| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
ARTICLE |
Andrology Laboratory (M.H., J.S., M.J., D.J.H., C.M.A.), ANZAC Research Institute, Concord Hospital, Concord, New South Wales 2139; and Department of Pathology (N.J.C.K.), University of Sydney, New South Wales 2006, Australia
Address all correspondence and requests for reprints to: Charles M. Allan, ANZAC Research Institute, Concord Hospital, Sydney, New South Wales 2139, Australia. E-mail: charles{at}med.usyd.edu.au.
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
ß.6 line, tg-FSH stimulated the maturation and proliferation (
2-fold) of Sertoli cells in hpg testes. Total Sertoli cell numbers were also significantly increased (1.5-fold) independently of FSH effects by T treatment alone. Selective FSH activity in ß.6 hpg testes increased total spermatogonia numbers 3-fold, which established a normal spermatogonia/Sertoli cell ratio. FSH also elevated meiotic spermatocyte numbers 7-fold, notably at pachytene (28-fold), but induced only limited numbers of postmeiotic haploid cells (absent in hpg controls) that arrested during spermatid elongation. In contrast, T treatment alone had little effect on postnatal spermatogonial proliferation but greatly enhanced meiotic progression with total spermatocytes increased 12-fold (pachytene 53-fold) relative to hpg testes, and total spermatid numbers 11-fold higher than tg-FSH hpg testes. Combining tg-FSH and T treatment had no further effect on Sertoli or spermatogonia numbers relative to FSH alone but had marked additive and synergistic effects on meiotic cells, particularly pachytene (107-fold more than hpg), to establish normal meiotic germ cell/Sertoli cell ratios. Furthermore, tg-FSH had a striking synergistic effect with T treatment on total spermatid numbers (19-fold higher than FSH alone), although spermatid to Sertoli cell ratios were not fully restored to normal, indicating elevated Sertoli cell numbers alone are insufficient to establish a maximal postmeiotic germ cell capacity. This unique model has allowed a detailed dissection of FSH in vivo activity alone or with T and provided compelling evidence that FSH effects on spermatogenesis are primarily via Sertoli and spermatogonial proliferation and the stimulation of meiotic and postmeiotic germ cell development in synergy with and dependent on T actions. |
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
-subunit and distinct ß-subunits, which are secreted as noncovalently bound heterodimers in response to a single hypothalamic hormone GnRH. In the testis, the responses of both FSH and LH ultimately converge onto Sertoli cells, which are the only cells within the seminiferous tubules that express receptors for FSH (2) and the androgens synthesized in response to LH (3). Consequently, both gonadotropins can act simultaneously on Sertoli cells, which may produce additive or synergistic effects on the development of associated germ cells. These intimate structural and functional requirements of FSH and LH have confounded previous investigations into the selective in vivo effects of each hormone in spermatogenesis. Classical approaches to examine gonadotropin function in vivo involve ablation-replacement analysis, such as hypophysectomy (4, 5, 6, 7, 8) or pharmacological suppression of GnRH activity (9, 10, 11) followed by replacement of exogenous FSH or LH/androgen. Although providing useful information, these models have inherent and unavoidable limitations that constrain their utility as experimental models. For instance, these studies all fall short of the ideal of specific, durable, or complete FSH deficiency and are restricted to postnatal effects of exogenous hormones on degenerating or regressed testicular backgrounds. In recent years, selective or targeted germline modifications of the mouse genome has created novel paradigms that offer new and incisive insights into the specific roles of FSH or LH in germ cell development. In earlier studies, our laboratory used the hypogonadal (hpg) mouse, which is functionally deficient in both FSH and LH/androgen activity because of a naturally occurring germline deletion in the GnRH gene (12). Research using this hpg paradigm provided definitive evidence that androgens alone, in the absence of FSH activity, were sufficient for qualitatively normal spermatogenesis in mice (13). These conclusions were verified by the emergence of similar mouse models that specifically lacked FSH receptor (14, 15) or subunit genes (16) and thereby also retained selective LH/androgen activity.
We recently created a novel model that exclusively expressed endogenous transgenic (tg)-FSH on the hpg background, in the absence of endogenous LH activity (17). This model provides several advantages over previous studies into the role of FSH in germ cell development. For example, the hpg background is selectively deficient in GnRH (18), whereas hypophysectomy removes other pituitary hormones that effect testicular function such as GH (19) and prolactin (20). Furthermore, immunoneutralization or pharmacological inhibition of GnRH, FSH, or Leydig cell function has variable efficacy and manifests incomplete suppression (10, 21, 22), and earlier studies using hpg mice were compromised by exogenous FSH preparations contaminated with LH (23) or immunogenicity toward postnatally administered recombinant human (rh)-FSH (24). In our tg-FSH model, endogenous expression of bioactive FSH is under the control of the rat insulin promoter II sequence (RIP) independently of the requirement of GnRH, LH secretion, and potential repression by androgens (25). The ability to study combined tg-FSH and androgen actions without confounding steroidal feedback effects on FSH secretion offers one distinct advantage of our transgenic strategy over similar mouse models with selective FSH activity after targeted disruption of the LH receptor gene (26, 27). The present strategy also allows the examination of longer-term FSH effects following tg-FSH expression during and after perinatal development, which is proposed to be an important period of FSH function (28). Therefore, this unique model provides the opportunity to characterize the definitive in vivo actions of FSH, and our present work reveals the cell types selectively stimulated by FSH alone or in combination with androgen within the seminiferous tubules of the rudimentary hpg testis.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
ß.6 line expressing bioactive human (h)-FSH was prepared as previously described (17) using the hpg strain (13, 29). The ß.6 line was selected for analysis because it expressed consistent serum levels of tg hFSH (tg-FSH) in a range previously found to induce a dose-dependent gonadal response in vivo (17). Mice expressing tg-FSH on a gonadotropin-deficient hpg background were obtained by cross-breeding animals heterozygote for the GnRH gene deletion, determined by detection of wild type and hpg or transgene PCR products as described (13, 17). Animals were housed under controlled conditions (12-h light, 12-h dark cycle, 19–22 C) with ad libitum access to food and water. All animal procedures were approved by the animal ethics committee and performed in accordance to the National Health and Medical Research Council code of practice for the care and use of animals and the NSW Animal Research Act (1985).
Animal treatments, serum, and tissue collection
For testosterone (T) treatment of hpg mice, anesthetized 21-d-old tg or non-tg littermates or age-matched males received a subdermal 1-cm SILASTIC brand (Dow Corning Corp., Midland, MI) implant containing crystalline T as described (13). After 6 wk of T treatment, animals were anesthetized and killed by terminal cardiac exsanguination and collected serum stored at -20 C. One fresh testis was removed, weighed, and stored at -20 C for intratesticular T assay and the remaining testis perfused and collected in Bouin’s fixative and embedded in hydroxymethylmethacrylate resin (Technovit 7100, Kulzer and Co., Friedrichsdorf, Germany) as described (17). Tissue sections were cut using a Polycut S microtome (Reichert Jung, Nossloch, Germany). Thin sections (3–5 µm) were stained with 0.5% toluidine blue, and thick sections (25 µm) for stereology were consecutively stained with periodic-acid-Schiff, hematoxylin and Scott’s blue solution. Control hpg and wild-type (non-hpg) testes and serum were obtained from non-tg littermates or age-matched males.
Quantitation of serum and intratesticular hormone levels
Serum levels of tg-FSH were determined using a two-site immunofluorometric hFSH assay kit (DELFIA, Wallac, Inc., Turku, Finland), which detects the presence of both human heterodimer subunits, with no significant cross-reactivity with endogenous mouse gonadotropins (17). Serum samples were assayed in duplicate, and the hFSH detection limit was 0.05 IU/liter. Serum and intratesticular testosterone levels were measured in duplicate by RIA as previously described (30), except samples were extracted twice in 10 volumes of hexane:ethyl acetate (3:2 vol/vol, pestiscan grade, Labscan, Dublin, Ireland).
Testicular cell flow cytometry analysis
Total testicular cell suspensions were derived from freshly isolated, decapsulated testes by mechanical dissociation and mincing of tissue with fine scissors in Ca2+/Mg2+-free Hanks’ buffer. Cell suspensions were filtered through gauze and diluted to 0.8 x 106 cells/ml in Ca2+/Mg2+-free Hanks’ containing 1% Triton X-100, ribonuclease (0.05 mg/sample), and propidium iodide (0.125 mg/sample). During flow cytometry, propidium iodide-stained DNA was detected as red fluorescence (585 nm) using a FACScan (Becton Dickinson and Co., Sydney, Australia) with an air-cooled argon ion laser for excitation (488 nm). Data for a minimum of 10,000 cells were collected for each sample and presented as histograms.
Stereological analysis
Testicular Sertoli and germ cell populations were quantified using the optical-dissector technique similar to that previously described for the rat testis (31). Briefly, fixed tissue sections (25 µm thick, three per testis) were derived from different regions along the longitudinal axis, and random uniform sampling of each section performed by light microscopy (x100/1.35 oil-immersion objective) using a microcator to monitor scanned depth, and cell numbers estimated using unbiased sample frames created by CAST grid software (Olympus Corp., Albertslund, Denmark) on images scanned directly to a real-time screen. Total Sertoli and germ cell numbers were extrapolated from calculated cell densities of the random sample volumes using respective testis weights and specific gravity of testis (d = 1.04 g/ml) (32). Each testicular cell type was identified using criteria described previously (32), which included shape, size, nuclear staining pattern, and when possible the distinct stages of the mouse spermatogenic cycle. Gonadotropin-independent germ cell development in hpg testes does not follow the defined stages of normal spermatogenesis; therefore, germ cell estimates were broadly grouped into cell types that enabled more complete comparisons between hpg testes and other experimental groups.
Data analysis
All statistical analysis was performed using SPSS (version 10.0, SPSS, Inc., Chicago, IL). Normally distributed data (Shapiro-Wilk test) were analyzed using one-way ANOVA with Tamhane post hoc tests. Nongaussian data were analyzed using a nonparametric (Mann Whitney) test. Differences were regarded significant when P was less than 0.05. All data in tables and figures are presented as mean ± SEM.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
ß.6 miceß.6, prepared using a tandem RIP-RIPß DNA transgene construct as previously described (17). This transgene construct was designed to direct simultaneous pancreatic expression of both hFSH subunits under the control of the RIP II sequence, which allowed androgen-insensitive expression of bioactive hFSH heterodimers on the hpg background. The expression of tg-FSH was readily differentiated from mouse FSH using an immunoassay specific for hFSH heterodimers. The ß.6 line expressed consistent circulating levels of tg-FSH (Table 1
), at levels over the required threshold and in the concentration range previously shown to induce a significant testicular response on the hpg background (17).
|
ß.6 hpg males with those of non-tg hpg littermates. Testes weights of adult tg-FSH hpg animals were increased 5-fold relative to non-tg controls (Fig. 1), consistent with our previous work (17). The testes of non-tg hpg mice treated with a 1-cm T implant, which results in a maximal androgen response (13), were approximately 33% of wild-type (non-hpg, non-tg) testis weight. In comparison with this maximal T effect, the addition of tg-FSH expression to the androgen response further increased testis size by 66% to reach 55% of wild type (Fig. 1).
|
). However, the total amount of testicular T (picomoles/testis) was significantly increased in tg-FSH hpg testes, compared with hpg controls, independently of T treatment. By comparison, non-tg and tg-FSH hpg males treated with 1-cm T implants had significantly increased serum and testicular T levels relative to untreated controls (Table 1).
Sertoli cell proliferation and maturation
The testes of nontransgenic hpg controls, compared with non-tg wild-type testes, exhibit seminiferous tubules with a disorganized germinal epithelium layer, without lumens, and immature-stage Sertoli cells lacking the characteristic tripartite nuclear structure (33) and basal localization of normal fully differentiated Sertoli cells (Fig. 2, A and D). In contrast, Sertoli cells of tg-FSH testes all contained tripartite nuclear structures indicative of Sertoli cell maturation (Fig. 2B). Stereological analysis revealed that the levels of tg-FSH expression in ß.6 hpg males promoted a 2-fold increase in absolute Sertoli cell numbers, compared with control hpg testes (Fig. 3). T-treated hpg (non-tg) testes also exhibited mature Sertoli cells with tripartite nuclear structures (Fig. 2C) and significantly increased total Sertoli cell numbers (49% above hpg controls), in the absence of tg-FSH expression (Fig. 3). The combination of T treatment and tg-FSH expression did not further increase Sertoli cells numbers relative to tg-FSH alone but did significantly increase the Sertoli cell population relative to T treatment alone (Fig. 3). This combined T treatment and tg-FSH effect resulted in the largest increase of the hpg Sertoli cell population, which remained significantly less than normal (P < 0.001), reaching approximately 70% of the wild-type population.
|
|
ß.6 tg-FSH hpg testes, there was a distinct population of haploid cells, which represented 16% of total testicular cells analyzed by flow cytometry, compared with a complete absence of haploid cells in non-tg hpg control testes (Fig. 4). In addition, the lack of a haploid shoulder demonstrated a deficiency of elongated spermatids in the tg-FSH testes. In contrast, the nontransgenic T-treated hpg mouse contained a larger proportion of haploid cells (52% of total testicular cells) relative to tg-FSH hpg testes (Fig. 4).
|
). In contrast, T treatment alone produced a nonsignificant (P = 0.085) 1.4-fold increase in spermatogonia, which was significantly lower than the specific FSH effect. Consistent with this reduced androgen effect on spermatogonia proliferation, there was no significant difference between spermatogonia numbers found in untreated or T-treated tg-FSH hpg testes (Fig. 5). Early primary spermatocyte numbers (preleptotene to zygotene, Pl-Z) were increased 5-fold by either circulating tg-FSH or androgen alone (P < 0.001), with no significant difference between either response. The combined FSHandrogen response further promoted Pl-Z spermatocyte proliferation (P < 0.0001), in which an additive effect of tg-FSH and androgen actions resulted in Pl-Z spermatocyte numbers equivalent to 66% of a normal non-tg testis (Fig. 5). Relative to the untreated hpg testis, the absolute levels of later-stage meiotic pachytene spermatocytes (PS) were markedly increased by either tg-FSH or androgen alone, 30-fold and 53-fold, respectively. The combined FSH-androgen response further elevated PS numbers 107-fold (equivalent to 61% of wild-type testis), indicative of a synergistic effect. The testes of tg-FSH hpg mice contained few postmeiotic germ cells, with the total numbers of round and elongated spermatids representing 3.9% and 0.5% of wild type, respectively. In contrast, T treatment alone had a greater impact on meiotic completion in hpg testes with round and elongated spermatids numbers reaching 26% of normal values (Fig. 5). The combined FSH-androgen response further increased round and elongated haploid germ cell numbers to 43% and 47% of normal values.
|
). For spermatogonial cells, the untreated or T-treated ß.6 hpg animals expressing tg-FSH produced a G/S ratio that was equivalent to the ratio found in a non-tg wild-type testis. For spermatogonia in non-tg hpg testes, the G/S ratio after T treatment alone remained significantly below the normal G/S ratio and was equivalent to hpg untreated controls. In contrast to the spermatogonia effects, tg-FSH, or T alone each significantly increased the G/S ratio of early meiotic spermatocytes (Pl-Z), and combined treatments further increased this G/S ratio to the level found in normal non-tg testes. Similarly, the G/S ratio for meiotic PS was significantly increased by either FSH or T treatment and restored to normal with the combined FSH-androgen response (Fig. 6). As germ cells progressed beyond meiosis, the impact of FSH alone was greatly diminished, compared with the G/S ratio of premeiotic germ cell populations. The combined actions of T treatment with the levels of tg-FSH expression in ß.6 males produced a marginal rise in the G/S ratios of both round and elongated spermatids but could not restore these ratios to wild-type levels (P < 0.05).
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The selective introduction of tg-FSH expression or T treatment onto the hpg background each individually increased testicular development at levels that were comparable with the gross gonadal effects reported in complementary loss-of-function mouse models using gene-specific knockouts. Testis weights of tg-FSH ß.6 hpg males were similar to the small testes in LH receptor-deficient models (12% and 18% of normal, respectively), which like our hpg paradigm also retained selective FSH activity without androgen effects (26, 27). Compared with these null-receptor models, which exhibit persistently elevated FSH levels, one advantage of our transgenic hpg approach is that the magnitude of the specific FSH response can be selected, independently of GnRH and LH actions or androgen feedback, using tg lines expressing higher or lower FSH levels (17). In the present study, we have specifically avoided the supraphysiological effects (e.g. gonadal hyperplasia) described in other models with excessive FSH activity (34, 35). A further benefit of our FSH-hpg model is the capacity to examine FSH effects combined with androgens without the problem of negative androgenic steroidal feedback regulation (36) that may inhibit FSH secretion and confound interpretation of LH receptor-null models during T treatment. Expression of tg-FSH stimulated greater testis growth, compared with postnatally administered exogenous rhFSH (23, 24), unmasking the more complete potential FSH effects. Any early fetal expression of tg-FSH conferred by the RIP (37) is unlikely to prematurely affect testis size because such expression precedes the appearance of functional FSH receptors (38). Selective T replacement in non-tg hpg males provided more substantial restoration of testis weight, compared with FSH alone, reaching between 30% and 40% of wild-type testis size. This is consistent with the testes weights of mice lacking the FSH ß-subunit (39) or receptor (14, 15) genes, thereby retaining selective LH/androgen activity, which were 30–50% of wild-type testis size.
We examined the specific in vivo effects of tg-FSH on Sertoli cell development in hpg testes. In contrast to a recent report that described no difference between the morphology of Sertoli cells in hpg and normal mouse testes (40), we observed morphological characteristics that showed hpg testes contain undifferentiated immature-like Sertoli cells, which included the presence of more irregularly shaped smaller nuclei and cells lacking the tripartite nuclear structures found in normal mature Sertoli cells (33). Our earlier studies showed that neonatal treatment of hpg mice with rhFSH produced a doubling of Sertoli cell numbers (30), whereas rhFSH treatment of weanling and adult hpg animals had no effect on final Sertoli cell numbers, with long-term actions compromised by the detection of antibodies to rhFSH (24). In the current study, tg-FSH expression in ß.6 hpg males stimulated Sertoli cell proliferation to a similar extent observed in hpg mice treated neonatally with exogenous rhFSH (30); however, the overall testis size was considerably greater in response to tg-FSH. This greater mitogenic response to tg-FSH is likely to reflect the previous limitations of exogenous rhFSH treatment, including the inability to deliver exogenous rhFSH prenatally and its postnatal immunogenicity (24, 30). The present effect of tg-FSH is even more striking when allowing for the moderate serum tg-FSH levels in ß.6 mice because higher tg-FSH levels in other tg lines further stimulated testis growth (17). The increased proliferation of immature Sertoli cells in response to perinatal exposure of tg-FSH provides strong evidence to support the proposal that the mitogenic role of FSH is maximal during the perinatal epoch of testicular development (28, 30).
Our present study revealed that androgen actions alone can stimulate Sertoli cell proliferation and maturation, in the absence of circulating tg-FSH. The Sertoli cells of T-treated testes resembled those observed in either FSH-stimulated or normal testes, which were found to contain the tripartite nuclear structure characteristic of mature Sertoli cells (33). It is unlikely that these androgen-stimulated Sertoli cell effects were due to endogenous FSH activity because mouse FSH remains undetectable following androgen treatment of hpg males (13). Therefore, the hpg paradigm can provide new insights into androgen-specific actions, which are more difficult to predict in classical models of pituitary disruption using antagonists or hypophysectomy because androgen-induced rises in circulating FSH are reported in some (9, 11) but not other (5, 6) rodent studies. The androgen-induced Sertoli cell maturation in hpg testes in the absence of FSH is also supported by observations in FSH receptor-deficient, androgen-replete mice of somatic Sertoli cells of normal appearance at 6–7 wk of age (15).
In contrast, previous work (24) did not find a significant mitogenic effect of androgens on Sertoli cells in hpg testes. Our reassessment of the independent mitogenic activity of T on Sertoli cells is likely to reflect the higher precision of the current optical dissector approach, which allowed more direct estimation of cell numbers in defined volumes of fixed testis sections, with less mathematical extrapolation than the previous physical dissector procedure (13) as well as increased animal numbers in each experimental group to reduce variance in the stereological analysis. It is noteworthy that combined FSH and androgen actions further increased Sertoli cell proliferation above the maximal androgenspecific effect but was not significantly greater than the effects of FSH alone. It is possible that addition neonatal T treatment together with the present postweaning administration of T may further increase the mitogenic effect of androgens on Sertoli cells. However, the total Sertoli cell numbers found in T-treated FSH-hpg testes were approximately 70% of normal testes, which are more likely to be limited by the tg-FSH levels in the ß.6 line because we recently showed testicular development was positively correlated with increasing circulating tg-FSH levels (17). Therefore, our present study has revealed that either FSH or androgen can independently stimulate Sertoli cell proliferation, although our findings suggest FSH has the dominant role as a perinatal Sertoli cell mitogen.
The effect of tg-FSH on distinct germ cell populations has greatly extended our previous observation that early germ cell development was enhanced after neonatal rhFSH treatment (24). The LH-independent FSH response significantly increased postnatal spermatogonial proliferation in hpg testes to a level that established a normal spermatogonia/Sertoli cell ratio. This developmental FSH effect on the mitotic germ cell population is consistent with the reduction of spermatogonia numbers after FSH-specific immunoneutralization (41). In contrast to FSH actions, T treatment had no significant effect on spermatogonia proliferation in hpg testes, either alone or in combination with tg-FSH, and the spermatogonia/Sertoli cell ratio remained unchanged. Classical ablation-replacement studies to examine androgen effects on early germ cell survival, on a background of degenerating spermatogenesis, have produced inconsistent conclusions that androgen actions are either involved (4, 5) or not sufficient (8, 31) for maintaining spermatogonial numbers. The reasons for such differences are unresolved but may be due to several variations between these spermatogenic regression models, including the efficiency or time course of hormonal deprivation and purity of hormone replacement. Our current paradigm is aimed at selective hormonal effects on spermatogenic development in rudimentary hpg testes. Although it is possible that additional neonatal androgen activity may be required to increase spermatogonia numbers, our earlier work (30) showed no significant effect of supplementary neonatal androgen treatment on spermatogonia numbers. Therefore, our findings indicate androgen actions have little effect on initiating early mitotic germ cell development in the postnatal hpg testes.
Germ cell development in the untreated gonadotropin-deficient hpg testis remains blocked at the pachytene stage of meiosis (13, 29), as it does in the androgen-resistant tfm phenotype (42). Transgenic FSH induced the maturation of meiotic spermatocytes in hpg testes. Selective FSH or T activity in the hpg testis each dramatically increased the absolute numbers of early (Pl-Z) spermatocytes to equivalent extents. Furthermore, their combined and apparently additive effects were able to fully establish a normal G/S ratio for these early meiotic cell types. Similar hormonal effects were observed for more advanced meiotic spermatocytes, noting the increased numbers of PS in T-treated hpg testes were 2-fold greater than the elevated PS level of androgen-deficient tg-FSH testes (of line ß.6 males). However, stimulation by both tg-FSH and T resulted in a marked synergistic effect on total PS numbers, which restored the G/S ratio for PS to normal. Thus, these findings show that both FSH and androgens have important roles in meiotic germ cell development, although androgens alone have a greater impact on later-stage meiosis. This contrasts with earlier work using exogenous rhFSH (24), which did not detect any additive germ cell effects during FSH and androgen-induced spermatogenesis. As mentioned above, this difference is likely to be due to the previous postnatal administration of cross-species gonadotropin, which may be constrained by the development of antibodies, as has been reported for both rhFSH (24) and human chorionic gonadotropin (43) studies.
The current observations that FSH actions in vivo, presumably via the Sertoli cell, can independently induce a small degree of postmeiotic haploid spermatid formation is supported by findings in mice with an inactivated LH receptor gene, which display arrested spermatogenesis at (27) or just beyond (26) the round spermatid stage despite dramatically reduced serum and testicular T levels. It is noteworthy that in all these mouse models, the testes would have experienced autonomous prenatal androgen production by Leydig cells independently of LH (44), providing a distinct developmental difference to another mouse model, the tfm male, with selective and permanent androgen insensitivity because of a mutated androgen receptor gene (42) and germ cell development arrested at first meiotic division (45). A related consideration in all these androgen-deficient models is the undescended testes. It is possible this may lead to an underestimation of postmeiotic tg-FSH actions in hpg testes if this incomplete descent was deleterious to testis development. However, cellular development in hpg testes (regardless of tg genotype) did not display the histopathological features of cryptorchidism such as degenerating Sertoli cells and spermatogonial arrest (46), and adult T-treated hpg testes develop qualitatively complete postnatal spermatogenesis at any age, suggesting there is no progressive effect of incomplete descent on the testicular developmental potential as would be expected for true cryptorchidism.
Although meiotic progression was minimal under FSH actions alone, T treatment alone stimulated significant levels of meiotic completion and qualitatively complete spermiogenesis in hpg testes. A combined FSH-androgen response had a marked synergistic effect on postmeiotic spermatid formation and development beyond a simple additive effect of the individual hormonal responses. Interestingly, this synergistic FSH-androgen effect increased during spermatogenesis as germ cells progressed from PS through to elongated spermatids and provides compelling evidence that FSH can have a significant impact on postmeiotic germ cell development. Although the combined FSH-androgen response was sufficient to restore the G/S ratio of all meiotic germ cells to wild-type levels, it was not able to fully restore the G/S ratios of spermatids to normal levels. Considering the prevailing view that a fixed germ cell carrying capacity exists for each Sertoli cell, it follows that this capacity was not reached for the spermatids in T-treated or combined tg-FSH/T-treated hpg animals. These reduced postmeiotic G/S ratios may be due to incomplete androgen actions during development because hpg animals were androgen deficient for 3 wk before T treatment at weaning, and our previous work (30) showed neonatal T administration significantly increased postnatal T effects on spermatid development in hpg testes. In addition, higher levels of tg-FSH may also increase the spermatid/Sertoli cell ratio because the tg-FSH levels of the ß.6 line were able to nonsignificantly increase the positive androgen effect on this ratio. Although the underlying Sertoli cell mechanism required to achieve a normal germ cell carrying capacity remains unknown, our study clearly demonstrates that increased Sertoli cell numbers alone are insufficient to generate the optimal postmeiotic germ cell support.
Quantification of serum and intratesticular T concentrations confirmed the T deficiency of the hpg mouse and the subsequent rise in both measurements following T treatment. Expression of tg-FSH did not alter serum or testicular T concentrations of hpg males. In addition, the serum T levels in hpg and tg-FSH hpg males were comparable with those reported in male LH receptor knockout mice (26, 27), although in contrast to the LH receptor knockout model, rescue by LH/human chorionic gonadotropin could be studied in our model. Furthermore, T treatment in the hpg mouse is not confounded by androgenic repression of endogenous FSH secretion because the tg-FSH promoter construct was chosen to be nonrepressible by androgens. Transgenic FSH expression significantly increased the testicular T content in hpg males, an effect that was not observed in previous studies that treated neonatal (30) or postnatal (23, 24) hpg males with rhFSH. However, the unaltered intratesticular T concentrations and the observed spermatogenic populations in Tg-FSH hpg testes were different (pre- vs. postmeiotic) from the effects of T administration, and the additive or synergistic effects of combined tg-FSH and T treatment on testicular and germ cell development indicated that FSH-stimulated actions were not due to changes to intratesticular T. The ability of FSH to stimulate the steroidogenic capacity of the hpg testis has previously been reported (23), and the present elevation in total T content may reflect perinatal tg-FSH actions or the continual presence of bioactive tg-FSH providing some undefined trophic support to Leydig cells. The testicular T content was not correlated with the levels of tg-FSH (Haywood, M., and C. M. Allan, unpublished data), and it remains to be determined whether this effect was due to increased Leydig cell numbers or volume in tg-FSH testes.
In summary, we have presented a detailed dissection of the specific and combined hormonal actions of FSH and androgen in male germ cell and Sertoli cell development using our novel transgenic FSH model (17). Our analysis has revealed that FSH has the dominant role in Sertoli cell development, although androgens can also independently stimulate Sertoli cell maturation and proliferation. This may explain the clinical observation that men with genetic defects in FSH action consistently demonstrate small testes because of a deficient complement of Sertoli cells (47), but as long as they retain normal androgen action, they can develop sufficient spermatogenesis to display fertility. By contrast, men congenitally lacking both LH and FSH have persistently small testes but spermatogenesis can be initiated by gonadotropin treatment (48). The present findings showed that the FSH response had the predominant effect on early mitotic germ cell (spermatogonia) proliferation, compared with androgen actions. Furthermore, we provide definitive evidence that FSH actions combine synergistically with androgens to have a marked effect on both meiotic and postmeiotic germ cell maturation. This unique in vivo experimental paradigm will provide a valuable foundation for future investigations directed at the molecular level to explore the underlying biochemical pathways that mediate these selective hormonal actions in spermatogenesis.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
Abbreviations: G/S ratio, Ratio of germ cells to Sertoli cells; h, human; hpg, hypogonadal; Pl-Z, preleptotene to zygotene; PS, pachytene spermatocyte; rh, recombinant human; RIP, rat insulin promoter; T, testosterone; tg, transgenic; tg-FSH, tg hFSH.
Received July 12, 2002.
Accepted for publication October 23, 2002.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
This article has been cited by other articles:
 |
|
 |
|
  M. H. Abel, P. J. Baker, H. M. Charlton, A. Monteiro, G. Verhoeven, K. De Gendt, F. Guillou, and P. J. O'Shaughnessy Spermatogenesis and Sertoli Cell Activity in Mice Lacking Sertoli Cell Receptors for Follicle-Stimulating Hormone and Androgen Endocrinology, July 1, 2008; 149(7): 3279 - 3285. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 F. De Luca, V. Mitchell, M. Wasniewska, T. Arrigo, M. F. Messina, M. Valenzise, L. de Sanctis, and N. Lahlou Regulation of spermatogenesis in McCune-Albright syndrome: lessons from a 15-year follow-up. Eur. J. Endocrinol., June 1, 2008; 158(6): 921 - 927. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. M. Ruwanpura, R. I. McLachlan, P. G. Stanton, and S. J. Meachem Follicle-Stimulating Hormone Affects Spermatogonial Survival by Regulating the Intrinsic Apoptotic Pathway in Adult Rats Biol Reprod, April 1, 2008; 78(4): 705 - 713. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
K. J. McTavish, M. Jimenez, K. A. Walters, J. Spaliviero, N. P. Groome, A. P. Themmen, J. A. Visser, D. J. Handelsman, and C. M. Allan Rising Follicle-Stimulating Hormone Levels with Age Accelerate Female Reproductive Failure Endocrinology, September 1, 2007; 148(9): 4432 - 4439. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 B. Furnes and J. Schimenti Fast forward to new genes in mammalian reproduction J. Physiol., January 1, 2007; 578(1): 25 - 32. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M.-Y. Tsai, S.-D. Yeh, R.-S. Wang, S. Yeh, C. Zhang, H.-Y. Lin, C.-R. Tzeng, and C. Chang Differential effects of spermatogenesis and fertility in mice lacking androgen receptor in individual testis cells PNAS, December 12, 2006; 103(50): 18975 - 18980. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 K. L. Matthiesson, R. I. McLachlan, L. O'Donnell, M. Frydenberg, D. M. Robertson, P. G. Stanton, and S. J. Meachem The Relative Roles of Follicle-Stimulating Hormone and Luteinizing Hormone in Maintaining Spermatogonial Maturation and Spermiation in Normal Men J. Clin. Endocrinol. Metab., October 1, 2006; 91(10): 3962 - 3969. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
L. Soriano-Guillen, V. Mitchell, J.-C. Carel, P. Barbet, M. Roger, and N. Lahlou Activating Mutations in the Luteinizing Hormone Receptor Gene: A Human Model of Non-Follicle-Stimulating Hormone-Dependent Inhibin Production and Germ Cell Maturation J. Clin. Endocrinol. Metab., August 1, 2006; 91(8): 3041 - 3047. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 K. L. Matthiesson and R. I. McLachlan Male hormonal contraception: concept proven, product in sight? Hum. Reprod. Update, July 1, 2006; 12(4): 463 - 482. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 R Rathi, A Honaramooz, W Zeng, R Turner, and I Dobrinski Germ cell development in equine testis tissue xenografted into mice. Reproduction, June 1, 2006; 131(6): 1091 - 1098. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C. M. Allan, A. Garcia, J. Spaliviero, M. Jimenez, and D. J. Handelsman Maintenance of Spermatogenesis by the Activated Human (Asp567Gly) FSH Receptor During Testicular Regression Due to Hormonal Withdrawal Biol Reprod, May 1, 2006; 74(5): 938 - 944. [Abstract] [Full Text] [PDF]
|
|
|
|
