Immunocytochemical localization of epidermal growth factor receptors in human testis from infertile subjects*

FERTILITY AND STERILITY Copyright 1994 The American Fertility Society Vol. 61. No.5, May 1994 Printed on acid-free paper in U. S. A. Immunocytochemical localization of epidermal growth factor receptors in human testis from infertile subjects* Carlo Foresta, M.D.t Alberto Varotto, M.D. Institute of Medical Semeiotics, Third Chair of Medical Pathology, University of Padua, Padua, Italy Objective: To investigate the immunolocalization of the epidermal growth factor receptor (EGFR) in normal and pathological human testis by immunocytochemical technique. Design: Cytologic specimens were obtained by bilateral fine needle aspiration (FNA) of the testis and stained in May Griinwald-Giemsa for the cytologic analysis; immunolocalization of EGFR was analyzed on duplicate slides from each testis using two anti-egfr monoclonal antibodies and peroxidase-antiperoxidase technique. Setting: Infertility center of an academic unit. Patients: A total of 42 infertile patients, affected by various testicular diseases. The control group was made up of 10 normal sperm patients with autoimmune infertility and cytologic picture of normal spermatogenesis. Interventions: Exogenous FSH was administered 75 IV 1M on alternate days for 3 months on 16 of the infertile patients who showed oligospermia and normal FSH plasma levels. Semen analysis and testicular FNA (and after cytologic and immunocytochemical studies) were repeated at 3 months of treatment. Main Outcome Measures: Luteinizing hormone and FSH plasma levels were determined by RIA methods; qualitative and quantitative parameters for the cytologic evaluation are reported in our previous works. Results: The cytologic analysis permitted identification of seven classes of infertile subjects, characterized by different cytologic pictures. Epidermal growth factor receptor immunostaining evidenced weak positivity on Sertoli and germ cells (with the exception of spermatozoa) in the presence of normal germ line and normal FSH plasma levels and strongly intense positivity in the presence of serious hypospermatogenesis, spermatogonial or spermatocytic arrest, and Sertoli cellonly syndrome. These conditions were characterized by higher FSH plasma levels than normal controls. All of the subjects who received exogenous FSH, with moderate hypospermatogenesis or spermatidic arrest, showed on Sertoli and germ cells a weak EGFR immunostaining before the treatment and intense immunostaining after the treatment. Conclusions: These results confirm recent demonstrations of EGFRs in human testis and evidence different EGFR immunostaining in the presence of various degrees of testicular damage, suggesting a role of this growth factor in growth and differentiation of the germ cells throughout spermatogenesis. The observation that intense EGFR immunostaining was found in subjects showing high FSH plasma levels and in all of the patients who received exogenous FSH, supports a possible role of this gonadotropin in the modulation of the EGFR expression. Fertil SterilI994;61:941-8 Key Words: EGF, EGF-Receptors, immunocytochemistry, cytology, human testis Received June 24, 1993; revised and accepted December 10, 1993. * Supported in part by grant 326/01/90 from Regione Veneto, Unita Locale Socio Sanitaria 21, Padova, Italy. t Reprint requests: Carlo Foresta, M.D., Istituto di Semeiotica Medica, Cattedra di Patologia Medica III, Via Ospedale Civile, 105,35128 Padova, Italy. Spermatogenesis is regulated by pituitary gonadotropins (1), but recently the involvement ofpolypeptide growth factors (GFs) in modulating spermatogenic process has been addressed (2). In fact, several evidences are consistent with the hypothesis that GFs, such as insulin-like growth factor 1(3), Vol. 61, No.5, May 1994 Foresta and Varotto EGFR in human testis 941

epidermal growth factor (EGF) (4), transforming growth factor-a (5), fibroblast growth factor (6), and interleukin-i (7) may playa key role in the proliferation, differentiation, and maintenance of germ cells throughout spermatogenesis, even if the mechanisms of such modulation and the testicular target cells still are not well known. Epidermal growth factor, a 53-amino acid polypeptide initially detected in mouse submaxillary gland extracts (8), is one such factor whose role in spermatogenesis has been highlighted recently. In mature male mice, it has been demonstrated that surgical sialoadenectomy causes a strong reduction of the sperm content in the epididymis and of the round and condensing spermatids in the testis, in the absence of concomitant variations of testosterone or FSH plasma levels (9). The replacement therapy with EGF is able to reverse this deleterious effect (9). Epidermal growth factor receptors (EGFRs) have been reported in cultured murine Leydig tumor cells (10) and, more recently, they were demonstrated by immunofluorescence technique on rat and monkey Leydig and Sertoli cells (11) and by immunocytochemical technique on mouse Leydig cells (12). Furthermore, in vitro studies suggested that EGF can modulate in immature rats characteristic functions of Leydig cells such as androgen biosynthesis (13) and of Sertoli cells such as androgen aromatization (14), lactate production (14), and inhibin secretion (15). Also, in humans it has been demonstrated that EGF may be involved in the regulation of growth and development of the testicular cells. High levels ofegf have been demonstrated in human seminal plasma (16), and EGF-like substance was detected in human male reproductive tissue (17). In a recent biochemical study, EGFRs have been evidenced in human testicular tissue and characterized by immunohistochemical method in the interstitial tissue (18). In our previous work in humans (19), EGFRs were demonstrated in interstitial, peritubular, and Sertoli cells by immunofluorescence technique, exhibiting different patterns between normal and pathological testicular pictures. In particular, an intense EGFR expression was found on Sertoli cells in the presence of serious testicular damage and high FSH plasma levels, suggesting a role of this gonadotropin in the regulation of this expression. However, the technical limits of the histologic examination and immunofluorescence technique did not allow us to discriminate whether also the germ cells possess EGFR and the developmental stages at which EGFR is expressed. Recently, we have proposed fine needle aspiration (FNA) of the testis as a parameter in the assessment of the testicular status in infertile subjects (20, 21). The cytologic analysis permits identification of the tubular cells in the different maturation steps and, as well as histologic preparations, characterization of the tubular damage (that is, Sertoli cell-only, hypospermatogenesis, maturation arrests). Therefore, using immunocytochemical methodology on testicular cells obtained by FNA, we have tried to better elucidate the pattern ofegfr expression on human testis. Furthermore, to clarify the hormonal regulation ofegfr expression, the study was performed before and after a treatment with exogenous FSH. MATERIALS AND METHODS We have studied 42 adult infertile males who were 20 to 34 years of age (mean, 26 ± 4). Seminal parameters of each subject were examined on two different occasions: 10 subjects were azoospermic, 22 were seriously oligospermic (counts < 10 X 10 6 cells/ml), and 10 patients, used as controls, were normospermic (count> 20 X 10 6 cells/ml, with normal percentages of forward motility and normal morphology) and infertility resulted because of the presence of high titres of antisperm autoantibodies (immunoglobulin [Ig] G binding to head, tail, and midpiece). Plasma FSH and LH were determined in each subject by RIAs, using 125 1 -labeled LH and FSH and a double monoclonal antibody (mab) provided by Ares-Serono (Milan, Italy). The interassay and intra-assay variations were 8.2% and 6.4%, respectively, for the LH assay and 4.2% and 2.7% for the FSH assay. The sensitivity of the assay was 0.15 miu /ml for LH and 0.25 miu /ml for FSH. Statistical significance was verified with Student's t-test for unpaired data. Probability values < 0.05 were regarded as statistically significant. Cell Preparation Our study was approved by the Hospital Ethical Committee, and informed consent was obtained from each patient. Human testicular cells were obtained by bilateral FN A of the testis, as described in detail in our previous works (20, 21). Briefly, 23-942 Foresta and Varotto EGFR in human testis Fertility and Sterility

gauge needles connected to 20-mL syringes were used for aspiration, and the cells were placed on two or more glasses for each testis and processed as follows: [1] one specimen of each testis was allowed to air dry for 24 hours at room temperature, stained in May-Griinwald-Giemsa, and examined under a light Orthoplan microscope (Wild Leitz, Wetzlar, Germany) for the cytologic evaluation, including qualitative and quantitative analysis; [2] in parallel, one or more duplicate slides of each testis were frozen within 30 minutes after aspiration and stored at -80 C until used for immunocytochemistry, as described below. Follicle-Stimulating Hormone Treatment of the Patients Human purified FSH (Metrodin; Serono, Rome, Italy), 75 IU 1M on alternate days, was administered for 3 months to those patients (n = 16) who had shown oligospermia and FSH plasma levels similar to our controls. In the same individuals, seminal and hormonal parameters were re-evaluated at 1 and 3 months of treatment, and testicular FNA was repeated at 3 months of treatment. Antibodies and Immunocytochemistry For immunolocalization ofegfrs, the study was performed using two different mabs, both purchased from Sigma Chemical Co. (St. Louis, MO). The first antibody specifically raises against the external carbohydrate portion (fragment 1005-1016) of both human and mouse EGFR (clone no. 29.1, mouse IgG 1 ); the second antibody specifically recognizes the intracellular domain (fragment 985-996) of this receptor (clone no. F4, mouse IgG 1 ). The immunocytochemical study was performed using peroxidase-antiperoxidase (PAP) staining kits (Histomune; Ortho Diagnostic Systems, Milan, Italy) for mouse antibodies, employing goat antimouse serum as linking reagent and peroxidase-labeled mouse serum as labeling reagent. Each slide was first fixed in acetone for 10 minutes, divided in adjacent areas, and then rehydrated in phosphate-buffered saline (PBS) for 15 minutes at room temperature. To reduce nonspecific background, cells were blocked with normal goat serum (provided with the PAP kits) for 20 minutes at room temperature. Endogenous peroxidase was not blocked by hydrogen peroxide to avoid removing tissues from the slides; this was identified using the negative control as a reference. After two washes in PBS, both EGFR antibodies, in a 1:100 dilution, were separately applied on adjacent areas without diffusion of reagents and left overnight at 4 C in a humidity chamber. The next steps of the immunocytochemical procedure were followed exactly as described by the manufacturer's instructions and reported in other studies (22). Control experiments included deletion of the primary or secondary antibody and replacement of the primary antibody with the nonimmune mouse serum (provided with the PAP kits) at the same dilution. Furthermore, for each experiment a "negative" and a "positive" internal control, employing a set of cells from a normal spermic subject and from a patient with Sertoli cell-only syndrome, respectively, were included. To assure an adequate comparison of the staining, the tissue specimens obtained before and after FSH therapy were stained on the same day. After the immunoreaction, the cells were counterstained in Mayer's hematoxylin (Ortho Diagnostic Systems, Milan, Italy), mounted in glycerine jelly, and examined under a light Orthoplan microscope (Wild Leitz, Wetzlar, Germany) at X125, X400, and X1,250 magnifications. Immunostaining was interpreted as absent, weak, or intense. RESULTS According to our previous works (20, 21), the cytologic analysis in the infertile subjects permitted identification of seven different pictures, corresponding to specific histologic pictures as follows: [1] normal germ line on both testes (autoimmune normospermic and obstructive azoospermic subjects); [2] serious or [3] moderate bilateral hypospermatogenesis; [4] serious unilateral hypospermatogenesis, with normal germ line on the contralateral testis; [5] maturation arrest at spermatogonial or spermatocytic level; [6] maturation arrest at spermatidic level; and [7] Sertoli cell-only syndrome (no germ cell observed). On the basis of these cytologic appearances, the infertile subjects were categorized in seven different groups, as summarized in Table 1. They were divided further into two subgroups (A and B) if normal-oligospermic or azoospermic, respectively. The hormonal findings are reported in Table 1. Mean LH and FSH plasma levels of normospermic controls were 2.0 ± 0.9 mlu jml (2.0 ± 0.9 IU jl) and 2.6 ± 0.9 mlujml (2.6 ± 0.9 IUjL), respectively. Plasma levels oflh were similar to the controls in subjects belonging to groups I, III, IV, V, Vol. 61, No.5, May 1994 Foresta and Varotto EGFR in human testis 943

Table 1 Cytological, Seminal, and Hormonal Patterns of the Infertile Subjects Cytological analysis Semen count Group Plasma LH Plasma FSH PlasmaFSH before therapy* before therapy after therapy miu/mlt miu/mlt miu/mlt Normal bilateral germ line 10 Normospermic I-A (complete spermatogenesis) 2 Azoospermic I-B Serious bilateral 3 Oligospermic II-A hypospermatogenesis 2 Azoospermic II-B Moderate bilateral 8 Oligospermic III hypospermatogenesis 1st testis: serious 7 Oligospermic IV hypospermatogenesis 2nd testis: normal germ line Spermatogonial or 3 Oligospermic V-A spermatocytic arrest 1 Azoospermic V-B 2.0 ± 0.9 2.6 ± 0.9 3.2 ± 1.3+ 13.6 ± 2.4 2.7 ± 1.0 3.8 ± 2.5 5.7 ± 3.2 2.6 ± 1.2 3.1 ± 1.1 5.3 ± 2.9 2.9 ± 1.4 10.8 ± 1.8 Spermatidic arrest 10ligospermic VI-A 1 Azoospermic VI-B Sertoli cell-only 4 Azoospermic VII * Values are means ± SD. t Conversion factor to SI unit, 1.0. 2.5 ± 1.0 3.3 ± 1.8 7.7 ± 2.8 26.6 ± 6.0 + P < 0.01 versus controls. P < 0.001 versus controls. 4.9 ± 2.6 and VI, whereas they were significantly higher in the presence of serious bilateral hypospermatogenesis (group II) and Sertoli cell-only syndrome (group VII); plasma levels of FSH were similar to the controls in subjects in groups I, III, IV, and VI, whereas they were significantly higher in the presence of severe bilateral hypospermatogenesis (group II), spermatogonial or spermatocytic arrest (group V), and Sertoli cell-only syndrome (group VII). After treatment with exogenous FSH, LH plasma levels remained unmodified (data not shown), and FSH plasma levels showed a weak, but not significant, increase (Table 1). No significant improvement of the seminal parameters (sperm count, motility, and morphology) and cytological pictures were observed after treatment in all of the treated patients (groups III, IV, and VI-A, data not shown). Immunocytochemistry for EGFR Figure 1 presents immunocytochemical staining of representative tissue specimens demonstrating cell immunolocalization of EGFR within the seminiferous epithelium. The nuclei were counterstained as blue in hematoxylin, which permits differentiation among the various germ cell subpopulations as well as in May-Griinwald-Giemsa stain. Cytologic tissue specimens, in fact, allow the analysis of single cells and of their own morphological features. In control incubations, no immunostaining was found when either primary or secondary antibody was omitted or was replaced by nonimmune serum (examples in Fig. lc and E). The results of the immunocytochemistry for the EGFR are summarized in Table 2. In subjects showing normal germ line on both testes (group I-A and B), a weak immunostaining was demonstrated in the cytoplasm of the Sertoli cells, easily identified by the presence of a round nucleus with a single nucleolus and a large cytoplasm. It was also demonstrated on the plasma membrane and in the cytoplasm of the germ cells, from spermatogonia to spermatids, whereas it was absent in mature spermatozoa. Similar weak immunostaining for the EGFR was also found in the cytoplasm of Sertoli cells, on the plasma membrane and in the cytoplasm of spermatogenic cells from patients showing moderate bilateral hypospermatogenesis (group III, Fig. la) and spermatidic arrest (group VI-A and B). In the presence of serious hypospermatogenesis (Fig. IB), on both testes (group II-A and B) or on a single testis (group IV) and of spermatogonial or spermatocytic arrest (group V-A and B), intense EGFR immunostaining was uniformly found in the cytoplasm of Sertoli cells, on the plasma membrane, and in the cytoplasm of germ cells that were seen in the immature stages of development. The rare spermatozoa encountered in these cases were always unstained. In group IV patients affected by 944 Foresta and Varotto EGFR in human testis Fertility and Sterility

Figure 1 (A), EGFR immunostaining of cells in a case of moderate hypospermatogenesis (slide from group III). (B), EGFR immunostaining of cells in a case of serious bilateral hypospermatogenesis (slide from group II). (C), Negative control in a case of normal spermatogenesis, in which primary antibody was omitted. (D), EGFR immunostaining of Sertoli cells and (E) corresponding negative control in which primary antibody was replaced with nonimmune serum, in a case of Sertoli cell-only (slides from group VII). Cells were identified by Mayer's hematoxilin counterstain (spermatogonia, -+; primary spermatocyte,==>; Sertoli cells, )1--) (Mayer's hematoxilin, magnification was X1,250). unilateral testicular damage, the contralateral normal t~stis exhibited a cell pattern of EG FR staining similar to that observed in normal spermic individuals (group I-A). In the presence of Sertoli cell-only syndrome (group VII, Fig. ld), intense EGFR immunolabeling was uniformly detected in the cytoplasm of all the Sertoli cells. In all of these specimens, identification of EG FR immunostaining of Sertoli cell plasma membrane was not possible because of the large cytoplasm with ill-defined borders; therefore, its positivity cannot be excluded. Moreover, no evaluation was possible on Leydig cells, only occasionally encountered in cytologic tissue specimens. Between the two mabs, no difference was observed in either the Vol. 61, No.5, May 1994 intensity or the distribution of EGFR immunostaining. After treatment with exogenous FSH, intense EGFR immunostaining was detectable in all of the treated subjects (groups III, IV, and VI-A) in the cytoplasm of the Sertoli cells, on the plasma membrane, and in the cytoplasm of the germ cells, with the exception of spermatozoa that appeared unstained. The strength of the immunoreaction was recorded as conspicuously stronger than that observed in the same subjects before therapy (Fig. 2) and similar to that observed in tissue specimens from patients showing serious testicular damage (groups II-A and B, V-A and B, and VII). In patients of group IV, the cells from the pathological Foresta and Varotto EGFR in human testis 945

Table 2 Different Patterns of EGFR Immunostaining in Germ and Sertoli Cells From the Patients Categorized in Seven Different Groups by Cytologic Pictures Immunostainil).g for the EGFR Before therapy Group 1-A 1-B II-A II-B III IV Normal testis Damaged testis V-A V-B VI-A VI-B VII After therapy* Germ cells Sertoli cells Germ cells Sertoli cells weak EGFR immunoreactivity in the presence of complete spermatogenesis (autoimmune normal spermic or obstructive azoospermic subjects) and a strongly intense EG FR immunostaining in the presence of severe testicular damage (serious hypospermatogenesis, spermatogonial or spermatocytic arrest, Sertoli cell-only syndrome). Furthermore, it demonstrated that EG FR immunostaining is also present on the plasma membrane and in the cytoplasm of germ cells in the different stages of development, with the exception of mature sperms. Similar to the Sertoli cells, the intensity of the germ immunoreaction was weak in normal testes and *For the patients of groups III, IV, and VI-A, the comparison of EG FR immunostaining between before and after FSH treatment is reported. testis did not show any change in the EG FR immunostaining, whereas the contralateral normal testis appeared intensely stained after therapy. DISCUSSION The staining pattern for the EGFR observed in this study is consistent with previous reports that have demonstrated that immunoreactive EGFR is detectable in mammalian and human testicular tissue. In rat and monkey testes, EG FR has been localized as prominent on Leydig and Sertoli cells (11). In human testes with normal histologic pictures, Stubbs et al. (18) observed that EGFR is present in the interstitial tissue, but other constituents of the testes, mainly germ and Sertoli cells, do not appear to immunostain positively for EG FR. The results of our previous study by immunofluorescence technique (19) confirmed these findings because a similar EG FR localization was detected on a testicular frozen section from individuals with normal sperm. However, in the same study, a bright EGFR immunofluorescence was demonstrated within the seminiferous tubules of patients affected by serious testicular damage. In the present study, the immunocytochemistry of the tissue specimens obtained from infertile subjects evidenced in the cytoplasm of Sertoli cells a 946 Foresta and Varotto EGFR in human testis Figure 2 Epidermal growth factor receptor immunostaining of tubular cells in a case of moderate bilateral hypospermatogenesis (slide from group III), before (cells on the left) and after (cells on the right) FSH treatment: (A and A 1 ), Sertoli cells. (B and B 1 ), Spermatogonium. (C and C 1 ), Primary spermatocyte. (D and D 1 ), Round spermatids and a few spermatozoa. The tissue specimens obtained before and after treatment were stained in the same experiment (Mayer's hematoxilin counterstain, magnification was X1,250). Fertility and Sterility

strong in the presence oftubular damage. It is possible that the absence of germ EGFR immunopositivity observed in other immunohistochemical studies that used normal testes as tissue may be because of a technical reason; in fact, in cytologic specimens, germ cells are seen as completely separated from each other and from Sertoli cells, and well-defined cytoplasm may favor greater deposition ofthe immunoreaction product, giving rise to a detectable positivity. Epidermal growth factor is a potent mitogenic agent in epidermal and nonepidermal tissues (23, 24); therefore, the presence of EGFRs in germ and Sertoli cells suggests a role of this factor in the regulation of growth and differentiation of these cells. Our evidence showing intense EGFR immunoreactivity in tubular cells obtained from subjects with serious tubular damage and high FSH plasma levels (groups II, V, and VII) furthermore suggests a possible role of this gonadotropin in the regulation of the EGFR expression. This hypothesis is consistent with other studies performed in rat granulosa cells (25) and is further supported by the FSH treatment that caused a strong change ofthe testicular EGFR staining pattern in patients in groups III, IV, and VI-A, showing moderate tubular alterations and normal FSH plasma levels. However, the analysis of EGFR immunoreactivity in subjects of groups IV affected by unilateral testicular damage, showing normal plasma FSH but different EGFR immunostaining between pathological and normal testes, demonstrated that in addition to FSH, local signals also may influence the EGFR expression, probably through paracrine mechanisms. The significance of the characteristic staining pattern for the EGFR in the presence of tubular damage and high FSH plasma levels remains to be clarified. The production of EGF-like peptides by Sertoli cells (17) suggests that EGF (or EGF-like peptides) might interact with its receptor to induce the proliferation of germ cells. Stimulating the expression of EGFR, FSH might exert its physiological influence on the spermatogenic process also through this mechanism. Therefore, the strong EGFR immunoreactivity observed in these testicular diseases (coupled with high FSH plasma levels) or after administration of exogenous FSH might reflect an attempt of this gonadotropin to activate the spermatogenic process via the EGFR. Alternatively, it might be a signal of an adverse change that occurred in the paracrine regulatory systems of the testes. This latter hypothesis is fur- ther supported by the lack of any improvements in terms of seminal parameters and the cytologic picture observed after therapy in all of the treated subjects. REFERENCES 1. Fritz 1. Sites of actions of androgens and follicle stimulating hormone on cells of the seminiferous tubule. In: Litwack G, editor. Biochemical actions of hormones. New York: Academic Press, 1978;249-78. 2. Bellve AR, Zheng W. Growth factors as autocrine and paracrine modulators of male gonadal functions. J Reprod Fertil 1989;85:771-93. 3. Naville D, Chatelain PG, Avallet 0, Saez JM. Control of production of insulin-like growth factor I by pig Leydig and Sertoli cells cultured alone or together. Cell-cell interactions. Mol Cell Endocrinol 1990;70:217-24. 4. Haneji T, Koide SS, Tajima Y, Nishimune Y. Differential effects of epidermal growth factor on the differentiation of type A spermatogonia in adult mouse cryptorchid testes in vitro. J Endocrinol 1991;128:383-8. 5. Skinner MK, Takacs K, Coffey RJ. Transforming growth factor-alpha gene expression and action in the seminiferous tubule: peritubular cell-sertoli cell interactions. Endocrinology 1990;124:845-54. 6. Jaillard C, Chatelain PG, Saez JM. In vitro regulation of pig Sertoli cell growth and function effects of fibroblast growth factor and somatomedin C. Bioi Reprod 1987;37:665-74. 7. Pollanen P, Soder 0, Parvinen M. Interleukin-l alpha stimulation of spermatogonial proliferation in vivo. Reprod Fertil Dev 1989;1:85-91. 8. Cohen S. Isolation of mouse submaxillary gland protein accelerating incisor eruption and eyelid opening in the newborn animal. J Bioi Chern 1962;237:1555-62. 9. Tsutsumi 0, Kurachi H, Oka T. A physiological role of epidermal growth factor in male reproductive function. Science 1986;233:975-7. 10. Ascoli M. Regulation of gonadotropin receptors and gonadotropin responses in a clonal strain of Leydig tumor cells by epidermal growth factor. J Bioi Chern 1981;256:179-83. 11. SUllrez-Quian CA, Dai M, Onoda M, Kriss RM, Dym M. Epidermal growth factor receptor localization in the rat and monkey testes. Bioi Reprod 1989;41:921-32. 12. Suarez-Quian CA, Niklinski W. Immunocytochemical localization of the epidermal growth factor in mouse testis. Bioi Reprod 1990;43:1087-97. 13. Verhoeven G, Cailleau J. Stimulatory effects of epidermal growth factor on steroidogenesis in Leydig cells. Mol Cell EndocrinoI1986;47:99-106. 14. Mallea LE, Machado AJ, Navaroli F, Rommerts FFG. Epidermal growth factor stimulates lactate production and inhibits aromatization in cultured Sertoli cells from immature rats. Int J AndroI1986;9:201-8. 15. Morris PL, Vale WW, Cappel S, Bardin CWo Inhibin production by primary Sertoli cell-enriched cultures: regulation by follicle-stimulating hormone, androgens, and epidermal growth factor. Endocrinology 1988;122:717-25. 16. Hirata Y, Uchinasi M, Hazama M, Fujita T. Epidermal Vol. 61, No.5, May 1994 Foresta and Varotto EGFR in human testis 947

growth factor in human seminal plasma. Horm Metab Res 1987;19:35-7. 17. Holmes SD, Spotts G, Smith RG. Rat Sertoli cells secrete a growth factor that blocks epidermal growti). factor (EGF) binding to its receptor. J Bioi Chern 1986;261:4076-80. 18. Stubbs SC, Hargreave TB, Habib FK. Localization and characterization of epidermal growth factor receptors on human testicular tissue by biochemical and immunohistochemical techniques. J Endocrinol 1990;125:485-92. 19. Foresta C, Caretto A, Varotto A, Rossato M, Scandellari C. Epidermal growth factor receptors (EGFR) localization in human testis. Arch Androl 1991;27:17-24. 20. Foresta C, Varotto A, Scandellari C. Assessment oftesticular cytology by fine needle aspiration as a diagnostic parameter in the evaluation of the azoospermic subject. Fertil Steril 1992;57:858-65. 21. Foresta C, Varotto A. Assessment of testicular cytology by fine needle aspiration as a diagnostic parameter in the evaluation of the oligospermic subject. Fertil Steril1992;58:1028-33. 22. Sternberger LA, Hardy PH, Cuculis JJ, Meyer HG. The unlabeled antibody enzyme method of immunohistochemistry. J Histochem Cytochem 1970;18:315-32. 23. Carpenter G, Cohen S. Epidermal growth factor. Annu Rev Biochem 1979;48:193-216. 24. Gospodarowicz D, Bialecki H. Fibroblast and epidermal growth factors are mitogenic agents for cultured granulosa cells of rodent, porcine and human origine. Endocrinology 1979;104:757-64. 25. Feng P, Knecht M, Catt K. Hormonal control of epidermal growth factor receptors by gonadotropins during granulosa cell differentiation. Endocrinology 1987;120:1121-6. 948 Foresta and Varotto EGFR in human testis Fertility and Sterility