the seminiferous epithelium have been described (46) and would provide the likely explanation for the seminiferous tubular damage observed in our model. On the other hand, the infertility of estrogen receptor ␣-knockout males was caused by abnormal fluid reabsorption in the epididymis, leading to disrupted spermatogenesis and seminiferous tubule organization through a back-pressure effect (47, 48). Because the hCG␣␤⫹ males presented either obstruction or dysfunction at several levels of the urinary tract, the spermatogenic defect may be a consequence of these, also causing obstruction of the vas deferens. Progressive changes associated with obstruction were seen in the epididymis, resembling granulomas seen in rodents after traumatic injury (49). Obstruction at the level of epididymis has been shown to cause degeneration of seminiferous tubules and vacuolization of seminiferous tubules in rats (50, 51). The accessory sex organs (prostate and seminal vesicles) were enlarged mainly due to distension of the glandular lumina, indicating either enhanced production of secretory material or impaired emptying of the glands. The former response would be in line with the elevated androgen levels in ERKO mice (47, 48). The distension of vas deferens and accumulation of sperm in distal vas as well as the pronounced distension of urinary bladder and enlarged kidneys clearly indicate infravesical (urethral) obstruction of, to date, unknown cause. No obvious signs of solid tissues blocking the urethral lumen were observed, suggesting that the obstruction was functional, probably due to defects in smooth muscle function or premature coagulation of secretory fluids. The presence of goblet-like cells in the prostatic epithelium may suggest altered secretory function. There was no evidence for LH receptor expression in kidneys from either wt or hCG␣␤⫹ mice that could explain a direct effect of hCG on this organ. The response is thus more probably a consequence of the severe steroid hormone imbalance in these mice. Although there are receptors of LH/hCG in the epididymis, seminal vesicles, and prostate (6 –9), we still consider most of the changes in the accessory sex glands of the hCG␣␤⫹ males to be due to hyperandrogenism. In contrast to the epididymis, prostate and seminal vesicle growth respond strongly to androgen excess (52). In summary, the present study provides in vivo evidence that the exposure of male mice to chronically elevated levels of hCG is not sufficient to promote testicular tumor formation. However, male reproductive function was severely affected at multiple sites, resulting in infertility. Chronic hCG

Endocrinology, November 2003, 144(11):4980 – 4990 4989

hyperstimulation led to apparent Leydig cell proliferation and hypertrophy and enhanced testicular steroidogenesis, despite dramatic LH receptor down-regulation. Progressive seminiferous tubule degeneration occurred as mice grew older, and their accessory reproductive organs and kidneys were also severely affected. It is possible that the abnormal function and obstructions of accessory sex glands and/or lower urinary tract due to adverse back-pressure are the main cause of the infertility observed. Acknowledgments The skillful technical assistance of N. Messner, J. Vesa, and T. Laiho is gratefully acknowledged. Received March 31, 2003. Accepted July 24, 2003. Address all correspondence and requests for reprints to: Ilpo Huhtaniemi, M.D., Ph.D., Institute of Reproductive and Developmental Biology, Imperial College London, Faculty of Medicine, Du Cane Road, London W12 ONN, United Kingdom. E-mail: ilpo.huhtaniemi@ imperial.ac.uk. This work was supported by grants from the Sigrid Juse´ lius Foundation, the Finnish Cancer Societies, the Academy of Finland, the Turku Graduate School of Biomedical Sciences, and the Center for International Mobility.

References 1. Gharib SD, Wierman ME, Shupnik MA, Chin WW 1990 Molecular biology of the pituitary gonadotropins. Endocr Rev 11:177–199 2. Narayan P, Gray J, Puett D 2002 Yoked complexes of human choriogonadotropin and the lutropin receptor: evidence that monomeric individual subunits are inactive. Mol Endocrinol 16:2733–2745 3. Ascoli M, Fanelli F, Segaloff DL 2002 The lutropin/choriogonadotropin receptor, a 2002 perspective. Endocr Rev 23:141–174 4. Hirakawa T, Galet C, Ascoli M 2002 MA-10 cells transfected with the human lutropin/choriogonadotropin receptor (hLHR): a novel experimental paradigm to study the functional properties of the hLHR. Endocrinology 143: 1026 –1035 5. Saez JM 1994 Leydig cells: endocrine, paracrine, and autocrine regulation. Endocr Rev 15:574 – 626 6. Tao YX, Lei ZM, Woodworth SH, Rao ChV 1995 Novel expression of luteinizing hormone/chorionic gonadotropin receptor gene in rat prostate. Mol Cell Endocrinol 111:R9 –R12 7. Reiter E, McNamara M, Closset J, Hennen G 1995 Expression and functionality of luteinizing hormone/chorionic gonadotropin receptor in the rat prostate. Endocrinology 136:917–923 8. Zhou HM, Zhang T, Liu YX 1996 Rat epididymis expresses luteinizing hormone receptor (LHR). Chinese Sci Bull 41:1608 –1610 9. Tao YX, Lei ZM, Rao ChV 1998 Seminal vesicles are novel sites of luteinizing hormone/human chorionic gonadotropin receptor gene expression. J Androl 19:343–347 10. Themmen APN, Huhtaniemi IT 2000 Mutations of gonadotropins and gonadotropin receptors: elucidating the physiology and pathophysiology of pituitary-gonadal function. Endocr Rev 21:551–583 11. Liu G, Duranteau L, Carel JC, Monroe J, Doyle DA, Shenker A 1999 Leydigcell tumors caused by an activating mutation of the gene encoding the luteinizing hormone receptor. N Engl J Med 341:1731–1736 12. Richter-Unruh A, Wessels HT, Menken U, Bergmann M, SchmittmannOhters K, Schaper J, Tappeser S, Hauffa BP 2002 Male LH-independent sexual precocity in a 3.5-year-old boy caused by a somatic activating mutation of the LH receptor in a Leydig cell tumor. J Clin Endocrinol Metab 87:1052–1056 13. Risma KA, Clay CM, Nett TM, Wagner T, Yun J, Nilson JH 1995 Targeted overexpression of luteinizing hormone in transgenic mice leads to infertility, polycystic ovaries, and ovarian tumors. Proc Natl Acad Sci USA 92:1322–1326 14. Rulli SB, Kuorelahti A, Karaer O, Pelliniemi LJ, Poutanen M, Huhtaniemi I 2002 Reproductive disturbances, pituitary lactotrope adenomas, and mammary gland tumors in TG female mice producing high levels of human chorionic gonadotropin. Endocrinology 143:4084 – 4095 15. Zhang FP, Poutanen M, Wilbertz J, Huhtaniemi I 2001 Normal prenatal but arrested postnatal sexual development of luteinizing hormone receptor knockout (LuRKO) mice. Mol Endocrinol 15:172–183 16. Lei ZM, Mishra S, Zou W, Xu B, Foltz M, Li X, Rao CV 2001 Targeted disruption of luteinizing hormone/human chorionic gonadotropin receptor gene. Mol Endocrinol 15:184 –200 17. Christensen AK, Peacock KC 1980 Increase in Leydig cell number in testes of

4990

18. 19. 20.

21.

22. 23. 24.

25. 26. 27.

28. 29. 30. 31. 32. 33. 34. 35.

Endocrinology, November 2003, 144(11):4980 – 4990

adult rats treated chronically with an excess of human chorionic gonadotropin. Biol Reprod 22:383–391 Nussdorfer GG, Robba C, Mazzocchi G, Rebuffat P 1980 Effects of human chorionic gonadotropins on the interstitial cells of the rat testis: a morphometric and radioimmunological study. Int J Androl 3:319 –332 Risbridger GP, Robertson DM, de Kretser DM 1982 The effects of chronic human chorionic gonadotropin treatment on Leydig cell function. Endocrinology 110:138 –145 Huhtaniemi IT, Warren DW, Catt KJ 1985 Regulation of infant and developing rat testicular gonadotropin and prolactin receptors and steroidogenesis by treatments with human chorionic gonadotropin, gonadotropin-releasing hormone analogs, bromocriptine, prolactin, and estrogen. Biol Reprod 32:721–732 Scott IS, Charlton HM, Cox BS, Grocock CA, Sheffield JW, O’Shaughnessy PJ 1990 Effect of LH injections on testicular steroidogenesis, cholesterol sidechain cleavage P450 mRNA content and Leydig cell morphology in hypogonadal mice. J Endocrinol 125:131–138 Gaytan F, Pinilla L, Romero JL, Aguilar E 1994 Differential effects of the administration of human chorionic gonadotropin to postnatal rats. J Endocrinol 142:527–234 Mendis-Handagama SM, Watkins PA, Gelber SJ, Scallen TJ 1998 The effect of chronic luteinizing hormone treatment on adult rat Leydig cells. Tissue Cell 30:64 –73 Pakarinen P, Vihko KK, Voutilainen R, Huhtaniemi I 1990 Differential response of luteinizing hormone receptor and steroidogenic enzyme gene expression to human chorionic gonadotropin stimulation in the neonatal and adult rat testis. Endocrinology 127:2469 –2474 Dufau ML 1996 The luteinizing hormone receptor. In: The Payne AH, Hardy MP, Russell LD, eds. Leydig cell. Vienna: Cache River Press; 333–350 Matzuk MM, Boime I 1988 The role of the asparagine-linked oligosaccharides of the ␣ subunit in the secretion and assembly of human chorionic gonadotrophin. J Cell Biol 106:1049 –1059 Van Casteren JI, Schoonen WG, Kloosterboer HJ 2000 Development of timeresolved immunofluorometric assays for rat follicle-stimulating hormone and luteinizing hormone and application on sera of cycling rats. Biol Reprod 62:886 – 894 Ding YQ, Huhtaniemi I 1989 Human serum LH inhibitor(s): behaviour and contribution to in vitro bioassay of LH using dispersed mouse Leydig cells. Acta Endocrinol (Copenh) 121:46 –54 Huhtaniemi IT, Rajaniemi HJ, Catt KJ 1986 Kinetic and autoradiographic studies on binding of hCG to the testicular LH receptors of neonatal rats. J Reprod Fertil 78:73– 82 Bradford MM 1976 A rapid and sensitive method for the quantitation of microgram quantities of protein utilizing the principle of protein-dye binding. Anal Biochem 72:248 –254 Bartke A, Steele RE, Musto N, Caldwell BV 1973 Fluctuations in plasma testosterone levels in adult male rats and mice. Endocrinology 92:1223–1228 McEwen BS, Lieberburg I, Chaptal C, Krey LC 1977 Aromatization: important for sexual differentiation of the neonatal rat brain. Horm Behav 21:249 –263 Hutchison JB 1997 Gender-specific steroid metabolism in neural differentiation. Cell Mol Neurobiol 17:603– 626 Barkley MS, Goldman BD 1977 The effects of castration and Silastic implants of testosterone on intermale aggression in the mouse. Horm Behav 9:32– 48 Sharpe R 1993 Experimental evidence for Sertoli-germ cell and Sertoli-Leydig cell interactions. In: Russell LD, Griswold MD, eds. The Sertoli cell. Clearwater, FL: Cache River Press; 391– 418

Rulli et al. • hCG-Overexpressing Transgenic Mice

36. Anderson RA, Sharpe RM 2000 Regulation of inhibin production in the human male and its clinical applications. Int J Androl 23:136 –144 37. Marchetti C, Hamdane M, Mitchell V, Mayo K, Devisme L, Rigot JM, Beauvillain JC, Hermand E, Defossez A 2003 Immunolocalization of inhibin and activin ␣ and ␤B subunits and expression of corresponding messenger RNAs in the human adult testis. Biol Reprod 68:230 –235 38. Young J, Couzinet B, Chanson P, Brailly S, Loumaye E, Schaison G 2000 Effects of human recombinant luteinizing hormone and follicle-stimulating hormone in patients with acquired hypogonadotropic hypogonadism: study of Sertoli and Leydig cell secretions and interactions. J Clin Endocrinol Metab 85:3239 –3244 39. Kinniburgh D, Anderson RA 2001 Differential patterns of inhibin secretion in response to gonadotropin stimulation in normal men. Int J Androl 24:95–101 40. Jin W, Arai KY, Herath CB, Kondo M, Ishi H, Tanioka Y, Watanabe G, Groome NP, Taya K 2001 Inhibin in the male Go¨ ttingen miniature pig: Leydig cells are the predominant source of inhibin B. J Androl 22:953–960 41. Matzuk MM, Finegold MJ, Su JG, Hsueh AJ, Bradley A 1992 ␣-Inhibin is a tumour-suppressor gene with gonadal specificity in mice. Nature 360:313–319 42. Hsueh AJ, Dufau ML, Catt KJ 1977 Gonadotropin-induced regulation of luteinizing hormone receptors and desensitization of testicular 3⬘:5⬘-cyclic AMP and testosterone responses. Proc Natl Acad Sci USA 74:592–595 43. Setchell BP, Sharpe RM 1981 Effect of injected human chorionic gonadotrophin on capillary permeability, extracellular fluid volume and the flow of lymph and blood in the testes of rats. J Endocrinol 91:245–254 44. Geesaman B, Villanueva-Meyer J, Bluestein D, Miller L, Mena I, Rajfer J 1992 Effects of multiple injections of HCG on testis blood flow. Urology 40:81– 83 45. van Vliet J, Rommerts FF, de Rooij DG, Buwalda G, Wensing CJ 1988 Reduction of testicular blood flow and focal degeneration of tissue in the rat after administration of human chorionic gonadotrophin. J Endocrinol 117: 51–57 46. Kerr JB, Sharpe RM 1989 Focal disruption of spermatogenesis in the testis of adult rats after a single administration of human chorionic gonadotrophin. Cell Tissue Res 257:163–169 47. Eddy EM, Washburn TF, Bunch DO, Goulding EH, Gladen BC, Lubahn DB, Korach KS 1996 Targeted disruption of the estrogen receptor gene in male mice causes alteration of spermatogenesis and infertility. Endocrinology 137:4796 – 4805 48. Hess RA, Bunick D, Lee KH, Bahr J, Taylor JA, Korach KS, Lubahn DB 1997 A role for oestrogens in the male reproductive system. Nature 390:509 –512 49. Itoh M, Xie Q, Miyamoto K, Takeuchi Y 1999 Major differences between the testis and epididymis in the induction of granulomas in response to extravasated germ cells. I. A light microsopical observation study in mice. Int J Androl 22:307–315 50. Altay B, Hekimgil M, KEFi A, Cikili N 2001 Histopathological examination of both ipsilateral and contralateral testes with different obstructive models in prepubertal and adult rats. Br J Urol Int 88:84 – 89 51. Flickinger CJ, Baran ML, Howards SS, Herr JC 1999 Degeneration of the seminiferous epithelium following epididymal obstruction in prepubertal rats. Anat Rec 254:76 – 86 52. Jones R 1977 Effects of testosterone, testosterone metabolites and anti-androgens on the function of the male accessory glands in the rabbit and rat. J Endocrinol 74:75– 88