Seminal transforming growth factor-ß in normal and infertile men

Bernadette Loras1, Florence Vételé1, Ahmed El Malki1, Jacques Rollet2, Jean-Claude Soufir3 and Mohamed Benahmed1,4

1 Institut National de la Santé et de la Recherche Médicale, INSERM U407, Faculté de Médecine Lyon-Sud, B.P. 12, F-69961 Oullins cedex, 2 Institut Rhône-Alpin, F-69400 Bron and 3 Laboratoire de Biologie Andrologique, CHU Bicêtre, F-94275 Le Kremlin-Bicêtre, France


    Abstract
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 References
 
Transforming growth factor-ß (TGFß) is a cytokine with autocrine and paracrine action in the testis and potent immunoregulatory and anti-inflammatory activities. In the present study, we examined the concentration of latent (acid-activatable) and free (active) TGFß in seminal plasma from normal subjects (n = 23) and infertile (n = 40) patients, by using a TGFß specific immunoenzymological assay, and a bioassay (CCL64 cell line growth inhibition) detecting any form of TGFß. Free TGFß1 was present in normal subjects at a concentration (1.82 ± 1.06 ng/ml) close to that known to give maximal stimulation in vitro. In pathological groups, the mean concentrations were not significantly different from the normal ones. Latent TGFß1 was present in normal seminal plasma at a high concentration (92.4 ± 29.2 ng/ml). In subjects with pathologies of both testis and genital apparatus, or with epididymal occlusion, mean latent TGFß1 concentrations were normal, whereas transferrin concentrations were lower. The concentrations found in the epididymal occlusion group indicate that TGFß1 is, for a large part, secreted by the genital tract. In the testicular pathology group, TGFß1 concentrations were 130.7 ± 61.2 ng/ml, a mean not statistically different from normal, although higher. No differences were found between patients with high and normal blood plasma follicle stimulating hormone, and this is consistent with the notion that most TGFß1 in seminal plasma is not of testicular origin. The TGFß bioassay ensured that immunologically detected TGFß was present in a bioactive or bioactivatable form. Furthermore, the values found in normal and pathological seminal plasmas were usually higher than those detected by the immunoassay, suggesting that other forms of TGFß might be present. Together, the present data show that very large amounts of TGFß are present in human seminal plasma. The TGFß ligand assay in the seminal plasma appears to indicate no differences between normal and infertile subjects.

Key words: human/infertility/seminal plasma/TGFß


    Introduction
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 References
 
Transforming growth factors-ß (TGFß) are members of a large family of peptidic growth factors. The TGFß functional system is complex as it includes at least three ligands: TGFß1, TGFß2, TGFß3, binding proteins, and two receptors. Secreted TGFß is non-covalently linked to different proteins, leading to an inactive form from which the ligands need to be released for binding to their cognate receptors. The binding proteins comprise LAP (latency associated protein) and the large latent TGFß1 complex where LAP binds to an additional protein. These two forms are produced by many cell types, and both can be activated to release mature TGFß1. In blood serum, mature TGFß1 released from these complexes is bound covalently to {alpha}2-macroglobulin, and is thus inactivatable. A small part of TGFß1 is linked by a non-covalent interaction with this protein and is thus activatable. In vitro, TGFß can be released in an active free form (`activated') from these complexes by a variety of treatments, and usually transient acidification is used. Physiologically, the activation is probably carried out by enzymes. TGFß have a very large spectrum of biological actions both in vivo and in vitro, influencing cell proliferation, differentiation and functional activity, usually by inhibiting these activities in epithelial cells, and stimulating them in cells of mesenchymal origin. Immunoregulatory, anti-inflammatory and healing activities, through the regulation of cellular matrix component formation, are also essential functions of the TGFß family (reviews: Massagué, 1990Go; Harpel et al., 1992Go; Sporn and Roberts, 1992Go; Matzuk, 1995Go).

In-vitro experiments conducted in several different species suggested that TGFß might be implicated in the local control of testicular development and functions. Indeed, TGFß as well as their receptors are expressed in the testis both in terms of mRNA and protein. TGFß have been shown to control the proliferation of mouse primordial germ cells and the functions of Leydig and Sertoli cells by inhibiting gonadotrophin effects, and to control peritubular myoid cells by stimulating extracellular matrix production (reviews: Benahmed, 1996Go; Gnessi et al., 1997Go). TGFß1 might also inhibit prostate growth both in vivo (Ilio et al., 1995Go; Nishi et al., 1996Go) and in vitro (Kim et al., 1996Go) and its secretion by rat prostatic cells was demonstrated in vitro (Steiner et al., 1994Go). Finally, in humans, both free and latent forms of TGFß1 were found on spermatozoa (Chu et al., 1996Go). In addition, TGFß may exert an immunosuppressive action in the male reproductive tract as in other tissues, but few data are available on this point (Pöllänen et al., 1993Go).

Following these different observations, our aim in the present study was to identify and measure TGFß in seminal plasma from normal men and infertile patients. Seminal plasma derives from secretions of the testes, epididymides and sex accessory glands.

TGFß has been demonstrated to be present in pools of human seminal plasma, by testing for inhibition of CCL64 cell multiplication (Lokeshwar and Block, 1992Go; Nocera and Chu, 1993Go), or by using an immunoassay (Nocera and Chu, 1995Go). Another immunoassay study involved three samples of seminal plasma from normal patients (Shrivastava et al., 1996Go). These data are based on a limited number of samples, and as far as we know, no data are available on human male reproductive pathology.

In this study, we quantified TGFß1 in human seminal plasma from 63 subjects by using a specific immunological assay, and checked that the TGFß were in an active or activatable form, by using a bioassay. We also investigated the relationship between seminal plasma TGFß and the normal or diseased status of the patients.


    Materials and methods
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 References
 
Subjects
A 3–5 day sexual abstinence was required from the subjects before semen collection. After liquefaction and centrifugation, seminal plasma samples were kept at –20°C until assayed.

Men were allocated into groups (Table IGo) according to clinical and biological data: testicular secretory pathology (15 cases), epididymal occlusion (six cases), and mixed pathology, i.e. patients with both secretory and non-obstructive genital tract pathology (19 cases). In 23 cases, no abnormalities were found; clinical examinations and spermiograms were normal (spermatozoa/ml >20x106; motility >40%; normal forms >60%), and couple infertility was related to the female partner. This group was considered as a control population, although the fertility of all subjects of this `control' population could not be demonstrated, since most couples went to the clinician for primary infertility.


View this table:
[in this window]
[in a new window]
 
Table I. Clinical data for 63 subjects involved in study of seminal transforming growth factor-ß
 
Spermiograms were carried out according to World Health Organization (1987) recommendations. As a part of the biological exploration, blood follicle stimulating hormone (FSH) was assayed in all subjects (normal values: 1–7 mU/ml, FSH-Coatria; BioMérieux, Marcy l'Etoile, France).

In most patients, all TGFß concentrations were measured by using the same seminal plasma sample. Due to small seminal plasma volumes, 11 patients were given only a free TGFß1 assay.

TGFß assays
Immunological assay

TGFß1 concentrations were measured in the seminal plasma of each patient by using a specific commercial kit (Quantikine, R & D systems, Abingdon, UK). This assay is based on the quantitative sandwich enzyme immunoassay technique, with soluble TGFß receptor type II and an enzyme-linked polyclonal TGFß1-specific antibody. According to the manufacturer, a 0.9% cross-reactivity was observed with TGFß3, but no cross-reactivity was observed with TGFß2.

Acidic activation of the samples was performed according to the manufacturer's directions for blood serum: addition of 1 volume of 2.5 N acetic acid/10 mol/l urea, 15 min incubation at room temperature, neutralization by addition of 1 volume 2.7 N NaOH/1 mol/l HEPES. In these samples, `activatable' TGFß1 was the sum of free TGFß1 and TGFß1 released from latent complexes. These acidified samples were diluted from 1/100 to 1/800 and measured at four different dilutions (chemicals were obtained from Sigma Chemical Co., St. Louis, MO, USA).

For the free active TGFß1 assay, the acidic activation step was omitted. The samples were diluted at 1/4, 1/8 and 1/16 and measured. In all free and activated samples, a correct parallelism with the standard curve was obtained.

Biological assay

The CCL64 mink lung epithelial cell line assay was used. This test is based on TGFß-induced inhibition of cell proliferation, as measured by a decrease of the incorporation of [3H]thymidine (Amersham Pharmacia Biotech, Buckingham, UK) into DNA when TGFß are present. The growth of these cells is inhibited by all forms of TGFß.

CCL64 cells were cultured in Dulbecco's modified essential medium, supplemented with 10% fetal calf serum (FCS) and penicillin–treptomycin. The cells were trypsinized for the testing, and 3x104 cells were seeded per well (1.75 cm2). Standard TGFß1 and diluted seminal plasma samples were incubated with the cells for 48 h in the same medium but with 3% FCS. [3H]Thymidine (250 nCi) was added for the last 15 h. The medium was then removed, and the cells were washed with PBS, trichloracetic acid, water, and methanol, to eliminate the free [3H]thymidine. The cells were then dissolved in 0.5 mol/l NaOH, and a liquid scintillation counter was used to measure the radioactivity.

Samples of seminal plasma were or were not acidified, as described for the immunoassay, and diluted in culture medium for the assay. With no other treatment, parallelism with the standard curve was obtained only with samples at high dilutions, due to the presence of interfering substances in seminal plasma. For example, this might be likely due to interactions between spermine in seminal plasma and FCS generating toxic free radicals. In an effort to overcome this problem, we treated seminal plasma samples with PEG (polyethylene glycol 6000, 6% final concentration) before dilution in culture medium. This procedure precipitated some proteins. Free and activatable TGFß1 present in the sample could be quantified in the supernatant fraction after PEG treatment. In treated samples, intra-assay reproducibility varied from 6 to 24% from high to low levels, and the overall inter-assay reproducibility was 20.2%.

Transferrin determination
Transferrin was also measured in the same samples, as a marker for Sertoli cell activity. Transferrin was radioimmunoassayed by using human holotransferrin (Sigma, L'Isle d'Abeau, France), both as a standard and for iodination, which was performed by the chloramine T method (Greenwood et al., 1963Go). Radiolabelled product and free iodide were separated on a column of G 50 Sephadex (Pharmacia, Uppsala, Sweden), eluted with phosphate-buffered saline. A human transferrin antiserum (Sigma, T6265, batch 041H-8814) was used. This antiserum did not recognize lactoferrin, also present in seminal fluid. The radioimmunoassay was performed for 48 h, at 4°C, and free and bound radioactivity were separated using g-globulins and PEG (12.5% final concentration). Each sample was assayed in triplicate, at three different dilutions.

Statistical analysis
Results are presented as means ± SD. The Mann–Whitney test was used to determine statistical differences between means, and linear correlation coefficients (r) were used for correlation calculations.


    Results
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 References
 
Immunological TGFß1 and transferrin in human seminal plasma
In normal subjects, activatable TGFß1 ranged from 43.0 to 143.3 ng/ml (Table IIGo). Free TGFß1 varied from <0.5 to 4.54 ng/ml. When the two assays were performed in the same seminal fluid specimen, the percentage of free to activatable TGFß1 was 1.98 ± 1.60% (<0.70 to 6.12%), and no correlation was found between the concentrations of activatable and free TGFß1 (r = –0.07, not significant).


View this table:
[in this window]
[in a new window]
 
Table II. Immunological TGFß1 and transferrin values in human seminal plasma (means ± SD and ranges are presented)
 
Transferrin concentrations in this group were 30.7 ± 11.3 µg/ml. No correlation was found between activatable TGFß1 and transferrin concentrations (r = 0.265, not significant), nor between free TGFß1 and transferrin concentrations (r = 0.379, not significant).

In pathological groups, although activatable TGFß1 concentrations tended to be higher in occlusive and mixed pathological groups than in normal subjects, the difference was not statistically significant. In the secretory pathological group, the mean value was 130.7 ng/ml, versus 92.9 ng/ml in fertile subjects, still higher than in the other pathological groups. However, the Mann–Whitney test did not give a significant result. In this group, no difference was found between the subgroup with normal blood FSH (n = 8, TGFß1 = 130.8 ± 58.0 ng/ml) and the subgroup with elevated blood FSH concentrations (n = 5, TGFß1 = 130.7 ± 73.1 ng/ml).

Free TGFß1 concentrations were found to be not statistically different between normal and any pathological group.

Transferrin concentrations were lower in secretory, mixed, and occlusive pathological groups, as expected since it originates mainly from the testis. No correlation was found between transferrin and free or activatable TGFß1 concentrations.

Biological TGFß
The samples of acidified seminal plasma inhibited CCL64 cell growth. At high dilutions, a dose-dependent effect was demonstrated. At low dilutions, the effect of TGFß was counterbalanced probably by some growth-stimulating substances present in the seminal plasma, or a substance inhibiting TGFß activity. In the non-acidified samples, the inhibitory effect was also demonstrated (Figure 1Go, top). These substances were either present in low quantities or active only at high concentrations, since the sample was inactive when diluted. The inhibitory activity was removed after PEG precipitation (Figure 1Go, bottom), suggesting that it was protein-based. Moreover, after PEG precipitation, free and activatable TGFß were quantifiable, the assays yielding data paralleling the standard curve (Figure 1Go).



View larger version (23K):
[in this window]
[in a new window]
 
Figure 1. Effect of standard transforming growth factor-ß (TGFß1) and seminal plasma on CCL64 cell proliferation inhibition, as detected by inhibition of [3H]thymidine incorporation in the cells. (Top) Recombinant TGFß ({circ}), serially diluted activated (•) and non-activated ({blacksquare}) human seminal plasma. Mean ± SD of a representative experiment. (Bottom) As for Top, after seminal plasma partial purification (i.e. PEG purification). In both cases, seminal plasma samples were first diluted in the culture medium according to the data from immunological assays, from 1/50 to 1/150 according to the samples. From this point onward, serial 1/2 dilutions were performed, and used in the biological assay.

 
In addition, free and activatable TGFß concentrations were higher, as a rule, than the values measured by the immunological assay, both in men with normal spermiograms and in infertile men (Table IIIGo).


View this table:
[in this window]
[in a new window]
 
Table III. Immunological and biological values of TGFß in human seminal plasma
 

    Discussion
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 References
 
The aim of the present study was (i) to evaluate seminal TGFß1 concentrations (immunoassay); (ii) to check the bioactivity of TGFß as well as to detect other forms of seminal TGFß (bioassay) and (iii) to determine whether there are differences in TGFß concentrations in seminal plasma from infertile men, compared to fertile men.

We found high concentrations of both free and activatable TGFß1 in seminal plasma from fertile and infertile men, as detected by the immunoassay. Parallelism of samples with the standard curve suggests that seminal plasma TGFß1 and the recombinant TGFß1 used as a standard in the assay are antigenically similar.

Due to the low percentage of free TGFß1 present before acidification, values obtained for activatable TGFß1 may be considered as representative of the latent stock form. In our study, normal activatable TGFß1 concentrations varied largely, as seminal plasma parameters do. The normal latent TGFß1 concentrations obtained in the present study (92.4 ± 29.2 ng/ml) are similar to those reported (Shrivastava et al., 1996Go) (71.0 ± 32.1 ng/ml in three samples of normal seminal plasma) but lower than those obtained by Nocera and Chu (Nocera and Chu, 1995Go) (238 ng/ml in four pools of seminal plasma from vasectomized subjects). As for free TGFß1, we found lower values than Shrisvastava et al. (Shrivastava et al., 1996Go) (9.2 ng/ml). It is of interest to note that both in the present and earlier studies, high concentrations of TGFß1 were found in seminal plasma. The presence of such high concentrations of TGFß1, 100-fold higher than is required to activate the receptors in vitro (which is 1–2 ng/ml; Benahmed et al., 1989Go; Esposito et al., 1991Go; Morera et al., 1992Go; Besset et al., 1994Go; Ilio et al., 1995Go; Kim et al., 1996Go) is intriguing. Even free (active) TGFß1 is present at concentrations that allow full activity of the peptide. The occurrence of such high concentrations of free and activatable TGFß1 raises the question as to whether the conversion of inactive to active peptide is a key regulatory step in seminal plasma, as has been suggested for different tissues (Harpel et al., 1992Go).

The biological estimate of TGFß was less accurate than the immunological determination. Nevertheless, the CCL64 cell assay has the advantage of detecting all forms of the ligand, thereby providing an estimate of its total activity in seminal plasma. The results obtained were the sum of TGFß1, TGFß2 and other forms, if present. This also allowed us to ascertain that the TGFß immunologically detected was not an inactive peptide in seminal plasma.

When secreted, latent TGFß is non-covalently linked to different proteins. Nocera and Chu (Nocera and Chu, 1993Go), in identifying TGFß as an immunosuppressive molecule in seminal plasma by gel filtration, found this activity in fractions of 100 to >440 kDa, i.e. complexed, since free mature TGFß has a mol. wt of 25 kDa. It is not known if complex(es) present in seminal fluid are identical or different from those found in blood. {alpha}2-Macroglobulin is a potential TGFß-binding protein, since it is present in seminal plasma at ~10 µg/ml (Glander et al., 1996Go).

TGFß concentrations in seminal plasma might be higher that those measured by the immunological assay, since the biological assay consistently gave higher values. Therefore other forms of TGFß, i.e. TGFß2, TGFß1–2 and TGFß3, may well be present in seminal plasma, and indeed TGFß2 was found in human seminal plasma (Nocera and Chu, 1995Go; Shrivastava et al., 1996Go). The functions of these other peptides, as demonstrated in vitro, are not very different from those of TGFß1 (reviews: Massagué, 1990Go; Sporn and Roberts, 1992Go; Matzuk, 1995Go), but mice with non-functional genes for TGFß1 or TGFß2 exhibit different phenotypes (Sanfort et al., 1997Go).

The absence of correlation between TGFß1 and transferrin concentrations, and the normal values obtained in cases of epididymal occlusion, where testicular secretions cannot contribute to seminal plasma formation, suggest that the testis and the epididymis contribute only partly to the final TGFß1 concentration in seminal plasma. It will be of interest to determine the extratesticular origin of the TGFß1 present in the seminal plasma. Furthermore, no differences were found in the secretory group between patients with high or normal blood plasma FSH, suggesting again that most seminal plasma TGFß1 was probably not under testicular hormonal control.

A role of this growth factor in regulating growth and functions of testicular somatic cells has been demonstrated in vitro (reviews: Benahmed, 1996Go; Gnessi et al., 1997Go). However, no relationships were observed in the present study between seminal TGFß1 concentrations and normal versus pathological testicular status, possibly because other regions of the reproductive tract secrete the ligand. Moreover, the TGFß ligands we measured here are only a small part of the TGFß system, which includes, in addition to the ligand, the binding protein(s), the membrane (and probably the soluble) receptors. Therefore, a possible difference between normal and pathological subjects might be found in the levels of these parameters. This observation suggests that it would be of interest to analyse these parameters in seminal plasma of infertile men. Finally, although we observed moderately elevated or decreased activatable TGFß1 concentrations in a few patients, we do not know at the present time whether these observations are relevant to their infertility. A higher number of patients would be necessary to draw a firm conclusion on that issue.

As we mentioned, normal free TGFß1 concentrations (<0.5 to 4.5 ng/ml) were of the same order of magnitude or higher than the concentrations giving maximal effect in vitro. This suggests a local action, possibly repair after local infection and inflammation (Noble et al., 1992Go). An immunosuppressive activity was demonstated in seminal plasma (James and Hargreave, 1984Go), and attributed to prostaglandins (Quaile et al., 1989Go) and to other local factors (Imade et al., 1997Go). The TGFß may also contribute to this activity (Nocera and Chu, 1993Go; Pöllänen et al., 1993Go).

As free and latent TGFß1 are found on ejaculated spermatozoa (Chu et al., 1996Go) one may postulate that spermatozoa are coated with locally secreted TGFß both produced during the spermatogenesis (Caussanel et al., 1997Go) and/or present in genital tract secretions. TGFß1 may play a critical role in preventing an immune reaction to spermatozoa in the female reproductive tract (Nocera and Chu, 1993Go; Chu et al., 1996Go). Also, it is possible that it may act on the spermatozoa itself as we have recently observed the presence of immunoreactive TGFß receptors on specific domains of the male gamete (V.Caussanel, S.Hamamah, A.Hazout, B.Loras, M.Benahmed, unpublished results).

In conclusion, the present data suggest that TGFß quantification in seminal plasma of infertile patients could be informative probably on the local gonadol immunological status of the patient or on the functional status of the accessory glands, which are responsible for most of the secretion of TGFß. Based on the published observations obtained from in-vitro studies, and more importantly in the transgenic models which indicate that TGFß and related peptides may play a key role in spermatogenesis, the analysis of testicular TGFß expression remains of major interest in oligozoospermic and azoospermic patients. To evaluate the potential critical role of TGFß in such cases, analysis of the signalling molecules in the testicular (tissues) biopsies is a more suitable approach.


    Acknowledgments
 
We thank Dr B.Laumon (INRETS, Lyon) for his help in statistical analysis, and Miss A.McLeer for a critical reading of the manuscript. Natural human platelet-extracted TGFß1 used in this study as a standard for biological tests was a gift of Dr J.-Cl.Hendrick, University of Liège, Belgium. This work was supported by Institut National de la Santé et de la Recherche Médicale (INSERM) and by Association pour la Recherche Clinique et Fondamentale Appliquée à la Reproduction (ARCEFAR).


    Notes
 
4 To whom correspondence should be addressed Back


    References
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 References
 
Benahmed, M., Sordoillet, C., Chauvin, M. A. et al. (1989) On the mechanisms involved in the inhibitory and stimulating actions of transforming growth factor-ß on porcine testicular steroidogenesis: an in vitro study. Mol. Cell. Endocrinol., 67, 155–164.[ISI][Medline]

Benahmed, M. (1996) Growth factors and cytokines in the testis. In Comhaire, F.H. (ed.), Male Infertility. Chapman & Hall, London, pp. 55–95.

Besset, V., Collette, J., Chauvin, M. A. et al. (1994) Effect of transforming growth factor-ß1 on the insulin-like growth factor system in cultured porcine Leydig cells. Mol. Cell. Endocrinol., 99, 251–257.[ISI][Medline]

Caussanel, V., Tabone, E., Hendrick, J.-C. et al. (1997) Cellular distribution of transforming growth factor betas 1, 2 and 3 and their types I and II receptors during postnatal development and spermatogenesis in both testis. Biol. Reprod., 56, 357–367.[Abstract]

Chu, T.M., Nocera, M.A., Flanders, K.C. et al. (1996) Localization of seminal transforming growth factor-ß1 on human spermatozoa: an immunocytochemical study. Fertil. Steril., 66, 327–330.[ISI][Medline]

Esposito, G., Keramidas, M., Mauduit, C. et al. (1991) Direct regulatory effects of transforming growth factor-ß1 on lactate production in cultured porcine Sertoli cells. Endocrinology, 128, 1441–1449.[Abstract]

Glander, H.J., Kraatzsch, J., Weisbrich, Ch. et al. (1996) Insulin-like growth factor and {alpha}2-macroglobulin in seminal plasma correlate with semen quality. Hum. Reprod., 11, 2454–2460.[Abstract]

Gnessi, L., Fabbri, A. and Spera, G. (1997) Gonadal peptides as mediators of development and functional control of the testis: an integrated system with hormones and local environment. Endocr. Rev., 18, 541–609.[Abstract/Free Full Text]

Greenwood, F.C., Hunter, W.M. and Glover, J.S. (1963) The preparation of 131I labelled growth hormone of high specific activity. Biochem. J., 89, 114–123.[ISI]

Harpel, J.G., Metz, C.N., Kojima, S. et al. (1992) Control of transforming growth factor-ß activity: latency vs activation. Prog. Growth Factor Res., 4, 321–335.[Medline]

Ilio, K.Y., Sensibar, J.A., Lee, C. (1995) Effect of TGF-ß1, TGF-{alpha}, and EGF on cell proliferation and cell death in rat ventral prostatic epithelial cells in culture. J. Androl., 16, 482–490.[Abstract/Free Full Text]

Imade, G.E., Baker, H.W.G., de Kretser, D.M. and Hedger, M.P. (1997) Immunosuppressive activities in the seminal plasma of infertile men: relationship to sperm antibodies and autoimmunity. Hum. Reprod., 12, 256–262.[Abstract]

James, K. and Hargreave, T. B. (1984) Immunosuppression by seminal plasma and its possible clinical significance. Immunol. Today, 5, 357–363.[ISI]

Kim, I.Y., Kim, J.H., Zelner, D.J. et al. (1996) Transforming growth factor-ß1 is a mediator of androgen-regulated growth arrest in an androgen-responsive prostatic cancer cell line, LNCaP. Endocrinology, 137, 991–999.[Abstract]

Lokeshwar, B.B. and Block, N.L. (1992) Isolation of a prostate carcinoma cell proliferation-inhibiting factor from human seminal plasma and its similarity to transforming growth factor ß. Cancer Res., 52, 5821–5825.[Abstract]

Massagué, J. (1990) The transforming growth factor-ß family. Annu. Rev. Cell. Biol., 6, 597–641.[ISI]

Matzuk, M.M. (1995) Functional analysis of mammalian members of the transforming growth factor-ß superfamily. Trends Endocrinol. Metab., 6, 120–127.[ISI]

Morera, A.M., Esposito, G., Ghiglieri, C. et al. (1992) Transforming growth factor-ß1 inhibits gonadotrophin action in cultured porcine Sertoli cells. Endocrinology, 130, 831–836.[Abstract]

Nishi, N., Oya, H., Matsumoto, K. et al. (1996) Changes in gene expression of growth factors and their receptors during castration-induced involution and androgen-induced regrowth of rat prostates. Prostate, 28, 139–152.[ISI][Medline]

Noble, N.A., Harper, J.R. and Border, W.A. (1992) In vivo interactions of TGF-ß and extracellular matrix. Prog. Growth Factor Res., 4, 369–382.[Medline]

Nocera, M.A. and Chu, T.M. (1993) Transforming growth factor-ß as an immunosuppressive protein in human seminal plasma. Am. J. Reprod. Immunol., 30, 1–8.[ISI][Medline]

Nocera, M.A. and Chu, T.M. (1995) Characterization of latent transforming growth factor-ß from human seminal plasma. Am. J. Reprod. Immunol., 33, 282–291.[ISI][Medline]

Pöllänen, P., von Euler, M., Jahnukainen, K. et al. (1993) Role of transforming growth factor ß in testicular immunosuppression. J. Reprod. Immunol., 24, 123–137.[ISI][Medline]

Quaile, A., Kelly, J., Hargreave, R.W. et al. (1989) Immunosuppression by human seminal prostaglandins. Clin. Exp. Immunol., 75, 387–391.[ISI][Medline]

Sanfort, L.P., Ormsby, I., Gittenberger-de Groot, A.C. et al. (1997) TGFß2 knockout mice have multiple developmental defects that are non-overlapping with other TGFß knockout phenotypes. Development, 124, 2659–2670.[Abstract/Free Full Text]

Shrivastava, M.D., Lippes, J. and Shrivastava, B.I. (1996) Cytokines of the human reproductive tract. Am. J. Reprod. Immunol., 36, 157–166.[ISI][Medline]

Sporn, M.B. and Roberts, A.B. (1992) Transforming growth factor-ß: recent progress and new challenges. J. Cell. Biol., 119, 1017–1021.[ISI][Medline]

Steiner, M.S., Zhou, Z.Z., Tonb, D.C. et al. (1994) Expression of transforming growth factor-ß1 in prostate cancer. Endocrinology, 135, 2240–2247.[Abstract]

World Health Organization (1987) Laboratory Manual for the Examination of Human Semen and Semen–Cervical Mucus Interaction, 2nd edn. Cambridge University Press, Cambridge.

Submitted on October 23, 1998; accepted on February 17, 1999.