- Email: [email protected]
Induction of meiosis in fetal mouse testis in vitro
DEVELOPMENTAL
BIOLOGY
Induction
52, 193-200 (1976)
of Meiosis ANNE
in Fetal Mouse Testis in Vitro
GRETE BYSKOV’
’ The Finsen Laboratory, The Finsenilnstitute, WI Department of Pathology,
AND LAURI
SAXBN~
Strandboulevarden 49,210O K@benhavn 8, Denmark; University of Helsinki, 0290 Helsinki, Finland
Accepted April
and
2,1976
Fetal mouse testes and ovaries with their urogenital connections were cultured singly or in pairs on Nuclepore filters. When a testis in which the sex was not yet morphologically detectable was cultured together with older ovaries containing germ cells which were progressing through the meiotic prophase, the male germ cells were triggered to enter meiosis. When older fetal testes in which the testicular cords have developed were cultured together with ovaries of the same age with germ cells in meiosis, the oocytes were prevented from reaching diplotene stage. It was concluded that the fetal male and female gonads secrete diffusable substances which influence germ cell differentiation. The male gonad secretes a “meiosis-preventing substance” (MPS) which can arrest the female germ cells within the meiotic prophase. The female gonad secretes a “meiosis-inducing substance” (MIS) which can trigger the nondifferentiated male germ cells to enter meiosis. INTRODUCTION
The mechanism which triggers the mammalian male and female germ cells to enter meiosis is not known, and experiments have failed to induce the onset of meiosis in the undifferentiated germ cell. In spontaneous hermaphrodites of mice, meiotic cells have been found in the fetal testes (Whitten, 1975). Also in fetal mouse chimeras, germ cells in meiotic prophase were observed in fetal testes (Mystkowska and Tarkowski, 1970; McLaren, Chandley, and Kosman-Aljaro, 1972). McLaren (1972) proposed that these germ cells might be XX as well as XY cells and that they were triggered to enter meiosis by surrounding somatic XX cells. The other germ cells which did not start the meiosis might be placed between noninducing somatic XY cells. One specific somatic cell system which is of mesonephric origin, the rete ovarii, seems to be necessary for the differentiation of the female gonad. The onset of meiosis (Byskov, 1974), as well as the follicular formation (Byskov et al., 1976), seems to be dependent on this urogenital connection during the early gonadal anlage.
In mammals the differentiation of a gonad into a recognizable testis occurs before the male germ cells enter meiosis. The testicular cords normally differentiate during fetal life but none of the male germ cells starts the meiotic prophase before early puberty. At the time when the testicular cords differentiate, the ovary of the same age and from the same species may not be recognizable as a female gonad by other criteria than that it is lacking the cord formations (Jost, 1965). Nevertheless, the meaning of the expression “sex differentiation” of the gonads works to cover that process of morphological differentiations of the testis which renders the gonads to be recognized as testes or ovaries. In some species, however, the meiosis in the female germ cells starts simultaneous with, or immediately after, the time when organization of the testicular cords in the male gonad occurs. In the mouse, for instance, sex differentiation of the testis starts on Day 12 (Odor and Blandau, 1969) as does the S-phase of the first meiotic division in the female germ cells (Peters, Levy, and Crone, 1962.) In the present work, we investigated 193
Copyright 0 1976 by Academic press, Inc. All rights of reproduction in any form reserved.
194
DEVELOPMENTAL
BIOLOGY
whether the female mouse urogenital complex, i.e., ovary together with rete ovarii, along with germ cells in meiosis could initiate the onset of meiosis in germ cells of fetal testes before and after gonadal sex differentiation. MATERIALS
AND
METHODS
Fetuses of the mouse strain BALBI cxCBA/TGTs were removed at the eleventh and the fourteenth day postcoitum (p.c.1. The urogenital complex consisting of the gonads together with the attached cranial part of the mesonephros, i.e., the rete system, were dissected free and glued with 1% agar on Nuclepore filters with a pore diameter of 1.0 or 0.1 pm (General Electric Co., Pleasanton, Calif.). The explants were placed either singly at one side of the filter or in pairs on opposite sides. One or several filters with explants were supported by a grid and placed in a culture dish with 2 ml of Eagle’s balanced salt solution with 10% fetal calf serum in a gasphase of 5% COz in air at 37°C. After a culture period of 4 or 7 days, the filters with the explants were fixed in Zenker, embedded in paraffin, sectioned serially at 5 pm, and stained with PAS and hematoxylin. The experimental setup, including the sexes and ages of the gonads in the cultures (Experiments A to K), are shown in Table 1. In Experiments A and B (paired gonads), a gonad was present at each side of the filter, forming a pair. Experiment A consisted of three dishes, each containing six pairs of gonads (18 filters). In B, there were two dishes, with three and four pairs of gonads (7 filters). In Experiments C to K, a single gonad was glued to the filter. C and D (single gonads) consisted each of one dish containing six fetal gonads from the eleventh day p.c., males together with females. E-I and K were culture controls. E and F consisted of three and two dishes, respectively, each with one single ll-dayold fetal gonad cultured for 4 days. G and H consisted of four and three dishes, each
VOLUME 52. 1976 TABLE
1
EXPERIMENTAL SETUP OF FETAL MOUSE GQNADS IN CULTURE Experiment Start of culture End of culture NUllIAge (days) and Age (days) and her of sex of gonads on sex of gonads on filters filters flkers Paired A gonads
116 iz+iG
B
Single gonads
119
148 149
l&3* 219
189 +m
18
21d 219**
7
C
116 -+iE
-
15c?* +m
6
D
116 -+iiF
-
188 -+m
6
Culture E controls F G
116
-
156
119
159
11C?
188
-
H
K
11P
189
146 -
216
149
* Meiosis-induced. vented.
219 ** Meiosis-arrested
3 2 4 3 6 6 or pre-
with one single 11-day-old fetal gonad, which was cultured for 7 days. In both Experiment I and K, there was one dish with six single gonads of lCday-old fetal males or females, respectively. The total number of filters was 66. In one of the dishes in Experiment A, the six filters had a pore diameter of 0.1 pm. The pore diameter in all other filters was 1.0 pm. Morphologically the sex in 11-day-old embryos cannot be determined, whereas this is possible at Day 14 p.c. The dishes, in which 11-day-old fetal gonads were paired with lCday-old ovaries (A), or were cultured singly on separate filters (C, D), therefore contained randomly both testes as well as ovaries. The sex of the ll-dayold gonads was determined morphologically at the end of the culture period. Go-
BYSKOV
AND SAXON
Znduction
nads of fetuses of ll-, 15, 18, and 21-dayold mice, i.e., controls, were sectioned and stained as in the experiments and their morphology was compared with the paired gonads, single gonads, and culture controls. RESULTS
Controls. In the 11-day-old gonad, in which the morphological sex differentiation has not yet occurred, many germ cells are dividing but no meiotic stages are seen. On Day 14 p.c., the gonads can be recognized as testes or ovaries. In the male where the testicular cords have formed, almost all of the gonocytes are in the “resting,” nondividing stage and no meiotic tigures are present. The testis on Days 15, 18, and 21 resembles the 14-day-old one in respect to the germ cells, which are still “resting” cells. In the lCday-old fetal ovary many oocytes are in leptotene and some are in zygotene. On Day 15, germ cells are also in pachytene. On Day 18 p.c., almost all germ cells have entered the meiotic prophase. Most are in pachytene and some have reached diplotene. By Day 21, which is usually the day of birth, all oocytes are in diplotene. Follicles have formed and growth of both oocytes and follicles have started. Culture controls. The 11-day-old fetal testes cultured for 4 days (Fig. 11) and 7 days (Fig. 10) (Experiments E and G) had testicular cords with germ cells in the resting stage. The morphology of the fetal ovaries of Experiments F and H was comparable to the control ovaries of Days 15 and 18, respectively. The developmental stage of the lCday-old fetal testis (Experiment I) and ovary (Experiment K) after 7 days in culture was similar to 21-day-old control testes and ovaries, respectively. Paired gonads. In Experiment A, eight of the 11-day-old fetal gonads cultured transfilter to fetal ovaries of Day 14 were identified as testes by the presence of gonocytes (Fig. 4) and/or testicular tubules
of Meiosis in Fetal Mouse Testes
195
(Fig. 1). Germ cells in different stages of the meiotic prophase were present in all these testes. No difference was noticed between the transfilters of l.O-pm pore size and the 0.1 pm. All germ cells in the testicular cords which were placed directly above the ovarian tissue had entered meiosis. Most germ cells were in zygotene and pachytene (Fig. 2). In one case, two diplotene stages were recognized (Fig. 4). However, in cords which were situated between 0.3 and 0.8 mm from the ovarian tissue, only about 20% of the germ cells had entered meiosis (Fig. 1). The remnant part of the germ cells in these cords were still in the resting phase. The 10 other explants of the 11-day-old fetal gonads of Experiment A were determined to be ovaries. The germ cells were distributed over a large area and revealed no signs of being enclosed within cords (Fig. 5). The germ cells were in zygotene and pachytene stages (Fig. 6). The 14-day-old fetal ovaries, which served as partners for all the 11-day-old fetal gonads in Experiment A, contained oocytes in diplotene stages after the 7 days in culture. The 14-day-old fetal ovaries combined with 11-day-old fetal ovaries had oocytes and follicles, which had started to grow (Figs. 5 and 7). Ovaries in combination with the testes, however, differed from the pure ovarian pairing: in all ovaries of Day 14, which were combined with ll-dayold fetal testes (Fig. 11, few of the oocytes had reached diplotene and few follicles had started to grow (Fig. 3). Many of the oocytes seem to be degenerating in early diplotene stage (Fig. 3). In Experiment B, the fetal testes of Day 14, which were cultured for 7 days transfilter to a lCday-old fetal ovary, had all germ cells in the resting, nondividing stage (Fig. 12). In the ovaries of these pairs, however, most oocytes were pyknotic (Fig. 13, arrows) or appeared to be degenerating pachytene stages. Few oocytes had reached the diplotene stage and no follicles had formed.
196
DEVELOPMENTAL
BIOLOGY
VOLUME 52, 1976
BYSKOV AND SAXON
Induction
Single gonads. Four of the six gonads in Experiment C were identified as testes. All these testes had cords in which about 20% of the germ cells were in leptotene (Figs. 8 and 91, whereas the remnant parts were still in the resting stage. The two ovaries in the dish contained germ cells in leptotene and zygotene. In Experiment D, there were three testes and three ovaries. After 7 days in culture together with ovaries, all the germ cells of these testes were in the resting stage. Germ cells in meiosis could not be recognized. The three ovaries in the culture were comparable to the 1% day-old controls. DISCUSSION
Three main results can be extracted from this study. First, germ cells of fetal mouse testes in which testicular cords are not yet formed are triggered to enter meiosis when cultured together with the female fetal mouse urogenital complex containing germ cells in meiosis. Second, the trigger, which influences the male germ cells to enter meiosis, is a diffusible substance. Third, the differentiation of fetal ovaries is inhibited when cultured together with fetal testis. The control mechanisms of the onset of meiosis in mammals are poorly understood. In cultures of ovaries of mice (Wolff, 1952; Baker and Neal, 1973), sheep
of Meiosis
in Fetal Mouse Testes
197
(Mauleon, 19731, and hamsters (Challoner, 19751, it has been shown that the germ cells will survive and progress beyond the leptotene stage, only in ovaries which are taken into culture after a certain critical stage of development. This critical stage seems to coincide with the time the somatic elements start to differentiate. The interaction between the somatic cells and the germ cells during gonadal differentiation has been reviewed by Short (1972). One of the somatic elements, the rete ovarii has been shown to be of importance for the onset of meiosis in the mouse (Byskov, 1974); the urogenital connection containing the rete ovarii was therefore preserved in all the experiments of the present study. In the described transfilters the meiosis was induced not only in filter systems, which should allow cellular contact (Experiment A with l.O-pm pore diameter), but also in the setup in which the small filter pores of 0. l-pm pore diameter probably will prevent cellular contact (Wartiovaara et al., 1974), as well as in the setup of single gonads of different sex cultured separately in the same dish (Experiment C). This indicates that a meiosis-inducing substance (MIS) secreted from the urogenital complex has reached the testes by diffusion. When the testes and ovaries are cul-
FIG. 1. A pair of gonads (11 d 114 P , Expt. Al cultured for 7 days on a Nuclepore filter. The testis upon the filter has germ cells in meiotic stages confined to testicular cords. Only a few meiotic cells (black dots) are present in the peripherally placed cord (thick arrow). The ovary below the filter contains oocytes in diplotene stage (thin arrow). x 250. FIG. 2. Testicular cords of the testis shown in Fig. 1. Most of the germ cells are in zygotene (Z) and in pachytene stage (P). x 1225. FIG. 3. Part of the ovary seen in Fig. 1. One of the oocytes seems to be degenerating in early diplotene stage (arrow). One is pyknotic and one is in diplotene stage (D). x 925. FIG. 4. Cross section of two testicular cords of a fetal testis cultured transfilter to a ll-day-old fetal ovary (11 d/14? , Expt. A). Germ cells in zygotene (Z) and diplotene (D) stages are present in one of the cords, whereas only undifferentiated gonocytes (arrows) are seen in the other cord. x 1225. FIG. 5. A pair of gonads (11 P/149 , Expt. A) cultured for 7 days on a Nuclepore filter. In the ovary (110 ) upon the filter the oocytes are distributed over a large area. In the older ovary below the filter several oocytes of different sizes are present. x 250. FIG. 6. Close-up of the ovary (ll? ) as seen in Fig. 5. Oocytes in zygotene (Z) and pachytene stage (P) characterize this ovary. One oocyte has reached early diplotene stage (D). x 1225. FIG. 7. Part of the ovary (14 P ) from Fig. 5. Small and growing oocytes all in diplotene stage populate the gonad. x 925.
198
DEVELOPMENTAL BIOLOGY
VOLUME 52, 1976
/
BYSICOV AND SAX&N
Induction
tured singly within the same dish for 4 days (Experiment C), only a fraction of the male germ cells enter meiosis. Also in Experiment A, only a part of the male germ cells entered meiosis when situated a certain distance from the ovarian partner. This reduced effect of triggering the male germ cells to enter meiosis in the systems where female urogenital complex and the testis are not in close apposition to each other could be a function of different possibilities: the trigger substance could be destroyed or might lose some of its activity when released into the culture medium, or it might be secreted in very small quantities, resulting in a short activity radius. A limited radius of functioning of secreted substances was proposed by Macintyre, Hunter, and Morgan (1960). In kidney grafts of fetal rat gonads, these researchers showed an inhibitory effect of a fetal testis on the ovarian development proportional with the distance between the implanted gonads. The inhibitory effect of a testis on ovarian differentiation was also seen in Experiment B where the 14-day-old fetal ovaries and testis were cultured in close apposition. In the testes no germ cells were triggered to enter meiosis, and in the ovary the germ cells were prevented from reaching the diplotene stage and no follicles were formed. The influence of a MIS on the gonocytes in the differentiated testes was apparently prevented. Simultaneously the normal meiotic events of the oocytes and follicular formation in the ovarian partner were disturbed. This could be a result of a
of Meiosis in Fetal Mouse Testes
199
testicular secretion of a diffusible meiosispreventing substance (MPS), which reacted in an antagonistic way to MIS. The lack of meiotic cells in the testes of Experiment D might also be explained by the ability of the differentiated testis to produce MPS, which could protect the gonocytes from being affected by the MIS which presumably is present in the culture medium. A similar meiotic-preventing substance might also be functioning on the germ cells in the female gonad. When the oocytes have reached diplotene stage, they stop their meiotic division, become enclosed in a follicle, and may stay here for a long period of time, blocked in the diplotene stage. A follicular meiosis-preventing substance was proposed by Tsafriri and Channing (1975). Their results indicated that granulosa cells of follicles inhibited the meiotic resumption of oocytes from medium-sized follicles. The chemical nature of the secreted MIS and MPS is unknown. Applications of different steroids to the developing mammalian gonads do not induce the onset of meiosis (for review see Burns, 1961). This apparently excludes equivalence of sexhormones to the meiosis-inducing substance. The meiosis-preventing substance might be chemically related to MIS. In fact, the failure of induction of meiosis in Experiments B and D could be a result of a catabolic metabolism of the MIS by the differentiating testis. The existence of a MPS produced by the morphological differentiated testis might
FIG. 8. A testis (11 d, Expt. C) with testicular cords (arrows) cultured for 4 days on a filter in a culture dish, which also contained ovaries of the same age. x 250. FIG. 9. Part of a testicular cord from the testis seen in Fig. 8 showing three germ cells in leptotene stage. x 1825. FIG. 10. Part of a testis with gonocytes (arrows) cultured singly for 7 days (116, Expt. G). x 1225. FIG. 11. Part of a testis with gonocytes (arrows) cultured singly for 4 days (118, Expt. E). x 1225. FIG. 12. A pair of gonads (146/14P, Expt. B) cultured for 7 days on a filter. Testicular tubules with germ cells in the resting stage are present in the testis upon the filter. The ovary below the filter contains numerous oocytes. X 450. FIG. 13. Part of the ovary seen in Fig. 12. Many of the oocytes are pyknotic (arrows) or are in pachytene stage (P). Only a few have reached diplotene stage (D) and no follicles have formed. x 1225.
200
DEVELOPMENTAL
BIOLI DGY
also account for the long ameiotic period which precedes the first appearance of meiotic germ cells in the testis as well as the maintenance of the pool of nonmeiotic male germ cells in the adult testis. The onset of meiosis in the female mammalian germ cells is possibly a response to a locally secreted MIS produced by the cells of the rete ovarii (Byskov, 1975). We wish to thank Anja Tuomi, Inga Larsen, and Anne Lise Mohr for the skillful technical assistance. REFERENCES BAKER, T. G., and NEAL, P. (1973). Initiation and control of meiosis and follicular growth in ovaries of the mouse. Ann. Biol. Anim. Biochim. Biophys. 13, 137-144. BURNS, R. K. (1961). Role of hormones in the differentiation of sex. In “Sex and Internal Secretions” (W. C. Young, ed.), Vol. 1, pp. 76-158. Bailliere, Tindall and Cox, London. BYSROV, A. G. (1974). Does the rete ovarii act as a trigger for the onset of meiosis? Nature (London) 252, 396-397. BYSKOV, A. G. (1975). The role of the rete ovarii in meiosis and follicle formation in the cat, mink and ferret. J. Reprod. Fert. 45, 201-209. BYSKOV, A. G., SKAKKEBAEK, N. E., STAFANGER, G., and PETERS, H. (1976). Influence of ovarian surface epithelium and rete ovarii on follicle formation. J. Anat., In press. CHALLONER, S. (1975). Studies of oogenesis and follicular development in the golden hamster. 2. Initiation and control of meiosis in uitro. J. Anat. 119, 149-156. JOST, A. (1965). Gonadal hormones in sex differentiation of the mammalian foetus. In “Organogenesis” (R. L. de Haan and H. Ursprung, eds.), pp. 611-628. Holt, Rinehart and Winston, New York. MYSTKOWSKA, E. T., and TARKOWSKI, A. K. (1970).
VOLUME 52, 1976
Behavior of germ cells and sexual differentiation in late embryonic and early postnatal mouse chimaeras. J. Embryol. Exp. Morph. 23, 395-405. MACINTYRE, M. N., HUNTER, J. E., and MORGAN, A. H. (19601. Spatial limits of activity of fetal gonad inductors in the rat. Anat. Rec. 138, 137-147. MAULSON, P. (1973). Modification experimentale de l’apparition et de l’evolution de la prophase meiotique dans l’ovarie d’embryon de brebis. Ann. Biol. Anim. Bioch. Bioph. 9, 89-102. MCLAREN, A. (1972). Germ cell differentiation in artifmal chimaeras of mice. In “The Genetics of Spermatozoon” (R. A. Beatty and S. GluecksonWaelsch, eds.), Proc. Int. Symp., pp. 313-324. MCLAREN, A., CHANDLEY, A. C., and KOFMAN-ALFARO, S. (1972). A study of meiotic germ cells in the gonads of foetal mouse chimaeras. J. Embryol. Exp. Morph. 27, 515-524. OWR, L., and BLANDAU, R. J. (1969). Ultrastructural studies on fetal and early postnatal mouse ovaries. I. Histogenesis and organogenesis. Amer. J. Anut. 124, 163-186. PETERS, H., LEVY, E., and CRONE, M. (1962). Deoxyribonucleic acid synthesis in oocytes of mouse embryos. Nature (London) 195, 915-916. SHORT, R. V. (1972). Germ cell sex. In “The Genetics of the Spermatozoon” (R. A. Beatty and S. Glueckson-Waelsch, eds.), Proc. Int. Symp., pp. 325-345. TSAFRIRI, A., and CHANNING, C. P. (1975). An inhibitory influence of granulosa cells and follicular fluid upon porcine oocyte meiosis in u&o. Endocrinology 96, 922-927. WARTIOVAARA, J., NORDLING, S., LEHTONEN, E., and SAXON, L. (1974). Transfilter induction of kidney tubules: Correlation with cytoplasmic penetration into Nuclepore filters. J. Embryol. Exp. Morph. 31, 667-682. WHITTEN, W. K. (1975). Chromosomal basis for hermaphrodism in mice. In 33rd Symp. Sot. Develop. Biol., Develop. Biol. Reprod., pp. 189-205. WOLFF, E. (1952). Sur la differentiation des gonades de souris explantees in vitro. C.R. Acud. Sci. 234, 1712-1714.
BIOLOGY
Induction
52, 193-200 (1976)
of Meiosis ANNE
in Fetal Mouse Testis in Vitro
GRETE BYSKOV’
’ The Finsen Laboratory, The Finsenilnstitute, WI Department of Pathology,
AND LAURI
SAXBN~
Strandboulevarden 49,210O K@benhavn 8, Denmark; University of Helsinki, 0290 Helsinki, Finland
Accepted April
and
2,1976
Fetal mouse testes and ovaries with their urogenital connections were cultured singly or in pairs on Nuclepore filters. When a testis in which the sex was not yet morphologically detectable was cultured together with older ovaries containing germ cells which were progressing through the meiotic prophase, the male germ cells were triggered to enter meiosis. When older fetal testes in which the testicular cords have developed were cultured together with ovaries of the same age with germ cells in meiosis, the oocytes were prevented from reaching diplotene stage. It was concluded that the fetal male and female gonads secrete diffusable substances which influence germ cell differentiation. The male gonad secretes a “meiosis-preventing substance” (MPS) which can arrest the female germ cells within the meiotic prophase. The female gonad secretes a “meiosis-inducing substance” (MIS) which can trigger the nondifferentiated male germ cells to enter meiosis. INTRODUCTION
The mechanism which triggers the mammalian male and female germ cells to enter meiosis is not known, and experiments have failed to induce the onset of meiosis in the undifferentiated germ cell. In spontaneous hermaphrodites of mice, meiotic cells have been found in the fetal testes (Whitten, 1975). Also in fetal mouse chimeras, germ cells in meiotic prophase were observed in fetal testes (Mystkowska and Tarkowski, 1970; McLaren, Chandley, and Kosman-Aljaro, 1972). McLaren (1972) proposed that these germ cells might be XX as well as XY cells and that they were triggered to enter meiosis by surrounding somatic XX cells. The other germ cells which did not start the meiosis might be placed between noninducing somatic XY cells. One specific somatic cell system which is of mesonephric origin, the rete ovarii, seems to be necessary for the differentiation of the female gonad. The onset of meiosis (Byskov, 1974), as well as the follicular formation (Byskov et al., 1976), seems to be dependent on this urogenital connection during the early gonadal anlage.
In mammals the differentiation of a gonad into a recognizable testis occurs before the male germ cells enter meiosis. The testicular cords normally differentiate during fetal life but none of the male germ cells starts the meiotic prophase before early puberty. At the time when the testicular cords differentiate, the ovary of the same age and from the same species may not be recognizable as a female gonad by other criteria than that it is lacking the cord formations (Jost, 1965). Nevertheless, the meaning of the expression “sex differentiation” of the gonads works to cover that process of morphological differentiations of the testis which renders the gonads to be recognized as testes or ovaries. In some species, however, the meiosis in the female germ cells starts simultaneous with, or immediately after, the time when organization of the testicular cords in the male gonad occurs. In the mouse, for instance, sex differentiation of the testis starts on Day 12 (Odor and Blandau, 1969) as does the S-phase of the first meiotic division in the female germ cells (Peters, Levy, and Crone, 1962.) In the present work, we investigated 193
Copyright 0 1976 by Academic press, Inc. All rights of reproduction in any form reserved.
194
DEVELOPMENTAL
BIOLOGY
whether the female mouse urogenital complex, i.e., ovary together with rete ovarii, along with germ cells in meiosis could initiate the onset of meiosis in germ cells of fetal testes before and after gonadal sex differentiation. MATERIALS
AND
METHODS
Fetuses of the mouse strain BALBI cxCBA/TGTs were removed at the eleventh and the fourteenth day postcoitum (p.c.1. The urogenital complex consisting of the gonads together with the attached cranial part of the mesonephros, i.e., the rete system, were dissected free and glued with 1% agar on Nuclepore filters with a pore diameter of 1.0 or 0.1 pm (General Electric Co., Pleasanton, Calif.). The explants were placed either singly at one side of the filter or in pairs on opposite sides. One or several filters with explants were supported by a grid and placed in a culture dish with 2 ml of Eagle’s balanced salt solution with 10% fetal calf serum in a gasphase of 5% COz in air at 37°C. After a culture period of 4 or 7 days, the filters with the explants were fixed in Zenker, embedded in paraffin, sectioned serially at 5 pm, and stained with PAS and hematoxylin. The experimental setup, including the sexes and ages of the gonads in the cultures (Experiments A to K), are shown in Table 1. In Experiments A and B (paired gonads), a gonad was present at each side of the filter, forming a pair. Experiment A consisted of three dishes, each containing six pairs of gonads (18 filters). In B, there were two dishes, with three and four pairs of gonads (7 filters). In Experiments C to K, a single gonad was glued to the filter. C and D (single gonads) consisted each of one dish containing six fetal gonads from the eleventh day p.c., males together with females. E-I and K were culture controls. E and F consisted of three and two dishes, respectively, each with one single ll-dayold fetal gonad cultured for 4 days. G and H consisted of four and three dishes, each
VOLUME 52. 1976 TABLE
1
EXPERIMENTAL SETUP OF FETAL MOUSE GQNADS IN CULTURE Experiment Start of culture End of culture NUllIAge (days) and Age (days) and her of sex of gonads on sex of gonads on filters filters flkers Paired A gonads
116 iz+iG
B
Single gonads
119
148 149
l&3* 219
189 +m
18
21d 219**
7
C
116 -+iE
-
15c?* +m
6
D
116 -+iiF
-
188 -+m
6
Culture E controls F G
116
-
156
119
159
11C?
188
-
H
K
11P
189
146 -
216
149
* Meiosis-induced. vented.
219 ** Meiosis-arrested
3 2 4 3 6 6 or pre-
with one single 11-day-old fetal gonad, which was cultured for 7 days. In both Experiment I and K, there was one dish with six single gonads of lCday-old fetal males or females, respectively. The total number of filters was 66. In one of the dishes in Experiment A, the six filters had a pore diameter of 0.1 pm. The pore diameter in all other filters was 1.0 pm. Morphologically the sex in 11-day-old embryos cannot be determined, whereas this is possible at Day 14 p.c. The dishes, in which 11-day-old fetal gonads were paired with lCday-old ovaries (A), or were cultured singly on separate filters (C, D), therefore contained randomly both testes as well as ovaries. The sex of the ll-dayold gonads was determined morphologically at the end of the culture period. Go-
BYSKOV
AND SAXON
Znduction
nads of fetuses of ll-, 15, 18, and 21-dayold mice, i.e., controls, were sectioned and stained as in the experiments and their morphology was compared with the paired gonads, single gonads, and culture controls. RESULTS
Controls. In the 11-day-old gonad, in which the morphological sex differentiation has not yet occurred, many germ cells are dividing but no meiotic stages are seen. On Day 14 p.c., the gonads can be recognized as testes or ovaries. In the male where the testicular cords have formed, almost all of the gonocytes are in the “resting,” nondividing stage and no meiotic tigures are present. The testis on Days 15, 18, and 21 resembles the 14-day-old one in respect to the germ cells, which are still “resting” cells. In the lCday-old fetal ovary many oocytes are in leptotene and some are in zygotene. On Day 15, germ cells are also in pachytene. On Day 18 p.c., almost all germ cells have entered the meiotic prophase. Most are in pachytene and some have reached diplotene. By Day 21, which is usually the day of birth, all oocytes are in diplotene. Follicles have formed and growth of both oocytes and follicles have started. Culture controls. The 11-day-old fetal testes cultured for 4 days (Fig. 11) and 7 days (Fig. 10) (Experiments E and G) had testicular cords with germ cells in the resting stage. The morphology of the fetal ovaries of Experiments F and H was comparable to the control ovaries of Days 15 and 18, respectively. The developmental stage of the lCday-old fetal testis (Experiment I) and ovary (Experiment K) after 7 days in culture was similar to 21-day-old control testes and ovaries, respectively. Paired gonads. In Experiment A, eight of the 11-day-old fetal gonads cultured transfilter to fetal ovaries of Day 14 were identified as testes by the presence of gonocytes (Fig. 4) and/or testicular tubules
of Meiosis in Fetal Mouse Testes
195
(Fig. 1). Germ cells in different stages of the meiotic prophase were present in all these testes. No difference was noticed between the transfilters of l.O-pm pore size and the 0.1 pm. All germ cells in the testicular cords which were placed directly above the ovarian tissue had entered meiosis. Most germ cells were in zygotene and pachytene (Fig. 2). In one case, two diplotene stages were recognized (Fig. 4). However, in cords which were situated between 0.3 and 0.8 mm from the ovarian tissue, only about 20% of the germ cells had entered meiosis (Fig. 1). The remnant part of the germ cells in these cords were still in the resting phase. The 10 other explants of the 11-day-old fetal gonads of Experiment A were determined to be ovaries. The germ cells were distributed over a large area and revealed no signs of being enclosed within cords (Fig. 5). The germ cells were in zygotene and pachytene stages (Fig. 6). The 14-day-old fetal ovaries, which served as partners for all the 11-day-old fetal gonads in Experiment A, contained oocytes in diplotene stages after the 7 days in culture. The 14-day-old fetal ovaries combined with 11-day-old fetal ovaries had oocytes and follicles, which had started to grow (Figs. 5 and 7). Ovaries in combination with the testes, however, differed from the pure ovarian pairing: in all ovaries of Day 14, which were combined with ll-dayold fetal testes (Fig. 11, few of the oocytes had reached diplotene and few follicles had started to grow (Fig. 3). Many of the oocytes seem to be degenerating in early diplotene stage (Fig. 3). In Experiment B, the fetal testes of Day 14, which were cultured for 7 days transfilter to a lCday-old fetal ovary, had all germ cells in the resting, nondividing stage (Fig. 12). In the ovaries of these pairs, however, most oocytes were pyknotic (Fig. 13, arrows) or appeared to be degenerating pachytene stages. Few oocytes had reached the diplotene stage and no follicles had formed.
196
DEVELOPMENTAL
BIOLOGY
VOLUME 52, 1976
BYSKOV AND SAXON
Induction
Single gonads. Four of the six gonads in Experiment C were identified as testes. All these testes had cords in which about 20% of the germ cells were in leptotene (Figs. 8 and 91, whereas the remnant parts were still in the resting stage. The two ovaries in the dish contained germ cells in leptotene and zygotene. In Experiment D, there were three testes and three ovaries. After 7 days in culture together with ovaries, all the germ cells of these testes were in the resting stage. Germ cells in meiosis could not be recognized. The three ovaries in the culture were comparable to the 1% day-old controls. DISCUSSION
Three main results can be extracted from this study. First, germ cells of fetal mouse testes in which testicular cords are not yet formed are triggered to enter meiosis when cultured together with the female fetal mouse urogenital complex containing germ cells in meiosis. Second, the trigger, which influences the male germ cells to enter meiosis, is a diffusible substance. Third, the differentiation of fetal ovaries is inhibited when cultured together with fetal testis. The control mechanisms of the onset of meiosis in mammals are poorly understood. In cultures of ovaries of mice (Wolff, 1952; Baker and Neal, 1973), sheep
of Meiosis
in Fetal Mouse Testes
197
(Mauleon, 19731, and hamsters (Challoner, 19751, it has been shown that the germ cells will survive and progress beyond the leptotene stage, only in ovaries which are taken into culture after a certain critical stage of development. This critical stage seems to coincide with the time the somatic elements start to differentiate. The interaction between the somatic cells and the germ cells during gonadal differentiation has been reviewed by Short (1972). One of the somatic elements, the rete ovarii has been shown to be of importance for the onset of meiosis in the mouse (Byskov, 1974); the urogenital connection containing the rete ovarii was therefore preserved in all the experiments of the present study. In the described transfilters the meiosis was induced not only in filter systems, which should allow cellular contact (Experiment A with l.O-pm pore diameter), but also in the setup in which the small filter pores of 0. l-pm pore diameter probably will prevent cellular contact (Wartiovaara et al., 1974), as well as in the setup of single gonads of different sex cultured separately in the same dish (Experiment C). This indicates that a meiosis-inducing substance (MIS) secreted from the urogenital complex has reached the testes by diffusion. When the testes and ovaries are cul-
FIG. 1. A pair of gonads (11 d 114 P , Expt. Al cultured for 7 days on a Nuclepore filter. The testis upon the filter has germ cells in meiotic stages confined to testicular cords. Only a few meiotic cells (black dots) are present in the peripherally placed cord (thick arrow). The ovary below the filter contains oocytes in diplotene stage (thin arrow). x 250. FIG. 2. Testicular cords of the testis shown in Fig. 1. Most of the germ cells are in zygotene (Z) and in pachytene stage (P). x 1225. FIG. 3. Part of the ovary seen in Fig. 1. One of the oocytes seems to be degenerating in early diplotene stage (arrow). One is pyknotic and one is in diplotene stage (D). x 925. FIG. 4. Cross section of two testicular cords of a fetal testis cultured transfilter to a ll-day-old fetal ovary (11 d/14? , Expt. A). Germ cells in zygotene (Z) and diplotene (D) stages are present in one of the cords, whereas only undifferentiated gonocytes (arrows) are seen in the other cord. x 1225. FIG. 5. A pair of gonads (11 P/149 , Expt. A) cultured for 7 days on a Nuclepore filter. In the ovary (110 ) upon the filter the oocytes are distributed over a large area. In the older ovary below the filter several oocytes of different sizes are present. x 250. FIG. 6. Close-up of the ovary (ll? ) as seen in Fig. 5. Oocytes in zygotene (Z) and pachytene stage (P) characterize this ovary. One oocyte has reached early diplotene stage (D). x 1225. FIG. 7. Part of the ovary (14 P ) from Fig. 5. Small and growing oocytes all in diplotene stage populate the gonad. x 925.
198
DEVELOPMENTAL BIOLOGY
VOLUME 52, 1976
/
BYSICOV AND SAX&N
Induction
tured singly within the same dish for 4 days (Experiment C), only a fraction of the male germ cells enter meiosis. Also in Experiment A, only a part of the male germ cells entered meiosis when situated a certain distance from the ovarian partner. This reduced effect of triggering the male germ cells to enter meiosis in the systems where female urogenital complex and the testis are not in close apposition to each other could be a function of different possibilities: the trigger substance could be destroyed or might lose some of its activity when released into the culture medium, or it might be secreted in very small quantities, resulting in a short activity radius. A limited radius of functioning of secreted substances was proposed by Macintyre, Hunter, and Morgan (1960). In kidney grafts of fetal rat gonads, these researchers showed an inhibitory effect of a fetal testis on the ovarian development proportional with the distance between the implanted gonads. The inhibitory effect of a testis on ovarian differentiation was also seen in Experiment B where the 14-day-old fetal ovaries and testis were cultured in close apposition. In the testes no germ cells were triggered to enter meiosis, and in the ovary the germ cells were prevented from reaching the diplotene stage and no follicles were formed. The influence of a MIS on the gonocytes in the differentiated testes was apparently prevented. Simultaneously the normal meiotic events of the oocytes and follicular formation in the ovarian partner were disturbed. This could be a result of a
of Meiosis in Fetal Mouse Testes
199
testicular secretion of a diffusible meiosispreventing substance (MPS), which reacted in an antagonistic way to MIS. The lack of meiotic cells in the testes of Experiment D might also be explained by the ability of the differentiated testis to produce MPS, which could protect the gonocytes from being affected by the MIS which presumably is present in the culture medium. A similar meiotic-preventing substance might also be functioning on the germ cells in the female gonad. When the oocytes have reached diplotene stage, they stop their meiotic division, become enclosed in a follicle, and may stay here for a long period of time, blocked in the diplotene stage. A follicular meiosis-preventing substance was proposed by Tsafriri and Channing (1975). Their results indicated that granulosa cells of follicles inhibited the meiotic resumption of oocytes from medium-sized follicles. The chemical nature of the secreted MIS and MPS is unknown. Applications of different steroids to the developing mammalian gonads do not induce the onset of meiosis (for review see Burns, 1961). This apparently excludes equivalence of sexhormones to the meiosis-inducing substance. The meiosis-preventing substance might be chemically related to MIS. In fact, the failure of induction of meiosis in Experiments B and D could be a result of a catabolic metabolism of the MIS by the differentiating testis. The existence of a MPS produced by the morphological differentiated testis might
FIG. 8. A testis (11 d, Expt. C) with testicular cords (arrows) cultured for 4 days on a filter in a culture dish, which also contained ovaries of the same age. x 250. FIG. 9. Part of a testicular cord from the testis seen in Fig. 8 showing three germ cells in leptotene stage. x 1825. FIG. 10. Part of a testis with gonocytes (arrows) cultured singly for 7 days (116, Expt. G). x 1225. FIG. 11. Part of a testis with gonocytes (arrows) cultured singly for 4 days (118, Expt. E). x 1225. FIG. 12. A pair of gonads (146/14P, Expt. B) cultured for 7 days on a filter. Testicular tubules with germ cells in the resting stage are present in the testis upon the filter. The ovary below the filter contains numerous oocytes. X 450. FIG. 13. Part of the ovary seen in Fig. 12. Many of the oocytes are pyknotic (arrows) or are in pachytene stage (P). Only a few have reached diplotene stage (D) and no follicles have formed. x 1225.
200
DEVELOPMENTAL
BIOLI DGY
also account for the long ameiotic period which precedes the first appearance of meiotic germ cells in the testis as well as the maintenance of the pool of nonmeiotic male germ cells in the adult testis. The onset of meiosis in the female mammalian germ cells is possibly a response to a locally secreted MIS produced by the cells of the rete ovarii (Byskov, 1975). We wish to thank Anja Tuomi, Inga Larsen, and Anne Lise Mohr for the skillful technical assistance. REFERENCES BAKER, T. G., and NEAL, P. (1973). Initiation and control of meiosis and follicular growth in ovaries of the mouse. Ann. Biol. Anim. Biochim. Biophys. 13, 137-144. BURNS, R. K. (1961). Role of hormones in the differentiation of sex. In “Sex and Internal Secretions” (W. C. Young, ed.), Vol. 1, pp. 76-158. Bailliere, Tindall and Cox, London. BYSROV, A. G. (1974). Does the rete ovarii act as a trigger for the onset of meiosis? Nature (London) 252, 396-397. BYSKOV, A. G. (1975). The role of the rete ovarii in meiosis and follicle formation in the cat, mink and ferret. J. Reprod. Fert. 45, 201-209. BYSKOV, A. G., SKAKKEBAEK, N. E., STAFANGER, G., and PETERS, H. (1976). Influence of ovarian surface epithelium and rete ovarii on follicle formation. J. Anat., In press. CHALLONER, S. (1975). Studies of oogenesis and follicular development in the golden hamster. 2. Initiation and control of meiosis in uitro. J. Anat. 119, 149-156. JOST, A. (1965). Gonadal hormones in sex differentiation of the mammalian foetus. In “Organogenesis” (R. L. de Haan and H. Ursprung, eds.), pp. 611-628. Holt, Rinehart and Winston, New York. MYSTKOWSKA, E. T., and TARKOWSKI, A. K. (1970).
VOLUME 52, 1976
Behavior of germ cells and sexual differentiation in late embryonic and early postnatal mouse chimaeras. J. Embryol. Exp. Morph. 23, 395-405. MACINTYRE, M. N., HUNTER, J. E., and MORGAN, A. H. (19601. Spatial limits of activity of fetal gonad inductors in the rat. Anat. Rec. 138, 137-147. MAULSON, P. (1973). Modification experimentale de l’apparition et de l’evolution de la prophase meiotique dans l’ovarie d’embryon de brebis. Ann. Biol. Anim. Bioch. Bioph. 9, 89-102. MCLAREN, A. (1972). Germ cell differentiation in artifmal chimaeras of mice. In “The Genetics of Spermatozoon” (R. A. Beatty and S. GluecksonWaelsch, eds.), Proc. Int. Symp., pp. 313-324. MCLAREN, A., CHANDLEY, A. C., and KOFMAN-ALFARO, S. (1972). A study of meiotic germ cells in the gonads of foetal mouse chimaeras. J. Embryol. Exp. Morph. 27, 515-524. OWR, L., and BLANDAU, R. J. (1969). Ultrastructural studies on fetal and early postnatal mouse ovaries. I. Histogenesis and organogenesis. Amer. J. Anut. 124, 163-186. PETERS, H., LEVY, E., and CRONE, M. (1962). Deoxyribonucleic acid synthesis in oocytes of mouse embryos. Nature (London) 195, 915-916. SHORT, R. V. (1972). Germ cell sex. In “The Genetics of the Spermatozoon” (R. A. Beatty and S. Glueckson-Waelsch, eds.), Proc. Int. Symp., pp. 325-345. TSAFRIRI, A., and CHANNING, C. P. (1975). An inhibitory influence of granulosa cells and follicular fluid upon porcine oocyte meiosis in u&o. Endocrinology 96, 922-927. WARTIOVAARA, J., NORDLING, S., LEHTONEN, E., and SAXON, L. (1974). Transfilter induction of kidney tubules: Correlation with cytoplasmic penetration into Nuclepore filters. J. Embryol. Exp. Morph. 31, 667-682. WHITTEN, W. K. (1975). Chromosomal basis for hermaphrodism in mice. In 33rd Symp. Sot. Develop. Biol., Develop. Biol. Reprod., pp. 189-205. WOLFF, E. (1952). Sur la differentiation des gonades de souris explantees in vitro. C.R. Acud. Sci. 234, 1712-1714.