Effects of human chorionic gonadotropin on testicular interstitial tissues in men with non-obstructive azoospermia

Transcription

1 ISSN: ORIGINAL ARTICLE Correspondence: Koji Shiraishi, Department of Urology, Yamaguchi University School of Medicine, Yamaguchi, Japan. Keywords: azoospermia, fibroblast, fibrosis, human chorionic gonadotropin, Leydig cell, micro-tese Received: 13-May-16 Revised: 18-Aug-16 Accepted: 12-Sep-16 Effects of human chorionic gonadotropin on testicular interstitial tissues in men with non-obstructive azoospermia S. Oka, K. Shiraishi and H. Matsuyama Department of Urology, Yamaguchi University School of Medicine, Yamaguchi, Japan doi:.1111/andr SUMMARY Non-obstructive azoospermia is a severe condition because spermatogenesis per se is disrupted. Although microdissection testicular sperm extraction is the standard for non-obstructive azoospermia, sperm retrieval is unsuccessful in approximately % of patients. For these patients, we conducted human chorionic gonadotropin-based salvage hormonal, and sperm retrieval was possible in % of patients. The objectives of this study were to assess the changes in interstitial lesions in patients with nonobstructive azoospermia and to evaluate the effects of human chorionic gonadotropin on these tissues. Testicular biopsy specimens were obtained from non-obstructive azoospermia patients who failed to obtain spermatozoa and from obstructive azoospermia patients. All non-obstructive azoospermia patients received salvage hormonal after microdissection testicular sperm extraction. Hematoxylin and eosin (H.E.) staining and immunohistochemical staining for steroidogenic acute regulatory protein antibody, the Leydig cell marker, and TE-7 antibody, the fibroblast marker, as well as picrosirius red staining to detect collagen fibers, were performed. We measured interstitial lesions, as Leydig cell area and the other area, with IMAGEJ software. Interstitial area, excluding Leydig cells, increased up to 12.% in non-obstructive azoospermia compared with 1.2% in obstructive azoospermia (p <.1), which was mainly because of fibrosis with TE-7-positive fibroblasts. The increase in interstitial lesions was correlated with Johnsen scores. Interstitial area, excluding the Leydig cells, decreased by 29% after salvage hormonal (p <.), indicating improvement of interstitial fibrosis in non-obstructive azoospermia. There were no significant difference in total Leydig cell area and size of each Leydig cells between obstructive azoospermia and non-obstructive azoospermia. After the salvage hormonal, a portion of the Leydig cells became hypertrophied and mean diameter of Leydig cell significantly increased (p <.1). This study showed anti-fibrotic effects of human chorionic gonadotropin and hypertrophic change of Leydig cells after human chorionic gonadotropin administration. INTRODUCTION Non-obstructive azoospermia (NOA) is diagnosed in approximately % of infertile men and is one of the most difficult conditions to treat. Microdissection testicular sperm extraction (micro-tese) is the most common procedure for NOA (Schlegel, 1999), however, 6% of the procedures failed to retrieve spermatozoa. For these patients, we conducted human chorionic gonadotropin (hcg)-based salvage hormonal, which consists of hcg and/or recombinant follicle-stimulating hormone (FSH), and spermatozoa were obtained from the % patients (Shiraishi et al., 12). The efficacy of this has been validated by a multi-institutional study, in which % of patients successfully obtained spermatozoa (Shiraishi et al., 16). We have reported the improvement of spermatogonial DNA synthesis, the elevation of intratesticular testosterone levels, and the increase in androgen receptor expression secondary to hormonal (Kato et al., 14; Shinjo et al., 13). However, the effects of hcg on testicular interstitial tissue were not considered in these reports; therefore, the influence of this hormonal on testicular interstitial lesions and human Leydig cells in NOA is still unclear. Many studies have focused on histological analysis regarding spermatogenesis because intratubular lesions are relevant to the success rate of micro-tese. In contrast, histopathological studies regarding testicular interstitial lesions in NOA are limited because testicular interstitial disorders do not directly affect spermatogenesis. However, there are some disorders associated with testicular interstitial lesions, such as mumps orchitis, and 232 Andrology, 17,, American Society of Andrology and European Academy of Andrology

2 AZOOSPERMIA AND TESTICULAR FIBROSIS interstitial fibrosis, that sometimes lead to irreversible damage of spermatogenesis (El-Fakahany et al., 14). Furthermore, testicular interstitial fibrosis after vasectomy results in impaired spermatogenesis that eventually progresses to irreversible dysfunction despite treatment with vasovasostomy (Shiraishi et al., 2). Awareness of this condition has contributed to the conservation of spermatogenesis (Shiraishi et al., 3). This information indicates that interstitial fibrosis has adverse effects on spermatogenesis. Although the effectiveness of for interstitial fibrosis remains unclear, we studied the effects of hcg on interstitial lesions. Additionally, NOA is also a significant risk factor for androgen deficiency (Bobjer et al., 12); however, limited information is available about the correlation between interstitial lesions and steroidogenesis. The purposes of this study were to compare testicular interstitial lesions between NOA and obstructive azoospermia (OA), which served as a control of normal spermatogenesis, and to assess the degree of interstitial fibrosis and the alteration of Leydig cells in NOA. Furthermore, we also studied the interstitial tissues of patients with NOA before and after salvage hormonal to elucidate the effects of hcg on interstitial fibrosis and the number and morphology of Leydig cells. MATERIALS AND METHODS Patients and samples We evaluated testicular biopsy specimens from NOA patients and OA patients with normal spermatogenesis who served as controls. Sections of NOA were obtained during micro- TESE before and after salvage hormonal. Testicular specimens were immediately fixed in Bouin s solution for 24 h. At least 6 months after the first micro-tese, patients began thrice weekly subcutaneous self-injections of IU of hcg (Gonadotropin; ASKA Pharmaceutical, Tokyo, Japan), which was continued for 3 months. Patients who continued to show high plasma FSH values were maintained on hcg treatment for 1 2 months prior to the second micro-tese. Patients who had low levels gonadotropin (decreased to <2 miu/l) also received recombinant human FSH (rhfsh) (GONAL-F; Merck Ser-ono, Geneva, Switzerland) at IU thrice weekly, for 2 months, prior to the second micro-tese. All procedures were performed by the same surgeon (K.S.) between April and March 13. This study was approved by the IRB of our institution, and written informed consent was obtained from all patients. Tissue processing and immunohistochemical staining We immunostained several sections for anti-steroidogenic acute regulatory protein (StAR) antibody (sc-286, 1 : ; Santa Cruz, Dallas, TX, USA), a marker of Leydig cells; anti-te-7 antibody (CBL271, 1 : ; Chemicon, Billerica, MA, USA), a marker of fibroblasts; and anti-proliferating cell nuclear antigen (PCNA) antibody (sc-6, 1 : ; Santa Cruz), a marker of cell proliferation. Sections were deparaffinized and rehydrated with xylene and alcohol. Endogenous peroxidases were blocked with min of incubation with 3% hydrogen peroxide. The sections were treated with microwave boiling for antigen retrieval with citrate buffer (ph 6) for StAR and PCNA antibodies and EDTA buffer (ph 8) for TE-7 antibody. The sections for TE-7 antibody were permeabilized with 1% Tween- for 3 min before blocking. After blocking with 1% bovine serum albumin, the sections were incubated with several primary antibodies overnight at 4 C. Secondary antiserum (Histofine Simple Stain MAX PO No. 4242; NICHIREI, Tokyo, Japan) was applied for 3 min, and detections were performed with DAB (No. 4172; NICHIREI). The sections were counterstained for 1 min with Mayer s hematoxylin and then mounted. Picrosirius red staining We performed picrosirius red staining (Picrosirius Red Stain Kit; Polysciences, Warrington, PA, USA) to evaluate fibrosis. Paraffin-embedded tissue blocks were sectioned at lm thickness. Sections were deparaffinized and hydrated with xylene and alcohol and then stained with picrosirius red. Quantitative analyses The area and conformation of the interstitial tissue were evaluated with H.E. staining and immunostaining for StAR and TE-7 antibodies and picrosirius red staining. We divided testicular interstitial tissue into a StAR-positive Leydig cell area and another area excluding Leydig cells and also measured total Leydig cell area and the remaining interstitial area. We randomly selected focal areas and marked them in different colors (magnification 9) and then used IMAGEJ software (ImageJ, version 1.48; NIH, Bethesda, MD, USA) to measure each area and area density per total testicular tissues including seminiferous tube. Interstitial vessels were not included in the marking. Furthermore, we randomly selected round-shaped Leydig cells per specimens and measured longitudinal axis of them with micrometer rage ruler attached in microscope (magnification 49). Johnsen s score was determined with H.E. staining. Two urologists (S.O. and K.S.) independently evaluated several sections and calculated the mean areas. Statistics Statistical analyses were performed with SPSS 12 (SPSS, Chicago, IL, USA). Data are expressed as means standard deviations (SDs). The two groups were compared with Mann Whitney rank-sum tests, and comparisons before and after salvage hormonal were performed with the Wilcoxon signed-rank test. A p-value <. was considered statistically significant. RESULTS Table 1 shows the characteristics of the patients in the OA and NOA groups. Testicular volume measured by punched-out orchidometer in patients with NOA was significantly smaller than in patients with OA. Significant differences in serum LH and FSH were also noted between the NOA and OA groups, while Table 1 Patient characteristics OA (n = ) NOA (n = ) p-value Age, years n.s. Left testis, ml <.1 Right testis, ml <.1 LH, IU/L <.1 FSH, IU/L <.1 Testosterone, ng/ml n.s. LH, luteinizing hormone; FSH, follicle-stimulating hormone; OA, obstructive azoospermia; NOA, non-obstructive azoospermia; NS, not significant. Data are expressed as mean standard deviations. 16 American Society of Andrology and European Academy of Andrology Andrology, 17,,

3 S. Oka, K. Shiraishi and H. Matsuyama serum testosterone was not significantly different between the groups. Initially, we evaluated the interstitial tissues with H.E. staining. In patients with NOA, the seminiferous tubules became thinner, and the numbers of sperm cells were decreased (Fig. 1B). The interstitial tissues in OA mainly consisted of Leydig cells (Fig. 1A); in contrast, the interstitial tissues in NOA showed increased numbers and types of cells (Fig. 1B). We definitively identified StAR-positive Leydig cells in the interstitial tissues (Fig. 1C and D). Interstitial area excluding Leydig cells in OA was 1.2% and that in NOA was 12.%, which was significantly different between groups (p <.1) (Fig. 1G). The interstitial tissues in NOA showed fibrosis, with picrosirius redpositive collagen fibers, and the seminiferous tubules were engulfed by thickened collagen fibers (Fig. 2A and B). The fibers could not be identified with H.E. staining. In the fibrotic lesions in NOA, the numbers of TE-7-positive fibroblasts were also increased, although these cells appeared more scattered (Fig. 2C and D). The area of StAR-positive Leydig cells in OA was 3.% and that in NOA was 6.6% (Fig. 1G) with no statistical difference (p =.). Leydig cell diameter in NOA was similar to that in OA (Fig. 4G). In contrast, subsets of Leydig cells in NOA showed micronodules (Fig. 1F). Morphologically, spindle-shaped Leydig cells were conspicuously identified in NOA (Fig. 1F). Serial sections immunostained for StAR and PCNA showed that spindleshaped Leydig cells and interstitial fibroblasts were immunostained for PCNA (Fig. 2H). There was a negative correlation between the area of interstitial lesions without Leydig cells and Johnsen scores; however, there was no correlation between the area of interstitial lesions and serum testosterone levels (Fig. 3A and B). Furthermore, the area of Leydig cells was not related to serum testosterone levels (Figure not shown). Compared with before salvage hormonal, the interstitial tissues decreased, and a portion of the Leydig cells became enlarged (Fig. 4A and B). The total interstitial area excluding Leydig cells significantly decreased by an average of 3.% after hormonal (p <.), however, the total Leydig cell area after the salvage hormonal was unchanged compared with that before the (Fig. 4E and F). After hormonal, hypertrophied Leydig cells were identified in clusters (Fig. 4A and B), and the mean diameter of Leydig cells after salvage hormonal was also significantly greater than those before the (1.12-fold, p <.1) (Fig. 4G). Lipid droplets were not identified by H.E. staining. DISCUSSION Previously, several animal studies have reported that chronic hcg treatments induced testicular interstitial alterations, such as Leydig cell hypertrophy and hyperplasia (Lamano-Carvalho et al., 1987; Andreis et al., 1989; Mendis-Handagama et al., 1998). However, there have been few studies involving human Leydig cells. Furthermore, very few attempts have been made to investigate the effects of hcg on testicular interstitial fibrosis. In this study, we investigated testicular interstitial fibrosis and Leydig cells in patients with NOA and evaluated the effects of hcg on interstitial tissues. To the best of our knowledge, this is the first report evaluating interstitial lesions in testicular tissue of infertile men after chronic hcg injections. Although the testicular interstitial tissues of normal men mainly consist of Leydig cells and vessels, the testicular histology of men with spermatogenic failure is often characterized by increased amounts of interstitial tissue and excessive fibrous connective tissue, in addition to Leydig cells and vessels. At present, the adverse effects of testicular interstitial fibrosis on spermatogenesis have not been elucidated. For example, fibrotic reactions, such as severe orchitis, cause extensive scarring associated with impairment of spermatogenesis (El-Fakahany et al., 14), indicating that fibrosis adversely affects spermatogenesis. Therefore, this study was conducted to evaluate interstitial fibrosis in patients with NOA. In this study, interstitial tissues in NOA were evaluated with H.E. staining, immunohistochemical staining for StAR and TE-7, and picrosirius red staining. The number of interstitial lesions, excluding Leydig cells, increased approximately -fold in NOA compared with OA. To investigate the distribution of fibrotic lesions, we used picrosirius red staining, which is used for detection of collagen fibers (Costa et al., ), and we determined that the testicular interstitium was mostly filled with collagen fibers and that the seminiferous tubules were enfolded by thick collagen fibers (Fig. 2B), indicating that the increased interstitial lesions in NOA were almost fibrous in nature. Using TE-7 antibody as a marker of both fibroblasts and myofibroblasts (Goodpaster et al., 7), we also identified many fibroblasts in the fibrotic lesions. These fibroblasts were also positive for PCNA antibody (Fig. 2H), indicating that fibroblast proliferation and collagen synthesis were active in NOA (Essers et al., ; Liu et al., 13). We also investigated the relationship between interstitial tissue and spermatogenesis. Testicular fibrosis was previously reported in NOA (McLachlan et al., 6). In this study, interstitial lesions were strongly and negatively correlated with Johnsen scores (Fig. 3A). This indicates that interstitial fibrosis is related to the severity of impaired spermatogenesis. Furthermore, we compared the numbers and morphologies of Leydig cells in OA and NOA. Previously, conditions causing spermatogenic failure have been linked to Leydig cell hyperplasia (Clegg et al., 1997); however, another report showed the patients with NOA have Leydig cell hypertrophy but not hyperplasia (Tash et al., 2), so this remains controversial. In this study, we used the StAR antibody to distinguish Leydig cells from other interstitial cells and investigated entire Leydig cell areas in several sections of tissue. StAR regulates the rate-limiting step of steroidogenesis and localizes to Leydig cells within the testes (Christenson & Strauss, ); therefore, we concluded that this antibody can be used as a Leydig cell marker. In this study, the area of StAR-positive Leydig cells did not change between OA and NOA (Fig. 1G). Furthermore, the diameter of each Leydig cells also did not change (Fig. 4G), indicating there is no definite alteration in Leydig cell number and size between OA and NOA. In contrast, we could identify that Leydig cell in NOA formed micronodules (Fig. 1F), which was reported as histological marker of spermatogenic failure (Holm et al., 3). Histological heterogeneity of Leydig cells in NOA was reported and the heterogeneity reflects the existence of cells at different stages in differentiation (Lottrup et al., 14). In this study, spindleshaped Leydig cells were notably identified in NOA (Fig. 1F). As the factor facilitating Leydig cell hyperplasia, exogenous hcg has been linked to Leydig cell proliferation (S oderstr om, 1986; Mendis-Handagama et al., 1998) and prolonged and elevated LH 234 Andrology, 17,, American Society of Andrology and European Academy of Andrology

4 AZOOSPERMIA AND TESTICULAR FIBROSIS Figure 1 H.E. staining (A and B) and immunohistochemical staining for steroidogenic acute regulatory protein (StAR), as a marker of Leydig cells (C F). Obstructive azoospermia (OA) specimens are presented as A, C, and E. Non-obstructive azoospermia (NOA) specimens are presented as B, D, and F. Arrows, arrow heads, and empty arrow point up testicular interstitial tissues, StAR-positive Leydig cells, and spindleshaped Leydig cells, respectively. Asterisk indicates Leydig cell micronodules. Testicular interstitial tissues increase and abundant cells are identified in NOA compared to that in OA (A and B). In the interstitial tissues, Leydig cells are clearly immunostained for StAR (C and D). Many spindle-shaped Leydig cells were identified in NOA (F). Interstitial area excluding Leydig cells and another Leydig cell are shown (G). Interstitial area without Leydig cells in NOA are significantly increased compared to that in OA. Data are expressed as mean standard deviations. **p <.1, Mann Whitney rank-sum tests. Bar = lm. HE ( 4) StAR ( 4) (A) (C) OA (B) (D) NOA (E) (F) * StAR ( ) * (G) 2 Interstitial area (%) OA Interstitial area without leydig cell Leydig cell area ** NOA levels in humans (Clegg et al., 1997) in previous reports. However, there was no correlation between LH levels and total Leydig cell areas in this study. We also investigated the relationship between interstitial lesions and serum testosterone levels. In this study, serum testosterone levels were not influenced by interstitial fibrosis. The correlation between interstitial fibrosis and steroidogenesis was slight compared to that between fibrosis and spermatogenesis (Fig. 3A and B). It is generally reported that serum testosterone levels are normal or decreased in patients with NOA (Bobjer et al., 12). Because Leydig cell numbers do not decrease in NOA, hypogonadism in NOA likely results from deterioration of the productivity of individual Leydig cells. In interstitial fibrosis, Leydig cell hyperplasia may compensate for the diminished production of testosterone. We also examined the effects of hcg with regard to interstitial fibrosis. Studies regarding the effects of hcg on the human testicular interstitium are limited, and the effects of hcg on fibrosis are also unclear. We determined that salvage hormonal abated the area of the interstitium excluding Leydig cells by 3.%, which was consistent with 29% reduction in interstitial volume (Fig. 4F). Because these interstitial tissues mainly consisted of fibrosis and many fibroblasts, this finding indicated improvement in fibrosis after salvage hormonal. It was previously reported that hcg for cryptorchidism decreased the interstitial collagen fiber (Favorito et al., ), which also supports the anti-fibrotic effects of hcg. However, the relationship between improvement in testicular fibrosis and spermatogenesis is still unclear. In animal studies, it has been shown that preventing testicular fibrosis after vasectomy with angiotensin II type 1 (AT1) inhibitors prevented the aggravation of spermatogenesis (Shiraishi et al., 3). Recently, anti-fibrotic 16 American Society of Andrology and European Academy of Andrology Andrology, 17,,

5 S. Oka, K. Shiraishi and H. Matsuyama Picrosirius red ( ) TE-7 ( ) (A) (C) OA (B) (D) NOA Figure 2 Picrosirius red staining (A and B) and immunohistochemical staining for TE-7 (C and D), proliferating cell nuclear antigen (PCNA) (E and F), and steroidogenic acute regulatory protein (StAR). Obstructive azoospermia (OA) specimens are presented as A, C, E, G, and I. Non-obstructive azoospermia (NOA) specimens are presented as B, D, F, H, and J. G and I are serial sections and H and J are also serial sections. Arrows heads, arrow, and empty arrow show the TE-7-positive fibroblasts, PCNA-positive fibroblasts, and Leydig cells, respectively. Many collagen fibers stained for picrosirius red are identified in interstitial lesions and seminiferous tubules are engulfed by thick collagen fibers in NOA (B). In the fibrotic lesions, many TE-7 positive fibroblasts are identified (D). PCNA is widely expressed in the increased interstitial cells in NOA (F). Many PCNA-positive fibroblasts were identified in NOA (H). Bar = lm in A F. Bar = 2 lming J. (E) (F) PCNA ( ) (G) (H) PCNA ( 4) (I) (J) StAR ( 4) drugs have been used for the treatment of fibrosis in other organs, such as the liver and lung, with favorable results (Azuma, ; Cohen-Naftaly & Friedman, 11), so the studies about testicular fibrosis and spermatogenesis are encouraging. The mechanisms of the anti-fibrotic effects of hcg are also unclear. Some reports indicate that testosterone has anti-inflammatory effects on human mast cells with androgen receptors (Guhl et al., 12). We previously reported that hcg hormonal induced high levels of intratesticular testosterone (Shinjo et al., 13), which may be involved in the improvements in testicular fibrosis. In addition, we investigated the effects of hcg on Leydig cells. After salvage hormonal, total Leydig cell area did not change compared with that before the, indicating hcg did not induce the increase of Leydig cells. Morphologically, a portion of Leydig cells became hypertrophied 236 Andrology, 17,, American Society of Andrology and European Academy of Andrology

6 AZOOSPERMIA AND TESTICULAR FIBROSIS Figure 3 Correlations between interstitial area without Leydig cells and Johnsen scores (A) and between interstitial area without Leydig cells and steroidogenesis (B). In patients with nonobstructive azoospermia (NOA), interstitial area is significantly correlated with Johnsen scores; in contrast, interstitial area is uncorrelated with serum testosterone levels. (A) Interstitial area without leydig cell (%) 3 R =.73 2 p =. Johnsen's score (B) Interstitial area without leydig cell (%) 3 2 R =.183 p = Testosterone (ng/ml) Figure 4 H.E. staining for interstitial tissues in non-obstructive azoospermia (NOA) before and after salvage hormonal (A D). Specimens of pre-hormonal are presented as A and C. Specimens of post-hormonal are presented as B and D. Interstitial area without Leydig cells significantly decreased after hormonal (E and F). A portion of Leydig cells become hypertrophy after salvage hormonal (D). Data are expressed as mean standard deviations. **p <.1, Wilcoxon signed-rank test. Bar = lm in A and B. Bar = 2 lm in C and D. HE ( 4) (A) Pre-hormonal (B) Post-hormonal (C) (D) HE ( 4) (E) Inters al area without leydig cell (%) 3 2 Pre-hormone Post-hormone (F) Inters al area without leydig cell (%) Pre-hormone Inters al area without leydig cell Leidig cell area * Post-hormone (G) Median diameter of leydig cell (μm) OA Pre-hormone * Post-hormone after salvage hormonal (Fig. 4D) and median diameter of Leydig cells also significantly became greater (Fig. 4G). These results showed that hcg induced Leydig cell hypertrophy but not hyperplasia. Interestingly, this enlargement was noted in several clusters (Fig. 4B). In an animal study, hypertrophic changes and accumulations of lipid 16 American Society of Andrology and European Academy of Andrology Andrology, 17,,

7 S. Oka, K. Shiraishi and H. Matsuyama droplets in the cytoplasm of Leydig cells were reported after chronic hcg injection (Andreis et al., 1989,199). Similarly, the hypertrophic changes noted in this study indicated that these hypertrophied Leydig cells were mainly involved in the production of testosterone in humans (Fig. 4D). Furthermore, it reported that testis in patients with hcg-producing tumor increased undifferentiated Leydig cells (Lottrup et al., 14). We affirmed that undifferentiated cells such as spindleshaped cells did not increase after salvage hormonal. For these reasons, we believe that hcg-based salvage hormonal does not induce unintended proliferation of Leydig cells. There were several potential limitations to our study. We did not directly measure fibrosis or fibroblast levels because of the difficulty of measuring collagen fibers and fibroblasts. Because we determined that interstitial lesions without Leydig cells mainly consisted of fibrosis and fibroblasts, we believe that this method of measuring the fibrosis is applicable for other studies. Another limitation was the small sample size as well as the potential for sample bias. We enrolled only patients with NOA who could not obtain spermatozoa via the first micro-tese. We must compare patients who could and could not obtain spermatozoa with the first micro-tese in the future. However, we think that the small simple size is adequate for evaluation of testicular interstitial lesions because individual differences in interstitial lesions are less comparable with differences in intratubular lesions. Furthermore, the correlation between improvement in interstitial fibrosis and spermatogenesis could not be evaluated in this study; therefore, further studies are needed to clarify the benefit of salvage hormonal in more samples. CONCLUSION We identified an unusual increase in testicular interstitial tissues showing fibrosis with fibroblasts in non-obstructive azoospermia; human chorionic gonadotropin-based salvage hormonal decreased the area of these interstitial tissues. Induction of human chorionic gonadotropin did not increase Leydig cells, but induced hypertrophic change in Leydig cells compared with that before the. These histological changes may not be involved in the deterioration of spermatogenesis. REFERENCES Andreis PG, Cavallini L, Mazzocchi G, Meneghelli V & Nussdorfer GG. (199) Lipid droplets in the secretory response of Leydig cells of normal and hcg-treated rats. J Submicrosc Cytol Pathol 22, Andreis PG, Cavallini L, Malendowicz LK, Belloni AS, Rebuffat P, Mazzocchi G & Nussdorfer GG. (1989) Morphological and functional responses of rat Leydig cells to a prolonged gonadotropins. J Submicrosc Cytol Pathol 21, Azuma A. () Pirfenidone: antifibrotic agent for idiopathic pulmonary fibrosis. Expert Rev Respir Med 4, Bobjer J, Naumovska M, Giwercman YL & Giwercman A. (12) High prevalence of androgen deficiency and abnormal lipid profile in infertile men with non-obstructive azoospermia. Int J Androl 3, Christenson LK & Strauss JF. () Steroidogenic acute regulatory protein (StAR) and the intramitochondrial translocation of cholesterol. Biochim Biophys Acta 29, Clegg ED, Cook JC, Chapin RE, Foster PMD & Daston GP. (1997) Leydig cell hyperplasia and adenoma formation: mechanisms and relevance to humans. Reprod Toxicol 11, Cohen-Naftaly M & Friedman SL. (11) Current status of novel antifibrotic therapies in patients with chronic liver disease. Therap Adv Gastroenterol 4, Costa SF, Costa WS, Sampaio FJ & Favorito LA. () Structural study of gubernaculum testis in fetuses with prune belly syndrome. J Urol 193, El-Fakahany H, Medhat WQ El-Ahwany A. (14) Microscopic testicular sperm extraction can be beneficial in nonobstructive azoospermia patients with history of mumps orchitis even if previous conventional biopsies were unsuccessful. Hum Androl 4, Essers J, Theil AF, Baldeyron C, van Cappellen WA, Houtsmuller AB, Kanaar R & Vermeulen W. () Nuclear dynamics of PCNA in DNA replication and repair. Mol Cell Biol 2, Favorito LA, Hidalgo A Jr, Pazos HM, Costa WS & Sampaio FJ. () Stereological and morphometric analysis of collagen and seminiferous tubules in testes of patients with cryptorchidism submitted or not to treatment with human chorionic gonadotropin. Int Braz J Urol 31, Goodpaster T, Legesse-Miller A, Hameed MR, Aisner SC, Randolph- Habecker J & Coller HA. (7) An immunohistochemical method for identifying fibroblasts in formalin-fixed, paraffin-embedded tissue. J Histochem Cytochem 6, Guhl S, Artuc M, Zuberbier T & Babina M. (12) Testosterone exerts selective anti-inflammatory effects on human skin mast cells in a cell subset dependent manner. Exp Dermatol 21, Holm M, Rajpert-De Meyts E, Andersson AM & Skakkebaek NE. (3) Leydig cell micronodules are a common finding in testicular biopsies from men with impaired spermatogenesis and are associated with decreased testosterone/lh ratio. J Pathol 199, Kato Y, Shiraishi K & Matsuyama H. (14) Expression of testicular androgen receptor in non-obstructive azoospermia and its change after hormonal. Andrology 2, Lamano-Carvalho TL, Favaretto AL, Silva AA & Antunes-Rodrigues J. (1987) Leydig cell hypertrophy and hyperplasia in adult rats treated with an excess of human chorionic gonadotropin (hcg). Gegenbaurs Morphol Jahrb 133, Liu XJ, Kong FZ, Wang YH, Zheng JH, Wan WD, Deng CL, Mao GY, Li J, Yang XM, Zhang YL, Zhang XL, Yang SL & Zhang ZG (13) Lumican accelerates wound healing by enhancing a2b1 integrin-mediated fibroblast contractility. PLoS ONE 8, e Lottrup G, Nielsen JE, Maroun LL, Moller LMA, Yassin M, Leffers H, Skakkebaek NE & Rajpert-De Meyts E. (14) Expression patterns of DLK1 and INSL3 identify stages of Leydig cell differentiation during normal development and in testicular pathologies, including testicular cancer and Klinefelter syndrome. Hum Reprod 29, McLachlan RI, Rajpert-De Meyts E, Hoei-Hansen CE, de Kretser DM & Skakkebaek NE. (6) Histological evaluation of the human testis approaches to optimizing the clinical value of the assessment: mini review. Hum Reprod 22, Mendis-Handagama S, Watkins PA, Gelber SJ & Scallen TJ. (1998) The effect of chronic luteinizing hormone treatment on adult rat Leydig cells. Tissue Cell 3, Schlegel PN. (1999) Testicular sperm extraction: microdissection improves sperm yield with minimal tissue excision. Hum Reprod 14, Shinjo E, Shiraishi K & Matsuyama H. (13) The effect of human chorionic gonadotropin-based hormonal on intratesticular testosterone levels and spermatogonial DNA synthesis in men with non-obstructive azoospermia. Andrology 1, Andrology, 17,, American Society of Andrology and European Academy of Andrology

8 AZOOSPERMIA AND TESTICULAR FIBROSIS Shiraishi K, Takihara H & Naito K. (2) Influence of interstitial fibrosis on spermatogenesis after vasectomy and vasovasostomy. Contraception 6, Shiraishi K, Yoshida K, Fujimiya T & Naito K. (3) Angiotensin II dependent testicular fibrosis and effects on spermatogenesis after vasectomy in the rat. JUrol17, Shiraishi K, Ohmi C, Shimabukuro T & Matsuyama H. (12) Human chorionic gonadotropin treatment prior to microdissection testicular sperm extraction in non-obstructive azoospermia. Hum Reprod 27, Shiraishi K, Ishikawa T, Watanabe N, Iwamoto T & Matsuyama H. (16) Salvage hormonal after failed microdissection testicular sperm extraction: S multi-institutional prospective study. Int J Urol 6, 496. S oderstr om KO. (1986) Leydig cell hyperplasia. Arch Androl 17, 7 6. Tash JA, McCallum S, Hardy MP, Knudsen B & Schlegel PN. (2) Men with nonobstructive azoospermia have Leydig cell hypertrophy but not hyperplasia. J Urol 168, American Society of Andrology and European Academy of Andrology Andrology, 17,,