- Email: [email protected]
Steroid production by Xenopus ovarian follicles at different developmental stages
DEVELOPMENTAL
BIOLOGY
99,502-509
(1983)
Steroid Production by Xenopus Ovarian Follicles at Different Developmental Stages’ J. E. FORTUNE’ Section of Physiology in the Division of Biological Sciences and Department of Physidogzl in the CoUege of Vete&mry Medicine, Cornell University, Ithaca, New York 14858 Received September 28, 1979; accepted in revised fm
May 16, 1986
Xaqpus ovarian follicles at different developmental stages were compared with respect to their capacity to produce and secrete steroids and to respond to gonadotropic hormones with changes in steroid production. Individual follicles were obtained by treating ovaries with collagenase and were incubated for 10 hr in incubation medium alone or in medium containing a Xenopus pituitary homogenate (FPH, 0.04 pituitary/ml). At 10 hr oocytes were scored for meiotic maturation and ovulation and incubation media and follicles were collected separately and frozen. Both tissue and media were measured for progesterone, testosterone, and estradiol by radioimmunoassay. The smallest follicles (Dumont Stages I and II) secreted very low or nondetectable levels of these three steroids in both the presence and absence of FPH. Medium-sized follicles (Stages III and IV) were active in secreting estradiol and FPH stimulated a twofold increase in the accumulation of estradiol in the incubation medium. In follicles larger than Stage IV, estradiol secretion decreased as follicular size increased. In contrast, large follicles (Stages V and VI) secreted much more testosterone and progesterone than medium-sized follicles and FPH significantly increased the accumulation of these two steroids. Overall, follicular secretion of testosterone was much greater than secretion of progesterone or estradiol. The concentration of the three steroids in the ovarian follicles at 10 hr of incubation followed patterns that were similar in general to the patterns of accumulation of steroids in the incubation medium. Ovine LH, but not FSH, was effective in stimulating secretion of the three steroids and in triggering meiotic maturation of oocytes in Stage VI follicles. Although many of the morphological and biochemical events of oogenesis and folliculogenesis in Xenqt~ have been described, this is the first report of the developmental pattern of steroid production. The data indicate that growth and development of Xenopua follicles are accompanied by shifts in steroidogenic abilities, with estradiol produced by medium-sized follicles and androgen and progesterone secreted by large follicles.
Despite the abundant evidence that steroids applied exogenously can influence oocyte growth and meiotic maturation, there have been few investigations of endogenous ovarian levels of steroids or the effects of gonadotropins on steroid production by anuran ovaries. Chieffi and Lupo (1963) detected progesterone, estradiol178, estriol, and estrone in the ovaries of BT&Ovulgaris. Fortune et al (1975) found that Xenopus ovaries contain progesterone and that gonadotropin preparations that trigger maturation and ovulation also significantly increase ovarian concentrations of progesterone. Licht and Crews (1976) reported that Rana luteinizing hormone greatly increased the secretion of progesterone by Ram ovaries incubated in vitro. These findings, coupled with reports that inhibitors that block progesterone synthesis also inhibit maturation and ovulation (Snyder and Schuetz, 1973; Fortune, unpublished observations), lend support to the hypothesis that progesterone is a necessary intermediate in the induction of maturation and ovulation. Several groups have reported that anuran ovaries and oocytes can convert exogenous radiolabeled steroid pre-
INTRODUCTION
It has been known for many years that exogenous application of gonadotropic or steroid hormones can exert effects on several aspects of ovarian development in amphibians. Treatment with gonadotropins or steroids can stimulate meiotic maturation and/or ovulation of full-grown anuran oocytes. Some evidence suggests that oocyte maturation is triggered via a two-step mechanism in which a gonadotropic hormone(s) stimulates follicle cells to produce a steroid which acts directly on the oocyte to cause meiotic maturation (Masui, 1967; Smith et CAL,1968). It is well established that exogenous estradiol can stimulate the synthesis of yolk proteins by the liver of male and female amphibians (see review by Wallace and Bergink, 1974). It has been suggested that gonadotropic hormones increase ovarian production of estrogen which in turn stimulates vitellogenesis. 1 This work was supported by National Research Service Award 5 F32 HD05049. *To whom all correspondence should be addressed: 823 Veterinary Research Tower, Cornell University, Ithaca, New York 14853. 502 0012-1606/83 $3.00 Copyright All rights
0 1983 by Academic Press, Inc. of reproduction in any form reserved.
J. E. FORTUNE
Ovarian
cursors to estrogens and androgens, as well as prowstins (Redshaw and Nicholls, 1971; Reynhout and Smith, 197% Sanyal and Sibre, 1973; Fouchet et al., 1975; ThibierFouchet et aL, 1976). More recently Fortune and Tsang (1981) found that Xenoms ovaries contain and secrete endogenous estradiol-17P and testosterone. Gonadotropin treatment in vitro stimulated 2- and lo-fold increases in ovarian estradiol and testosterone concentrations, respectively. Testosterone concentrations in gonadotropin-treated ovaries were much greater than the concentrations of progesterone or estradiol. This finding raised the question of the role of androgens in Xenopus ovaries. The current experiments were designed to determine the developmental pattern of steroid production by Xenopus ovarian follicles. When Xenopus are maintained in the laboratory all stages of follicular development are usually present within the ovary simultaneously and therefore, follicles at various developmental stages can be isolated from the ovaries of one frog. Follicles at five different stages have been isolated and incubated with gonadotropins and the secretion of progesterone, testosterone, and estradiol has been measured by radioimmunoassay. MATERIALS
AND
METHODS
1. Animals and incubations. Sexually mature Xenopus laevis females were obtained from the South African Snake Farm (Fish Hoek, South Africa), maintained in tap water, and fed a diet of chopped beef heart and liver. Frogs were killed by decapitation and the pituitary was removed and homogenized in 2 ml O-R2 solution (Wallace et al., 1973). The ovaries were removed, rinsed with O-R2 solution, weighed, and cut into small pieces of about 30 mg. Individual follicles were obtained from the ovarian pieces by dissolving the interfollicular connections with collagenase. About 2 g of ovarian pieces were placed in each of several 25-ml Erlenmeyer flasks containing approximately 10 ml of a 0.15% solution of collagenase (Sigma Chemical Co.) in O-R2 solution. The flasks were placed in a Dubnoff metabolic shaker at 26°C and shaken at about 80 strokes/min for 20-60 min. This procedure freed many individual follicles from the ovarian pieces, but did not disrupt the association between oocytes and their surrounding follicle cells. The time needed for dissociation varied considerably among animals. In some experiments, pieces that had not completely dissociated into follicles were collected and exposed to collagenase a second time. After dissociation the tissue was rinsed several times with O-R2 solution and individual follicles that appeared intact and healthy were sorted into size classes under a dissecting microscope. Oocytes that had
Sikm3id.s in Xen0pu.s
503
been separated from their follicular tissue or follicles that appeared damaged by collagenase treatment were discarded. Follicles were classified according to the criteria of Dumont (1972). Follicles at Dumont Stages I and II are small (50-300 and 300-450 pm in diameter, respectively) and previtellogenic. Vitellogenesis begins at Stage III (450-600 pm) and continues during Stage IV (600-1000 pm), when animal and vegetal hemispheres differentiate. Vitellogenesis is completed during Stage V (1000-1200 pm) and Stage VI follicles are full-grown (1200-1300 pm) and capable of undergoing meiotic maturation and ovulation. Follicles were incubated at room temperature in the wells of plastic cluster dishes (Costar, Cambridge, Mass.) in 500 ~1 O-R2 solution containing polyvinylpyrrolidone (PVP). PVP reduced the tendency of follicles treated with collagenase to stick to the incubation dishes. In initial experiments incubation wells contained 20 Stage VI follicles, 25 Stage V, 30 Stage IV, 40 Stage II, or 50 Stage I-II. Later experiments focused on certain of these stages; the number of follicles used in these experiments will be described in the Results section. Follicles were incubated in O-R2 solution alone (controls) or O-R2 solution containing the frog pituitary homogenate (FPH), highly purified ovine luteinizing hormone (LH; LER-1056-C2), or highly purified ovine follicle stimulating hormone (FSH; LER-1976-A2). The dose of FPH used throughout these experiments (0.04 pituitary/ ml) stimulates a maximal response in terms of percentage of Stage VI oocytes that ovulate and exhibit meiotic maturation (Fortune et aL, 1975). After 10 hr of incubation follicles were examined under a dissecting microscope for meiotic maturation and ovulation. A lo-hr incubation period was chosen because ovulation and maturation usually occur within 10 hr in vitro and because our previous results indicated that when whole ovarian pieces, containing all stages of follicular development, were incubated with FPH, medium and tissue concentrations of estradiol, progesterone, and testosterone were either increasing slightly or had reached a plateau by 10 hr (Fortune and Tsang, 1981; Fortune et aL, 1975). Maturation was scored by the presence of Roux’ spot, a white spot that appears at the animal pole during maturation. Ovulation was scored by counting the number of oocytes that were no longer enclosed within the follicular wall. After dishes were scored for maturation and ovulation, the medium and tissue were collected separately from each incubation well and frozen. 2. Steroid assays. The ovarian follicles collected from the incubation wells were homogenized in 1 ml O-R2 solution and the homogenates were extracted with diethyl ether and methanol as described for Xenopus ovarian pieces (Fortune and Tsang, 1981). Progesterone,
504
DEVELOPMENTAL BIOLOGY
testosterone, and estradiol were measured in aliquots of these extracts and in unextracted aliquots of incubation medium by radioimmunoassays (RIA) that have been described previously (Fortune and Armstrong, 1978; Fortune and Eppig, 1979). The estradiol antiserum is highly specific for estradiol and its low cross-reactivities with other steroids have been reported previously (Korenman et aL, 1974). The progesterone antiserum cross reacts 11.64% with 5/3-dihydroprogesterone, ~10% with other progestins tested, and ~0.1% with the androgens and estrogens tested. The testosterone antiserum (Gay and Kerlan, 1978) is not specific for testosterone and cross-reacts with 5a-dihydrotestosterone (DHT; 69%), 3cr-androstanediol (14%), and 3@-androstanediol (22%); cross-reactivities with estrogens, progestins, corticoids, and other androgens tested were 1% or less. In previous experiments testosterone was isolated chromatographically from extracts of ovarian pieces (containing all stages of follicular development) and incubation medium (Fortune and Tsang, 1981). Since almost all of the androgen measured by the testosterone antiserum before chromatographic separation was recovered in the testosterone fraction after chromatography, extracts were not subjected to chromatographic purification in the current experiments and the steroid measured by the radioimmunoassay is referred to as “testosterone.” Accumulation of steroids in the incubation medium (or net secretion) and concentration of steroids in the tissue are both expressed per follicle incubated. Significant differences were determined by analysis of variance, Scheffe’s method for comparisons of means, and Duncan’s multiple range test. RESULTS
I. Production of Progesterone, Testosterone, and Estradiol-l7@ by FoUicles at D&m& Developmental Stages In the first series of experiments follicles were sorted into five different size classes and incubated for 10 hr in the presence or absence of frog pituitary homogenate (FPH; 0.04 pituitary/ml). Each treatment was applied in duplicate to tissue from four frogs. After 10 hr follicles were scored for maturation and ovulation and the incubation medium and tissue were collected, frozen, and later measured for steroids by RIA. The lower panels of Figs. l-3 show the accumulation of progesterone, testosterone, and estradiol in the medium of follicles of different sizes and the upper panels depict steroid concentrations in the follicles. The secretion of testosterone and progesterone followed similar developmental patterns (Figs. 1 and 2, lower panels). In control dishes of small (Stages I-II) or medium-sized follicles (Stages III and IV) accumulation of these two steroids was either very low or below
VOLUME 99. 1983
z.z : 12C
n
CONTROL
q
TISSUE
+I
$8 =.0 :
z
FPH
4
STAGE
OF FOLLICULAR
DEVELOPMENT
FIG. 1. Progesterone concentration (pg/follicle + SEM, ?z = 8 incubation dishes) in follicular tissue and incubation medium of ovarian follicles incubated in control medium or medium containing a pituitary homogenate (FPH) for 10 hr.
the sensitivity of the assays. Only follicles that had reached Stage V or VI secreted substantial quantities of progesterone and testosterone and Stage VI follicles secreted slightly more of these two steroids than did Stage V. Treatment with FPH increased the secretion of testosterone and progesterone by Stage VI and Stage V follicles by approximately 400% (P < 0.001) and by Stage IV follicles by approximately 1500% (P < 0.001). Both control and FPH-treated follicles secreted about 10 times more testosterone than progesterone. The concentrations of testosterone in follicular tissue (Fig. 2, upper panel) were similar both in pattern and amount to the concentrations in the culture medium. In contrast, progesterone secretion (i.e., accumulation in the medium) increased much more dramatically with follicular development than did the tissue concentrations of progesterone (Fig. 1, upper panel) and FPH had a far greater stimulatory effect on secretion of progesterone than on follicular concentrations. The developmental pattern of estradiol secretion was quite different from that of progesterone and testosterone. Medium-sized follicles (Stages III and IV) were most active in secreting estradiol, while smaller or larger follicles produced much lower quantities (Fig. 3, lower panel). FPH stimulated a twofold increase (P < 0.01) in estradiol secretion by Stages V, IV, and III, but
Ovarian
J. E. FORTUNE
s
FJ +I 320 al
n
CONTROL
q
FPH TISSUE
MEDIUM
F
STAGE
OF FOLLICULAR
DEVELOPMENT
FIG. 2. Testosterone concentration (pg/follicle f SEM, cubation dishes) in follicular tissue and incubation medium follicles incubated in control medium or medium containing homogenate (FPH) for 10 hr.
n = 8 inof ovarian a pituitary
the gonadotropin treatment only slightly increased (P < 0.05) estradiol secretion by Stage VI or Stages III. At most developmental stages the amounts of estradiol in follicular tissue were greater than the quantities secreted into the medium, but the overall developmental pattern and the responses to FPH were very similar (Fig. 3, upper panel). In these experiments FPH treatment also stimulated the maturation and ovulation of Stage VI, and occasionally Stage V, follicles. The percentage maturation for Stage VI oocytes ranged from 10 to 98% for the four experiments and the mean was 51.2 f 12.5% (n = 8 incubation wells with 20 follicles/well). The mean percentage maturation was 8.5 + 2.2% for Stage V oocytes (n = 8 incubation wells with 25 follicles/well). FPH triggered ovulation in only one experiment in which 95% of Stage VI follicles and 12% of Stage V follicles ovulated.
2. The Effects of LH and FSH on Xenopus Ovarian FoUicle.s. Since an LH-like and an FSH-like gonadotropin have been isolated from pituitaries of the bullfrog Rana cutesbeiana (Licht and Papkoff, 1974; Papkoff et al, 1976), it was of interest to test the effects of LH and FSH on steroidogenesis by Xewpus follicles. Purified ovine LH and FSH were used because purified amphibian gonadotropins were not available and ovine LH was reported to be active in stimulating ovulation and meiotic mat-
Steroidsin
505
Xenopua
uration in Xenqs (Licht and Papkoff, 1976; Mulner et aZ., 1978). Figures 4-6 show steroid secretion when follicles were exposed to various concentrations of ovine LH or FSH. Isolated follicles were obtained from the ovaries of four frogs and sorted into size classes. Since Stage VI and Stage V follicles were the most active in secreting progesterone and testosterone (Figs. 1,2), these stages were used to assessthe effects of LH and FSH on progesterone and testosterone secretion. And since medium-sized follicles were the most active in secreting estradiol (Fig. 3), Stage III and Stage IV follicles were exposed to LH and FSH to test the effects of these gonadotropins on estradiol production. In each experiment, incubation dishes containing 23 follicles (10 Stage VI and 13 Stage V) were treated in duplicate with various concentrations of LH or FSH. After 10 hr the oocytes were scored for maturation and ovulation and the incubation medium was collected and later assayed for progesterone and testosterone. At the concentrations used, FSH had very little effect on the secretion of progesterone or testosterone by Stage VI and Stage V follicles (Figs. 4 and 5). In contrast, LH stimulated significant increases in the secretion of these
n 5 :
CONTROL
q
44
FPH TISSUE
f
6 5 2
12
;
4 VI STAGE
V IV OF FOLLICULAR
III I+11 DEVELOPMENT
FIG. 3. Estradiol-1’i’P concentration (pg/follicle L SEM, n = 8 incubation dishes) in follicular tissue and incubation medium of ovarian follicles incubated in control medium or medium containing a pituitary homogenate (FPH) for 10 hr.
506
DEVELOPMENTAL BIOLOGY
VOLUME 99, 1983
z! 6 E CONCENTRATION
OF LH OR FSH
[gg/ml)
FIG. 4. Progesterone secretion (pg/follicle f SEM; n = 8 incubation dishes) by ovarian follicles (10 Stage VI + 13 Stage V per dish) incubated in medium containing various concentrations of ovine LH or ovine FSH for 10 hr.
two steroids (P < 0.01). The minimum effective dose of LH was 0.5 Kg/ml for progesterone and 0.1 pg/ml for testosterone. When groups of 35 follicles (15 Stage IV and 20 Stage III) were treated with various concentrations of LH or FSH, LH, but not FSH, significantly increased (P < 0.01) estradiol secretion by the Stage IV and Stage III follicles (Fig. 6). The minimum effective dose of LH was 0.5 &g/ml (P < 0.01). In each experiment replicate dishes were also treated with a combination of LH and FSH at 0.5 and 1 Mg/ml and with FPH. Simultaneous exposure of follicles to both LH and FSH had the same effect (P > 0.05) on progesterone, testosterone and estradiol secretion as exposure to LH alone (Table 1). The highest dose of LH (1 pg/ml) was as effective in stimulating testosterone secretion as was FPH, but was not as effective (P < 0.05) as FPH in stimulating progesterone and estradiol secretion. Treatment with FSH or with the two lowest doses of LH did not stimulate meiotic maturation (data not shown). Although maturation was observed in dishes exposed to 0.5 or 1.0 gg LH/ml, FPH treatment resulted in significantly more maturation (P < 0.01) than did the highest dose of LH (Table 1).
O
I 0.01
I
I
0.1 0.5 CONCENTRATION OFLH ORFSH k/ml]
, 1.0
FIG. 5. Testosterone secretion (pg/follicle k SEM; n = 8 incubation dishes) by ovarian follicles (10 Stage VI + 13 Stage V per dish) incubated in medium containing various concentrations of ovine LH or ovine FSH for 10 hr.
It seems clear that Xenopus follicles undergo developmental changes in steroid synthesizing abilities. The smallest ovarian follicles (Stage I and II) appeared incapable of secreting any of the three steroids measured (Figs. l-3). It is possible that these small follicles secreted very low levels of the steroids that were not detected by the assay methods. The lower limits of sensitivity of the assays for progesterone, estradiol, and testosterone were 1.25, 0.5, and 1.25 pg/follicle, respectively. Medium-sized follicles (Stage III and IV) secreted primarily estradiol and its secretion was significantly enhanced by FPH (Fig. 3). Stage IV follicles were the most active in secreting this steroid, but Stage III follicles were equally active if estradiol secretion is expressed
DISCUSSION
Previous reports have indicated that Xenopus ovaries synthesize the three major classes of sex steroids from endogenous precursors and that gonadotropins that stimulate meiotic maturation, ovulation, and follicular growth also increase the production of these steroids (Fortune et al, 1975; Fortune and Tsang, 1981). However, this is the first report of the developmental pattern of steroid production.
20
I 0.01 CONCENTRATION
I 1 ( 0.1 0.5 1.0 OF LH OR FSH [ug/ml]
FIG. 6. Estradiol-17P secretion (pg/follicle + SEM; n = 8 incubation dishes) by ovarian follicles (15 Stage IV + 20 Stage III per dish) incubated in medium containing various concentrations of ovine LH or ovine FSH for 10 hr.
J.
TABLE THE
EFFECTS
SECRETION
GONADOTROPIN OOCYTE MATURATION
(n
= 8 INCUBATION
Steroid Gonadotropin treatment
Progesterone”
secretion
FORTUNE
Ovarian
1
OF SEVERAL
AND
E.
PREPARATIONS OF Xenqpus
ON STEROID FOLLICLES
DISHES) (&follicle
+ SEM)
Testosterone’ 3
6.4 f 0.8
0
13.2 + 2.8
218 f 27
16.7 + 3.0
79 f 16
1 fig/ml 0.5 pg/ml
8.0 + 1.7 5.6 + 1.2
189 k 22 152 + 22
12.6 + 2.2 9.5 + 1.9
46 k 16 28 k 11
+ FSH 1 pg/ml each 0.5 pg/ml each
8.9 f 2.1 5.6 k 1.7
229 k 31 148 k 26
13.3 + 2.8 10.7 f 1.9
53 + 17 21+ 7
Control FPH
1.1 f 0.4
25f
Estradiolb
Percentage maturation zk SEM”
LH
LH
a Progesterone and testosterone were assayed in the medium of incubation dishes containing 10 Stage VI and 13 Stage V follicles. Percentage maturation was assessed at 10 hr and is expressed as a percentage of Stage VI follicles. bEstradiol was assayed in the medium of incubation dishes containing Stage IV and 20 Stage III follicles.
15
on the basis of follicular size. The secretion of estradiol began to taper off at Stage V and became negligible as the follicles reached full size. These results are in agreement with those of Mulner et al. (1978), who found that Stage IV follicles were much more active than Stage VI in aromatizing radioactive androstenedione to estradiol. The injection of estradiol stimulates yolk protein synthesis by the livers of male or female frogs (Wallace and Bergink, 1974). The present results indicate that endogenous estradiol secretion begins when follicles first become vitellogenic (Stage III), is greatest when yolk accumulation is most rapid (Stage IV), and tapers off as vitellogenesis is completed (Stage V). Therefore, it seems likely that gonadotropins applied in vivo promote vitellogenesis through the mediation of estradiol synthesized by medium-sized follicles. However, gonadotropins also promote uptake of yolk proteins by vitellogenic oocytes and this effect does not appear to be mediated by estradiol (Wallace and Bergink, 1974). Progesterone was secreted primarily by large Stage V and VI follicles and its secretion was increased by FPH (Fig. 1). Exogenous progesterone stimulates meiotic maturation (Schuetz, 1967a,b; Subtelny et al, 1968; Smith and Ecker, 1971; Morrill and Bloch, 1977) and inhibitors of progesterone synthesis, such as cyanoketone, aminoglutethimide, and estradiol, also block maturation (Snyder and Schuetz, 19’73;Fortune, unpublished results; Spiegel et aZ., 1978). Therefore, it seems likely that the progesterone produced by Stage VI follicles is important for oocyte maturation. The finding that Stage VI follicles are responsible for a substantial portion of ovarian progesterone production confirms an earlier report by For-
Steroids
in
Xenopus
507
tune et al. (1975) and is in accord with the work of Mulner et al. (1978), who showed that Stage VI follicles converted more radioactive pregnenolone to progesterone than did Stage IV follicles. In the present experiments progesterone secretion by Stage V follicles did not differ significantly from secretion by Stage VI follicles, yet very few Stage V follicles underwent meiotic maturation. Thus, it appears that the follicle develops the capacity to produce progesterone before the oocyte develops the ability to respond to it by maturing. Marot et al. (1977) reported that pretreatment with small doses of progesterone facilitated meiotic maturation in response to larger doses. Therefore, it is possible that the progesterone synthesized by Stage V follicles produces the same sort of priming effect in vivo. We had been surprised to find that Xenopus ovaries contain and synthesize much more testosterone than progesterone or estradiol (Fortune and Tsang, 1981). Although the testosterone antibody used in the radioimmunoassay cross-reacts 69% with the nonaromatizable androgen DHT, chromatographic separation of testosterone and DHT showed that most of the androgen measured by the assay consisted of testosterone (Fortune and Tsang, 1981). Thus, it seemed possible that testosterone might function simply as a precursor for estradiol synthesis. The current experiments show that testosterone is secreted primarily by large Stage V and VI follicles, rather than the medium-sized follicles that are most active in estradiol production (Fig. 2). Since the developmental pattern of testosterone secretion followed that of progesterone and since exogenous testosterone stimulates maturation (Morrill and Bloch, 1977; Schuetz, 1967a; Smith and Ecker, 1971), it is possible that progesterone may trigger maturation by serving as a precursor for synthesis of testosterone which would then stimulate maturation. Schatz and Morrill (1979) reported that Rana pipiens follicles rapidly metabolized radiolabelled progesterone, but not progesterone derived from radiolabelled acetate. Although these data seem inconsistent with the finding that Xenopua follicles produce considerable amounts of testosterone, it is possible that progesterone synthesized from exogenous precursors is compartmentalized and metabolized differently from progesterone synthesized from endogenous precursors. Species differences do not seem likely since we have recently found that Rana pipiens follicles synthesize considerable quantities of testosterone (Fortune and Wigglesworth, unpublished data). Tissue concentrations of estradiol and testosterone mimicked the pattern of secretion of these steroids at different developmental stages and in response to FPH (Figs. 2 and 3, upper panels). In contrast, progesterone accumulation in the medium showed more dramatic changes as a result of development and FPH treatment
508
DEVELOPMENTAL
BIOLOGY
than did concentrations in follicular tissue. These results suggest that measurement of steroids in the medium of incubated Xm follicles provides an adequate assessment of follicular steroidogenesis. This finding is of some practical importance because measurements of steroids in medium are easier, quicker, and more reliable than measurements on tissue which require solvent extraction. These experiments were not designed to localize the follicular sites of synthesis of the three steroids. The findings of Masui (1967), Smith et ak, (1968), Mulner et al (1978), and Thibier-Fouchet et al. (1976) have implicated the follicle cells in progesterone and estradiol synthesis. However, since oocytes can also metabolize exogenous steroid precursors (Reynhout and Smith, 1973; Fouchet et al, 1975; Thibier-Fouchet et c& 1976; Sanyal and Sibre, 1973), it is not clear that the follicle cells are the primary site of steroid synthesis. The data have been expressed here per follicle. It might be more appropriate to express them on the basis of follicular volume or surface area, since these parameters change over the course of development. In the absence of information on the follicular site of synthesis of each steroid, it is difficult to select the proper correction factor. However, it is clear that the changes that occur in steroid secretion during follicular development do not simply mirror the increases in follicular size and volume. This is evident for estradiol, since large follicles produce less estradiol than medium-sized follicles. In contrast, increases in the ability of follicles to produce progesterone in response to FPH roughly parallel increases in follicular surface area. During development from Stages I-II to Stage VI progesterone production increased about 29X while follicular surface area increased about 17X. On the other hand, FPH-stimulated testosterone production increased about 170~ during development, while surface area increased 17~ and volume 72X. During development from Stage IV to Stage V follicular testosterone production increased 11X, while surface area increased about twofold and volume 2.6X. Thus, changes in estradiol and testosterone production appear to be developmental changes in the regulation of the steroidogenic pathway. Ovine LH, but not FSH, increased the secretion of progesterone, testosterone, and estradiol by Xenop~.~ follicles and promoted meiotic maturation in a dosedependent fashion (Figs. 4-6; Table 1). The effect of LH was not as great as that of Xenopus gonadotropins (FPH) except on testosterone secretion. However, it is possible that higher doses of LH would have produced a response equivalent to the response to FPH, since the highest dose of LH was maximal only for progesterone secretion. Recent experiments with mammalian follicles and follicle cells have determined that. while both FSH and
VOLUME
99, 1983
LH promote steroid biosynthesis, the two gonadotropins act at different points in the steroidogenic pathway (see review by Armstrong and Dorrington, 1977). LH stimulates the production of androgens (Fortune and Armstrong, 1977), while FSH promotes the aromatization of androgens to estrogen (Moon et al, 1975; Dorrington et al, 1975; Fortune and Armstrong, 1978; Erickson and Hsueh, 1978; Fortune and Eppig, 1979). In terms of this mammalian model, ovine LH seems to exert both LHlike and FSH-like effects on Xenopu.s steroidogenesis. Thus, it is not clear what role FSH plays in Xenopus ovarian function. Since steroids were measured only in the incubation medium in the experiments with FSH and LH, it is possible that FSH promoted steroid synthesis in the tissue, but not secretion into the medium. However, later measurement of steroids in control tissue and tissue treated with the highest dose of FSH provided no evidence for an FSH effect on steroid synthesis (data not shown). Elucidation of the relative roles of the FSH and LH in X&opus must await the availability of purified Xenopus gonadotropins. In conclusion, Xen0pu-s ovarian follicles appear to pass through three developmental stages in terms of steroid secretion: small follicles secrete very little progesterone, testosterone or estradiol; medium-sized, vitellogenic follicles are active in producing estradiol; and large follicles shift to production of progesterone and large amounts of testosterone. The role of testosterone in Xenopus ovarian function is an intriguing question that remains to be elucidated. I am grateful to S. Vincent, Y. Eisner, and P. Tsang for excellent technical assistance, to L. E. Reichert (Albany Medical College, Albany, N. Y.) for providing purified ovine LH and FSH, and to H. Dormady and P. Koos for typing this manuscript. REFERENCES ARMSTRONG, D. T., and DORRINGTON, J. H. (1977). Estrogen biosynthesis in the ovaries and testes. Advan Sex Hcnm Res. 3, 217-25’7. CHIEFFI, G., and LUPO, C. (1963). Identification of sex hormones in the ovarian extracts of Torpedo marmorata and B$o vulgaris. Ga Camp. Endocrinol 3. 149-152. DORRINGTON, J. H., MOON, Y. S., and ARMSTRONG, D. T. (1975). Estradiol170 biosynthesis in cultured granulosa cells from hypophysectomiwd immature rats; stimulation by follicle-stimulating hormone. Endocrinology 97, 13281331. DUMONT, J. N. (1972). Oogenesis in Xenqpus .!a.euti (Daudin). I. Stages of oocyte development in laboratory-maintained animals. J. lMorphd 136, 153-180. ERICKSON, G. F., and HSUEH, A. J. W. (1978). Stimulation of aromatase activity by follicle stimulating hormone in rat granulosa cells in viva and in vitro. Emhcrinology 102, 1275-1282. FORTUNE, J. E., and ARMSTRONG, D. T. (1977). Androgen production by theta and granulosa isolated from proestrous rat follicles. Endocrinology 100, 1341-1347. FORTUNE, J. E., and ARMSTRONG, D. T. (1978). Hormonal control of 17@estradiol biosynthesis in proestrous rat follicles: Estradiol pro-
Ovarian
J. E. FORTUNE duction 235.
by isolated
theca
versvs
granulosa.
Endocrinology
102,227-
J. E., and EPPIG, J. J. (1979). Effects of gonadotropins on steroid secretion by infantile and juvenile mouse ovaries in vitro. Endocrinology 105, ‘760-768. FORTUNE, J. E., and TSANG, P. C. (1981). Production of androgen and estradiol-170 by Xenopus ovaries treated with gonadotropins in vitro. Gen Camp. EndocGwL 43, 234-242. FORTUNE, J. E., CONCANNON, P. W., and HANSEL, W. (1975). Ovarian progesterone levels during in vitro oocyte maturation and ovulation in Xenopus laevis. BioL Reprod 13, 561-567. FOUCHET, C., SERRES, C., BELL& R., and OZON, R. (1975). Mechanism of action of progesterone on amphibian oocytes. Uptake and metabolism of progesterone by isolated oocytes of Pleurodeles waltlii and Xenopus laeuis. Camp. Biochem Physiol 52B, 205-210. GAY, V. L., and KERLAN, J. T. (1978). Serum LH and FSH following passive immunization against circulating testosterone in the intact male rat and in orchidectomized rats bearing subcutaneous silastic implants of testosterone. Arch. AndroL 1, 239-240. KORENMAN, S. G., STEVENS, R. H., CARPENTER, L. A., ROBB, M., NISWENDER, G. D., and SHERMAN, B. M. (1974). Estradiol radioimmunoassay without chromatography: Procedure, validation and normal values. J. Clin Endocrinol MetaboL 38, 718-720. LICHT, P., and CREWS, D. (1976). Gonadotropin stimulation of in vitro progesterone production in reptilian and amphibian ovaries. Gev~ Camp. EndocrinoL 29, 141-151.
FORTUNE,
LICHT, P., and PAPKOFF, H. (1974). Separation of two distinct gonadotropins from the pituitary gland of the bullfrog Rana catesbeiana. Endocrinology 94, 1587-1595. LICHT, P., and PAPKOFF, H. (1976). Species specificity in the response of an in vitro amphibian (Xm &is) ovulation assay to mammalian luteinizing hormones. Gen Camp. Endocrinol 29, 552-555. MAROT, J., BELL& R., and OZON, R. (1977). In vitro facilitation of Xelzqpus oocyte maturation by subthreshold doses of progesterone. Dev. Biol 59, 91-95. MASUI, Y. (1967). Relative roles of the pituitary, follicle cells and progesterone in the induction of oocyte maturation in Rana p&em. J. Exp. Zoo1 166, 365-376. MOON, Y. S., DORRINGTON, J. H., and ARMSTRONG, D. T. (1975). Stimulatory action of follicle-stimulating hormone on estradiol-17p secretion by hypophysectomized rat ovaries in organ culture. Endcc7%wlogy 97, 244-247. MORRILL, G. A., and BLOCH, E. (1977). Structure-function relationships of various steroids relative to induction of nuclear breakdown and ovulation in isolated amphibian oocytes. J. Ster&a’Bioc~ 8,133139. MULNER,
O., THIBIER,
C., and OZON,
R. (1978).
Steroid
biosynthesis
by
Steroids
in
Xenopus
509
ovarian follicles of Xenopus lamis in vitro during oogenesis. Gen. Cmp. EndocrinoL 34, 287-295. PAPKOFF, H., FARMER, S. W., and LICHT, P. (1976). Isolation and characterization of luteinizing hormone from amphibian (Rana Catesbeiana) pituitaries. Life Sti 18, 245-250. REDSHAW, M. R., and NICHOLLS, T. J. (1971). Oestrogen biosynthesis by ovarian tissue of the South African clawed toad, Xenopus 1aevi.s Daudin. Gen. Cmp. Endocrinol 16, 85-96. REYNHOUT, J. K., and SMITH, L. D. (1973). Evidence for steroid metabolism during the in vitro induction of maturation in oocytes of Rana pipiens. Dew. BioL 30.392-402. SANYAL, M. K., and SIBRE, E. R. (1973). Breakdown of germinal vesicle of frog ooeytes with 5a-reduced products of progesterone in vitro. Proc. Sac. Exp. BioL Med. 144,483-486. SCHATZ, F., and MORRILL, G. A. (1979). Studies on the relative roles of pituitary and progesterone in the induction of meiotic maturation in the amphibian oocyte. Dz#iientiatim 13, 89-99. SCHUETZ, A. W. (1967a). Effect of steroids on the germinal vesicle of oocytes of the frog (Rana pipiens) in vitro. Proc. Sot. Exp. BioL Med. 124, 1307-1310. SCHUETZ, A. W. (1976b). Action of hormones on germinal vesicle breakdown in frog (Rana pipiens) oocytes. J. Exp. ZooL 166, 347354. SMITH, L. D., and ECKER, R. E. (1971). The interactions of steroids with Rana pipiens oocytes in the induction of maturation. Dev. BioL 25, 232-247. SMITH, L. D., ECKER, R. E. and SUBTELNY, S. (1968). In vitro induction of physiological maturation in Rana pipiens oocytes removed from their ovarian follicles. Deu. BioL 17, 627-643. SNYDER, B. W., and SCHUETZ, A. W. (1973). In vitro evidence of steroidogenesis in the amphibian (Ram pipiens) ovarian follicle and its relationship to meiotic maturation and ovulation. J. Exp. ZooL 183, 333-342. SPIEGEL, J., JONES, E., and SNYDER, B. W. (1978). Estradiol-170 interference with meiotic maturation in Ram pipiens ovarian follicles: Evidence for inhibition of 3&hydroxysteroid dehydrogenase. J. Exp. ZooL 204, 187-192. SUBTELNY, S., SMITH, L. D., and ECKER, R. E. (1968). Maturation of ovarian frog eggs without ovulation. J. Exp. ZooL 168, 39-47. THIBIER-FOUCHET, C., MLJLNER, O., and OZON, R. (1976). Progesterone biosynthesis and metabolism by ovarian follicles and isolated oocytes of Xenopus laevis. BioL Reprod 14, 317-326. WALLACE, R. A., and BERGINK, G. W. (1974). Amphibian vitellogenin: properties, hormonal regulation of hepatic synthesis and ovarian uptake, and conversion to yolk proteins. Amer. ZooL 14.1159-1175. WALLACE, R. A., JARED, D. W., DUMONT, J. N., and SEGA, M. W. (1973). Protein incorporation by isolated amphibian oocytes. III. Optimum incubation conditions. J. Exp. ZooL 184, 321-334.
BIOLOGY
99,502-509
(1983)
Steroid Production by Xenopus Ovarian Follicles at Different Developmental Stages’ J. E. FORTUNE’ Section of Physiology in the Division of Biological Sciences and Department of Physidogzl in the CoUege of Vete&mry Medicine, Cornell University, Ithaca, New York 14858 Received September 28, 1979; accepted in revised fm
May 16, 1986
Xaqpus ovarian follicles at different developmental stages were compared with respect to their capacity to produce and secrete steroids and to respond to gonadotropic hormones with changes in steroid production. Individual follicles were obtained by treating ovaries with collagenase and were incubated for 10 hr in incubation medium alone or in medium containing a Xenopus pituitary homogenate (FPH, 0.04 pituitary/ml). At 10 hr oocytes were scored for meiotic maturation and ovulation and incubation media and follicles were collected separately and frozen. Both tissue and media were measured for progesterone, testosterone, and estradiol by radioimmunoassay. The smallest follicles (Dumont Stages I and II) secreted very low or nondetectable levels of these three steroids in both the presence and absence of FPH. Medium-sized follicles (Stages III and IV) were active in secreting estradiol and FPH stimulated a twofold increase in the accumulation of estradiol in the incubation medium. In follicles larger than Stage IV, estradiol secretion decreased as follicular size increased. In contrast, large follicles (Stages V and VI) secreted much more testosterone and progesterone than medium-sized follicles and FPH significantly increased the accumulation of these two steroids. Overall, follicular secretion of testosterone was much greater than secretion of progesterone or estradiol. The concentration of the three steroids in the ovarian follicles at 10 hr of incubation followed patterns that were similar in general to the patterns of accumulation of steroids in the incubation medium. Ovine LH, but not FSH, was effective in stimulating secretion of the three steroids and in triggering meiotic maturation of oocytes in Stage VI follicles. Although many of the morphological and biochemical events of oogenesis and folliculogenesis in Xenqt~ have been described, this is the first report of the developmental pattern of steroid production. The data indicate that growth and development of Xenopua follicles are accompanied by shifts in steroidogenic abilities, with estradiol produced by medium-sized follicles and androgen and progesterone secreted by large follicles.
Despite the abundant evidence that steroids applied exogenously can influence oocyte growth and meiotic maturation, there have been few investigations of endogenous ovarian levels of steroids or the effects of gonadotropins on steroid production by anuran ovaries. Chieffi and Lupo (1963) detected progesterone, estradiol178, estriol, and estrone in the ovaries of BT&Ovulgaris. Fortune et al (1975) found that Xenopus ovaries contain progesterone and that gonadotropin preparations that trigger maturation and ovulation also significantly increase ovarian concentrations of progesterone. Licht and Crews (1976) reported that Rana luteinizing hormone greatly increased the secretion of progesterone by Ram ovaries incubated in vitro. These findings, coupled with reports that inhibitors that block progesterone synthesis also inhibit maturation and ovulation (Snyder and Schuetz, 1973; Fortune, unpublished observations), lend support to the hypothesis that progesterone is a necessary intermediate in the induction of maturation and ovulation. Several groups have reported that anuran ovaries and oocytes can convert exogenous radiolabeled steroid pre-
INTRODUCTION
It has been known for many years that exogenous application of gonadotropic or steroid hormones can exert effects on several aspects of ovarian development in amphibians. Treatment with gonadotropins or steroids can stimulate meiotic maturation and/or ovulation of full-grown anuran oocytes. Some evidence suggests that oocyte maturation is triggered via a two-step mechanism in which a gonadotropic hormone(s) stimulates follicle cells to produce a steroid which acts directly on the oocyte to cause meiotic maturation (Masui, 1967; Smith et CAL,1968). It is well established that exogenous estradiol can stimulate the synthesis of yolk proteins by the liver of male and female amphibians (see review by Wallace and Bergink, 1974). It has been suggested that gonadotropic hormones increase ovarian production of estrogen which in turn stimulates vitellogenesis. 1 This work was supported by National Research Service Award 5 F32 HD05049. *To whom all correspondence should be addressed: 823 Veterinary Research Tower, Cornell University, Ithaca, New York 14853. 502 0012-1606/83 $3.00 Copyright All rights
0 1983 by Academic Press, Inc. of reproduction in any form reserved.
J. E. FORTUNE
Ovarian
cursors to estrogens and androgens, as well as prowstins (Redshaw and Nicholls, 1971; Reynhout and Smith, 197% Sanyal and Sibre, 1973; Fouchet et al., 1975; ThibierFouchet et aL, 1976). More recently Fortune and Tsang (1981) found that Xenoms ovaries contain and secrete endogenous estradiol-17P and testosterone. Gonadotropin treatment in vitro stimulated 2- and lo-fold increases in ovarian estradiol and testosterone concentrations, respectively. Testosterone concentrations in gonadotropin-treated ovaries were much greater than the concentrations of progesterone or estradiol. This finding raised the question of the role of androgens in Xenopus ovaries. The current experiments were designed to determine the developmental pattern of steroid production by Xenopus ovarian follicles. When Xenopus are maintained in the laboratory all stages of follicular development are usually present within the ovary simultaneously and therefore, follicles at various developmental stages can be isolated from the ovaries of one frog. Follicles at five different stages have been isolated and incubated with gonadotropins and the secretion of progesterone, testosterone, and estradiol has been measured by radioimmunoassay. MATERIALS
AND
METHODS
1. Animals and incubations. Sexually mature Xenopus laevis females were obtained from the South African Snake Farm (Fish Hoek, South Africa), maintained in tap water, and fed a diet of chopped beef heart and liver. Frogs were killed by decapitation and the pituitary was removed and homogenized in 2 ml O-R2 solution (Wallace et al., 1973). The ovaries were removed, rinsed with O-R2 solution, weighed, and cut into small pieces of about 30 mg. Individual follicles were obtained from the ovarian pieces by dissolving the interfollicular connections with collagenase. About 2 g of ovarian pieces were placed in each of several 25-ml Erlenmeyer flasks containing approximately 10 ml of a 0.15% solution of collagenase (Sigma Chemical Co.) in O-R2 solution. The flasks were placed in a Dubnoff metabolic shaker at 26°C and shaken at about 80 strokes/min for 20-60 min. This procedure freed many individual follicles from the ovarian pieces, but did not disrupt the association between oocytes and their surrounding follicle cells. The time needed for dissociation varied considerably among animals. In some experiments, pieces that had not completely dissociated into follicles were collected and exposed to collagenase a second time. After dissociation the tissue was rinsed several times with O-R2 solution and individual follicles that appeared intact and healthy were sorted into size classes under a dissecting microscope. Oocytes that had
Sikm3id.s in Xen0pu.s
503
been separated from their follicular tissue or follicles that appeared damaged by collagenase treatment were discarded. Follicles were classified according to the criteria of Dumont (1972). Follicles at Dumont Stages I and II are small (50-300 and 300-450 pm in diameter, respectively) and previtellogenic. Vitellogenesis begins at Stage III (450-600 pm) and continues during Stage IV (600-1000 pm), when animal and vegetal hemispheres differentiate. Vitellogenesis is completed during Stage V (1000-1200 pm) and Stage VI follicles are full-grown (1200-1300 pm) and capable of undergoing meiotic maturation and ovulation. Follicles were incubated at room temperature in the wells of plastic cluster dishes (Costar, Cambridge, Mass.) in 500 ~1 O-R2 solution containing polyvinylpyrrolidone (PVP). PVP reduced the tendency of follicles treated with collagenase to stick to the incubation dishes. In initial experiments incubation wells contained 20 Stage VI follicles, 25 Stage V, 30 Stage IV, 40 Stage II, or 50 Stage I-II. Later experiments focused on certain of these stages; the number of follicles used in these experiments will be described in the Results section. Follicles were incubated in O-R2 solution alone (controls) or O-R2 solution containing the frog pituitary homogenate (FPH), highly purified ovine luteinizing hormone (LH; LER-1056-C2), or highly purified ovine follicle stimulating hormone (FSH; LER-1976-A2). The dose of FPH used throughout these experiments (0.04 pituitary/ ml) stimulates a maximal response in terms of percentage of Stage VI oocytes that ovulate and exhibit meiotic maturation (Fortune et aL, 1975). After 10 hr of incubation follicles were examined under a dissecting microscope for meiotic maturation and ovulation. A lo-hr incubation period was chosen because ovulation and maturation usually occur within 10 hr in vitro and because our previous results indicated that when whole ovarian pieces, containing all stages of follicular development, were incubated with FPH, medium and tissue concentrations of estradiol, progesterone, and testosterone were either increasing slightly or had reached a plateau by 10 hr (Fortune and Tsang, 1981; Fortune et aL, 1975). Maturation was scored by the presence of Roux’ spot, a white spot that appears at the animal pole during maturation. Ovulation was scored by counting the number of oocytes that were no longer enclosed within the follicular wall. After dishes were scored for maturation and ovulation, the medium and tissue were collected separately from each incubation well and frozen. 2. Steroid assays. The ovarian follicles collected from the incubation wells were homogenized in 1 ml O-R2 solution and the homogenates were extracted with diethyl ether and methanol as described for Xenopus ovarian pieces (Fortune and Tsang, 1981). Progesterone,
504
DEVELOPMENTAL BIOLOGY
testosterone, and estradiol were measured in aliquots of these extracts and in unextracted aliquots of incubation medium by radioimmunoassays (RIA) that have been described previously (Fortune and Armstrong, 1978; Fortune and Eppig, 1979). The estradiol antiserum is highly specific for estradiol and its low cross-reactivities with other steroids have been reported previously (Korenman et aL, 1974). The progesterone antiserum cross reacts 11.64% with 5/3-dihydroprogesterone, ~10% with other progestins tested, and ~0.1% with the androgens and estrogens tested. The testosterone antiserum (Gay and Kerlan, 1978) is not specific for testosterone and cross-reacts with 5a-dihydrotestosterone (DHT; 69%), 3cr-androstanediol (14%), and 3@-androstanediol (22%); cross-reactivities with estrogens, progestins, corticoids, and other androgens tested were 1% or less. In previous experiments testosterone was isolated chromatographically from extracts of ovarian pieces (containing all stages of follicular development) and incubation medium (Fortune and Tsang, 1981). Since almost all of the androgen measured by the testosterone antiserum before chromatographic separation was recovered in the testosterone fraction after chromatography, extracts were not subjected to chromatographic purification in the current experiments and the steroid measured by the radioimmunoassay is referred to as “testosterone.” Accumulation of steroids in the incubation medium (or net secretion) and concentration of steroids in the tissue are both expressed per follicle incubated. Significant differences were determined by analysis of variance, Scheffe’s method for comparisons of means, and Duncan’s multiple range test. RESULTS
I. Production of Progesterone, Testosterone, and Estradiol-l7@ by FoUicles at D&m& Developmental Stages In the first series of experiments follicles were sorted into five different size classes and incubated for 10 hr in the presence or absence of frog pituitary homogenate (FPH; 0.04 pituitary/ml). Each treatment was applied in duplicate to tissue from four frogs. After 10 hr follicles were scored for maturation and ovulation and the incubation medium and tissue were collected, frozen, and later measured for steroids by RIA. The lower panels of Figs. l-3 show the accumulation of progesterone, testosterone, and estradiol in the medium of follicles of different sizes and the upper panels depict steroid concentrations in the follicles. The secretion of testosterone and progesterone followed similar developmental patterns (Figs. 1 and 2, lower panels). In control dishes of small (Stages I-II) or medium-sized follicles (Stages III and IV) accumulation of these two steroids was either very low or below
VOLUME 99. 1983
z.z : 12C
n
CONTROL
q
TISSUE
+I
$8 =.0 :
z
FPH
4
STAGE
OF FOLLICULAR
DEVELOPMENT
FIG. 1. Progesterone concentration (pg/follicle + SEM, ?z = 8 incubation dishes) in follicular tissue and incubation medium of ovarian follicles incubated in control medium or medium containing a pituitary homogenate (FPH) for 10 hr.
the sensitivity of the assays. Only follicles that had reached Stage V or VI secreted substantial quantities of progesterone and testosterone and Stage VI follicles secreted slightly more of these two steroids than did Stage V. Treatment with FPH increased the secretion of testosterone and progesterone by Stage VI and Stage V follicles by approximately 400% (P < 0.001) and by Stage IV follicles by approximately 1500% (P < 0.001). Both control and FPH-treated follicles secreted about 10 times more testosterone than progesterone. The concentrations of testosterone in follicular tissue (Fig. 2, upper panel) were similar both in pattern and amount to the concentrations in the culture medium. In contrast, progesterone secretion (i.e., accumulation in the medium) increased much more dramatically with follicular development than did the tissue concentrations of progesterone (Fig. 1, upper panel) and FPH had a far greater stimulatory effect on secretion of progesterone than on follicular concentrations. The developmental pattern of estradiol secretion was quite different from that of progesterone and testosterone. Medium-sized follicles (Stages III and IV) were most active in secreting estradiol, while smaller or larger follicles produced much lower quantities (Fig. 3, lower panel). FPH stimulated a twofold increase (P < 0.01) in estradiol secretion by Stages V, IV, and III, but
Ovarian
J. E. FORTUNE
s
FJ +I 320 al
n
CONTROL
q
FPH TISSUE
MEDIUM
F
STAGE
OF FOLLICULAR
DEVELOPMENT
FIG. 2. Testosterone concentration (pg/follicle f SEM, cubation dishes) in follicular tissue and incubation medium follicles incubated in control medium or medium containing homogenate (FPH) for 10 hr.
n = 8 inof ovarian a pituitary
the gonadotropin treatment only slightly increased (P < 0.05) estradiol secretion by Stage VI or Stages III. At most developmental stages the amounts of estradiol in follicular tissue were greater than the quantities secreted into the medium, but the overall developmental pattern and the responses to FPH were very similar (Fig. 3, upper panel). In these experiments FPH treatment also stimulated the maturation and ovulation of Stage VI, and occasionally Stage V, follicles. The percentage maturation for Stage VI oocytes ranged from 10 to 98% for the four experiments and the mean was 51.2 f 12.5% (n = 8 incubation wells with 20 follicles/well). The mean percentage maturation was 8.5 + 2.2% for Stage V oocytes (n = 8 incubation wells with 25 follicles/well). FPH triggered ovulation in only one experiment in which 95% of Stage VI follicles and 12% of Stage V follicles ovulated.
2. The Effects of LH and FSH on Xenopus Ovarian FoUicle.s. Since an LH-like and an FSH-like gonadotropin have been isolated from pituitaries of the bullfrog Rana cutesbeiana (Licht and Papkoff, 1974; Papkoff et al, 1976), it was of interest to test the effects of LH and FSH on steroidogenesis by Xewpus follicles. Purified ovine LH and FSH were used because purified amphibian gonadotropins were not available and ovine LH was reported to be active in stimulating ovulation and meiotic mat-
Steroidsin
505
Xenopua
uration in Xenqs (Licht and Papkoff, 1976; Mulner et aZ., 1978). Figures 4-6 show steroid secretion when follicles were exposed to various concentrations of ovine LH or FSH. Isolated follicles were obtained from the ovaries of four frogs and sorted into size classes. Since Stage VI and Stage V follicles were the most active in secreting progesterone and testosterone (Figs. 1,2), these stages were used to assessthe effects of LH and FSH on progesterone and testosterone secretion. And since medium-sized follicles were the most active in secreting estradiol (Fig. 3), Stage III and Stage IV follicles were exposed to LH and FSH to test the effects of these gonadotropins on estradiol production. In each experiment, incubation dishes containing 23 follicles (10 Stage VI and 13 Stage V) were treated in duplicate with various concentrations of LH or FSH. After 10 hr the oocytes were scored for maturation and ovulation and the incubation medium was collected and later assayed for progesterone and testosterone. At the concentrations used, FSH had very little effect on the secretion of progesterone or testosterone by Stage VI and Stage V follicles (Figs. 4 and 5). In contrast, LH stimulated significant increases in the secretion of these
n 5 :
CONTROL
q
44
FPH TISSUE
f
6 5 2
12
;
4 VI STAGE
V IV OF FOLLICULAR
III I+11 DEVELOPMENT
FIG. 3. Estradiol-1’i’P concentration (pg/follicle L SEM, n = 8 incubation dishes) in follicular tissue and incubation medium of ovarian follicles incubated in control medium or medium containing a pituitary homogenate (FPH) for 10 hr.
506
DEVELOPMENTAL BIOLOGY
VOLUME 99, 1983
z! 6 E CONCENTRATION
OF LH OR FSH
[gg/ml)
FIG. 4. Progesterone secretion (pg/follicle f SEM; n = 8 incubation dishes) by ovarian follicles (10 Stage VI + 13 Stage V per dish) incubated in medium containing various concentrations of ovine LH or ovine FSH for 10 hr.
two steroids (P < 0.01). The minimum effective dose of LH was 0.5 Kg/ml for progesterone and 0.1 pg/ml for testosterone. When groups of 35 follicles (15 Stage IV and 20 Stage III) were treated with various concentrations of LH or FSH, LH, but not FSH, significantly increased (P < 0.01) estradiol secretion by the Stage IV and Stage III follicles (Fig. 6). The minimum effective dose of LH was 0.5 &g/ml (P < 0.01). In each experiment replicate dishes were also treated with a combination of LH and FSH at 0.5 and 1 Mg/ml and with FPH. Simultaneous exposure of follicles to both LH and FSH had the same effect (P > 0.05) on progesterone, testosterone and estradiol secretion as exposure to LH alone (Table 1). The highest dose of LH (1 pg/ml) was as effective in stimulating testosterone secretion as was FPH, but was not as effective (P < 0.05) as FPH in stimulating progesterone and estradiol secretion. Treatment with FSH or with the two lowest doses of LH did not stimulate meiotic maturation (data not shown). Although maturation was observed in dishes exposed to 0.5 or 1.0 gg LH/ml, FPH treatment resulted in significantly more maturation (P < 0.01) than did the highest dose of LH (Table 1).
O
I 0.01
I
I
0.1 0.5 CONCENTRATION OFLH ORFSH k/ml]
, 1.0
FIG. 5. Testosterone secretion (pg/follicle k SEM; n = 8 incubation dishes) by ovarian follicles (10 Stage VI + 13 Stage V per dish) incubated in medium containing various concentrations of ovine LH or ovine FSH for 10 hr.
It seems clear that Xenopus follicles undergo developmental changes in steroid synthesizing abilities. The smallest ovarian follicles (Stage I and II) appeared incapable of secreting any of the three steroids measured (Figs. l-3). It is possible that these small follicles secreted very low levels of the steroids that were not detected by the assay methods. The lower limits of sensitivity of the assays for progesterone, estradiol, and testosterone were 1.25, 0.5, and 1.25 pg/follicle, respectively. Medium-sized follicles (Stage III and IV) secreted primarily estradiol and its secretion was significantly enhanced by FPH (Fig. 3). Stage IV follicles were the most active in secreting this steroid, but Stage III follicles were equally active if estradiol secretion is expressed
DISCUSSION
Previous reports have indicated that Xenopus ovaries synthesize the three major classes of sex steroids from endogenous precursors and that gonadotropins that stimulate meiotic maturation, ovulation, and follicular growth also increase the production of these steroids (Fortune et al, 1975; Fortune and Tsang, 1981). However, this is the first report of the developmental pattern of steroid production.
20
I 0.01 CONCENTRATION
I 1 ( 0.1 0.5 1.0 OF LH OR FSH [ug/ml]
FIG. 6. Estradiol-17P secretion (pg/follicle + SEM; n = 8 incubation dishes) by ovarian follicles (15 Stage IV + 20 Stage III per dish) incubated in medium containing various concentrations of ovine LH or ovine FSH for 10 hr.
J.
TABLE THE
EFFECTS
SECRETION
GONADOTROPIN OOCYTE MATURATION
(n
= 8 INCUBATION
Steroid Gonadotropin treatment
Progesterone”
secretion
FORTUNE
Ovarian
1
OF SEVERAL
AND
E.
PREPARATIONS OF Xenqpus
ON STEROID FOLLICLES
DISHES) (&follicle
+ SEM)
Testosterone’ 3
6.4 f 0.8
0
13.2 + 2.8
218 f 27
16.7 + 3.0
79 f 16
1 fig/ml 0.5 pg/ml
8.0 + 1.7 5.6 + 1.2
189 k 22 152 + 22
12.6 + 2.2 9.5 + 1.9
46 k 16 28 k 11
+ FSH 1 pg/ml each 0.5 pg/ml each
8.9 f 2.1 5.6 k 1.7
229 k 31 148 k 26
13.3 + 2.8 10.7 f 1.9
53 + 17 21+ 7
Control FPH
1.1 f 0.4
25f
Estradiolb
Percentage maturation zk SEM”
LH
LH
a Progesterone and testosterone were assayed in the medium of incubation dishes containing 10 Stage VI and 13 Stage V follicles. Percentage maturation was assessed at 10 hr and is expressed as a percentage of Stage VI follicles. bEstradiol was assayed in the medium of incubation dishes containing Stage IV and 20 Stage III follicles.
15
on the basis of follicular size. The secretion of estradiol began to taper off at Stage V and became negligible as the follicles reached full size. These results are in agreement with those of Mulner et al. (1978), who found that Stage IV follicles were much more active than Stage VI in aromatizing radioactive androstenedione to estradiol. The injection of estradiol stimulates yolk protein synthesis by the livers of male or female frogs (Wallace and Bergink, 1974). The present results indicate that endogenous estradiol secretion begins when follicles first become vitellogenic (Stage III), is greatest when yolk accumulation is most rapid (Stage IV), and tapers off as vitellogenesis is completed (Stage V). Therefore, it seems likely that gonadotropins applied in vivo promote vitellogenesis through the mediation of estradiol synthesized by medium-sized follicles. However, gonadotropins also promote uptake of yolk proteins by vitellogenic oocytes and this effect does not appear to be mediated by estradiol (Wallace and Bergink, 1974). Progesterone was secreted primarily by large Stage V and VI follicles and its secretion was increased by FPH (Fig. 1). Exogenous progesterone stimulates meiotic maturation (Schuetz, 1967a,b; Subtelny et al, 1968; Smith and Ecker, 1971; Morrill and Bloch, 1977) and inhibitors of progesterone synthesis, such as cyanoketone, aminoglutethimide, and estradiol, also block maturation (Snyder and Schuetz, 19’73;Fortune, unpublished results; Spiegel et aZ., 1978). Therefore, it seems likely that the progesterone produced by Stage VI follicles is important for oocyte maturation. The finding that Stage VI follicles are responsible for a substantial portion of ovarian progesterone production confirms an earlier report by For-
Steroids
in
Xenopus
507
tune et al. (1975) and is in accord with the work of Mulner et al. (1978), who showed that Stage VI follicles converted more radioactive pregnenolone to progesterone than did Stage IV follicles. In the present experiments progesterone secretion by Stage V follicles did not differ significantly from secretion by Stage VI follicles, yet very few Stage V follicles underwent meiotic maturation. Thus, it appears that the follicle develops the capacity to produce progesterone before the oocyte develops the ability to respond to it by maturing. Marot et al. (1977) reported that pretreatment with small doses of progesterone facilitated meiotic maturation in response to larger doses. Therefore, it is possible that the progesterone synthesized by Stage V follicles produces the same sort of priming effect in vivo. We had been surprised to find that Xenopus ovaries contain and synthesize much more testosterone than progesterone or estradiol (Fortune and Tsang, 1981). Although the testosterone antibody used in the radioimmunoassay cross-reacts 69% with the nonaromatizable androgen DHT, chromatographic separation of testosterone and DHT showed that most of the androgen measured by the assay consisted of testosterone (Fortune and Tsang, 1981). Thus, it seemed possible that testosterone might function simply as a precursor for estradiol synthesis. The current experiments show that testosterone is secreted primarily by large Stage V and VI follicles, rather than the medium-sized follicles that are most active in estradiol production (Fig. 2). Since the developmental pattern of testosterone secretion followed that of progesterone and since exogenous testosterone stimulates maturation (Morrill and Bloch, 1977; Schuetz, 1967a; Smith and Ecker, 1971), it is possible that progesterone may trigger maturation by serving as a precursor for synthesis of testosterone which would then stimulate maturation. Schatz and Morrill (1979) reported that Rana pipiens follicles rapidly metabolized radiolabelled progesterone, but not progesterone derived from radiolabelled acetate. Although these data seem inconsistent with the finding that Xenopua follicles produce considerable amounts of testosterone, it is possible that progesterone synthesized from exogenous precursors is compartmentalized and metabolized differently from progesterone synthesized from endogenous precursors. Species differences do not seem likely since we have recently found that Rana pipiens follicles synthesize considerable quantities of testosterone (Fortune and Wigglesworth, unpublished data). Tissue concentrations of estradiol and testosterone mimicked the pattern of secretion of these steroids at different developmental stages and in response to FPH (Figs. 2 and 3, upper panels). In contrast, progesterone accumulation in the medium showed more dramatic changes as a result of development and FPH treatment
508
DEVELOPMENTAL
BIOLOGY
than did concentrations in follicular tissue. These results suggest that measurement of steroids in the medium of incubated Xm follicles provides an adequate assessment of follicular steroidogenesis. This finding is of some practical importance because measurements of steroids in medium are easier, quicker, and more reliable than measurements on tissue which require solvent extraction. These experiments were not designed to localize the follicular sites of synthesis of the three steroids. The findings of Masui (1967), Smith et ak, (1968), Mulner et al (1978), and Thibier-Fouchet et al. (1976) have implicated the follicle cells in progesterone and estradiol synthesis. However, since oocytes can also metabolize exogenous steroid precursors (Reynhout and Smith, 1973; Fouchet et al, 1975; Thibier-Fouchet et c& 1976; Sanyal and Sibre, 1973), it is not clear that the follicle cells are the primary site of steroid synthesis. The data have been expressed here per follicle. It might be more appropriate to express them on the basis of follicular volume or surface area, since these parameters change over the course of development. In the absence of information on the follicular site of synthesis of each steroid, it is difficult to select the proper correction factor. However, it is clear that the changes that occur in steroid secretion during follicular development do not simply mirror the increases in follicular size and volume. This is evident for estradiol, since large follicles produce less estradiol than medium-sized follicles. In contrast, increases in the ability of follicles to produce progesterone in response to FPH roughly parallel increases in follicular surface area. During development from Stages I-II to Stage VI progesterone production increased about 29X while follicular surface area increased about 17X. On the other hand, FPH-stimulated testosterone production increased about 170~ during development, while surface area increased 17~ and volume 72X. During development from Stage IV to Stage V follicular testosterone production increased 11X, while surface area increased about twofold and volume 2.6X. Thus, changes in estradiol and testosterone production appear to be developmental changes in the regulation of the steroidogenic pathway. Ovine LH, but not FSH, increased the secretion of progesterone, testosterone, and estradiol by Xenop~.~ follicles and promoted meiotic maturation in a dosedependent fashion (Figs. 4-6; Table 1). The effect of LH was not as great as that of Xenopus gonadotropins (FPH) except on testosterone secretion. However, it is possible that higher doses of LH would have produced a response equivalent to the response to FPH, since the highest dose of LH was maximal only for progesterone secretion. Recent experiments with mammalian follicles and follicle cells have determined that. while both FSH and
VOLUME
99, 1983
LH promote steroid biosynthesis, the two gonadotropins act at different points in the steroidogenic pathway (see review by Armstrong and Dorrington, 1977). LH stimulates the production of androgens (Fortune and Armstrong, 1977), while FSH promotes the aromatization of androgens to estrogen (Moon et al, 1975; Dorrington et al, 1975; Fortune and Armstrong, 1978; Erickson and Hsueh, 1978; Fortune and Eppig, 1979). In terms of this mammalian model, ovine LH seems to exert both LHlike and FSH-like effects on Xenopu.s steroidogenesis. Thus, it is not clear what role FSH plays in Xenopus ovarian function. Since steroids were measured only in the incubation medium in the experiments with FSH and LH, it is possible that FSH promoted steroid synthesis in the tissue, but not secretion into the medium. However, later measurement of steroids in control tissue and tissue treated with the highest dose of FSH provided no evidence for an FSH effect on steroid synthesis (data not shown). Elucidation of the relative roles of the FSH and LH in X&opus must await the availability of purified Xenopus gonadotropins. In conclusion, Xen0pu-s ovarian follicles appear to pass through three developmental stages in terms of steroid secretion: small follicles secrete very little progesterone, testosterone or estradiol; medium-sized, vitellogenic follicles are active in producing estradiol; and large follicles shift to production of progesterone and large amounts of testosterone. The role of testosterone in Xenopus ovarian function is an intriguing question that remains to be elucidated. I am grateful to S. Vincent, Y. Eisner, and P. Tsang for excellent technical assistance, to L. E. Reichert (Albany Medical College, Albany, N. Y.) for providing purified ovine LH and FSH, and to H. Dormady and P. Koos for typing this manuscript. REFERENCES ARMSTRONG, D. T., and DORRINGTON, J. H. (1977). Estrogen biosynthesis in the ovaries and testes. Advan Sex Hcnm Res. 3, 217-25’7. CHIEFFI, G., and LUPO, C. (1963). Identification of sex hormones in the ovarian extracts of Torpedo marmorata and B$o vulgaris. Ga Camp. Endocrinol 3. 149-152. DORRINGTON, J. H., MOON, Y. S., and ARMSTRONG, D. T. (1975). Estradiol170 biosynthesis in cultured granulosa cells from hypophysectomiwd immature rats; stimulation by follicle-stimulating hormone. Endocrinology 97, 13281331. DUMONT, J. N. (1972). Oogenesis in Xenqpus .!a.euti (Daudin). I. Stages of oocyte development in laboratory-maintained animals. J. lMorphd 136, 153-180. ERICKSON, G. F., and HSUEH, A. J. W. (1978). Stimulation of aromatase activity by follicle stimulating hormone in rat granulosa cells in viva and in vitro. Emhcrinology 102, 1275-1282. FORTUNE, J. E., and ARMSTRONG, D. T. (1977). Androgen production by theta and granulosa isolated from proestrous rat follicles. Endocrinology 100, 1341-1347. FORTUNE, J. E., and ARMSTRONG, D. T. (1978). Hormonal control of 17@estradiol biosynthesis in proestrous rat follicles: Estradiol pro-
Ovarian
J. E. FORTUNE duction 235.
by isolated
theca
versvs
granulosa.
Endocrinology
102,227-
J. E., and EPPIG, J. J. (1979). Effects of gonadotropins on steroid secretion by infantile and juvenile mouse ovaries in vitro. Endocrinology 105, ‘760-768. FORTUNE, J. E., and TSANG, P. C. (1981). Production of androgen and estradiol-170 by Xenopus ovaries treated with gonadotropins in vitro. Gen Camp. EndocGwL 43, 234-242. FORTUNE, J. E., CONCANNON, P. W., and HANSEL, W. (1975). Ovarian progesterone levels during in vitro oocyte maturation and ovulation in Xenopus laevis. BioL Reprod 13, 561-567. FOUCHET, C., SERRES, C., BELL& R., and OZON, R. (1975). Mechanism of action of progesterone on amphibian oocytes. Uptake and metabolism of progesterone by isolated oocytes of Pleurodeles waltlii and Xenopus laeuis. Camp. Biochem Physiol 52B, 205-210. GAY, V. L., and KERLAN, J. T. (1978). Serum LH and FSH following passive immunization against circulating testosterone in the intact male rat and in orchidectomized rats bearing subcutaneous silastic implants of testosterone. Arch. AndroL 1, 239-240. KORENMAN, S. G., STEVENS, R. H., CARPENTER, L. A., ROBB, M., NISWENDER, G. D., and SHERMAN, B. M. (1974). Estradiol radioimmunoassay without chromatography: Procedure, validation and normal values. J. Clin Endocrinol MetaboL 38, 718-720. LICHT, P., and CREWS, D. (1976). Gonadotropin stimulation of in vitro progesterone production in reptilian and amphibian ovaries. Gev~ Camp. EndocrinoL 29, 141-151.
FORTUNE,
LICHT, P., and PAPKOFF, H. (1974). Separation of two distinct gonadotropins from the pituitary gland of the bullfrog Rana catesbeiana. Endocrinology 94, 1587-1595. LICHT, P., and PAPKOFF, H. (1976). Species specificity in the response of an in vitro amphibian (Xm &is) ovulation assay to mammalian luteinizing hormones. Gen Camp. Endocrinol 29, 552-555. MAROT, J., BELL& R., and OZON, R. (1977). In vitro facilitation of Xelzqpus oocyte maturation by subthreshold doses of progesterone. Dev. Biol 59, 91-95. MASUI, Y. (1967). Relative roles of the pituitary, follicle cells and progesterone in the induction of oocyte maturation in Rana p&em. J. Exp. Zoo1 166, 365-376. MOON, Y. S., DORRINGTON, J. H., and ARMSTRONG, D. T. (1975). Stimulatory action of follicle-stimulating hormone on estradiol-17p secretion by hypophysectomized rat ovaries in organ culture. Endcc7%wlogy 97, 244-247. MORRILL, G. A., and BLOCH, E. (1977). Structure-function relationships of various steroids relative to induction of nuclear breakdown and ovulation in isolated amphibian oocytes. J. Ster&a’Bioc~ 8,133139. MULNER,
O., THIBIER,
C., and OZON,
R. (1978).
Steroid
biosynthesis
by
Steroids
in
Xenopus
509
ovarian follicles of Xenopus lamis in vitro during oogenesis. Gen. Cmp. EndocrinoL 34, 287-295. PAPKOFF, H., FARMER, S. W., and LICHT, P. (1976). Isolation and characterization of luteinizing hormone from amphibian (Rana Catesbeiana) pituitaries. Life Sti 18, 245-250. REDSHAW, M. R., and NICHOLLS, T. J. (1971). Oestrogen biosynthesis by ovarian tissue of the South African clawed toad, Xenopus 1aevi.s Daudin. Gen. Cmp. Endocrinol 16, 85-96. REYNHOUT, J. K., and SMITH, L. D. (1973). Evidence for steroid metabolism during the in vitro induction of maturation in oocytes of Rana pipiens. Dew. BioL 30.392-402. SANYAL, M. K., and SIBRE, E. R. (1973). Breakdown of germinal vesicle of frog ooeytes with 5a-reduced products of progesterone in vitro. Proc. Sac. Exp. BioL Med. 144,483-486. SCHATZ, F., and MORRILL, G. A. (1979). Studies on the relative roles of pituitary and progesterone in the induction of meiotic maturation in the amphibian oocyte. Dz#iientiatim 13, 89-99. SCHUETZ, A. W. (1967a). Effect of steroids on the germinal vesicle of oocytes of the frog (Rana pipiens) in vitro. Proc. Sot. Exp. BioL Med. 124, 1307-1310. SCHUETZ, A. W. (1976b). Action of hormones on germinal vesicle breakdown in frog (Rana pipiens) oocytes. J. Exp. ZooL 166, 347354. SMITH, L. D., and ECKER, R. E. (1971). The interactions of steroids with Rana pipiens oocytes in the induction of maturation. Dev. BioL 25, 232-247. SMITH, L. D., ECKER, R. E. and SUBTELNY, S. (1968). In vitro induction of physiological maturation in Rana pipiens oocytes removed from their ovarian follicles. Deu. BioL 17, 627-643. SNYDER, B. W., and SCHUETZ, A. W. (1973). In vitro evidence of steroidogenesis in the amphibian (Ram pipiens) ovarian follicle and its relationship to meiotic maturation and ovulation. J. Exp. ZooL 183, 333-342. SPIEGEL, J., JONES, E., and SNYDER, B. W. (1978). Estradiol-170 interference with meiotic maturation in Ram pipiens ovarian follicles: Evidence for inhibition of 3&hydroxysteroid dehydrogenase. J. Exp. ZooL 204, 187-192. SUBTELNY, S., SMITH, L. D., and ECKER, R. E. (1968). Maturation of ovarian frog eggs without ovulation. J. Exp. ZooL 168, 39-47. THIBIER-FOUCHET, C., MLJLNER, O., and OZON, R. (1976). Progesterone biosynthesis and metabolism by ovarian follicles and isolated oocytes of Xenopus laevis. BioL Reprod 14, 317-326. WALLACE, R. A., and BERGINK, G. W. (1974). Amphibian vitellogenin: properties, hormonal regulation of hepatic synthesis and ovarian uptake, and conversion to yolk proteins. Amer. ZooL 14.1159-1175. WALLACE, R. A., JARED, D. W., DUMONT, J. N., and SEGA, M. W. (1973). Protein incorporation by isolated amphibian oocytes. III. Optimum incubation conditions. J. Exp. ZooL 184, 321-334.