Expression of extracellular matrix proteins and vimentin in testes of azoospermic man: an immunohistochemical and morphometric study

Kemal Hakan Glkesen¹, Tibet Erdoğru², Canan Figen Sargin³, Glten Karpuzoğlu⁴

1Department of Biostatistics, 2Department of Urology, 3Department of Molecular Biology and Genetics, 4Department of Pathology, Akdeniz University Medical School, Antalya, Turkey

Asian J Androl  2002 Mar; 4:  55-60

Keywords: male infertility; azoospermia; testis; seminiferous tubules; extracellular matrix; vimentin; fibronectins; collagen; laminin; immunohistochemistry

Abstract

1 Introduction

Seminiferous tubules are surrounded by lamina propria, which is composed of basement membrane (BM) and outer three to six layers of myoid cells, connective tissue, and fibroblasts, and the outmost one or two layers are composed of mainly fibroblasts [1]. Type I and IV collagens and laminin are present in the BM and the thickness of the BM increases by 50 % in the aged [2]. The type IV collagen and laminin are produced by the Sertoli cells, while type I and IV collagens and fibronectin are secreted by the myoid cells [3].

Fibronectin is present in the peritubular cells and absent in the Sertoli cells as identified by the immunofluorescence in the rat testicular cell culture [4]. Another study indicated that in monoculture, fibronectin and type IV collagen were positive in the rat peritubular cells, but laminin was negative; in Sertoli cells type VI collagen and laminin were positive and fibronectin was negative. Co-cultures of peritubular cells and Sertoli cells resulted in increased formation and deposition of ECM components [5]. In case of impaired spermatogenesis, the lamina propria was frequently thickened by an increase in the ECM components with differentiation of the inner myoid cells towards fibroblastic nature [6].

In the Sertoli-cell-only syndrome, the localization of laminin and type IV collagen is different from that in the normal testis. In the normal testis both laminin and type IV collagen are localized at the epithelial BM and around the myoid cells. However in the Sertoli-cell-only syndrome, the epithelial BM and the first myoid cell layer are separated by a wide homogenous layer negative for both laminin and type IV collagen, resulting in the appearance of two concentric rings around the tubular lumen: an inner ring representing the BM and the outer ring, the myoid cell layers [7]. This homogenous layer is composed of collagen fibres and deposits of amorphous substance [8] and contains collagen types I and III [9]. In the testes of cryptorchid men, two weakly stained layers of type IV collagen are present; the inner layer occupies the innermost part of the lamina propria and the outer layer surrounds the periphery of lamina propria [10]. A thickened BM is observed in atrophied testis [11]. Normal or pathologic, all testicular tissue demonstrates vimentin positivity in the Sertoli cells [12-14]. Immu-nostaining of myoid cells with anti-vimentin antibodies is weak in normal or near-to-normal tubules, and intense in hyalinised tubules [8,15].

The aim of this study was to investigate the expression of ECM proteins and vimentin and the morphology in the testis of Turkish azoospermic men.

2 Materials and methods

Testicular biopsy specimens, harvested from 88 Turkish males, aged 19-48 years, who had non-obstructive azoospermia and normal karyotype were used for this study. They were divided into three groups according to their histopathologic picture: Group 1 (Sertoli-cell-only syndrome) 22 cases, Group 2 (maturation arrest at spermatocyte level) 20 cases and Group 3 (hyposper-matogenesis with the presence of elongated spermatids and/or spermatozoa) 46 cases. Testis biopsy materials were obtained by open biopsy and fixed for 2-12 hours in Bouin's solution before embedding in paraffin. Four mm sections were prepared for H & E, masson trichrome, PAS, and immunohistochemistry.

After deparaffinization and rehydratation, antigen retrieval was performed in a microwave oven. A water-filled container was heated to 90 and then slides in 10 mmol/L citrate buffer pH 6.0 were placed in the container. Subsequent procedures were carried by the MSIPE protocol by Dako Optimised Staining System and Techmate 500 Plus software (Dako, Glostrup, Denmark) using collagen type IV (1:50, monoclonal mouse anti-human, clone CIV 22, Dako, Glostrup, Denmark), vimentin (1:25, monoclonal mouse anti-human, clone V9, Dako, Glostrup, Denmark) and fibronectin (1:500, polyclonal rabbit anti-human, Ig fraction, Dako, Glostrup, Denmark) antibodies. After counterstaining with H & E, the sections were dehydrated in graded ethanol rinses, cleared in xylene, and coverslipped with Permount. Negative control tests run without primary or secondary antibody.

The 4 mm tissue sections were deparaffinized and rehydrated according to the standard protocol. Antigen retrieval was performed by incubating for 15 minutes with trypsin solution. After peroxidase treatment for 10 minutes, human serum was applied for five minutes. Then, sections were incubated for 30 min at room temperature with anti-laminin antibody (1:30, polyclonal rabbit anti-human, Dako, Glostrup, Denmark) and the antibody was visualized with an avidin-biotin horseradish peroxidase complex using a Universal LSAB kit (Dako, Glostrup, Denmark). After counterstaining with H&E, the sections were dehydrated in graded ethanol rinses, cleared in xylene, and coverslipped with Permount. Negative control tests were run without the primary or secondary antibody.

The slides were examined under a light microscope. The seminiferous tubular diameter and the connective matrix thickness in the tubular wall were measured using a vernier ocular and expressed as mm. The connective matrix zone (CMZ) thickness in the tubular wall is measured as the thickness of the zone between the BM and the peritubular cells of the seminiferous tubular wall.

Seminiferous tubules were typed according to the thickness of their connective matrix in the lamina propria: (a) no apparent connective matrix, (b) the thickness of the connective matrix was 1-5 µm; (c) thickness of connective matrix was more than 5 µm, and (d) hyalinised tubule. This scoring system is modified from Santamaria et al. [8], who subdivided the stage of sclerosis into five types: (a) tubules with normal lamina propria thickness ( 7 µm), (b) tubules with slightly thickened lamina propria (7.1-10 µm), (c) tubules with initial sclerosis (10.1-20 µm), (d) tubules with advanced sclerosis (20.1-35 µm), and (e) tubules with complete sclerosis (>35 µm). The reason of modifying this scoring system is our emphasis on the thickness of the connective matrix rather than the whole seminiferous tubular wall. The best two sections were selected for each case and at least 30 tubules per section were measured for determining the mean tubular diameter and the thickness of the connective matrix. Measured tubules were classified into above described four tubule types, in which the most frequent tubule type for each patient is defined as predominant tubule type. Scoring was performed according to Johnsen [16] and Silber et al. [17].

Data were expressed in meanSD. Chi-square tests, Kruskal-Wallis variance analysis and Mann-Whitney U test were used for statistical analysis performed with the help of SPSS 10.0 software. Correlation tests and binary logistic regression analysis were also used. P< 0.05 was set as significant.

3 Results

The data related to the age, the mean tubular diameter and the mean Johnsen score are summarized in Table 1.

Table 1. Age (year), tubular diameter (µm) and Johnsen score in the patients. Group I: Sertoli-cell-only, Group II: Maturation arrest, Group III: Hypospermatogenesis.

Immunostaining for laminin revealed that it was present in the BM and peritubular cells of seminiferous tubules. Two different staining patterns were observed in the BM of the tubules. In most cases laminin demonstrated a diffuse staining in BM, but in some testes, the laminin deposits formed invaginations into the seminiferous epithelium. A total of 22.5 % of the patients demonstrated laminin deposits and no significant difference was observed between the patient groups. The peritubular cells in Type a tubules were negative or weakly positive with laminin immunohistochemistry, but the peritubular cells in Type b and c tubules were usually positive. The Sertoli cells were uniformly negative. In all patients, most of the Leydig cells demonstrated a granular staining with laminin. In all the patients at least 50 percent of the Leydig cells were positive. The vascular walls were also positive.

With immunostaining, collagen type IV was present in the BM and peritubular cells of the seminiferous tubules. The collagen type IV expression in the peritubular cells disappeared in degenerated tubules. A granular staining is observed in 3322 % of the Leydig cells.

In all the patients the peritubular cells and the interstitial cells were positive with fibronectin immunohisto-chemistry. The Sertoli cells were faintly or weakly posi-tive.

The peritubular cells, the Sertoli cells, the Leydig cells and the interstitial cells were positive with vimentin immunohistochemistry in all the patients.

As can be seen from Table 2, the predominant tubule type was significantly different between the 3 groups (Group 1 vs. 2 P<0.05; Group 1 vs. 3 P<0.01; Group 2 vs. 3 P<0.05).

Both laminin and type IV collagen were localized at the epithelial BMs and the peritubular cells. However, in most of the tubules, the epithelial BM and the first peri-tubular cell layer were separated by a wide homogenous layer (CMZ) negative for both laminin and type IV collagen, resulting in the appearance of two concentric rings around the tubular lumen, an inner ring representing the BM and the outer ring the peritubular cell layers. The peritubular cells were also positive for fibronectin and vimentin. The layer between the BM and the peritubular cell was also negative for fibronectin and vimentin. It appeared as a homogenous blue zone with masson trichrome, and as a homogenous red zone with PAS (Figure 1). Occasionally the peritubular cells were observed near the BM or in the CMZ. Table 3 lists the data related to the minimum and maximum thickness of the CMZ in the seminiferous tubules. When compared according to the presence of CMZ, Group 3 was better than Group 1 (P<0.01) and Group 2 (P<0.01). There was no significant difference between Group 1 and 2. There was a negative correlation between the minimum CMZ thickness and the Silber score (r=-0.288, P<0.01) and the Johnsen score (r=-0.431, P<0.01). The maximum CMZ thickness is higher in Group 1 than in Group 3 (P<0.01). There was a negative correlation between the maximum CMZ thickness and the Silber score (r= -0.320, P<0.01) and the Johnsen score (r=-0.493, P< 0.01).

Binary logistic regression analysis was performed for the presence of elongated spermatids and/or spermatozoa. Morphometric variables (mean tubular diameter, predominant tubule type, minimum CMZ thickness, maximum CMZ thickness) were entered into the equation and only the predominant tubule type revealed a significant association (P<0.01).

Table 3 Number of patients as per minimum and maximum thickness of connective matrix zone. Group I: Sertoli-cell-only, Group II: Maturation arrest, Group III: Hypospermatogenesis.

Figure 1. Photomicrographs (1000) of seminiferous tubular wall. a-d: types of tubule; I: PAS; II: Masson trichrome; III: Vimentin immunohistochemistry; IV: Collagen type IV immunohistochemistry; V: Fibronectin immunohistochemistry.

In this study, we examined the relationship between the morphometric parameters and the state of spermatogenesis and the characteristics of laminin, type IV colla-gen, fibronectin, and vimentin expression in primary infertile azoospermic males.

The mean tubular diameter is 18030 µm in normal testes [1]. In the present study, mean tubular diameter is 14038 µm in Group 3. This finding supports the previous studies reporting a decreased tubular diameter in men suffering from hypospermatogenesis [10,15]. The mean tubular diameter in Group 3 was higher than in other 2 groups, while it was not significantly different between Group 1 and 2. In certain cases, the simultaneous occurrence of complete hyalinisation in some tubules and presence of spermiogenesis in other tubules were observed. Fifty percent of Group 3 patients carried some tubules scored 1 according to Johnsen. These findings demonstrate that in case of infertility, one should not expect a all-encompassing disturbance in the histopathologic picture [18].

In the study of Skinner et al. [5], the cells in the rat peritubular cell culture are negative with laminin. In our study, the peritubular cells were negative with laminin immunohistochemistry when there was no connective matrix in the tubular wall. The tubules with connective matrix demonstrated negative, positive, partial positive or weak staining with laminin immunohistochemistry. The peritubular cells in the end stage tubules were also laminin negative. According to these findings, the peritubular cells in normal testis do not have active laminin synthesis, but through the direct or indirect effect of stimuli deteriorating the testicular structure, the peritubular cells starts laminin synthesis; this synthetic process diminishes in the end stage tubule.

Both the BM and the peritubular cells were positive with collagen type IV immunohistochemistry. The collagen type IV expression in the peritubular cells disappeared in the degenerated tubules. A close relationship between spermatogenesis and the morphology of testes was observed. In testes with no or low sperm production, the diameter of the seminiferous tubules decreases, the thickness of the seminiferous tubule wall increases, a connective matrix is formed between the peritubular cells and the BM, and the thickness of this matrix increases with deteriorating testicular histology. In case of absence of spermatids, there were always tubules with CMZ. Possibly, the presence of the connective matrix in the tubule wall decreases the interchange between the spermatogenic epithelium and the blood and thus disturbs spermatogenesis; disturbed spermatogenesis will cause increased synthesis of connective matrix. The mechanism of this process is not completely defined up to date.

Santamaria et al. [19,20] performed experimental studies to enlighten the physiopathology of this process. In these studies, they created artificial testis ischemia with intra-scrotal injection of epinephrine. In the control testes, BrdU-labelled nuclei (proliferating cells) were observed only in spermatogonia and some primary spermatocytes, whereas testes from epinephrine-treated rats showed BrdU labelling in some of the spermatogonia and peritubular cells. The percentage of peritubular cells that were immunopositive for the proliferating cell nuclear antigen (PCNA) also increased. The absolute volume occupied by laminin and fibronectin immunostaining decreased from the 3rd to the 8th week of treatment, and increased from the 8th to the 11th week. These changes, associated with germ cell depletion and tubular fibrosis, suggest that tubular ischemic atrophy caused by epinephrine alters the peritubular myoid cells, which change their immunophenotype and increase the secretion of the extracellular matrix components producing tubular fibrosis. The mechanism of this alteration may involve direct effects on the peritubular cells or the changes may be secondary to germ cell and/or Sertoli cell lesions.

The relationship between the morphometric data and the success of assisted reproduction awaits further study [21]. On the other hand, the molecular pathogenesis of diminished spermatogenesis in testes should be more clearly defined and possible early diagnostic and therapeutic approaches should be investigated.

In conclusion, a clear relation is present between the testicular morphology and its function. The Seminiferous tubular diameter decreases and the thickness of tubular wall increases in the testis of azoospermic men. The main component responsible of this increase is the CMZ, which is located between the BM and the peritubular cells. The CMZ does not contain laminin, fibronectin, collagen type IV or vimentin. The most important morphologic predictive factor for spermiogenesis is the predominant tubule type.

This research is supported by Akdeniz University Research Foundation.

References

[1] Trainer TD. Testis and excretory duct system. In: Sternberg SS, editor. Histology for Pathologists. 2nd ed. Philadelphia: Lipincott-Raven Publishers; 1997. p 1019-37.
[2] Xi YP, Nette EG, King DW, Rosen M. Age-related changes in normal human basement membrane. Mech Aging Dev 1982; 19: 315-24.
[3] Maekawa M, Kamimura K, Nagano T. Peritubular myoid cells in the testis: their structure and function. Arch Histol Cytol 1996; 59: 1-13.
[4] Tung PS, Skinner MK, Fritz IB. Fibronectin synthesis is a marker for peritubular cell contaminants in Sertoli cell-enriched cultures. Biol Reprod 1984; 30: 199-211.
[5] Skinner MK, Tung PS, Fritz IB. Cooperativity between Sertoli cells and testicular peritubular cells in the production and deposition of extracellular matrix components. J Cell Biol 1985; 100: 1941-7.
[6] Davidoff MS, Breucker H, Holstein AF, Seidl K. Cellular architecture of the lamina propria of human seminiferous tubules. Cell Tissue Res 1990; 262: 253-61.
[7] Pöllanen PP, Kallajoki M, Risteli L, Risteli J, Suominen JJ. Laminin and type IV collagen in the human testis. Int J Androl 1985; 8: 337-47.
[8] Santamaria L, Martin R, Nistal M, Paniagua R. The peritubular myoid cells in the testes from men with varicocele: an ultrastructural, immunohistochemical and quantitative study. Histopathology 1992; 21: 423-33.
[9] Takaba H. A morphological study of the testes in patients with idiopathic male infertility: immunohistochemical analysis of collagens and laminin in human testes. Hinyokika Kiyo 1990; 36: 1173-80.
[10] Santamaria L, Martinez-Onsurbe P, Paniagua R, Nistal M. Laminin, type IV collagen, and fibronectin in normal and cryptorchid human testes. An immunohistochemical study. Int J Androl 1990; 13: 135-46.
[11] Virtanen I, Lohi J, Tani T, Korhonen M, Burgeson RE, Lehto VP, et al. Distinct changes in the laminin composition of basement membranes in human seminiferous tubules during development and degeneration. Am J Pathol 1997; 150: 1421-31.
[12] Aumuller G, Steinbruck M, Krause W, Wagner HJ. Distribution of vimentin-type intermediate filaments in Sertoli cells of the human testis, normal and pathologic. Anat Embryol (Berl) 1988; 178: 129-36.
[13] Steger K, Rey R, Kliesch S, Louis F, Schleicher G, Bergmann M. Immunohistochemical detection of immature Sertoli cell markers in testicular tissue of infertile adult men: a preliminary study. Int J Androl 1996; 19: 122-8.
[14] Bergmann M, Kliesch S. The distribution pattern of cytokeratin and vimentin immunoreactivity in testicular biopsies of infertile men. Anat Embryol (Berl) 1994; 190: 515-20.
[15] Martin R, Santamaria L, Nistal M, Fraile B, Paniagua R. The peritubular myofibroblasts in the testes from normal men and men with Klinefelter's syndrome. A quantitative, ultrastruc-tural, and immunohistochemical study. J Pathol 1992; 168: 59-66.
[16] Johnsen SG. Testicular biopsy score count--a method for registration of spermatogenesis in human testes: normal values and results in 335 hypogonadal males. Hormones 1970; 1: 2-25.
[17] Silber SJ, Rodriguez-Rigau LJ. Quantitative analysis of testicle biopsy: determination of partial obstruction and prediction of sperm count after surgery for obstruction. Fertil Steril 1981; 36: 480-5.
[18] Silber SJ. Sertoli cell only revisited. Hum Reprod 1995; 10: 1031-2.
[19] Santamaria L, Martin R, Codesal J, Ramirez R, Paniagua R. Immunohistochemical quantitative study of the peritubular lamina propria after induction of testicular atrophy induced by epinephrine. Int J Androl 1995; 18: 295-306.
[20] Santamaria L, Martin R, Codesal J, Paniagua R. Myoid cell proliferation in rat seminiferous tubules after ischaemic testicular atrophy induced by epinephrine. Morphometric and immunohistochemical (bromo-deoxyuridine and PCNA) studies. Int J Androl 1995; 18: 13-22.
[21] Seo JT, Ko WJ. Predictive factors of successful testicular sperm recovery in non-obstructive azoospermia patients. Int J Androl 2001; 24: 306-10.

home

Correspondence to: Dr. K. Hakan Glkesen, Akdeniz Üniversitesi Tıp Fakltesi, Biyoistatistik AD, 07059 Arapsuyu, Antalya, TURKEY.
Tel: +90-532-775 7910, Fax: +90-242-227 4482
E-mail: gulkesen@med.akdeniz.edu.tr
Received 2002-01-14      Accepted 2002-02-28