Estrogens and male reproduction: a new concept.

The mammalian testis serves two main functions: production of spermatozoa and synthesis of steroids; among them estrogens are the end products obtained from the irreversible transformation of androgens by a microsomal enzymatic complex named aromatase. The aromatase is encoded by a single gene (cyp19) in humans which contains 18 exons, 9 of them being translated. In rats, the aromatase activity is mainly located in Sertoli cells of immature rats and then in Leydig cells of adult rats. We have demonstrated that germ cells represent an important source of estrogens: the amount of P450arom transcript is 3-fold higher in pachytene spermatocytes compared to gonocytes or round spermatids; conversely, aromatase activity is more intense in haploid cells. Male germ cells of mice, bank voles, bears, and monkeys express aromatase. In humans, we have shown the presence of a biologically active aromatase and of estrogen receptors (alpha and ss) in ejaculated spermatozoa and in immature germ cells in addition to Leydig cells. Moreover, we have demonstrated that the amount of P450arom transcripts is 30% lower in immotile than in motile spermatozoa. Alterations of spermatogenesis in terms of number and motility of spermatozoa have been described in men genetically deficient in aromatase. These last observations, together with our data showing a significant decrease of aromatase in immotile spermatozoa, suggest that aromatase could be involved in the acquisition of sperm motility. Thus, taking into account the widespread localization of aromatase and estrogen receptors in testicular cells, it is obvious that, besides gonadotrophins and androgens, estrogens produced locally should be considered to be physiologically relevant hormones involved in the regulation of spermatogenesis and spermiogenesis.

Estrogens and male reproduction: a new concept

The mammalian testis serves two main functions: production of spermatozoa and synthesis of steroids; among them estrogens are the end products obtained from the irreversible transformation of androgens by a microsomal enzymatic complex named aromatase. The aromatase is encoded by a single gene (cyp19) in humans which contains 18 exons, 9 of them being translated. In rats, the aromatase activity is mainly located in Sertoli cells of immature rats and then in Leydig cells of adult rats. We have demonstrated that germ cells represent an important source of estrogens: the amount of P450arom transcript is 3-fold higher in pachytene spermatocytes compared to gonocytes or round spermatids; conversely, aromatase activity is more intense in haploid cells. Male germ cells of mice, bank voles, bears, and monkeys express aromatase. In humans, we have shown the presence of a biologically active aromatase and of estrogen receptors (alpha and ß) in ejaculated spermatozoa and in immature germ cells in addition to Leydig cells. Moreover, we have demonstrated that the amount of P450arom transcripts is 30 percent lower in immotile than in motile spermatozoa. Alterations of spermatogenesis in terms of number and motility of spermatozoa have been described in men genetically deficient in aromatase. These last observations, together with our data showing a significant decrease of aromatase in immotile spermatozoa, suggest that aromatase could be involved in the acquisition of sperm motility. Thus, taking into account the widespread localization of aromatase and estrogen receptors in testicular cells, it is obvious that, besides gonadotrophins and androgens, estrogens produced locally should be considered to be physiologically relevant hormones involved in the regulation of spermatogenesis and spermiogenesis.

Aromatase in testis: expression and role in male reproduction.

The mammalian testis serves two main functions: production of spermatozoa and synthesis of steroids, among them estrogens are the end products obtained from the irreversible transformation of androgens by aromatase (P450arom). In the rat the pattern of P450arom expression differs among the testicular somatic cell types according to age; in addition, we have shown that gonocytes, spermatogonia, spermatocytes (preleptotene, pachytene), spermatids and spermatozoa, represent an important source of estrogens; the expression of aromatase is three-fold higher in pachytene spermatocyte (PS) compared to gonocytes. In man both Leydig cells and immature germ cells (PS and round spermatids, RS) as well as ejaculated spermatozoa expressed a biologically active aromatase revealed as a single band of 49 kDa on western blots. Up today P450arom has been demonstrated in male germ cells of all mammals so far studied (mice, bank vole, bear and monkey). The aromatase gene is highly conserved and is unique in humans; its expression is regulated in a cell-specific manner via the alternative use of various promoters located in the first exon. Nevertheless, data concerning the regulation of P450arom especially in germ cells are scarce. We have demonstrated that TGFbeta inhibits the expression of Cyp19 in PS and RS via the SMAD pathway although TNFalpha exerts a stimulatory role in PS, which is amplified in presence of dexamethasone. It is noteworthy that dexamethasone alone exerts a positive effect on Cyp19 expression in PS and a negative one in RS. Cyclic AMP is also a positive regulator of P450arom gene expression in germ cells. In addition, we have shown that androgens and estrogens modulate Cyp19 gene expression, whatever the testicular cell type studied, which favored the presence of androgens and estrogens responsive elements on the Cyp19 gene promoter(s). Moreover, in presence of seminiferous tubules conditioned media, the amount of aromatase transcripts is increased in Leydig cells, therefore, suggesting that other locally produced modulators are involved in the regulation of the aromatase gene expression and among them the liver receptor homolog-1 (LRH-1) from germ cells origin is concerned. Using RACE-PCR we have confirmed that promoter II directs the expression of aromatase gene, whatever the testicular cell type studied in the rat but the involvement of another promoter, such as PI.4 is suggested. Finally, the aromatase gene is constitutively expressed both in somatic and germ cells of the testis and the identification of the promoter(s) concerned as well as their detailed regions which direct(s) the expression of Cyp19 gene is obviously very important but largely unknown especially according to the ontogeny of the male gonad.

Regulation of aromatase gene expression in Leydig cells and germ cells.

The ability of the testis to convert irreversibly androgens into estrogens is related to the presence of a microsomal enzymatic complex named aromatase. Although somatic cells and germ cells (GC) have the capacity to produce estrogens the regulation of the CYP19 gene expression in adult rat testicular cells and specially in freshly purified Leydig cells, pachytene spermatocytes (PS) and round spermatids (RS) is not fully understood. In the present study we have analyzed the putative effects of steroid hormones, transforming growth factor beta (TGFbeta), cytokine (tumor necrosis factor alpha, TNFalpha) and dexamethasone (Dex) on CYP19 expression in these purified testicular cells from adult rat. In parallel the biological role of seminiferous tubules and Sertoli cells conditioned media on the expression of aromatase was studied. Using a highly specific quantitative competitive RT-PCR we established that testosterone (T) enhances CYP19 gene expression in Leydig cells and germ cells, and augments the estradiol outputs. The non-aromatizable androgen 5alpha-DHT induces the same effect as T on P450 aromatase (P450arom) gene expression but was inefficient on the estradiol output. In PS and RS an inhibitory effect on CYP19 gene transcription was observed with TGFbeta (1 ng/ml) alone or in combination with T. Conversely, the addition of TNFalpha (20 ng/ml) increases the P450arom transcription in PS although an inhibitory effect is observed in RS. Together with T, TNFalpha decreases the amount of P450arom mRNA in PS and RS. In PS we found that Dex regulates positively CYP19 expression and negatively in RS. Furthermore in PS a synergistic effect of Dex and TNFalpha on P450arom mRNA expression was observed whereas an additive one was recorded for RS. Therefore in germ cells TNFalpha likely enhances expression of aromatase through promoter PI.4 in PS, possibly via an AP1 site upstream the GAS element, while in RS TNFalpha requires glucocorticoids as a co-stimulator to increase CYP19 gene expression. Finally in presence of seminiferous tubules or Sertoli cell conditioned media, the amount of aromatase transcripts is increased in both Leydig cells and germ cells therefore suggesting that other locally produced modulators, yet unknown, but from Sertoli cell origin, are concerned in the regulation of the aromatase gene expression in rat testicular cells. In summary, using an in vitro model of mature rat Leydig cells, pachytene spermatocytes and round spermatids, we have shown that several factors direct the expression of the aromatase gene and it is obvious that not only promoter PII but also promoter PI.4 are concerned.

Steroids control the aromatase gene expression in purified germ cells from the adult male rat.

The ability of the testis to convert irreversibly androgens into estrogens is related to the presence of a microsomal enzymatic complex named aromatase. In rodents, germ cell production of estrogens is known, although the regulation of the cytochrome p450 aromatase (p450 arom) gene expression is not completely elucidated. In the present study, we have investigated the putative effects of steroids (testosterone, 5alpha-dihydrotestosterone (DHT) and estradiol) on Cyp19 gene expression in purified adult rat pachytene spermatocytes (PSs) and round spermatids (RSs). Using a highly specific quantitative competitive RT-PCR method we established that testosterone enhances in a dose- and time-related manner aromatase gene expression in PSs and RSs; 5alpha-DHT induces the same effect. Furthermore, testosterone increases the estradiol output in both germ cell populations whereas 5alpha-DHT was inefficient, therefore suggesting that the effect of androgens on p450 arom gene transcription was independent of estrogen formation. In fact estradiol inhibits the Cyp19 gene expression in PSs and RSs. ICI 182780, an estrogen receptor antagonist, has no effect on testosterone-stimulated aromatase expression in PSs and RSs. By contrast, ICI 182780 suppresses the inhibitory effect of estradiol on p450 arom mRNA expression in PSs and RSs. Similarly, nilutamide, a non-steroidal anti-androgen specific for androgen receptors, abolishes the testosterone-stimulated aromatase expression in PSs and RSs. These observations show that androgens up-regulate aromatase gene expression in purified adult rat germ cells whereas estrogens exert an opposite effect, which may suggest the presence of androgen and estrogen responsive elements on the aromatase promoter(s).

Regulation of aromatase gene expression in purified germ cells of adult male rats: effects of transforming growth factor beta, tumor necrosis factor alpha, and cyclic adenosine 3',5'-monosphosphate.

Estrogens are key regulators of sexual differentiation and development in vertebrates. The P450 aromatase (P450arom) is the steroidogenic enzyme responsible for the synthesis of estrogens from androgens. In the adult rat testis, aromatase transcripts and activity have been observed in somatic cells and germ cells, including pachytene spermatocytes (PS) and round spermatids (RS), but little is known concerning regulation of the aromatase gene expression, especially in germ cells. The quality of germ cell preparations was assessed by the absence of androgen-binding protein and stem cell factor transcripts, two specific markers for Sertoli cells. By employing a competitive quantitative reverse transcriptase-polymerase chain reaction technique, we confirmed that germ cells contained P450arom transcripts and demonstrated that the aromatase gene was up-regulated by cAMP. Conversely, transforming growth factor (TGF) beta1 inhibited Cyp19 gene expression in a dose- and a time-dependent manner in both PS and RS. The addition of tumor necrosis factor (TNF) alpha to purified germ cells induced an increase of the amount of P450arom mRNA in PS, although an inhibitory effect was observed in RS. When PS were treated with dexamethasone (Dex), a similar enhancement of the aromatase transcript level was observed, whereas an inhibitory effect was recorded for RS. Furthermore, in either TGFbeta1- or TNFalpha-treated germ cells, the addition of Dex stimulated the aromatase gene transcription. Experiments using 5' rapid amplification of cDNA ends suggested that promoter PII is mainly concerned in the regulation of the aromatase gene expression in germ cells of adult male rats; however, the presence of other promoters could not be excluded.

Complementary approaches demonstrate that cellular aromatization in the bank vole testis is related to photoperiod.

A growing body of evidence indicates that germ cells, at least in several mammalian species, are responsible for estrogen formation since they possess active aromatase. In seasonally breeding rodent, the bank vole, the length of photoperiod seems to be the primary environmental factor regulating annual changes in the reproductive activity. However, in this species gonadal steroidogenesis is still not well understood, neither the site of aromatization in testicular cells. In the bank vole testis, aromatase visualized by immunohistochemistry was found in Leydig cells, Sertoli cells, and germ cells: especially in spermatocytes and spermatids. Moreover, in the immuno-electron microscopic study, gold particles indicating aromatase were observed over the cytoplasm of elongated spermatids. The presence of aromatase and the activity of this enzyme were found in microsomal preparations of the whole testes and those of seminiferous tubules. This was measured by means of Western blot and the biochemical assay with tritiated androstenedione, respectively. Additionally, using radioimmunological assays testosterone and estradiol concentrations in homogenates were detected. All the studied parameters revealed close correlation with the length of photoperiod being evidently higher in animals kept in the long day conditions when compared with those from short light cycles.

Immunolocalization and activity of aromatase in the bank vole testes.

A direct approach to identify the cellular source of P450 aromatase in the bank vole testes (seasonally breeding rodents) is the use of immunohistochemistry with a specific antibody that recognizes this enzyme. To confirm the presence of functional aromatase, its activity was measured in microsomal preparations of whole testes and of seminiferous tubules by means of biochemical assay with tritiated androstenedione. The assay was validated using increasing concentrations of both microsomal preparations. Immunoreactive aromatase was found in Leydig cells, Sertoli cells, and germ cells, especially in spermatocytes and spermatids. The aromatase activity was present in microsomal fractions of whole testis and seminiferous tubules. The immunolocalization of P450 aromatase and aromatase activity have been found as photoperiod-dependent.

Aromatase expression in male germ cells.

The cytochrome P450 aromatase (P450arom) is the terminal enzyme responsible for the formation of estrogens from androgens. According to the age, aromatase activity has been measured in immature and mature rat Leydig cells, as well as in Sertoli cells whereas in pig, ram and human the aromatase is mainly present in Leydig cells. In the rat testis, we have immunolocalised the P450arom not only in Leydig cells but also in germ cells and especially in elongated spermatids. Related to the stage of germ cell maturation, we have shown that the level of P450arom mRNA transcripts decreases, it is much more abundant in younger than in mature germ cells whereas the aromatase activity is two- to four-fold greater in spermatozoa when compared to the two other enriched-germ cell preparations. Moreover, we have reported the existence of alternative splicing events of P450arom mRNA in pachytene spermatocytes and round spermatids giving rise to two isoforms lacking the last coding exon which, therefore, cannot encode functional aromatase molecules. In rat germ cells, the aromatase gene expression is not only under androgen control but also subjected to cytokine (TNFalpha) and growth factor (TGFbeta) regulation. In the bank-vole testis, we have evidenced a synchronisation between a fully developed spermatogenesis and a strong positive immunoreactivity for both P450arom and estrogen receptor (ERbeta) in spermatids. Therefore, the aromatase gene expression and its translation in a fully active protein in rodent germ cells evidence an additional site for estrogen production within the testis. Our recent data showing that human ejaculated spermatozoa expressed specific transcripts for P450arom reinforced the observations reported in germ cells of other mammalian species. Together with the widespread distribution of ERs in testicular cells these data bring enlightenments on the hormonal regulation of male reproductive function. Indeed these female hormones (or the ratio androgens/estrogens) do play a physiological role (either directly on germ cells or via testicular somatic cells) in the maintenance of male gonadal functions and obviously, several steps are concerned particularly the spermatid production and the epididymal sperm maturation.