| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Discipline of Obstetrics and Gynaecology, Research Centre for Reproductive Health, University of Adelaide, Adelaide, South Australia 5005, Australia
Address all correspondence and requests for reprints to: Associate Professor Sarah A. Robertson, Discipline of Obstetrics and Gynaecology, University of Adelaide, Adelaide, South Australia 5005, Australia. E-mail: sarah.robertson{at}adelaide.edu.au.
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Androgens are the principal steroid hormones regulating male reproductive function, with roles in spermatogenesis, secondary sex organ function, and sexual behavior. Androgens are synthesized in the testis from cholesterol through a series of enzymatically controlled chemical reactions (2). In vitro culture experiments indicate that TGFß1 can modulate androgen synthesis in Leydig cells (3, 4, 5) through regulation of LH/human chorionic gonadotropin (hCG) receptor expression (4, 6) and modulating steroidogenic enzymes including P450 17
-hydroxylase/C17–20-lyase (6, 7) and 3ß-hydroxysteroid dehydrogenase (8). Within the seminiferous tubules of the testis, TGFß1 and its receptors are expressed by Sertoli cells and germ cells (9, 10, 11, 12). The pattern of synthesis of TGFß1 and the necessity for interaction between germ cells and Sertoli cells in regulating Tgfb1 mRNA expression has led to the postulate of a role for this cytokine in spermatogenesis (11, 13, 14).
An opportunity to study the precise physiological significance of TGFß1 in male fertility in vivo is provided by availability of mice with a null mutation in the Tgfb1 gene. TGFß1-deficient mice suffer massive multifocal inflammatory lesions and do not survive longer than 3 wk (15, 16). However, rendering the mice immunocompromised, by treatment with dexamethasone (17) or anti-CD11 (18), mutation in the ß2-microglobulin (19) or major histocompatability class II (20) genes, or mutation in the Prkdc gene to yield severe combined immunodeficiency (scid) (18), prevents inflammation and permits Tgfb1 null mutant mice to develop to healthy adulthood.
Reproductive function in male Tgfb1 null mutant mice has not previously been the subject of formal investigation. The aim of this study was to examine reproductive competence in male Tgfb1 null mutant mice bred on a Prkdcscid mutant background to allow their survival to reproductive age (18). This study provides new insights into the role of TGFß1 in male fertility, showing that mice deficient in this cytokine suffer complete infertility caused by impaired LH secretion, resulting in reduced testosterone levels, acting in concert with testosterone-independent sexual dysfunction.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
To genotype mice, tail DNA was analyzed by PCR to detect the intact and disrupted Tgfb1 gene, as previously described (21). Briefly, the forward primer 5'-GAGAAGAACTGCTGTGTGCG was used together with reverse primers 5'-GTGTCCAGGCTCCAAATATAGG to detect the intact Tgfb1 gene or 5'-CTCGTCCTGCAGTTCATTCA to detect the disrupted Tgfb1 gene containing a neomycin resistance gene (Neor) inserted into exon 6. Over the 3-yr duration of the project, a total of 632 male progeny from approximately 250 litters produced by Tgfb1+/– breeding pairs were genotyped, yielding 184 Tgfb1+/+, 321 Tgfb1+/–, and 127 Tgfb1–/– male mice (ratio +/+ to +/– to –/– of 1.00:1.74:0.69).
Assessment of male fertility
Individual adult male Tgfb1–/– and Tgfb1+/± mice were each housed with two normal adult B10.BR females (ARC, Perth, Australia) for a period of 3 wk, from 7 to 10 wk of age. The females were checked daily for the presence of vaginal plugs or sperm in vaginal smears indicating a mating event. Vaginal smears were used to track estrous cycles in the females by analysis of cellular content under phase contrast microscopy (22). Mated females were removed from males and checked over the ensuing 3 wk for overt signs of pregnancy and delivery of pups.
Histological analysis of reproductive organs
Adult male Tgfb1–/– and Tgfb1+/± mice were killed at 10 wk of age and the reproductive organs (testis, seminal vesicle, and penis) were weighed and fixed in 10% buffered formalin for histological analysis. Other organs were also dissected and weighed.
After fixation, reproductive tissues were washed in four changes of PBS (pH 7.4) over 2 d and processed and embedded in paraffin. Transverse sections (5 µm) were cut from the medial region of the penis, seminal vesicle, and testis and then stained with hematoxylin and eosin for histological analysis using a BH2 light microscope (Olympus, Tokyo, Japan). For analysis of testis, data from 30 tubules from each mouse were collected, comprised of 10 randomly selected round seminiferous tubules from each of three regions separated in the transverse plane by approximately 500 µm. Spermatogonia, spermatocytes, round spermatids and elongated spermatids were identified on the basis of characteristic morphological features defined according to Oakberg (23). Each tubule was evaluated for the presence of each of these four stages of germ cell differentiation, and the percent of tubules containing less than four stages, as well as the percent of tubules containing immature germ cells sloughed off into the tubule lumen, were calculated. The diameter of each tubule was determined as the mean of two orthogonal diameters measured with a Video Image Analysis system using Video Pro 32 software (Leading Edge, Adelaide, Australia), and mean tubule diameter values were calculated for each mouse.
Analysis of sperm number and viability
Sperm were squeezed from the cauda epididymis of 10-wk-old males and dispersed by incubation for 15 min in G-100 (Vitrolife, Gothenburg, Sweden) at 37 C in 5% CO2, counted and used in in vitro fertilization studies to assess viability. Oocytes were collected from 3- to 4-wk-old CBA x C57BL/6 F1 female mice (University of Adelaide, Central Animal House) after superovulation with FSH (Folligon, Intervet, Boxmeer, The Netherlands) (5 IU, ip) followed 50 h later with hCG (Chorulon, Intervet) (5 IU, ip). Ovulated oocyte cumulus complexes were collected 14 h after hCG from the oviduct and exposed to capacitated sperm (106 sperm/ml) for 4 h in pregassed rS1 media (Vitrolife) at 37 C in 5% CO2. Oocytes were washed and placed in 30 µl droplets of rS1 under mineral oil (Sigma, St. Louis, MO) for the first 48 h, assessed for cleavage, and then transferred to pregassed 30 µl droplets of SQC media (Vitrolife). Blastocyst formation was assessed on d 5 after initiation of culture.
Analysis of hypothalamo-pituitary-gonadal axis
Serum was collected by cardiac puncture at the time the animals were killed from Tgfb1–/– and Tgfb1+/± male mice at 6 and 10 wk of age. Testes were decapsulated and homogenized in ice-cold 1 ml methanol, centrifuged, and the supernatant collected for intratesticular testosterone measurement. The samples were dried and resuspended in 1 ml PBS.
Testosterone, androstenedione, and estradiol were measured using commercial RIA kits DSL-4100, DSL-3800, and DSL-4400, respectively (Diagnostic Systems Laboratories, Webster, TX) according to the manufacturer’s instructions at the National Association of Testing Authorities (Sydney, Australia)-accredited Reproductive Medicine Laboratories (Adelaide, Australia). Assays were performed in duplicate, and all samples were measured in a single assay. The lower limits of detection were 0.05 ng/ml testosterone, 0.03 ng/ml androstenedione, and 4.7 pg/ml estradiol. The within-assay coefficients of variation were 10.1% in the testosterone assay, 5.7% in the androstenedione assay, and 10.0% in the estradiol assay.
To measure serum FSH and LH, Tgfb1–/– and Tgfb1+/± male mice at 8 wk of age were exposed to an adult cycling female for 15 min to induce an endogenous LH surge (24), and blood was collected by cardiac puncture. Serum FSH and LH were quantified by RIA using reagents kindly provided by Dr. A. F. Parlow [National Institute of Diabetes and Digestive Kidney Diseases (NIDDK) National Hormone and Peptide Program, Torrance, CA]. Iodinated preparations and antisera used in the FSH and LH RIAs were rFSH-I-9 and anti-rFSH-S-11 and rLH-I-9 and anti-rLH-S-10, respectively. Results are expressed in terms of NIDDK mFSH-RP (AFP-5308D) and NIDDK mLH-RP (AFP-5306A). The iodination preparations were iodinated using Iodogen reagent (Pierce, Rockford, IL). Goat antirabbit (GAR12; Monash Institute of Reproduction and Development, Monash University, Melbourne, Australia) was used as second antibody for the FSH and LH RIAs. Assays were performed in duplicate on 20-µl serum samples, and all samples were measured in a single assay. The lower limits of detection were 0.78 ng/ml FSH and 0.10 ng/ml LH. The within-assay coefficients of variation were 8.1% in the FSH assay and 7.3% in the LH assay. Data points below the assay threshold were assigned the threshold values for data analysis.
To analyze testis responsiveness to gonadotropins, Tgfb1–/– and Tgfb1+/± male mice at 8 wk of age were administered hCG (Chorulon; 5 IU/10 g, ip) 1 h before the animals were killed, and blood was collected by cardiac puncture for serum testosterone measurement.
Evaluation of testosterone replacement on fertility
To determine whether sexual dysfunction in Tgfb1 null mutant male mice was caused by neonatal and/or adult testosterone deficiency, exogenous testosterone was administered to Tgfb1–/– and Tgfb1+/± mice by three different protocols. Some Tgfb1–/– males were given SILASTIC brand implants (Dow Corning, Midland, MI) containing 20 µl of 150 mg/ml testosterone (Sigma) in oil at 5 wk of age. The implant was soaked in sterile PBS overnight and inserted under the skin on the hind rump under 2% halothane anesthetic. Alternatively, testosterone propionate (TP) (Upjohn, Bridgewater, NJ) (200 µg in 100 µl) was administered to male Tgfb1 null mutant mice every second day from the age of 5 wk for a period of at least 3 wk. A third approach aimed to androgenize the neonatal brain, using a single sc injection of TP (100 µg in 20 µl) administered to 1- or 2-d-old pups (25). To maintain serum testosterone concentration in adulthood in this group, TP was given daily from 5 wk of age, by sc injection as described above. Mice treated with testosterone were housed with normally cycling B10 females for a period of 3 wk (from 7 to 10 wk of age), with females checked daily for the presence of a plug or sperm-positive vaginal smear.
Evaluation of sexual behavior and erectile response
Experiments to evaluate sexual behavior and mating competence were carried out with sexually inexperienced, virgin Tgfb1–/– and Tgfb1+/± males at 8–10 wk of age. For 1 wk before the experiment, males were exposed to the scent of females by housing adjacent to adult female cages, with bedding transferred from the female cage to the male cage. The experiment used superovulated 4-wk-old B10.BR females, induced into estrus by treatment with FSH (Folligon, 5 IU, ip, at 1000 h on d 1) followed by hCG (Chorulon, 5 IU, ip, at 1000 h on d 3). A single superovulated female was added to the male’s cage at 2200 h on d 3 of the superovulation protocol, and behavior was recorded with a video camera for the following 2 h under red light. The presence of a plug or sperm-positive vaginal smear the following morning indicated a mating event. Superovulation and thus sexual receptivity in the female was confirmed by the presence of ovulated oocytes in the oviduct. If the female failed to ovulate, data were excluded and the experiment was repeated using a new female the following evening.
Male sexual behavior was quantified by blinded analysis of the video tape footage. Parameters of sexual behavior were quantified for each male, including: 1) the amount of time spent in anogenital investigation during the first 10 min after introduction of a female; 2) latency (the time interval between introduction of the female and first mount); 3) the number and duration of mounting events; 4) the number and duration of intromission events (defined as mount with thrusting); and 5) the occurrence of ejaculation. Each behavior was identified according to characteristic features as previously described (26).
In additional experiments, the erectile response of Tgfb1–/– and Tgfb1+/± mice were evaluated by electrical stimulation of penile reflexes using a rectal electroejaculation probe according to a previously published protocol (27). Adult males at 8 wk of age were anesthetized with Avertin before insertion into the rectum of a custom-made probe (provided by Tecirlioglu and Trounson, Monash University, Melbourne, Australia). Briefly, three cycles of computer-generated sine waves were administered, with each cycle consisting of a sequence of 3-sec pulses of incrementally increasing voltage (0.5–3.0 V in 0.5 V steps), with -sec rest interval between pulses. Each cycle was separated by a-sec rest period, and the entire procedure lasted approximately 4 min. An erectile response was evidenced by occurrence of characteristic changes in the penis including cups (intense flaring of the penis tip) and flips (lateral movement of penis at the joint) (28). Mice were killed immediately after electrical stimulation by cervical dislocation.
Statistical analysis
Data were analyzed by independent-samples t test or one-way ANOVA with post hoc Tukey’s test where appropriate. When Shapiro-Wilk’s test showed data sets were not normally distributed, data were analyzed by nonparametric Kruskal Wallis H test followed by Mann-Whitney U test. Significance was inferred at P < 0.05.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
). Over a similar period, 11 of 12 (92%) Tgfb1+/± males mated with both females and pups were later delivered by all of the mated females. Analysis of vaginal smears revealed that all females housed with Tgfb1–/– males were cycling normally with an estrus event every 4–6 d, and no evidence of pseudopregnancy, potentially indicative of sterile mating, was observed.
|
). Between the ages of 5 and 10 wk, Tgfb1–/– males consistently weighed approximately 20% less than age-matched controls. To evaluate the effect of TGFß1 deficiency on the weight, gross structure, and morphology of reproductive and other organs, male mice of both genotypes were killed at the age of 10 wk and male sexual organs as well as other tissues were dissected. Relative to body weight, no difference in the wet weight of the testis, seminal vesicle, or penis was observed in Tgfb1–/– mice (Fig. 2B). However, there were differences in the weight of other organs. The lungs of male Tgfb1 null mutant males weighed 50% more than littermate controls relative to body weight. Peritoneal fat and spleen mass were decreased by 40 and 60%, respectively. The weight of the liver, heart, kidney, quadricep muscle, and brain were comparable between Tgfb1–/– and Tgfb1+/± littermates relative to body weight.
|
, A and B), and the urethra, corpus cavernosum, and epithelium as well as other tissue compartments were present and of normal dimensions and appearance in the penis (Fig. 3, C and D).
|
). When the two groups were compared, a significant reduction in sperm count was observed in Tgfb1–/– mice (P = 0.047). When the three severely affected individuals were eliminated from the statistical analysis, a modest reduction in epididymal sperm count remained evident in Tgfb1 null mutant mice (P = 0.054).
|
); however, individuals with low numbers of epididymal sperm were associated with abnormal testes histology (Fig. 5C). When parameters of testis pathology were quantified in testis sections from Tgfb1–/– and Tgfb1+/± males, differences in spermatogenesis in situ were observed in some but not all the tissues from TGFß1-deficient mice. One of eight Tgfb1–/– males showed a large proportion of tubules containing only spermatogonia or spermatocytes and no mature elongated spermatids (Fig. 6A). Similar proportions of males showed evidence of immature germ cells sloughed off into the lumen of the seminiferous tubules irrespective of genotype (Fig. 6B). A small but significant reduction in mean tubule diameter was detected in the testes of Tgfb1–/– males (P = 0.028) (Fig. 6C), but there was no consistent relationship between small tubule diameter and quality of spermatogenesis (data not shown). In the Tgfb1–/– group, sperm count and disrupted spermatogenesis was highly correlated with body weight (R = 0.82, n = 14, P < 0.0001), with the greatest pathology evident in the most severely growth-impaired males.
|
|
).
|
). This deficiency was not due to delayed sexual maturation because a similar scale of deficiency remained evident at 10 wk (P < 0.0001) (Fig 7A). There was no effect of Tgfb1 gene dosage on testosterone levels because serum testosterone concentrations were similar in 10-wk-old wild-type (Tgfb1+/+) and heterozygous (Tgfb1+/–) male mice (16.3 ± 3.6 and 16.5 ± 3.2 ng/ml, respectively). Intratesticular testosterone was similarly affected by genotype, with a 95% reduction in mean levels in Tgfb1–/– males at 10 wk of age (Fig. 7B). TGFß1 deficiency also resulted in a 64% reduction in mean serum androstendione concentration (Fig. 7C), whereas serum estradiol was unaffected by genotype in 10-wk-old adult males (Fig. 7D). There was no correlation between serum or intratesticular testosterone concentration and epididymal sperm count or abnormal testicular morphology in Tgfb1–/– males, with a similar testosterone content in mice with greater than 107 or less than 107 sperm recovered from the epididymis (data not shown). There was no correlation between serum or intratesticular testosterone concentration and body weight (data not shown).
|
). Serum FSH was also significantly reduced, albeit to a lesser extent, in Tgfb1–/– males (11.4 ± 0.6 ng/ml, compared with 14.5 ± 0.3 ng/ml for Tgfb1+/± males) (P = 0.0001) (Fig. 8B).
|
).
Effect of testosterone replacement on mating ability of Tgfb1 null mutant males
To determine whether testosterone depletion was the cause of infertility in Tgfb1–/– males, the effect of exogenous testosterone administration in neonates and during adulthood was examined. Both SILASTIC brand testosterone implants and sc TP injections were found to result in normal levels of serum testosterone in Tgfb1–/– males at 10 wk of age (15.3 ± 3.7 and 17.3 ± 5.0 ng/ml, respectively). However, both treatments failed to alleviate the infertility phenotype because when treated Tgfb1–/– males were housed for 3 wk (at age 7–10 wk) with naturally cycling adult B10.BR females, vaginal plugs or sperm-positive vaginal smears were not detected (Table 2). Similarly, when neonatally androgenized Tgfb1–/– males were treated with sc TP injections, despite normal levels of serum testosterone (17.7 ± 3.1 ng/ml), males remained unable to mate with females. Daily vaginal smears indicated that all females housed with testosterone-treated Tgfb1–/– males cycled normally and evidence of pseudopregnancy was not observed.
|
, and movies 1 and 2 for examples of male sexual behavior from Tgfb1+/+ and Tgfb1–/– males, respectively). All of five Tgfb1+/± males showed the expected characteristic sequence of behavioral events, with initial anogenital investigation of the female, followed by mounting, intromission, and finally ejaculation behavior. There was a considerable range in the time course of response to females as well as the number of mounts and intromissions before successful ejaculation; however, each male attained ejaculation within the 2-h test period and thereafter did not show further sexual interest in the female. The presence of a plug or sperm-positive vaginal smear in the female on the following day confirmed that ejaculation had occurred, and this was evident in receptive females housed with all control males.
|
). In both cases the intromission events were shorter in duration than intromission behavior displayed by controls and were not sustained to ejaculation, with none of the Tgfb1–/– males ejaculating during the test period. Tgfb1–/– mice continued to display mounting behavior for the full duration of the test period, maintaining a rate of approximately one mount every 2 min, comparable with that seen in Tgfb1+/± before ejaculation. This resulted in a mean total of 48 mounts per Tgfb1–/– male over the 2-h test period, none of which led to ejaculation. The presence of a plug or sperm-positive vaginal smear was not seen in any of the females housed with Tgfb1–/– males. There was no evidence of any link between body weight and incidence of mounting or intromission, with even the smallest Tgfb1 null mutant mice showing avid interest in females and energetic mounting activity.
Effect of electrical stimulation on penile function
To investigate whether erection could be achieved with electrical stimulation in Tgfb1–/– males, anesthetized mice were treated with a rectal electroejaculation probe. Electrical stimulation elicited erectile responses in Tgfb1–/– and Tgfb1+/± males, as evidenced by occurrence of characteristic changes in the penis including cups (intense flaring of the penis tip) and flips (lateral movement of penis at the joint) (28). These responses were observed in approximately 50% of males regardless of genotype (Table 4), with similar volumes of seminal fluid expelled. Motile sperm (>104/µl) were identified by microscopic evaluation of seminal fluids from two Tgfb1–/– males, confirming ejaculation had indeed taken place.
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Both serum androstenedione and testosterone levels were reduced in Tgfb1 null mutant mice. This was attributed to impaired endocrine signaling, with serum LH reduced substantially and FSH to a lesser extent in Tgfb1 null mutant males. Administration of exogenous gonadotropin to TGFß1-deficient mice induced normal levels of circulating testosterone, showing that the effect of TGFß1 deficiency on testosterone is secondary to disrupted LH signaling and supporting a requirement for TGFß1 in normal hypothalamic or pituitary function. Male mice with a mutation in the LH receptor or LH ß-subunit are infertile with hypogonadism and defects in testosterone synthesis and spermatogenesis (29, 30, 31), showing the absolute requirement for this hormone in normal steroidogenesis and testes growth and function. LH was detectable at low levels in most of the TGFß1-deficient mice, indicating that synthesis must be impaired or dysregulated, rather than completely absent. LH secretion in Tgfb1 null mutant females is impaired to a similar extent, adversely influencing development and ovulation of ovarian follicles (21). FSH is also required for normal testosterone synthesis in mice (32), and although FSH was moderately reduced in Tgfb1 null mutant males, complete recovery of normal testosterone levels after hCG administration showed that FSH deficiency did not contribute to the steroidogenic deficit. Absence of local TGFß1 signaling might also contribute to disrupted testicular steroidogenesis, although the current result of reduced testosterone is difficult to reconcile with reports from in vitro studies showing that endogenous TGFß1 acts in an autocrine manner to inhibit Leydig cell steroidogenesis (6).
Relative to the 20% smaller body weight of TGFß1-deficient mice, the testes were of the expected weight, were normally descended and were functionally competent. This is consistent with observations that quantitatively complete spermatogenesis can continue in the presence of very low testosterone. In LH receptor knockout mice, LH-independent steroidogenesis is sufficient to maintain spermatogenesis despite the reduction in intratesticular testosterone to 2% of wild-type levels (29), and mice with a null mutation in pituitary adenylate cyclase-activating peptide show serum testosterone levels diminished by more than 90% in association with normal testicular structure and unimpaired spermatogenesis (33). Further analysis of the morphology of the testes revealed some differences between TGFß1-deficient and replete males. A modest reduction in the average diameter of seminiferous tubules was evident in Tgfb1 null mutant males; however, this appeared to have little impact on spermatogenesis. In the majority of males, normal sperm numbers were recovered from the epididymis, and these sperm were viable and developmentally competent as assessed by in vitro fertilization of oocytes, with development of blastocysts occurring at the expected rate. In a small number of males, sperm were all but absent from the epididymis and spermatogenesis was severely impaired, with a large proportion of seminiferous tubules containing only spermatogonia or spermatocytes and no mature spermatids. There was no evidence for greater testosterone deficiency in those few mice with impaired spermatogenesis.
Body weight can be a determinant of fertility in rodents as in other species (34) and systemic growth restriction might have contributed to impaired spermatogenesis. The Tgfb1 null mutant males showed a retarded growth trajectory, averaging approximately 20% smaller body mass, compared with wild-type litter mates. The Tgfb1 null mutant males that exhibited substantially reduced epididymal sperm numbers were also those with the lowest body weight. However, growth impairment can be excluded as a cause of the testosterone deficiency and profound inability to copulate seen in all Tgfb1 null mutant males. There was no correlation between body weight and testosterone levels or mating parameters. Several of the largest Tgfb1 null mutants had body weights equal to or higher than the smallest fully fertile, wild-type males of the same age, and these larger mice showed no evidence of less severe mating phenotype or relatively higher testosterone. This is consistent with studies showing that only severe nutritional impairment ablates fertility in mice, with no effect on testosterone or loss of reproductive function in mice exposed to chronic food restriction resulting in similar or more severe growth impairment than observed in the Tgfb1 null mutants (35, 36).
Among Tgfb1 null mutant mice with epididymal sperm counts approximating those of wild-type males, a small reduction in mean sperm count was evident in the Tgfb1 null mutant mice. Whereas this was likely attributable to smaller absolute testes size, a contribution of abnormal pituitary gonadotropins cannot be discounted. A severe infertility phenotype with reduction in seminiferous tubule diameter together with arrest of spermatogenesis at the spermatid stage is observed in mice deficient in LH receptor (31). Young mice deficient in the LH ß-subunit also show narrow seminiferous tubules and arrest in spermatogenesis at the round spermatid stage (30), although this phenotype is alleviated with advancing age (29). The slightly decreased serum FSH concentration in TGFß1 mice potentially also contributed to disrupted spermatogenesis, as has been reported for the FSH receptor null mouse (32).
Our results are not consistent with previous studies implicating local intratesticular TGFß1 signaling in spermatogenesis and in contrast show that testicular TGFß1 synthesis is not essential for morphogenesis of competent sperm. Male gametes synthesize TGFß1 with greatest abundance detected in spermatocytes and early round spermatids (11), suggesting a role for TGFß1 in regulating spermatocyte proliferation and differentiation. Our data conflict with conclusions from in vitro culture experiments that TGFß1 might act to impair spermatocyte progression through the second meiotic division (37) and/or facilitate apoptosis of male germ cells at puberty (38). This does not preclude a role for TGFß cytokines in spermatogenesis because TGFß2 and TGFß3 as well as other members of the TGFß superfamily have overlapping functions and might exert compensatory effects in TGFß1-deficient mice.
Whether the lesion in gonadotropin synthesis seen in Tgfb1 null mutant mice originates in the pituitary or the hypothalamus is unclear. TGFß1 and its receptors can be found in both the anterior pituitary (39, 40) and the hypothalamus (41), suggesting some functions of both tissues may be dependent on this cytokine. TGFß1 is known to regulate prolactin release from lactotrophs in the pituitary (42, 43) and has been found to inhibit FSH mRNA expression in primary murine gonadotropes (44). In the hypothalamus, TGFß1 is the active factor produced by hypothalamic astrocytes that stimulates LHRH secretion (45) and also acts to regulate microglial recruitment and function (46).
It was somewhat surprising that testosterone replacement failed to even partially restore fertility in TGFß1-deficient males. We used several methods of androgen replacement in an attempt to mimic normal testosterone availability during development and in adult mice. Testosterone implants increased serum testosterone levels to approximate levels in normal mice, but consistent with previous reports (47), we observed delayed wound healing after implant insertion, which might have adversely affected sexual function. Testosterone propionate given with or without prior neonatal androgenization, a protocol that successfully restored mating ability in LH-deficient hypogonadal mice (25), was also unsuccessful in restoring mating ability in Tgfb1 null males. Together, these studies showed that testosterone deficiency is not the primary cause of infertility in Tgfb1 null males and led us to evaluate other aspects of their sexual behavior.
The act of mating in mice comprises a characteristic sequence of defined behavioral events, with initial anogenital investigation of the female, followed by mounting, intromission and finally ejaculation (26). Tgfb1 null mutant males were incapable of proceeding beyond the earliest stages of this characteristic behavioral sequence. This was not the result of reduced motivation; Tgfb1 null mutant males consistently showed normal sexual interest as indicated by anogenital investigatory behavior, followed by attempts to grasp and mount females. However, only some males were capable of full mounting activity and only rarely did mounting lead to intromission. When evidence of intromission did occur, it was sustained for less than half the normal duration and did not proceed to ejaculation. Vaginal plugs or evidence of sperm deposition was never seen in the female mice caged with null mutant males. Furthermore, brief intromission without ejaculation was insufficient to induce cervical stimulation in females because the characteristic signs of pseudopregnancy were not evident in females housed with Tgfb1 null mutant males.
Our initial attempts have not identified an underlying mechanism for this sexual dysfunction. TGFß1 is believed to have an inhibitory role in the regulation of penile growth during puberty (48), and null mutation might have been expected to result in abnormal growth of the penis or secondary sex organ development. However, no overt structural defects were evident in the penis of null mutant mice, and when organ weight was normalized to body weight, the penis and seminal vesicle glands were of expected weight and dimensions. Furthermore, the appropriate tissue structures and compartments were seen in the penis and seminal vesicle glands at the histological level.
Androgens are critically important in programing sexual behavior and the ability to perform the mating act (49), acting centrally to stimulate the brain to respond to sexual stimuli, and peripherally to facilitate erection via induction of nitric oxide synthase (NOS) activity (50). Consistent with this, administration of testosterone can improve sexual function in several testosterone-deficient mouse models (25, 51, 52). However, because testosterone failed to alleviate the mating incompetence in TGFß1-deficient mice, a behavioral or physiological lesion independent of reduced testosterone synthesis is implicated.
Other obvious potential explanations for mating incompetence can also be discounted. Despite their smaller size, Tgfb1 null mutant males maintained a steady weight and other overt signs of good health during the course of the mating study. Their physical capacity to grasp and mount females was unimpaired, with the Tgfb1 null mutant males generally exhibiting considerable persistence and stamina in interacting with females, on average mounting just as frequently and several times more over the course of the experiment than their wild-type littermates.
The observation of intromission failure led us to consider the possibility of erectile dysfunction in Tgfb1 null mutant male mice. Failure to achieve or sustain an erection is the most likely explanation for impaired intromission and ejaculation. Penile erection is regulated by a variety of mechanisms that are both psychological and physical, whereby central and peripheral neurovascular networks act in concert to result in relaxation of the corpora cavernosum smooth muscle and allow engorgement (53). Good erectile responses were observed in Tgfb1 null males given electrical stimulation, showing that the penis is capable of responding to an external impulse. However, electrical stimulation is a nonphysiological excitatory stimulus, and the response to electroejaculation does not rule out a deficit in the neural pathways regulating the erection response. In view of the known actions of TGFß in regulating NOS enzymes (54, 55, 56), a central or peripheral imbalance in activity of NOS or other neurotransmitters that regulate sexual function warrants investigation.
Our findings of normal relative mass and morphology of reproductive tissues contrasts with some nonreproductive organs, which were relatively altered in size in Tgfb1 null mutant mice. The lungs were increased as a result of TGFß1 deficiency, and fat and spleen weight were decreased. Our data are consistent with the known roles of TGFß1 in inhibition of lung morphogenesis (57, 58). A vital role for TGFß1 in the immune system is already demonstrated by the autoimmune pathology observed in immune competent TGFß1-deficient mice (15, 16). Their reduced spleen weight presumably reflects reduced lymphohemopoiesis or impaired lymphocyte proliferation. Reduced fat deposits in the absence of TGFß1 might be due to perturbations in proliferation of adipocyte precursor cells (59) or reduced feed intake due to leptin dysregulation (60).
The findings reported herein have implications for understanding the pathophysiology of human sexual dysfunction. Increased penile expression of Tgfb1 has been associated with penile fibrosis associated with Peyronie’s disease (61, 62), and elevated plasma levels of TGFß1 have been reported in men with vasculogenic causes of erectile dysfunction, particularly when linked with smoking (63). Our studies imply that TGFß1 deficiency might also be associated with sexual dysfunction and/or disturbances in the hypothalamic-pituitary-gonadal axis and indicate that perturbations in the TGFß1 signaling pathway warrant investigation as a cause of sexual disturbance in men. Variation in TGFß1 activity in humans is known to result from endogenous and exogenous factors contributing to homeostasis in TGFß synthesis and bioavailability. These include polymorphisms in genes regulating the TGFß signaling pathway (64) as well as environmental and lifestyle factors such as diet, stress, smoking, and infection (63, 65, 66).
In conclusion, this study demonstrates the absolute requirement for TGFß1 in male reproductive function. TGFß1 is identified as essential for hypothalamic-pituitary-gonadal axis activity and LH regulation of testosterone synthesis. In addition, TGFß1 is necessary for normal sexual competence, with deficiency resulting in insufficient erectile function for sustained intromission and ejaculation, despite normal libido. Further studies are required to elucidate the mechanism by which hypothalamic-pituitary activity and erectile function might be regulated at the local and/or systemic level by TGFß1 signaling and to investigate the extent to which perturbed TGFß1 bioavailability might have a role in human reproductive disorders.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
Disclosure Statement: The authors have nothing to disclose.
First Published Online May 3, 2007
Abbreviations: hCG, Human chorionic gonadotropin; NOS, nitric oxide synthase; scid, severe combined immunodeficiency; TP, testosterone propionate.
Received January 23, 2007.
Accepted for publication April 20, 2007.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
