Developmental stages of fetal-type Leydig cells in prepubertal rats

Transcription

1 Development 107, (1989) Printed in Great Britain The Company of Biologists Limited Developmental stages of fetal-type Leydig cells in prepubertal rats T. KUOPIO 1 *, J. TAPANAINEN 2, L. J. PELLINIEMI 3 and I. HUHTANIEMI 4 Departments of Anatomy^ and Physiology*, and Laboratory of Electron Microscopy*, University of Turku, Turku, and Department of Obstetrics and Cynecology 2, University of Oulu, Oulu, Finland * Address for reprints: Department of Anatomy, University of Turku, Kiinamyllynkatu 10, SF Turku, Finland Summary Fetal Leydig cells were studied in rats during and after the perinatal-neonatal period by comparing changes in morphology, number and volume with changes in testicular steroids and serum luteinizing hormone (LH) concentration. Stereologic examination indicated regression of fetal Leydig cells in testis by showing that their total volume as well as the average cell volume decreased between prenatal day 20 and postnatal day 3. The total number and total volume of cells both increased between postnatal days 3 and 11 but the average cell volume did not change during the same time period. Determination of serum LH showed a close correlation between an increase in LH concentration and increases in total number and volume of cells. The combined number of fetal- and adult-type Leydig cells on day 20 was more than 20 times the number of fetal cells at 3 days of age. Electron microscopic analysis showed that fetal Leydig cells after birth formed conspicuous clusters, which were surrounded by a layer of envelope cells and extracellular material. Occasional dividing fetal Leydig cells and possible precursors of fetal or adult Leydig cells were observed. Mitoses of spindle-shaped pericordal cells were frequent during the neonatal period. During and after the second postnatal week fetal Leydig cells again showed signs of regression, indicated by disintegration of the cell clusters, a decrease in cell size, accumulation of collagen between the cells and a decrease in steroid content per cell. The cytoplasm showed no degenerative changes, but the shape of the nuclei changed from spherical into irregular. The present results suggest that the perinatal regression of fetal Leydig cells is followed by an LHinduced phase of growth. This growth period precedes a second phase of regression that coincides with early development of adult-type cells reported earlier. Contrary to the present concept of a biphasic pattern of Leydig cell development in rat, we suggest three consecutive stages: fetal (fetal cells in fetal testis), early juvenile (fetal cells during neonatal-early juvenile life) and juvenile-adult (adult cells before and after puberty). The regression of the fetal cells suggests that they have only a minor role in testicular hormone production after the appearance of adult-type Leydig cells prior to puberty. Key words: morphometry, ultrastructure, steroids, Leydig cells, testis, fetal, neonatal, prepubertal, rat. Introduction Morphologically and functionally distinct populations of fetal and adult Leydig cells are present in testis of developing rats (Christensen, 1975; Huhtaniemi et al. 1984; de Kretser and Kerr, 1988). The cells of the populations are referred to as fetal and adult Leydig cells, because they differentiate and start their hormone production during fetal life and before and after puberty, respectively. It has been suggested that fetal Leydig cells decrease in number and finally disappear from the interstitium as a result of cell death or dedifferentiation during postnatal life (Gondos, 1977). However, recent studies have indicated that the total number of the cells in testis does not change markedly immediately after birth even though the number of the cells per unit volume of testis decreases (Mendis- Handagama et al. 1987; Zirkin and Ewing, 1987; Kerr and Knell, 1988; Kuopio et al. 1989a). Moreover, Kerr and Knell (1988) recently reported their intriguing finding that fetal Leydig cells persist as a distinct population in adult testis after the growth of adult-type cells. Earlier we showed that testicular steroid content per Leydig cell is highest during fetal life, suggesting that fetal Leydig cells are steroidogenically more active than adult-type cells (Tapanainen et al. 1984). Before birth, the steroid content per cell decreases by an unknown mechanism but still remains higher than that in adults through the first and second postnatal weeks (Tapanainen et al. 1984). The present study extends our earlier findings (Tapanainen et al. 1984; Kuopio et al. 1989a,b,c) of morphological and functional differences in fetal Ley-

2 214 T. Kuopio and others dig cells before adult-type cells begin to dominate. Our results explain the discrepancy between the old concept of early Leydig cell regression and recent studies reporting a constant number of cells in pre- and postnatal rats. Moreover, our results suggest a luteinizing hormone-induced proliferation of the fetal cells followed by their structural and functional regression after the early neonatal period. Materials and methods Animals and specimen preparation Male Wistar rats between fetal day 20 and postnatal day 23 were used for this study. The animals were kept in controlled temperature (22 C) and photoperiod (14L:10D). Laboratory animal chow and water were available ad libitum. The rats were killed by decapitation, blood was collected for serum luteinizing hormone (LH) measurements and testes were dissected out, weighed and prepared for morphologic and stereologic analysis. The testes were fixed by immersion in 5% glutaraldehyde in 0-16% moll" 1 2-, 4-, 6-collidine-HCl buffer (ph7-4) and further processed as described earlier (Kuopio et al ). Sections (l^rn) were stained with toluidine blue for morphometry at the light microscopic level (day 20 of gestation and days 3, 11, 15 and 20 postnatal). Sections for electron microscopy (day 20 of gestation to day 23 postnatal) were stained with uranyl acetate and lead citrate. Specimens were collected for analysis of testicular steroids at days 3, 11 and 15 after birth as described earlier (Tapanainen etal. 1984). Stereologic methods Stereologic measurements included only the fetal type Leydig cells except at 20 days postnatal, when adult-type cells were included to analyse the total number of Leydig cells of both types per testis. Cells were identified according to criteria described earlier (Mendis-Handagama et al. 1987; Zirkin and Ewing, 1987; Kerr and Knell, 1988; Kuopio etal. I989a,b,c). The stereologic methods of the present study have been described earlier (Kuopio et al. 1989a). Briefly, the volume densities of Leydig cells (V V LC> fraction of testis volume occupied by the cells) and Leydig cell nuclei (V VN uc) were estimated using the point-counting method (Weibel, 1979). The numerical densities of Leydig cells (N V LC. number of cells per unit volume of testis) were calculated from V VN uc and from the number of nuclear profiles of Leydig cells per unit area of section (N ANU c) using the method of Weibel and Gomes where N VL c = l//3*(n 3/2 ANUC / V VNUC 1/2 )- The coefficient ys is a dimensionless shape coefficient, which, in the case of nearly spherical particles, =1-38 (Weibel, 1979). To calculate the total volume and number of Leydig cells per testis, VVLC and NVLC were multiplied by testis volume, respectively. Testis volume was obtained directly from weight measurements because the specific gravity of testis does not considerably differ from l-0gcm~ 3 (Mori and Christensen, 1980). Testicular weights and numbers of animals used in the stereologic analysis are shown in Table 1. The average volume of Leydig cells was derived by dividing V V LC by N V LC- Measurements were made from three randomly selected blocks from each animal and from one randomly selected l imi section from each block. The whole section area was analyzed using a microscope with a 40 x objective lens and an eyepiece grid. The grid covered 0-02 mm 2 at a time and had 625 line intersections. The average number of analysed fields per animal was 100. Table 1. The number of animals and the weights of the testes (mean ± S.E.M.) used in the stereologic measurements Number of animals Testis weight (mg) f20* * days of fetal life (f). 1-5 ± ± ± ±5 Determination of serum LH concentration and steroid content per Leydig cell Serum LH was measured with homologous radioimmunoassay kits provided by The National Pituitary Agency and NIADDK (Bethesda, MD) as described earlier (Huhtaniemi et al. 1986). Results are expressed in terms of the RP-1 standard. To estimate the total steroids per Leydig cell the steroid content per testis was calculated from measurements of individual steroid concentrations (testosterone, 5a'-dihydrotestosterone, progesterone, 17-hydroxy progesterone, pregnenolone, androstenedione and 5<r-androstane-3a-, 17/3- diol) published previously (Tapanainen etal. 1984). Values of the steroid content per testis were then divided by the number of cells per testis (Fig. 3) from randomly selected pairs in respective age groups of animals. The obtained individual values were averaged and analyzed statistically. Statistics Statistical comparisons of the age groups were made using Newman-Keuls multiple range test (N-K). Results Leydig cell stereology The total volume of fetal Leydig cells in testis (Fig. 1) and the average Leydig cell volume (Fig. 2) decreased (P<0-05) in the perinatal period. The total number of cells did not significantly change during the same time (Fig. 3). Between days 3 and 11 after birth, the total number (Fig. 3) and volume (Fig. 1) of cells increased significantly (P<0-05). The combined number of the f Fig. 1. Total volume of fetal Leydig cells from fetal day 20 (f) to postnatal day 15. Each bar is the mean ± S.E.M. of 4 to 5 animals. Statistically significant differences were observed between groups indicated by different letters -05, N-K).

3 Prepubertal fetal Leydig cells fetal- and adult-type Leydig cells on day 20 was more than 20 times the number of the fetal cells at the age of 3 days (Fig. 3). No significant change was seen in average cell volume between ages of 3 and 15 days (Fig. 2). Leydig cell morphology After birth, fetal Leydig cells formed conspicuous, GO, smoothly delineated clusters in which the cells were tightly packed and intercellular spaces were narrow. The clusters were surrounded by a layer of envelope cells and extracellular material which separated them from the surrounding interstitium. No other cell types Or blood vessels were found inside the clusters (Fig. 4). Cell number (cells x 104/testis VZQ) (cells x 105/testisCZ]) Fig. 2. Average fetal Leydig cell volume from fetal day 20 to postnatal day 15. Each bar is the mean ± S.E.M. of 4 to 5 animals. Statistically significant differences were observed between the groups indicated by different letters (f<0-05, N-K) Fig. 3. Total number of fetal Leydig cells in testis from fetal day 20 to postnatal day 20. Each bar is the mean ± S.E.M. of 4 to 5 animals. Statistically significant differences were observed between the groups indicated by different letters (P<0-05, N-K). VfWli. L fv L Fig. 4. Electron micrograph of a cluster of fetal Leydig cells (L), age 8 days. Cells in the cluster are closely attached to each other. The cluster is surrounded by an envelope cell (e). The space between the Leydig cells and the envelope cell contains positively staining collagen fibers (thin arrows). x4500.

4 216 T. Kuopio and others c % * Fig. 5. Electron micrograph of two pericordal cells in mitosis (thick arrows), age 5 days. The cell on the left is immediately adjacent to the testicular cord (c) while the one on the right is separated from the cord by one or two cell layers. X4500. Mitoses of spindle-shaped pericordal cells were frequent during the early days of-life (Figs 5 and 6). Occasional mitoses of fetal Leydig cells (Fig. 7) and immature cells without typical features of Leydig cells or mesenchymal cells (Fig. 8) were also observed in the interstitium. During and after the second postnatal week, some clusters of fetal Leydig cells started to disintegrate. The cells in the clusters were separated from each other by intercellular spaces which increased in size and became filled with negatively staining collagen fibers (Fig. 9). The expansion of the space between the cells probably occurred as a consequence of reduced cell size. There were no degenerative changes in the cytoplasm, but the characteristically spherical nuclei (Fig. 4) became irregular in shape (Fig. 10). Interstitial macrophages were often associated with clusters that showed this type of disintegration (Fig. 10). Serum LH concentration and steroid content per Leydig cell A significant increase in serum LH concentration was seen between the days 3 and 11 (Fig. 11). The change in the LH concentration significantly (P< 0-001) correlated with an increase in total number (r = ) and volume (r = ) of fetal Leydig cells during the same period (days 3 and 15), suggesting a cause-andeffect relationship. Steroid content per Leydig cells decreased significantly between days 11 and 15 (Fig. 12). The correlation between steroid content per cell and serum LH concentration was negative (r = ) and statistically significant (P<0-02). Discussion Fetal Leydig cells are generally believed to decrease in number and finally to disappear from the interstitium during a process of regression after birth (Gondos, 1977). The present study and other recent reports, however, suggest that the total number of fetal Leydig cells in testis does not change significantly between the end of pregnancy and the early days of postnatal life (Mendis-Handagama et al. 1987; Zirkin and Ewing, 1987; Kerr and Knell, 1988; Kuopio etal. 1989a). At the same time, the present results confirm and extend the estimate of Roosen-Runge and Anderson (1959) who reported that total Leydig cell volume decreases between the late fetal period to the fourth postnatal day. If the total number of cells does not change significantly during the perinatal period, then the decrease in total volume of cells must be due to a reduction in size of individual cells. This conclusion has been supported by recent observations of Zirkin and Ewing (1988) and confirmed by us here. Earlier reports of drastic Leydig cell regression after birth (Roosen-Runge and Anderson, 1959) may be explained by a marked decrease in numerical density, or the number of the cells per unit volume of testis (Lording and de Kretser, 1972; Tapanainen et al. 1984; Zirkin and Ewing, 1987; Kerr

5 Prepubertal fetal Ley dig cells 217 Fig. 6. Light micrograph of several testicular cords with two pericordal cells in mitosis (thick arrows), age 5 days. c, testicular cord. x310. Fig. 7. Light micrograph of a fetal Leydig cell (arrowhead), and a pericordal cell (thick arrow) in mitosis, age 1 day. c, testicular cord. x610. Fig. 8. Electron micrograph of a group of interstitial cells which are considered immature fetal or adult Leydig cells, age 5 days. X4500. and Knell, 1988) as well as a reduction in individual cell volume. In addition, the reorganization and clustering of fetal Leydig cells soon after birth (Roosen-Runge and Anderson, 1959; Kuopio et al. 1989c) may give an impression of regression. The decrease in total volume of fetal Leydig cells during the perinatal period is followed by a postnatal phase of regrowth, also observed by Roosen-Runge and Anderson (1959) and Mendis-Handagama et al. (1987). According to the present study, this growth phase is represented by an increase in the number of Leydig cells, not from changes in cell size as is true during the perinatal phase of regression. Similarly, MendisHandagama et al. (1987) found an increase in the number of the cells between the days 5 and 10, whereas Kerr and Knell (1988) reported no comparable change during the first and second postnatal weeks. Another discrepancy in earlier literature concerns the ratio of Leydig cell number just after birth to that during and after the third postnatal week. Our earlier report of a large decrease in steroid content of Leydig cells from perinatal period to adulthood was based on measurements showing a 20-fold increase in Leydig cell number between days 2 and 3 versus 3 weeks of age (Tapanainen etal. 1984). Recently, Zirkin and Ewing (1987) reported only a fivefold increase in cell number during the corresponding period, a result that would negate our earlier conclusions reporting differences in steroid production of fetal versus adult cells (Tapanainen et al. 1984). The findings in the present study, however, are in accord with our earlier results (Tapanainen et al. 1984) and also supported by data of the Leydig cell numbers from the other laboratories (Mori and Christensen, 1980; Mendis-Handagama et al. 1987; Kerr and Knell, 1988). The direct correlation between an increase in serum LH and the number of cells per testis suggests that the postnatal growth phase of the fetal Leydig cells may be regulated by a physiological rise in serum LH during the first week of life (Lee et al. 1975; Ketelslegers et al. 1978; Ramaley, 1979). This idea receives support from our earlier study which showed a marked and rapid increase in the number of fetal Leydig cells in newborn rats after exogenous hcg administration (Kuopio et al. 1989a). The presence of mitoses in fetal Leydig cells of untreated rats and earlier in rats treated with hcg (Kuopio et al. 1989a) suggests that cell proliferation can contribute to the increase in cell number. Fouquet and Kann (1987) emphasise that convincing evidence for mitoses in Leydig cells in young rats is not available because the reports are based on observations on paraffin sections from which identification of cells is not

6 218 T. Kuopio and others -/ -i^\ 10 Fig. 9. Electron micrograph of a cluster of fetal Leydig cells (L), age 15 days. The cells are separate from each other, and negatively staining collagen fibers (thin arrows) appear in the intercellular spaces. X4400. Fig. 10. Electron micrograph of a regressing cluster of fetal Leydig cells (L), age 23 days. The cytoplasm of the cells is reduced and the intercellular spaces are large. A macrophage (m) is in a typical location close to the regressing cluster, c, seminiferous tubule. x3800.

7 Prepubertal fetal Leydig cells Fig. 11. Serum LH concentration in rats at ages 3, 11 and 15 days of postnatal life. Each bar is the mean ± S.E.M. of 4 to 5 animals. Statistically significant differences were observed between groups indicated by different letters (><0-05, N-K) Fig. 12. Mean content of total steroids per fetal Leydig cells at the ages of 3, 11 and 15 days of postnatal life. Each bar is the mean ± S.E.M. of 3 to 5 values. Statistically significant difference was observed between the groups indicated by different letters (P<0-05, N-K). reliable. The same criticism cannot be applied to the present observations because mitotic figures can be reliably identified in 1,um plastic sections. In addition to cell proliferation, differentiation from immature precursors may increase the Leydig cell number as well. At this time, however, it is not possible to differentiate between precursors of fetal Leydig cells versus adult Leydig cells. Therefore, the contribution in numerical growth of the two cell types made by differentiation of immature interstitial cells and that made by mitotic cells with an elongated shape in the pericordal position remains to be clarified. The postnatal increase in fetal Leydig cell number is soon followed by a phase of regression. This is indicated by disintegration of cell clusters, accumulation of collagen between the cells and a decline in the steroid content per cell. Earlier we showed a decrease in steroids per cell during the last days of fetal life and again after day 20 (Tapanainen et al. 1984). The latter decline coincided with replacement of fetal Leydig cells by an adult cell population and therefore probably represented a difference in the steroidogenic capacity of the two cell populations (Tapanainen et al. 1984). The present analysis, directed in more detail to the first and second postnatal weeks, showed in addition, a decline in steroids per cell preceding the shift between the two populations. This decline coincides with a small decrease in serum LH concentration during the second postnatal week observed previously (Lee et al. 1975; Ketelslegers et al. 1978; Ramaley, 1979), but not in the present material. Deprivation of LH would be consistent with functional changes in fetal cells. However, in vitro testosterone production per testis in the presence of maximally stimulating amounts of hcg decreases between days 5 and 10 (Huhtaniemi et al. 1982) despite a simultaneous increase in total cell number. This suggests that changes in serum LH concentration between the days 11 and 15 per se may not be responsible for the observed in vivo decrease in the steroids per cell. Moreover, LH receptor measurements (Huhtaniemi et al. 1982) do not indicate that changes in number of receptors can explain decreased steroid production per cell during the second postnatal week. The physiologic role of the small population of fetal Leydig cells possibly persisting after puberty (Kerr and Knell, 1988) remains to be investigated. As far as hormone production is concerned, they appear to be an unimportant minority when compared with the adult-type cells, which may be more than 200- to 500-fold greater in number (Mori and Christensen, 1980; Tapanainen et al. 1984; Mendis-Handagama et al. 1987; Kerr and Knell, 1988). Our study showing disintegration of Leydig cell clusters and accumulation of collagen fibers in expanding intercellular spaces is consistent with an early histologic study (Roosen-Runge and Anderson, 1959), in which the Leydig cell clusters were seen to be dispersed into individual cells separated by a dense network of fibers. Separation of tightly clustered fetal Leydig cells and expansion of intercellular spaces may result from a reduction in cell size after the first week of life. The observed changes in the average cell size determined stereologically were not statistically significant in the present material between ages 11 and 15 days; however, Mendis-Handagama et al. (1987) observed a 50 % reduction in the cell size during the third week of life. Perhaps this decrease begins earlier, at least in some of the cells. Taken together, the reduction of cell size seems to be associated with a decline in steroid production of fetal cells during both the perinatal and early juvenile periods (for classification of sexual development, see Adams and Steiner, 1988). Our observations of an increase in collagen between the Leydig cells together with a simultaneous decrease in the cell size and steroid production suggests that accumulation of collagen is related to regression of Leydig cells in developing rats. The change in amount of interstitial collagen (Hatakeyama, 1965; Pelliniemi et al. 1980; Schulze, 1984) as well as in the basement membranes of Leydig cells (Kuopio and Pelliniemi, 1989; Kuopio et al ,c) in physiological, experimental and pathological conditions is further evidence that the extracellular matrix may be involved in regulation of Leydig cells. Our results indicate that a second, prepubertal phase

8 220 T. Kuopio and others of growth occurs after the initial regression of fetal Leydig cells in perinatal animals. This growth period is followed by another phase of regression during the second postnatal week, coinciding with the development of adult-type cells (Mendis-Handagama et al. 1987). Instead of the commonly accepted biphasic pattern of Leydig cell growth in laboratory rodents (de Kretser and Kerr, 1988), there are in fact three consecutive stages of Leydig cell development in rats: fetal (fetal cells in fetal testis), early juvenile (fetal cells during neonatal-early juvenile life) and juvenile-adult (adult cells before and after puberty). This pattern resembles the situation in the pig, which no longer can be considered an exception (Dierichs et al. 1973; de Kretser and Kerr, 1988) among the species. Supported by grants from the Sigrid Jus61ius Foundation and from the Academy of Finland. References ADAMS, L. A. & STEINER, R. A. (1988). Puberty. In Oxford Reviews of Reproductive Biology, vol 10 (ed. J. R. Clarke), pp Oxford: Oxford University Press. CHRISTENSEN. A. K. (1975). Leydig cells. In Handbook of Physiology, vol 5 (ed. D. W. Hamilton & R. O. Greep), pp Washington D.C.: American Physiological Society. DE KRETSER, D. M. & KERR, J. B. (1988). The cytology of the testis. In The Physiology of Reproduction, vol. 1 (ed. E. Knobil, J. D. Neill, Ewing L. L., Greenwald G. S., Markert G, Pfaff D. W.), pp New York: Raven Press, Ltd. DIERICHS, R., WROBEL, K.-H. & SCHILLING, E. (1973). Licht- und elektronenmikroskopische Untersuchungen an den Leydigzellen des Schweines wahrend der postnatalen Entwicklung. Z. Zellforsch. mikrosk. Anat. 143, FOUQUET, J.-P. & KANN, M-L. (1987). Mitotic activity in monkey and rat Leydig cells. Reprod. Nutr. Develop. 27, GONDOS, B. (1977). Testicular development. In The Testis, vol 4. Advances in Physiology, Biochemistry and Function (ed. A. D. Johnson & W. R. Gomes), pp New York: Academic Press. HATAKEYAMA, S. (1965). A study on the interstitial cells of the human testis, especially on their fine-structural pathology. Ada Path. Jap HUHTANIEMI, I. T, NEVO, N., AMSTERDAM, A. & NAOR, Z. (1986). Effect of postnatal treatment with gonadotropin-releasing hormone antagonist on sexual maturation of male rats. Biol. Reprod. 35, 50L-507. HUHTANIEMI. I. T., Nozu, K., WARREN, D. W., DUFAU, M. L. & CATT, K. J. (1982). Acquisition of regulatory mechanisms for gonadotrophin receptors and steroidogenesis in the maturing rat testis. Endocrinology 111, HUHTANIEMI. I. T., WARREN, D. W. & CATT, K. J. (1984). Functional maturation of rat testis Leydig cells. Ann. N.Y. Acad. Sci. 438, HUHTANIEMI, I. T.. WARREN, D. W. & CATT, K. J. (1985). Regulation of infant and developing rat testicular gonadotropin and prolactin receptors and steroidogenesis by treatments with human chorionic gonadotropin, gonadotropin-releasing hormone analogs, bromocriptine, prolactin, and estrogen. Biol. Reprod. 32, KERR, J. B. & KNELL, C. M. (1988). The fate of fetal Leydig cells during the development of the fetal and postnatal rat testis. Development KETELSLEGERS, J.-M., HETZEL, W. D., SHERINS, R. J. & CATT, K. J. (1978). Developmental changes in testicular gonadotropin receptors: plasma gonadotropins and plasma testosterone in the rat. Endocrinology 103, KUOPIO, T., PARANKO, J. & PELLINIEMI, L. J. (1989c). Basement membrane and epithelial features of fetal-type Leydig cells in rat and human testis. Differentiation (in press). KUOPIO, T. & PELLINIEMI, L. J. (1989). Patchy basement membranes of rat Leydig cells shown by ultrastructural immunolabeling. Cell Tissue Res. 256, KUOPIO, T., PELLINIEMI, L. J. & HUHTANIEMI, I. (1989a). Rapid Leydig cell proliferation and LH receptor replenishment after a single injection of human chorionic gonadotropin in the neonatal rat testis. Biol. Reprod. 40, KUOPIO, T., SAVOURAS, P., PELLINIEMI, L. J. & HUHTANIEMI, I. (19896). Transplantation of newborn rat testis under the kidney capsule of adult host as a model to study the structure and function of Leydig cells. J. Androl. (in press). LEE, V. W. K., DE KRETSER, D. M., HUDSON, B. & WANG, C. (1975). Variations in serum FSH, LH and testosterone levels in male rats from birth to sexual maturity. J. Reprod. Fertil. 42, LORDING, D. W. & DE KRETSER, D. M. (1972). Comparative ultrastructural and histochemical studies of the interstitial cells of the rat testis during fetal and postnatal development. J. Reprod. Fertil. 29, MENDIS-HANDAGAMA, S. M. L. C, RISBRIDGER, G. P. & DE KRETSER, D. M. (1987). Morphometric analysis of the components of the neonatal and the adult rat testis interstitium. Int. J. Androl. 10, MORI, H. & CHRISTENSEN, A. K. (1980). Morphometric analysis of Leydig cells in the normal rat testis. J. Cell Biol. 84, PELLINIEMI, L. J., DYM, M., CRIGLER, J. F., RETIK, A. B. & FAWCETT, D. W. (1980). Development of Leydig cells in human fetuses and in patients with androgen insensitivity. In Testicular Development, Structure and Function (ed. A. Steinberger & E. Steinberger), pp New York: Raven Press. RAMALEY, J. A. (1979). Development of gonadotropin regulation in the prepubertal mammal. Biol. Reprod. 20, ROOSEN-RUNGE, E. C. & ANDERSON, D. (1959). The development of the interstitial cells in the testis of the albino rat. Acta Anat. 37, SCHULZE, C. (1984). Sertoli cells and Leydig cells in man. Advances in Anatomy, Embryology and Cell Biology, vol 88. pp Berlin Heidelberg New York: Springer. TAPANAINEN, J., KUOPIO, T., PELLINIEMI, L. J. & HUHTANIEMI, I. (1984). Rat testicular endogenous steroids and number of Leydig cells between the fetal period and sexual maturity. Biol. Reprod. 31, WEIBEL, E. R. (1979). Stereological methods. In Practical Methods for Biological Morphometry, vol. 1. London: Academic Press. ZIRKIN, B. R. & EWING, L. L. (1987). Leydig cell differentiation during maturation of the rat testis: A stereological study of cell number and ultrastructure. Anat. Rec. 219, (Accepted 5 July 1989)