BIOLOGY OF REPRODUCTION 71, 104–116 (2004) Published online before print 25 February 2004. DOI 10.1095/biolreprod.103.024604
Dysplastic Development of Seminiferous Tubules and Interstitial Tissue in Rat Hypogonadic (hgn/hgn) Testes1 Hiroetsu Suzuki,2 Mio Yagi, Kenichi Saito, and Katsushi Suzuki Department of Veterinary Physiology, Nippon Veterinary and Animal Science University, Musashino-shi, Tokyo 180-8602, Japan ABSTRACT
controlled primarily by endocrine factors. From case reports of human disorders and phenotypic analyses of testicular function knockout mice, it has been shown that normal spermatogenesis during adulthood requires all these developmental steps to progress normally . Knockout mice generated by targeting strategies have enhanced our understanding of the molecular mechanisms regulating embryonic gonad differentiation and the endocrine mechanisms controlling prepubertal and pubertal testicular maturation [2, 3]. In contrast, the insufficiency of genetic animal models with mutations in the factors controlling testicular growth and differentiation at the late embryonic and early postnatal stages has hindered our understanding of the molecular mechanisms controlling testicular development at these stages. During the late embryonic to early postnatal stages in normal male rats, the primitive sex cords differentiate into mature seminiferous tubules. Morphological changes in developing seminiferous tubules include the mitosis of primordial germ cells (PGCs), the mitotic quiescence of PGCs into gonocytes, the proliferation of supporting (pre-Sertoli) cells, and the formation of well-organized, single-cell layers of Sertoli cells along the basement membrane [4, 5]. This is followed by the reinitiation of gonocyte mitosis, the movement of gonocytes to the basement membrane, the differentiation of gonocytes into spermatogonia, and the formation of tight junctions between neighboring Sertoli cells [4, 5]. On the other hand, the cytological differentiation of peritubular myoid cells , the maturation of peritubular tissue including myoid cells and connective tissue [7, 8], and the differentiation of adult-type Leydig cells from peritubular cells  occur in the testicular peritubular and interstitial tissue during postnatal development. The molecular regulation of these events remains to be elucidated. Hypogonadic rats of the HGN inbred strain are characterized by male sterility [10, 11], reduced female fertility , and renal hypoplasia [13–15], all controlled by a single recessive allele (hgn) located on chromosome 10 . No other mutant animal or human disease showing a phenotype similar to that of the hgn/hgn rat has been reported. The gonads of the male hgn/hgn fetus are thought to produce rat Sry, Mu¨llerian inhibiting substance (MIS), and testosterone; these males have descended testes and all of the male reproductive accessory organs but no female reproductive organs. However, the weight of an adult hgn/hgn testis is almost equal to that of a normal female ovary and about 1% that of a normal rat testis . In male hgn/hgn rats, the seminiferous tubules are present in neonatal testes, but the postnatal growth of the tubules is severely affected . The population of Sertoli cells is very small, and most of the gonocytes of hgn/hgn rats fail to differentiate into spermatogonia and are likely to have degenerated before
The hypogonadic rat is characterized by male sterility, reduced female fertility, and renal hypoplasia controlled by a single recessive allele (hgn) on chromosome 10. Plasma testosterone is low and levels of gonadotropins are high in adult male hgn/hgn rats, indicating that the cause of hypogonadism lies within the testis itself. We found that the postnatal growth of the seminiferous tubules was severely affected. Here we describe the details of postnatal testicular pathogenesis of the hgn/ hgn rats. In these rats, gonadal sex determination and initial differentiation of each type of testicular cell occur, but proliferation, differentiation, and maturation of these cells during postnatal testicular development is severely affected. Postnatal pathological changes include reduced proliferation and apoptotic cell death of Sertoli cells, abnormal mitosis and cell death of gonocytes, reduced deposition of extracellular matrix proteins into the basal lamina, lack of the formation of an outer basal lamina, formation of multiple layers of undifferentiated peritubular cells, and the delayed appearance and islet conformation of adult-type Leydig cells. Apoptotic cell death of Sertoli cells and disappearance of FSH receptor mRNA expression indicate that this mutant rat is a useful model for Sertoli cell dysfunction. The abnormalities listed above might be caused by defective interactions between Sertoli cells and other types of testicular cells. Because the results presented here strongly indicate that a normal allele for hgn encodes a factor playing a critical role in testicular development, the determination of the gene responsible for hgn and the analysis of early alterations of gene expression caused by mutations in this gene would provide important information on the mechanisms of testicular development.
apoptosis, interstitial cells, Leydig cells, Sertoli cells, testis
Testicular development begins with gonadal sex determination and differentiation, induced by genes downstream of transcription factor Sry. This is followed by late embryonic and early postnatal testicular development, which is controlled primarily by paracrine factors, and prepubertal growth and pubertal maturation of the testes, which are Supported in part by a grant-in-aid for Scientific Research to H.S. (12760204) and to Ka.S. (13660309) from the Ministry of Education, Science and Culture of Japan. 2 Correspondence: Hiroetsu Suzuki, Department of Veterinary Physiology, Nippon Veterinary and Animal Science University, 1-7-1 Kyonan-cho, Musashino-shi, Tokyo 180-8602, Japan. FAX: 81 422 33 2094; e-mail: [email protected]
Received: 7 December 2003. First decision: 5 January 2004. Accepted: 24 February 2004. Q 2004 by the Society for the Study of Reproduction, Inc. ISSN: 0006-3363. http://www.biolreprod.org
TESTICULAR PATHOGENESIS OF HYPOGONADIC RAT
entering meiosis [17, 18]. In hgn/hgn testes, peritubular cells form multiple layers around the seminiferous tubules at Postnatal Day (PND) 3–12 [17, 18], when the differentiation of single layers of alkaline phosphatase (AP)-positive myoid cells is observed in normal testes . Adult hgn/ hgn males display abnormal islet distribution of Leydig cells surrounding the seminiferous tubules as well as low levels of testosterone and high levels of gonadotropins in plasma [10, 11]. The phenotype of the hypogonadic mutant rats strongly suggests that a normal allele for hgn encodes a factor playing critical roles in testis development and function [10, 11, 16]. We presumed that the postnatal development of dysplastic hgn/hgn testes includes the reduction of cell proliferation, the acceleration of cell death, and/or defects in cell differentiation arising from a defect in a single gene, hgn. To characterize the physiological function of the normal allele for hgn in postnatal testicular development, we describe the pathogenesis of hgn/hgn dysplastic testes by immunohistochemical staining of cytoskeletal proteins, extracellular matrix (ECM) proteins, and proliferating cell nuclear antigen (PCNA), as well as by terminal deoxynucleotidyl transferase-mediated dNTP nick end labeling (TUNEL) staining for apoptotic cells and histochemical detection of 3b-hydroxysteroid dehydrogenase (3b-HSD) activity. Using semiquantitative reverse transcription-polymerase chain reaction (RT-PCR) analysis, we assayed the expression levels of follicle stimulating hormone receptor (FSHR). Our results show that, despite the initial differentiation of testicular cells in the fetal gonad, early postnatal proliferation and differentiation of the cells are defective in hgn/hgn rats. MATERIALS AND METHODS
Animals We utilized male hgn/hgn rats and their phenotypically normal littermates (1/?; 1/hgn or 1/1) from the HGN strain maintained in our department . Sister-brother mating of the strain has continued for more than 40 generations. Rats were killed at PND 0, 1, 3, 7, 12, 18, 21, 40, and 80 by an overdose of ether. The testes were removed, weighed to distinguish affected rats from normal ones , and processed. All rats used in this study were fed a certified commercial diet (CR-LPF; Oriental Yeast Co. Ltd., Tokyo, Japan) and kept in a clean, conventional animal room under controlled light (14L:10D), temperature (22 6 18C), and humidity (55% 6 5%). The experimental procedure and care of animals were in accordance with the guidelines of the Animal Care and Use Committee of Nippon Veterinary and Animal Science University.
Preparation of Paraffin Sections and Cryostat Sections Paraffin sections were prepared for azan staining; hematoxylin and eosin (H-E) staining; immunostaining of fibronectin (Fn), laminin (Lm), type IV collagen (IVc), and PCNA; and for the TUNEL technique. The testes were fixed in Bouin solution, with the time of immersion varying according to testicular size (normal testis: 1 h for PND 1, 1.5 h for PND 3, 2 h for PND 7, 4 h for PND 12, and overnight for PND 21, 40, and 80; hypogonadic testis, less than 4 h for all days of age examined). The paraffin sections were prepared as previously described . Sections were deparaffinized in xylene, hydrated in graded alcohol, and immersed in 0.01 M phosphate buffered saline (PBS; pH 7.4) before immunohistochemical analysis. Cryostat sections were used for immunostaining of cytokeratin (Ck), vimentin (Vm), and alpha-smooth muscle isoactin (Sma) and for histochemical detection of 3b-HSD activity. Cryostat sections were prepared as previously described . After being mounted on poly-L-lysinecoated slides, the sections were fixed in cold acetone-ethanol fixative (1: 1 v/v) for 20 min at 2208C. The sections were rinsed in PBS before immunohistochemistry or enzyme detection. The stained sections were examined under a light microscope (BX50; Olympus Co., Tokyo, Japan). The images were obtained by a Penguin 600CL digital camera system (Pixcera Co., Osaka, Japan) attached to the microscope.
Immunohistochemistry of Cytoskeletons and Extracellular Matrix Proteins After paraffin and cryostat sections had been immersed in methanol containing 3% periodic acid to inactivate internal peroxidase, the sections were soaked in PBS containing 5% skim milk to block nonspecific antigen-antibody reactions. The polyclonal antibodies used in this study were anti-rat Fn (1:1000 dilution; Chemicon International Inc., Temecula, CA), anti-rat Ln (1:500 dilution; Chemicon), and anti-mouse IVc (1:500 dilution; LSL Co. Ltd., Tokyo, Japan). Monoclonal antibodies were anti-Vm (1:3 dilution; clone V9; Roche Applied Science, Indianapolis, IN), antiCks 8, 18, and 19 (1:3 dilution, clone 2A4; Biohit, Helsinki, Finland), anti-Sma (1:1 dilution, clone 1A4; Nichirei Co., Tokyo, Japan), and antiPCNA (1:750 dilution, clone PC10; Oncogene Science Inc., Cambridge, MA). Following overnight incubation at 48C, the sections were rinsed in PBS, and the primary antibodies were detected with streptavidin and biotin complex (Histofine SAB-PO kit for monoclonal primary antibodies and SAB-MO kit for polyclonal primary antibodies; Nichirei) or by the labeled polymer method (Histofine Simple Stain PO [MULTI], Nichirei). The slides were incubated with 3,39-diaminobenzidine tetrahydrochloride (DAB) for 2–7 min, and the reaction was stopped by immersion of the sections in tap water. The sections were counterstained with hematoxylin. For negative controls, the primary antibodies were replaced by normal mouse or rabbit serum. Before immunostaining with anti-PCNA antibody, the sections were processed in a microwave (3 min 3 4) in 0.01 M citric acid buffer (pH 6.0) to reactivate their antigenicity, soaked in water, immersed in 2 N HCl for 20 min, and soaked in PBS.
Histochemistry of 3b-HSD 3b-HSD was detected by the tetrazolium salt reaction method . Cryostat sections were fixed in cold acetone for 5 min. The incubation medium for the enzyme reaction contained 2 mM dehydroepiandrosterone as substrate, nicotinamide adenine dinucleotide as cofactor, tetranitroblue tetrazolium as H1 acceptor, 0.01 M potassium cyanide, 0.05 M magnesium chloride, and 0.1 M Tris buffer, pH 7.4. The sections were incubated for 45 min at 378C in the dark in a humidity chamber. After being rinsed in tap water, the sections were counterstained with Nuclear Fast Red. In some cases, the substrate was omitted in the negative control to test specificity (data not shown).
Testicular Cell Counts on Histological Sections Simple morphometric experiments were performed on the paraffin sections stained with H-E. The testes used in this experiment were derived from three 1/? males and three hgn/hgn males at PND 1 and 3. Ten almost-round sections of tubules were randomly selected from histological sections of each testis, and the numbers of Sertoli cells and germ cells in the tubules were counted under a microscope. The diameters of the tubules used for cell counts were also measured using an ocular micrometer. The values were averaged and compared between the 1/? and hgn/hgn males. A Student t-test was used for statistical analysis.
Counts of Total Number of Testicular Cells Total number of testicular cells was estimated by counting the number of cells in an aliquot of cellular suspension from digested testicular tissue. Digestions were performed as described . In brief, the testes were removed and put into PBS, and the surrounding tissues and tunica albuginea were trimmed under a stereoscopic microscope. Testes of both sides were combined and chopped in calcium- and magnesium-free phosphate buffered saline (CMF-PBS). Tissue samples were digested for 20 min at 328C under gentle agitation in CMF-PBS containing 200 mg/ml dispase, 500 mg/ml collagenase (types I and IV), 30 mg/ml soybean trypsin inhibitor, and 35 mg/ml DNase I. The cells were washed in PBS containing 1% bovine serum albumin and suspended in minimum essential medium, and the number of cells was counted on a hemacytometer by trypan blue exclusion. Large cells (those with diameters greater than 15 mm) were designated as gonocytes, and the other cells as testicular somatic cells . Representative values were estimated from 10 identical experiments. Average cell numbers were calculated from 5 hgn/hgn males and 5 1/? males at PND 0 and 3. A Student t-test was used for statistical analysis of the data.
TUNEL Apoptotic cells were detected with an in situ apoptosis detection kit (TaKaRa Bio Inc., Shiga, Japan). After digestion with proteinase K (27
SUZUKI ET AL.
mg/ml) for 20 min, the sections were immersed in 3% periodic acid to block internal peroxidase activity. Fluorescein isothiocyanate (FITC)-conjugated dCTPs were incorporated into nick DNA for 90 min at 378C in the presence of terminal deoxynucleotidyl transferase (TdT). The sections were rinsed in PBS and incubated with HRP-conjugated anti-FITC antibody for 30 min at 378C. The sections were incubated in DAB solution for 15 min, and the reactions were stopped by immersion in tap water. The sections were counterstained in hematoxylin. For the negative control, TdT was omitted.
RNA Extraction and Reverse Transcription-Polymerase Chain Reaction Total RNA was extracted from rat testes using ISOGEN reagent (Wako Pure Chemicals, Osaka, Japan). The RNA samples were reverse transcribed to cDNA for 20 min at 428C using an RT-PCR High Kit (Toyobo, Osaka, Japan), with each 20-ml RT reaction mixture containing 1 mg total RNA, 1.25 mM random primer, 1 mM of each dNTP, 10 units RNAse inhibitor, and 20 units M-MLV reverse transcriptase. The samples were heated at 958C for 5 min to inactivate the reverse transcriptase. Each PCR reaction was carried out in 25 ml of a reaction mixture containing 5 ml cDNA, 1.5 mM MgCl2, 0.1 mM of each primer, 0.25 mM of each dNTP, and 0.6 U rTaq polymerase (Toyobo), in the reaction buffer supplied by the manufacturer. The primers used were those for FSH receptor, 59GCTGATGCAGAAAGAAAGTCGG-39 (sense) and 59-CTCCGGGTTG ATGTACAGAAGA-39 (antisense), which generated a 380-base pair (bp) product, and for G3PDH, 59-ACCACAGTCCATGCCATCAC-39 (sense) and 59-TCCACCACCCTGTTGCTGTA-39 (antisense), which generated a 450-bp product. The amplification protocol consisted of 20–35 cycles of denaturation for 30 sec at 948C, annealing for 30 sec at 608C, and extension for 90 sec at 728C, performed in a Thermal Cycler PTC-100 (MJ Research, Waltham, MA), with the number of cycles optimized for semiquantitative analysis of PCR products. The PCR products were electrophoresed on 2% agarose gels and visualized by ethidium bromide staining. Gel images were obtained using a digital imaging system (Printgraph-CX; ATTO, Tokyo, Japan). The PCR products were subsequently cloned into pGEM T Easy Vector (Promega Co., Madison, WI) and sequenced to confirm the amplification of target cDNA.
Histology of Dysplastic hgn/hgn Testis Stained by Azan Technique
To examine the postnatal development of the seminiferous tubules and interstitial tissue of hgn/hgn testes, we stained testicular sections by the azan technique and examined their histology by light microscopy. This staining method showed not only the cell morphology in the seminiferous tubules but also the fibrous status of the interstitial tissue of the hgn/hgn testes (Figs. 1 and 2). The central area of the testicular sections lacked seminiferous tubules in both genotypes at birth (data not shown). The area was occupied by growing seminiferous tubules in the 1/? testis from PND 3 onward (Fig. 1, A, C, E, and G). In the hgn/ hgn testes at PND 3, the growth of the tubules was severely defective and interstitial mesenchymal tissue remained in the central area of the testes (Fig. 1B). These interstitial tissues became fibrous and rich in collagen fibers after PND 12 (Fig. 1, D, F, and H). In the 1/? testis in the early postnatal stage, the numbers of Sertoli cells increased and the cells formed a palisadelike distribution (Figs. 1A and 2A). Most gonocytes were arrested at the mitotic interphase and located in the centers of the tubules at PND 3 (Fig. 2A). By PND 7, these cells had moved to the basal compartment and differentiated into spermatogonia (data not shown). Some of the germ cells had entered meiosis by PND 12, and spermatocytes were observed (Fig. 2C). In the hgn/hgn testis, the Sertoli cells failed to form a palisade-like structure along the basement membrane, primarily because their number was very small (Figs. 1B and 2B). Gonocytes occupied the insides of the
tubules in the early postnatal stages and often showed abnormal mitotic metaphase (Fig. 2, B and D). We also found some abnormally large gonocytes containing multinuclei (Fig. 2B). In the 1/? testes, by PND 18, each seminiferous tubule had formed a lumen (Figs. 1E and 2E) and many spermatocytes were observed in the tubules (Fig. 2E), and on PND 40, spermatids were observed (Fig. 2G). In the hgn/hgn testes at PND 18, Sertoli cells and indistinctive germ cells were observed in the tubules (Fig. 2F), and there was no lumen (Figs. 1F and 2F). No spermatogenesis was observed at PND 40, and the tubules included abnormal fibrous material (Figs. 1H and 2H). In the peritubular tissue of the 1/? testes, one or two layers of peritubular cells surrounded the tubules at PND 3 (Fig. 2A). A single layer of peritubular cells had established continuous contact with the basement membrane at PND 7 (data not shown), and the outer basal lamina appeared outside a single layer of peritubular cells around PND 12 (Fig. 2, C, E, and G). The peritubular cells, which had become flattened, were located in the space between two basal laminae (Fig. 2, C, E, and G). In the hgn/hgn testis, by contrast, peritubular cells formed three or more layers around the tubules (Fig. 2, B and D). Their peritubular tissue lacked the outer basal lamina of the tubules at PND 12 and 18 (Fig. 2, D and F), and the peritubular cells had not become flattened but were still round (Fig. 2, D and F). These cells, which had become thin, were at times located between the inner and incomplete outer basal laminae at PND 40 (Fig. 2H). In the 1/? testes, some of the interstitial peritubular cells that were separated from the myoid cells by the outer basal lamina had changed their spindle configurations to a round shape by PND 12 and 18 (Fig. 2, C and E). Nearly mature adult Leydig cells were observed at PND 40 (Fig. 2G). In the hgn/hgn testis, the cells forming multiple layers around the tubules did not change their shape, keeping their spindle configuration at PND 12 and 18 (Fig. 2, D and F). At PND 40, however, the cells forming islets around the tubules showed the round shape characteristic of Leydig cells (Fig. 2H). These results suggest that, in the hgn/hgn testis, the proliferation and differentiation of both intra- and extratubular cells were severely defective in the postnatal stages. Immunostaining of Cytoskeletal Proteins
To examine the differentiation and distribution of somatic cells in hgn/hgn testes, cryostat sections were immunostained with antibodies for cell-type-specific cytoskeletal proteins. Ck is one of the transit markers of immature Sertoli cells . In agreement with previous results , Ck was located in the cytoplasm under the nucleus of each Sertoli cell in the 1/? testis at PND 3. Immunostaining for Ck revealed a palisade-like distribution of Sertoli cells along the basement membrane (Fig. 3A). In the hgn/hgn rats, Ck-positive cells were found in the tubules, but there were too few of these cells to form a palisade-like distribution along the basement membrane. Moreover, a considerable number of Ck-positive Sertoli cells were detached from the tubular basement membrane in the hgn/hgn testes (Fig. 3B). Because Ck filaments disappeared with increasing age, we immunostained the sections of adult rat testes with antibody to Vm. This protein, which is not expressed in germ cells, was found to be localized to the testicular somatic cells . In the 1/? tubules, Vm immunostaining was localized around the nuclei of the Sertoli cells, extending toward the apical portions of their cytoplasm (Fig. 3C).
TESTICULAR PATHOGENESIS OF HYPOGONADIC RAT
FIG. 1. Histology of 1/? (A, C, E, G) and hgn/hgn (B, D, F, H) testes at PND 3 (A, B), 12 (C, D), 18 (E, F), and 40 (G, H). Postnatal growth of seminiferous tubules is severely affected in the hgn/hgn testis. Interstitial mesenchymal tissue has remained in the central area of the hgn/hgn testis at PND 3 (B) and has become fibrous tissue including abundant collagen fibers (D, F, H). Degenerated seminiferous tubules surrounded by an islet conformation of adulttype Leydig cells are seen in the fibrous interstitial tissue of the hgn/hgn testis (H). i, Interstitial tissue; t, seminiferous tubule; v, blood vessels; fl, fetal-type Leydig cell; al, adult-type Leydig cells. A–H) Original magnification 360.
In the hgn/hgn testes, immunostaining for Vm revealed that the indistinctive cells remaining in the tubules were not germ cells but somatic cells because they had filaments stained with anti-Vm antibody. The distribution of these filaments was irregular and their extensions were not straight (Fig. 3D). To confirm the differentiation of peritubular myoid cells, we stained testicular sections with antibody to Sma, a cytodifferential marker for peritubular myoid cells . At PND 3, most of the single layer of peritubular cells in contact with the basement membrane was immunostained for Sma in the testes from both normal and hgn/hgn rats, although the peritubular cells forming multiple layers around the hgn/hgn seminiferous tubules were not (Fig. 3, E and
F). At PND 40, intense immunostaining of Sma was detected in the peritubular myoid cells associated with the basement membrane of 1/? testes (Fig. 3G). In contrast, at PND 40, weak but continuous immunostaining of Sma was detected in the single layer of peritubular myoid cells of hgn/hgn testes (Fig. 3H). The islet conformation of multiple layers of round cells around tubules was not stained with the anti-Sma antibody (Fig. 3H). Immunolocalization of ECM Proteins
Because the azan technique revealed abnormal peritubular and fibrous interstitial tissue in the hgn/hgn testis, we examined the distribution of some ECM proteins that con-
SUZUKI ET AL.
FIG. 2. Histology of 1/? (A, C, E, G) and hgn/hgn (B, D, F, H) testes at PND 3 (A, B), 12 (C, D), 18 (E, F), and 40 (G, H). There are few Sertoli cells, which fail to make a palisade-like distribution along the basement membrane in the hgn/hgn testis (B). The insides of the hgn/hgn tubules are full of gonocytes that often show abnormal mitotic metaphase (arrowheads) (B, D). Abnormally large gonocytes including two or three nuclei (*) are also observed (B). Single layers of flattened peritubular myoid cells are located between two layers of the basal lamina (arrows) of the 1/? testis (C, E, G). Peritubular cells of the hgn/hgn testis form multiple layers around the tubules and lack an outer basal lamina (D, F). Adult-type Leydig cells form multiple cellular conformations in the peritubular tissue of the adult hgn/hgn testis (H). In some portions of the basement membrane, flattened peritubular cells (myoidlike cells) are seen between two layers of incomplete basal lamina (arrows) (H). v, Blood vessels; fl, fetal-type Leydig cell; al, adult-type Leydig cells; g, gonocyte; s, Sertoli cell; p, peritubular mesenchymal cell; sg, spermatogonia; sc, spermatocytes; pm, peritubular myoid cells; sp, spermatid cells. A– H) Original magnification 3300.
stitute the basal lamina of the seminiferous tubules. In confirmation of previous findings [22, 23], at PND 3, we found that IVc was localized mainly to a single layer of peritubular cells in contact with the basement membrane and to the basal lamina in 1/? testes (Fig. 4A). In hgn/hgn testes, considerable immunostaining for IVc was observed in the interstitial tissue, and weak staining of IVc was detected in a single layer of peritubular cells associated with the basal lamina. In these animals, however, the outer multiple layers of peritubular cells were negative for IVc staining (Fig. 4B). Fn is a marker of peritubular myoid cells and a component of the basal lamina of peritubular tissue [23, 24]. In 1/? testes, immunostaining of Fn was observed in both the peritubular and interstitial tissues at PND 7 (Fig. 4C). In
hgn/hgn testes, however, multiple layers of peritubular cells were not stained for Fn (Fig. 4D). At PND 18, intense Fn staining was seen in the basal lamina of 1/?, but not in that of hgn/hgn (data not shown). Lm is an important component of the basal lamina and is deposited in the inner basal lamina in the early stages of testicular development . We were able to detect accumulation of Lm in the peritubular cells at PND 3 (data not shown) and basal lamina at PND 18 (Fig. 4E) of 1/? testes. In the hgn/hgn testes, however, staining for Lm was poor in both the peritubular cells and basal lamina in hgn/hgn testes (Fig. 4F). In the 1/? testis at PND 18, IVc accumulation was observed in both the inner and outer basal laminae (Fig. 4G). In contrast, in the hgn/hgn testis at PND 18, staining of the outer
TESTICULAR PATHOGENESIS OF HYPOGONADIC RAT
FIG. 3. Immunostaining of Ck (A, B, PND 3), Vm (C, D, PND 80), and Sma (E, F, PND3; G, H, PND 40) in cryostat sections of 1/? (A, C, E, G) and hgn/hgn (B, D, F, H) testes. Ck-positive Sertoli cells form a palisade-like distribution along the basement membrane (A). There are few Ck-positive cells, and these fail to form a palisade-like structure in the hgn/hgn testis (B). Although some Ck-positive cells have lost contact with the basement membrane in the testes of both genotypes, more cells are detached in the hgn/hgn testes (arrow) (A, B). There is intense immunostaining of Vm around the nuclei of indistinctive cells remaining in the hgn/hgn tubules (D). In hgn/hgn testes, peritubular cells in contact with the outer surface of seminiferous tubules are weakly immunostained for Sma (F, H). Islet conformation of Sma-negative cells around the tubules represents Leydig cells (H). i, Interstitial tissue; t, seminiferous tubule; al, adult-type Leydig cells. A– H) Original magnification 3120.
basal lamina with antibody to IVc was not detected in the peritubular tissues. Although immunostaining of IVc in the inner basal lamina was observed, the intensity was very weak (Fig. 4H). Taken together, these results indicate that the distribution of ECM proteins in the peritubular and interstitial tissues was abnormal in developing hgn/hgn testes. Histochemical Localization of 3b-HSD
To characterize the Leydig cells, we performed histochemical assays on cryostat sections for the activity of 3bHSD. This is a key enzyme in the biosynthesis of androgen and is detected in both fetal- and adult-type Leydig cells [9, 19]. We detected 3b-HSD activity in fetal-type Leydig
cells of both the neonatal 1/? and hgn/hgn testes (Fig. 5, A and B). From PND 18 onward, 3b-HSD-positive adult-type Leydig cells were found in the interstitial tissue of 1/? testes (Fig. 5, C and E). However, the multiple layers of peritubular cells around the seminiferous tubules were negative for 3b-HSD activity during the postnatal period in hgn/hgn testes (Fig. 5D). At PND 40 in the latter, cells forming an islet conformation around the seminiferous tubules were positively stained for 3b-HSD (Fig. 5F). Testicular Cell Count
To obtain statistical evidence showing that the hgn/hgn phenotype alters the number of Sertoli and germ cells in
SUZUKI ET AL.
FIG. 4. Immunostaining of IVc (A, B, PND3; G, H, PND 18), Fn (C, D, PND7), and Lm (E, F, PND 18) in 1/? (A, C, E, G) and hgn/hgn (B, D, F, H) testes. In 1/? testes, immunostaining of IVc, Fn, and Lm are primarily localized to the peritubular tissue, including the tubular basement membrane and single layers of peritubular myoid cells (A, C, E). In the hgn/hgn testis, single layers of peritubular cells associated with tubular basement membrane are stained with the antibody to IVc (B) but not with antibodies to Fn (D) and Lm (F). Although the interstitial tissue is strongly stained with antibody to the ECM proteins, multiple layers of peritubular cells showed little or no immunostaining (B, D, F). Two layers of basal lamina surrounding the seminiferous tubules are intensely immunostained with antibody to IVc at PND 18 (arrow) (G). Although there is weak staining associated with the inner basal lamina (arrowhead) in the hgn/hgn testis, no outer basal lamina stained with antibody to IVc was observed (H). p, Peritubular mesenchymal cell; pm, peritubular myoid cells; i, interstitial tissue; t, seminiferous tubule; v, blood vessels. A–F) Original magnification 3120; (G, H) original magnification 3300.
the seminiferous tubules, we counted Sertoli cells and gonocytes at PND 1 and 3. Although we found that the diameters of the tubular sections were comparable in both genotypes, the number of Sertoli cells in the seminiferous tubules of the hgn/hgn rats was about half that in the tubules of the 1/? rats, but the number of gonocytes was about three times higher (Table 1). To estimate the total numbers of somatic and germ cells in these testes, we enzymatically dispersed the testicular tissue at PND 0 and 3 and counted the number of somatic and germ cells. We found that the numbers of both cells were significantly smaller in the hgn/hgn than in the 1/? testes at both time points examined (Fig. 6). Whereas at
birth, the number of somatic cells in hgn/hgn testes was about one third that in 1/? testes, it had declined to about one eighth at PND 3. In contrast, the number of germ cells in hgn/hgn testes remained about half that of 1/? testes at both time points. Thus, the proliferation of somatic cells was severely defective, and the ratio of gonocytes to somatic (Sertoli) cells was abnormally high in hgn/hgn testes, findings consistent with our previous observation  that gonocytes constituted most of the tubules in hgn/hgn testes at PND 1 and 3. In addition, about two thirds of the gonocytes in hgn/hgn testes were dead on dye-exclusion testing, whereas most of the gonocytes in 1/? testes were alive.
TESTICULAR PATHOGENESIS OF HYPOGONADIC RAT
FIG. 5. Histochemical localization of 3bHSD activity in 1/? (A, C, E) and hgn/hgn (B, D, F) testis at PND 0 (A, B, original magnification 3120), 18 (C, D, original magnification 3300), and 40 (E, F, original magnification 3120). Clusters of 3bHSD positive fetal-type Leydig cells are observed in the interstitial tissue (A, B). 3b-HSD positive adult Leydig cells are present in the interstitial tissue of the 1/? testis at PND 18 (C). Multiple layers of peritubular cells around the hgn/hgn tubules are negative for 3b-HSD (D). Adulttype Leydig cells forming islet conformations in the hgn/hgn testis show 3b-HSD activity at PND 40 (F). i, Interstitial tissue; t, seminiferous tubule; fl, fetal-type Leydig cell; al, adult-type Leydig cells; p, peritubular mesenchymal cell.
Cell Proliferation Activity and Apoptosis
some of which were at mitotic metaphase, were stained with anti-PCNA antibody in the hgn/hgn testis (Fig. 7B). In the interstitial tissue of hgn/hgn testes, multiple layers of peritubular mesenchymal cells showed immunostaining for PCNA (Fig. 7B). To confirm that apoptosis is related to a reduction in numbers of Sertoli cells and increased cell death of enzymatically separated gonocytes, we assayed the apoptosis of these cells by the TUNEL method. No apoptotic cells were detected in the 1/? testes at PND 1 (Fig. 7C), whereas a considerable number of TUNEL-positive cells was observed in the seminiferous tubules in the hgn/hgn testes (Fig. 7D). Most of the apoptotic cells in the latter appeared to be Sertoli cells because they had small nuclei and were located in the basal region of the tubules. At PND 7, phys-
Our cell counts suggested that there is a defect in the regulation of cell proliferation in the perinatal hgn/hgn testis. To assay cell proliferation activity, we performed immunostaining of PCNA. PCNA is an auxiliary protein of DNA polymerase that is required for DNA replication during the S phase and therefore is exclusively expressed in proliferating cells [25, 26]. In 1/? testes at PND 3, Sertoli cells showed intense immunostaining of PCNA, whereas gonocytes arrested at mitotic interphase were negative for PCNA staining (Fig. 7A). In hgn/hgn testes, the nuclei of most Sertoli cells were flattened or irregular and stained with PCNA. However, the intensity of the staining varied, and some of the Sertoli cells of hgn/hgn testes showed weak immunostaining for PCNA (Fig. 7B). The gonocytes,
TABLE 1. Number of Sertoli cells and gonocytes in the seminiferous tubule of hgn/hgn and 1/? testis at postnatal day (PND) 1 and 3.* PND 1 Diameter of tubule (mm) Number of Sertoli cells Number of germ cells
65.4 6 4.4 9.1 6 0.2a 8.8 6 1.4a
60.2 6 1.8 21.3 6 1.8 3.3 6 0.9
68.4 6 2.2 9.4 6 1.3a 8.5 6 2.1a
59.9 6 5.8 22.6 6 1.8 2.3 6 0.1
* Each value indicates mean 6 SD (n 5 3). a Significant difference between hgn/hgn and 1/? at P , 0.05 (Student t-test).
SUZUKI ET AL.
as indicated by their reduced number and increased apoptosis. The FSH receptor, which is expressed by Sertoli cells, is necessary for normal postnatal proliferation and differentiation of the cells [27, 28]. We analyzed the expression of FSHR mRNA in postnatal testes (Fig. 8). Expression of FSHR mRNA was detected at all ages examined in the 1/? testes. In the hgn/hgn testes, the level of expression at birth was comparable with that of 1/? testes. The expression gradually decreased with age and had completely disappeared by PND 12. These results suggest that the reduced mitotic activity and increased apoptosis of Sertoli cells in the early postnatal stages in hgn/hgn testes are independent of the defective expression of FSHR.
FIG. 6. Total number of Sertoli (A) and germ (B) cells in hgn/hgn and 1/? testes at PND 0 and 3. Each value represents the mean from five animals, with the vertical bars representing standard deviations. There are significantly fewer types of both cells in hgn/hgn than in 1/? testes at both ages. Black column, Total number of cells; clear column, number of viable cells. Significant difference in each value between hgn/hgn and 1/? is labeled as * (P , 0.01) and ** (P , 0.05).
iological apoptosis of germ cells was rarely observed in 1/? testes. TUNEL-positive cells were merely detected in hgn/hgn testes at PND 7 (data not shown), although abnormal mitotic metaphases of gonocytes were often observed. Expression Pattern of FSH Receptor
In the early postnatal pathogenesis of hgn/hgn dysplastic testes, a distinct early defect found in somatic cells is the failure of proliferation and differentiation of Sertoli cells,
DISCUSSION During male cord formation of fetal gonad development, Ck-positive epithelial cells (precursor Sertoli cells) differentiate from mesenchymal cells . Precursor Sertoli cells proliferate rapidly and form a palisade-like distribution along the basement membrane, enclosing the gonocytes at the perinatal stage . Although our finding of Ck-positive Sertoli cells at PND 3 in hgn/hgn testes indicated that initial differentiation of these cells occurs in the fetal testis, they were sparsely distributed and failed to make normal distribution because there were few precursor cells. Moreover, in these testes, a considerable number of Ck-positive cells have lost contact with the basement membrane and are located inside the seminiferous tubules. This disruption of the cell-matrix adhesion machinery between Sertoli cells and the basement membrane may be involved in apoptosis of Sertoli cells and/or poorly formed basement membranes. Although Sertoli cells proliferate from the embryonic to the prepubertal stage , their highest mitotic activity is observed during the perinatal stage , indicating that the increase in numbers of testicular cells at this stage is due primarily to the proliferation of Sertoli cells. Our cell-count experiment showed that the number of Sertoli cells was significantly lower in hgn/hgn than in normal tubules at both PND 1 and 3 and that the perinatal proliferation of somatic (Sertoli) cells is defective in hgn/hgn testes. The establishment of a normal population of Sertoli cells before puberty is necessary for normal spermatogenesis and normal levels of sperm production because each Sertoli cell has the capacity to support only a fixed number of germ cells . Although the minimum number of Sertoli cells needed to establish a base for spermatogenesis is not known, there may be too few in the hgn/hgn seminiferous tubules. We found that the intensity of PCNA staining of Sertoli cells was varied in hgn/hgn testes at PND 3. We detected some Sertoli cells showing poor PCNA staining. This observation indicates that the proliferative activity of Sertoli cells is decreased in the hgn/hgn testis. PCNA is also involved in cell cycle arrest and DNA repair [32, 33]. Thus, it is possible that the intense PCNA staining in the irregular nuclei of hgn/hgn Sertoli cells might be involved in accumulation of the protein associated with cell cycle arrest in the apoptotic process. This is supported by our observation that the apoptotic bodies of Sertoli cells were strongly stained for PCNA in the hgn/hgn testis. There were a considerable number of TUNEL-positive Sertoli cells in the hgn/hgn tubules at PND 1, and most of these cells were also positive for Vm by double-staining methods (data not shown). We were unable to find any reports describing similar apoptosis of Sertoli cells during the perinatal period. Because Sertoli cells play a central role in testicular devel-
TESTICULAR PATHOGENESIS OF HYPOGONADIC RAT
FIG. 7. Immunostaining of PCNA in 1/? (A) and hgn/hgn (B) testes at PND 3 (original magnification 3300) and TUNEL staining for apoptotic cells in 1/? (C) and hgn/ hgn (D) testes at PND 1 (original magnification 3230). The nuclei of Sertoli cells, forming a palisade-like distribution along the basement membrane, show intense staining of PCNA with constant intensity. Gonocytes are negative for PCNA (A). In the hgn/hgn testis (B), there are few Sertoli cells and the intensity of PCNA staining is varied among them. Some Sertoli cells show weak staining (arrowhead). An apoptotic body is also strongly stained for PCNA (*). Most of the gonocytes, including the cells showing mitotic metaphase or having three nuclei (arrow), are positive for PCNA. Peritubular cells are highly stained for PCNA in the hgn/hgn testis (B). TUNEL-positive cells are not detected in the 1/? testis (C). A considerable number of TUNEL-positive cells (arrowheads) are observed in the hgn/hgn testis (D).
opment and spermatogenesis, their failure to proliferate and differentiate results in disorders of testicular function [1, 28]. However, we were unable to identify any animal model other than the hgn/hgn rat that showed distinct disorders in the perinatal proliferation and differentiation of Sertoli cells. Although postnatal dysplastic development of testes has been reported in male Dhh-null and DAX1-deficient mice, the primary pathological alterations are found in the peritubular tissues [34–37]. In these mice, intratubular abnormalities are likely to be secondary events. The dysfunction of Sertoli cells in hgn/hgn is one of the reasons why these rats show more severely dysplastic testes. FSHR is expressed in Sertoli cells, and FSH has been shown to increase the rate of proliferation of Sertoli cells [38–40]. Plasma levels of gonadotropins are consistently low during the early postnatal stages in normal rats. In hgn/ hgn rats, however, FSH levels are elevated from PND 3 onward [11, 17]. Our ability to detect FSHR mRNA expression in early postnatal hgn/hgn testes indicates that the reduced proliferation of Sertoli cells is not directly caused by defective expression of FSHR. However, the disappearance of FSHR mRNA expression beginning at PND 12 indicates that Vm-positive Sertoli cells in the degenerated hgn/hgn tubules lose their expression of FSH receptor. There are many paracrine factors that may be involved in the regulation of Sertoli cell proliferation and differentiation [28, 41–47]. To date, no factor has yet been identified that is expressed exclusively in the testes and that induces Sertoli cell-specific proliferation during the late embryonic
and early postnatal periods. Because Sertoli cells fail to proliferate and differentiate into mature Sertoli cells, the defect in hgn/hgn rats may include the defective expression of certain critical factor(s) regulating Sertoli cell proliferation and differentiation.
FIG. 8. Expression of FSH receptor mRNA in 1/? and hgn/hgn testes. An FSH receptor-specific fragment was amplified by RT-PCR from total RNA samples extracted from 1/? and hgn/hgn testes at PND 0, 1, 3, 7, 12, and 18. G3PDH cDNA fragments were also amplified from the same RNA samples as internal controls. Representative results from triplicate experiments are shown.
SUZUKI ET AL.
Following the active proliferation of PGCs in the sex cord during the period from Embryonic Day (ED) 14–17, these cells experience an arrest of mitosis, which continues for about 8 days during rat testicular development . Around PND 3, the gonocytes resume their mitotic activity, relocate toward the basal compartments, and differentiate into spermatogonia [4, 5]. In female rats, on the other hand, PGCs enter meiotic prophase and arrest in the dictyate stage until a few days after birth . Somatic cells in the male gonad are thought to prevent PGCs from entering meiosis and direct them to spermatogenesis . On histological examination, neonatal hgn/hgn PGCs showed a cytology typical of gonocytes, suggesting that fetal male hgn/hgn gonads have the capacity to prevent PGCs from entering meiosis. However, the gonocytes of these animals often showed abnormal mitotic metaphases. We found that the ratio of gonocytes to Sertoli cells in the tubules was extremely high in hgn/hgn testes. Because gonocytes are functionally coupled with Sertoli cells via gap junction and cell adhesion molecules , this elevated ratio may be related to the abnormal proliferation and degeneration of gonocytes in hgn/hgn testes. In addition, we often observed multinuclear germ cells in the perinatal hgn/hgn tubules. Resumption of gonocyte proliferation produces daughter cells connected by intercellular bridges during early postnatal development of normal testes [50, 51]. It is therefore likely that the appearance of multinuclear germ cells would result from the premature proliferation of gonocytes in the absence of an appropriate support of Sertoli cells. During the perinatal period in the normal testis, gonocytes are arrested in the G0/G1 stage and are negative for immunostaining of PCNA . In hgn/hgn testes, however, we detected PCNA-positive gonocytes at PND 0 (data not shown) and 3. Because the relocation of gonocytes and their resumption of mitosis are thought to be independent of each other , there is a possibility that gonocytes could restart mitosis before being enclosed in a basal compartment under such abnormal conditions. It has been proposed that, if PGC meiosis is inhibited in the male gonad environment, then PGC mitotic arrest is automatically accomplished by a mechanism inherent to the germ cells . We found, however, that abnormal mitosis of hgn/hgn gonocytes was present as early as ED 19 (data not shown), suggesting that there may be a mechanism preventing gonocytes from entering mitosis that is defective in the late embryonic and early postnatal hgn/hgn testis. Because the defect in the hgn/hgn phenotype originates in somatic rather than germ cells, the interaction between Sertoli cells and PGCs may regulate the mitotic arrest of gonocytes. Testicular somatic cells produce mitogens to accelerate their proliferation during the period when PGCs arrest mitosis. Thus, it is possible that the palisade-like distribution of Sertoli cells would enclose the PGCs, preventing their access to an environment rich in mitogens, and that compact clustering of PGCs surrounded by Sertoli cells  would provide contact inhibition for mitosis of PGCs. This morphological milieu, however, could not be established in neonatal hgn/hgn testes. In contrast with our histological results, we found that the total number of gonocytes was smaller in hgn/hgn than in 1/? testes at birth and that the increase in the number of gonocytes during the period from PND 0 to PND 3 was also smaller in hgn/hgn than in 1/?. This inconsistency is due to accelerated cell death of gonocytes in the degenerated hgn/hgn tubules. In our dye-exclusion experiments, we found that most neonatal gonocytes had progressed to cell
death in hgn/hgn testes, although TUNEL staining did not show distinct evidence of gonocyte apoptosis. Some reports have suggested that the physiological apoptosis of germ cells occurs at the same time as peak mitosis and that the death of gonocytes during their arrest phase occurs by a process resembling necrosis [54, 55]. Therefore, gonocyte death is likely to progress through necrosis in hgn/hgn testes because of the incomplete support provided by the defective Sertoli cells. Although we did not utilize a gonocyte-specific marker in this study, it is a potential hypothesis that the gonocytes of hgn/hgn testes failed to arrest mitosis, failed to differentiate into spermatogonia, and degenerated through a necrotic process during postnatal testicular development. Our findings suggest that these defects may be related to the functional failure of Sertoli cells. Although testicular cords are formed in the absence of primordial germ cells , interaction between Sertoli cells and peritubular myoid cells is necessary for testicular cord formation. Peritubular myoid cells differentiate from progenitor cells that have migrated from the mesonephros [57, 58]. Immature peritubular myoid cells in contact with the outer surface of the seminiferous tubular basal lamina are thought to differentiate into mature myoid cells during postnatal testicular development [6, 7]. The initial differentiation of peritubular myoid cells should occur in hgn/hgn gonads because testicular cords lined with continuous basement membrane were observed at birth. In our previous report, however, we showed that the seminiferous tubules were surrounded by multiple layers of AP-negative peritubular cells . In the present study, we showed that Sma-positive peritubular myoid cells were in contact with the basement membrane in hgn/hgn testes at PND 3. Thus, the differentiation of Sma-positive peritubular myoid cells occurred at the time programmed, although the appearance of a single layer of mature AP-positive myoid cells was delayed in hgn/hgn testes . In normal testes, the peritubular myoid cells became flattened and located in the space between the inner and outer basal laminae at PND 12. This organization of peritubular tissue may be important for normal maturation of peritubular myoid cells because a continuous layer of AP activity and intense immunostaining of desmin appear in peritubular myoid cells at about the same stage in normal testicular development [6, 24]. In hgn/ hgn testes, however, these cytodifferential markers were not expressed in the peritubular myoid cells during early postnatal development. Beginning at PND 21, AP activity was detected in a single layer of peritubular cells, but even in the hgn/hgn testis at this stage, the peritubular cells became flattened and the outer basal lamina was partly formed outside a single layer of peritubular cells. Therefore, it may be important for the differentiation of mature myoid cells that a single layer of peritubular cells in contact with the basement membrane produces ECM proteins to effect a separation from the outer peritubular cells by establishing an outer basal lamina. Peritubular myoid cells cooperate with Sertoli cells to produce ECM proteins, which compose the basement membrane of the cord. The reduced number of Sertoli cells and the defective differentiation of Sertoli cells and peritubular myoid cells in hgn/hgn testes may be associated with a reduction of ECM protein production and deposition into the basal lamina. Lm is produced by both Sertoli and peritubular myoid cells and is an important component of the inner basal lamina . Lm has been reported to promote the adhesion of Sertoli cells to the basement membrane and the differentiation of premature Sertoli cells . There-
TESTICULAR PATHOGENESIS OF HYPOGONADIC RAT
fore, the degeneration of Sertoli cells in hgn/hgn testes may be involved in the reduced production of Lm. Peritubular myoid cells are a major source of Fn [61, 62], which is deposited primarily in the outer basal lamina . Lack of Fn staining in the basement membrane and peritubular tissue may be involved in developmental and functional defects of peritubular myoid cells in hgn/hgn rats. IVc is synthesized by both Sertoli and myoid cells . In normal rat testes, IVc was continuously deposited in the inner and outer basal laminae at PND 18. In contrast with other ECM proteins, IVc was localized in a single layer of the peritubular myoid cells in hgn/hgn perinatal testes. However, the lack of IVc accumulation in the outer basal lamina of postnatal hgn/hgn tubules supports the crucial relationship between the maturation of myoid cells and the deposition of IVc in the outer basal lamina. Fetal-type Leydig cells also originate from progenitor cells derived from mesonephric tissue . During postnatal testicular development, fetal-type Leydig cells are replaced by adult-type Leydig cells as a major source of testosterone, where the adult-type cells differentiate from peritubular cells . The initial differentiation of fetal Leydig cells should have occurred in hgn/hgn gonads because 3bHSD-positive cells are found in early postnatal testes. However, the appearance of adult-type Leydig cells was delayed in the hgn/hgn testis. The multiple layers of peritubular cells around the seminiferous tubules were negative for 3bHSD at PND 18, a point at which adult-type Leydig cells show activity in normal testes. However, we were unable to detect any distinct expression of myoid cell-specific marker proteins (AP , Sma, and desmin [data not shown]) in these multilayered cells. Further, these cells showed intense immunostaining for PCNA but little or no immunostaining for ECM proteins such as Fn, Lm, IVc, and Ic (data not shown). These results suggest that the multilayered cells may be indifferent peritubular cells that have deviated from the normal process of differentiation, derived from an excess of peritubular cells resulting from growth retardation of the hgn/hgn seminiferous tubules. Dhh is a signaling molecule expressed by Sertoli cells. The expression of Dhh is necessary for normal development of adulttype Leydig cells and peritubular tissues [34, 35]. Some pathological alterations in the hgn/hgn testis appear to be similar to those of Dhh-null mice. It is reasonable to propose that insufficient secretion of Dhh because of dysfunction of the Sertoli cells might result in the defective differentiation of peritubular tissues and the delayed appearance of adult-type Leydig cells in the hgn/hgn testis. Further investigations are required to confirm this possibility.
6. 7. 8. 9. 10.
12. 13. 14.
ACKNOWLEDGMENTS 23. We thank Ms. Junko Takahashi for her help in cell-count experiments. We thank Mr. Eijiro Nakamiya and Ms. Kumi Daigo for their time-consuming efforts to maintain the HGN strain. 24.
REFERENCES 1. Skakkebaek NE, Rajpert-De Meyts E, Main KM. Testicular dysgenesis syndrome: an increasingly common developmental disorder with environmental aspects. Hum Reprod 2001; 16:972–978. 2. Burns KH, Matzuk MM. Minireview: genetic models of the study of gonadotropin actions. Endocrinology 2002; 143:2823–2835. 3. Cotinot C, Pailhoux E, Jaubert F, Fellous M. Molecular genetics of sex determination. Semin Reprod Med 2002; 20:157–168. 4. Mackay S. Gonadal development in mammals at the cellular and molecular levels. Int Rev Cytol 2000; 200:47–99. 5. Orth JM, Jester WF, Li LH, Laslett AL. Gonocyte-Sertoli cell inter-
actions during development of the neonatal rodent testis. Curr Top Dev Biol 2000; 50:103–124. Palombi F, Farini D, Salanova M, de Grossi S, Stefanini M. Development and cytodifferentiation of peritubular myoid cells in the rat testis. Anat Rec 1992; 233:32–40. Lesson CR, Lesson TS. The postnatal development and differentiation of the boundary tissue of the seminiferous tubule of the rat. Anat Rec 1963; 147:243–260. Ross MH. The fine structure and development of the peritubular contractile cell component in the seminiferous tubules of the mouse. Am J Anat 1967; 121:523–557. Mendis-Handagama SM, Ariyaratne HB. Differentiation of the adult Leydig cell population in the postnatal testis. Biol Reprod 2001; 65: 660–671. Suzuki K, Hakamata Y, Hamada A, Kikukawa K, Wada MY, Imamichi T. Male hypogonadism as a candidate of deficiency of postnatal testicular growth or differentiating factor(s): a new autosomal recessive mutation in the rat. J Hered 1988; 79:54–58. Hakamata Y, Kikukawa K, Kamei T, Suzuki K, Taya K, Sasamoto S. A new male hypogonadism mutant rat (hgn/hgn): concentrations of testosterone (T), luteinizing hormone (LH), and follicle-stimulating hormone (FSH) in the serum and the responsiveness of accessory sex organs to exogenous T, FSH, human chorionic gonadotropin, and luteinizing hormone-releasing hormone. Biol Reprod 1988; 38:1145– 1153. Suzuki H, Hakamata Y, Kamei T, Kikukawa K, Suzuki K. Reproduced fertility in female homozygotes for hgn (male hypogonadism) selected by hgn-associated hypoplastic kidney. Cong Anom 1992; 32:167–178. Suzuki K, Suzuki H, Hakamata Y, Kamei K, Kikukawa K. Genetic analysis and histology of hypoplastic kidneys in the male hypogonadic mutant (hgn/hgn) rat. Cong Anom 1991; 31:305–314. Suzuki H, Suzuki K. Pathophysiology and postnatal pathogenesis of hypoplastic kidney (hpk/hpk) in the male hypogonadic mutant rat (hgn/hgn). J Vet Med Sci 1995; 57:891–897. Suzuki H, Suzuki K. Rat hypoplastic kidney (hpk/hpk) induces renal anemia, hyperparathyroidism, and osteodystrophy at the end stage of renal failure. J Vet Med Sci 1998; 60:1051–1058. Suzuki H, Kokado M, Saito K, Kunieda T, Suzuki K. A locus responsible for hypogonadism (hgn) is located on rat chromosome 10. Mamm Genome 1999; 10:1106–1107. Suzuki K, Hakamata Y, Suzuki H, Taya K, Sasamoto S. Postnatal alterations of testicular morphology, testicular testosterone contents and plasma hormone levels in male hypogonadic rat (hgn/hgn). J Reprod Dev 1993; 39:333–346. Suzuki H, Inaba M, Suzuki K. Temporal and spatial distribution of alkaline phosphatase activity in male hypogonadic rat (hgn/hgn) testis during postnatal development. J Vet Med Sci 1998; 60:671–679. Haider SG, Servos G. Ultracytochemistry of 3b-hydroxysteroid dehydrogenase in Leydig cell precursors and vascular endothelial cells of the postnatal rat testis. Anat Embryol 1998; 198:101–110. De Felici M, McLaren A. In vitro culture of mouse primordial germ cells. Exp Cell Res 1983; 144:417–427. Paranko J, Kallajoki M, Pelliniemi LJ, Lehto VP, Virtanen I. Transient coexpression of cytokeratin and vimentin in differentiating rat Sertoli cells. Dev Biol 1986; 117:35–44. Enders GC, Kahsai TZ, Lian G, Funabiki K, Killen PD, Hudson BG. Developmental changes in seminiferous tubule extracellular matrix components of the mouse testis: alpha 3(IV) collagen chain expressed at the initiation of spermatogenesis. Biol Reprod 1995; 53:1489–1499. Yazama F, Esaki M, Sawada H. Immunocytochemistry of extracellular matrix components in the rat seminiferous tubule: electron microscopic localization with improved methodology. Anat Rec 1997; 248: 51–62. Galdieri M, Ricci G. Characterization of different cell populations isolated from rat testis peritubular cells. Differentiation 1998; 63:13– 19. Prelich G, Tan CK, Kostura M, Mathews MB, So AG, Downey KM, Stillman B. Functional identity of proliferating cell nuclear antigen and a DNA polymerase-delta auxiliary protein. Nature 1987; 326: 517–520. Hall PA, Levison DA, Woods AL, Yu CC, Kellock DB, Watkins JA, Barnes DM, Gillett CE, Camplejohn R, Dover R, Waseem N, Lane D. Proliferating cell nuclear antigen (PCNA) immunolocalization in paraffin sections: an index of cell proliferation with evidence of deregulated expression in some neoplasms. J Pathol 1990; 162:285–294. Krishnamurthy H, Babu PS, Morales CR, Sairam MR. Delay in sexual
28. 29. 30. 31.
32. 33. 34.
37. 38. 39. 40.
41. 42. 43. 44.
SUZUKI ET AL.
maturity of the follicle-stimulating hormone receptor knockout male mouse. Biol Reprod 2001; 65:522–531. Sharpe RM, McKinnell C, Kivlin C, Fisher JS. Proliferation and functional maturation of Sertoli cells, and their relevance to disorders of testis function in adulthood. Reproduction 2003; 125:769–784. Clermont Y, Perey B. Quantitative study of the cell population of the seminiferous tubules in immature rats. Am J Anat 1965; 100:241– 267. Orth JM. Proliferation of Sertoli cells in fetal and postnatal rats: a quantitative autoradiographic study. Anat Rec 1982; 203:485–492. Orth JM, Gunsalus GL, Lamperti AA. Evidence from Sertoli celldepleted rats indicates that spermatid number in adults depends on numbers of Sertoli cells produced during perinatal development. Endocrinology 1988; 122:787–794. Shivji KK, Kenny MK, Wood RD. Proliferating cell nuclear antigen is required for DNA excision repair. Cell 1992; 69:367–374. Hirose Y, Yoshimi N, Makita H, Hara A, Tanaka T, Mori H. Early alterations of apoptosis and cell proliferation in azoxymethane-initiated rat colonic epithelium. Jpn J Cancer Res 1996; 87:575–582. Clark AM, Garland KK, Russell LD. Desert hedgehog (Dhh) gene is required in the mouse testis for formation of adult-type Leydig cells and normal development of peritubular cells and seminiferous tubules. Biol Reprod 2000; 63:1825–1838. Pierucci-Alves F, Clark AM, Russell LD. A developmental study of the desert hedgehog-null mouse testis. Biol Reprod 2001; 65:1392– 1402. Jeffs B, Meeks JJ, Ito M, Martinson FA, Matzuk MM, Jameson JL, Russell LD. Blockage of the rete testis and efferent ductules by ectopic Sertoli and Leydig cells causes infertility in Dax1-deficient male mice. Endocrinology 2001; 142:4486–4495. Meeks JJ, Crawford SE, Russell TA, Morohashi K, Weiss J, Jameson JL. DAX1 regulates testis cord organization during gonadal differentiation. Development 2003; 130:1029–1036. Orth JM. The role of follicle-stimulating hormone in controlling Sertoli cell proliferation in testes of fetal rats. Endocrinology 1984; 115: 1248–1255. Orth JM. FSH-induced Sertoli cell proliferation in the developing rat is modified by b-endorphin produced in the testis. Endocrinology 1986; 119:1876–1878. Griswold MD, Solari A, Tung PS, Fritz IB. Stimulation by folliclestimulating hormone of DNA synthesis and of mitosis in cultured Sertoli cells prepared from testes of immature rats. Mol Cell Endocrinol 1977; 7:151–165. Levine E, Cupp AS, Miyashiro L, Skinner MK. Role of transforming growth factor-alpha and the epidermal growth factor receptor in embryonic rat testis development. Biol Reprod 2000; 62:477–490. Petersen C, Boitani C, Froysa B, Soder O. Transforming growth factor-alpha stimulates proliferation of rat Sertoli cells. Mol Cell Endocrinol 2001; 181:221–227. Hu J, Shima H, Nakagawa H. Glial cell line-derived neurotropic factor stimulates Sertoli cell proliferation in the early postnatal period of rat testis development. Endocrinology 1999; 140:3416–3421. Boitani C, Stefanini M, Fragale A, Morena AR. Activin stimulates
46. 47. 48. 49. 50. 51. 52. 53. 54. 55. 56. 57. 58. 59.
Sertoli cell proliferation in a defined period of rat testis development. Endocrinology 1995; 136:5438–5444. Fragale A, Puglisi R, Morena AR, Stefanini M, Boitani C. Age-dependent activin receptor expression pinpoints activin A as a physiological regulator of rat Sertoli cell proliferation. Mol Hum Reprod 2001; 7:1107–1114. Colvin JS, Green RP, Schmahl J, Capel B, Ornitz DM. Male-to-female sex reversal in mice lacking fibroblast growth factor 9. Cell 2001; 104:875–889. Ricci G, Catizone A, Galdieri M. Pleiotropic activity of hepatocyte growth factor during embryonic mouse testis development. Mech Dev 2002; 118:19–28. Beaumont HM, Mandl AM. A quantitative study of primordial germ cells in the male rat. J Embryol Exp Morphol 1963; 11:715–740. Adams IR, McLaren A. Sexual dimorphic development of mouse primordial germ cells: switching from oogenesis to spermatogenesis. Development 2002; 129:1155–1164. Huckins C, Clermont Y. Evolution of gonocytes in rat testis during late embryonic and early post-natal life. Arch Anat Histol Embryol 1968; 51:341–354. Gondos B. Intercellular bridges and mammalian germ cell differentiation. Differentiation 1973; 1:177–182. McGuinness MP, Orth JM. Reinitiation of gonocyte mitosis and movement of gonocytes to the basement membrane in testes of newborn rats in vivo and in vitro. Anat Rec 1992; 233:527–537. Matsui Y. Developmental fates of the mouse germ cell line. Int J Dev Biol 1998; 42:1037–1042. Wang RA, Nakane PK, Koji T. Autonomous cell death of mouse male germ cells during fetal and postnatal period. Biol Reprod 1998; 58: 1250–1256. Boulogne B, Olaso R, Levacher C, Durand P, Habert R. Apoptosis and mitosis in gonocytes of the rat testis during foetal and neonatal development. Int J Androl 1999; 22:356–365. Mintz B, Russel ES. Gene induced embryological modification of primordial germ cells in the mouse. J Exp Zool 1957; 134:207–237. Tilmann C, Capel B. Mesonephric cell migration induces testis cord formation and Sertoli cell differentiation in the mammalian gonad. Development 1999; 126:2883–2890. Nishino K, Yamanouchi K, Naito K, Tojo H. Characterization of mesonephric cells that migrate into the XY gonad during testis differentiation. Exp Cell Res 2001; 267:225–232. Gelly JL, Richoux JP, Leheup BP, Grignon G. Immunolocalization of type IV collagen and laminin during rat gonadal morphogenesis and postnatal development of the testis and epididymis. Histochemistry 1989; 93:31–37. Mackay S, Booth SH, MacGowan A, Smith RA. Ultrastructural studies demonstrate that epithelial polarity is established in cultured mouse pre-Sertoli cells by extracellular matrix components. J Electron Microsc 1999; 48:159–165. Maekawa M, Kamimura K, Nagano T. Pertitubular myoid cells in the testis: their structure and function. Arch Histol Cytol 1996; 59:1–13. Skinner MK, Stallard B, Anthony CT, Griswold MD. Cellular localization of fibronectin gene expression in the seminiferous tubule. Mol Cell Endocrinol 1989; 66:45–52.