Plasma and follicular tissue steroid levels in the elasmobranch fish, Torpedo marmorata

GENERAL

AND

COMPARATIVE

ENDOCRINOLOGY

85,

327-333 (1992)

Plasma and Follicular Tissue Steroid Levels in the Elasmobranch Fish, Torpedo marmorata’ S. FASANO, M. D’ANTONIO, R. PIERANTONI,~ AND G. CHIEFFI Dipartimento di Fisiologia Umana e Funzioni Biologiche Integrate “Filippo Bottazzi”, I Facoltd di Medicina e Chirurgia, Via Costantinopoli 16, 80138 Napoli and Stazione Zoologica, Villa Comunale, 80121 Napoli, Italy Accepted May 14, 1991 Steroid concentrations in plasma and follicular tissues (theta plus granulosa layers) were determined by radioimmunoassay in the aplacental viviparous ray, Torpedo marmorata, during various stages of the reproductive cycle. Steroids in the uterine fluid of pregnant animals and in preovulatory atretic follicles were also measured. In the follicular tissue of cyclic animals, levels of progesterone were always lower than those of estradiol-17p and androgens (testosterone plus So-dihydrotestosterone). EstradioLl7R and androgen levels increased as the animals approached the ultimate maturational stage before ovulation. Androgens were not detectable in plasma, while estradioLl7P increased dramatically before ovulation. In pregnant animals, only small ovarian follicles (~5 mm in diameter) were observed, and these had hormone concentrations that were similar to those of the small follicles of cyclic animals. Progesterone was the only steroid detected in the uterine fluid of pregnant animals. In completely sclerotic atretic follicles of pregnant animals, steroids were not detected. Progesterone was the main hormone in atretic follicles undergoing yolk resorption. This suggests that the latter may contribute to the elevated plasma progesterone concentrations of pregnant animals. 8 1992 Academic press. 1~.

Among the most interesting aspects of reproduction in female elasmobranchs is that, among them, there are examples of oviparity, ovoviviparity, and viviparity (Dodd, 1975; Chieffi and Pierantoni, 1987). However, direct measurements of steroid hormones, and information concerning endocrine correlates, are available for only a few species. Steroid hormone levels and morphological parameters have been correlated during the sexual cycle in the aplacental viviparous Squalus acanthias (Tsang and Callard, 1987a,b), and in the oviparous Ruju erinuceu (Koob and Callard, 1986). In the oviparous Scyliorhinus cuniculu (Sumpter, 1977; Sumpter and Dodd, 1979; Jenkins and Dodd, 1980) and in Squulus

ucunthius (Tsang

and Callard, 1987a,b) stage-dependent differences in ovarian sensitivity to gonadotropin stimulation were observed. In the aplacental viviparous ray, Torpedo murmorutu, no data had been available on the endocrine correlates of reproduction. In previous reports, Chieffr and co-workers have provided information on ovarian steroid identification and steroidogenie capability, and on the histology of the hypophysis and gonads (Chiefft and Pierantoni, 1987). In this study we describe steroid concentration measurements in follicular tissue (theta plus granulosa layers) and plasma during the reproductive cycle of Torpedo murmorutu. MATERIALS

AND METHODS

Female Torpedo marmorata (n = 33) were captured in the Gulf of Naples in 1986 and 1987. Total body length and weight were recorded routinely for each

’ Supported by Grants MPI (40% and 60%) and 87.02371.74 CNR. * To whom correspondence should be addressed. 327

0016~6480/92 $1.50 Copyri&t B 1992 by Academic Press. Inc. AU rights of reproduction in any form reserved.

328

FASANO

animal. Fish were anesthetized using MS-222 (Sigma Chemical Co.), and blood samples were collected by cardiac puncture, placed into heparinized tubes, and centrifuged at 15OOg for 10 min. Plasma was stored at -80°C until hormone analysis. After bleeding, fish were killed by spinal pithing and ovarian follicles were removed and deprived of yolk and some of them were fixed in Bouin’s fluid to be prepared for histological observations after staining with hemahtm plus eosin. Ovarian follicles were classified into one of three classes according to their diameters: (1) l-5 mm, (2) 6-10 mm, and (3) 1l-30 mm. Only those atretic follicles of preovulatory origin were collected because of the difficulty of obtaining sufficient quantities of postovulatory follicles (Chiefft, 1961). Atretic follicles were classified into groups (according to Chieffi, 1961) which included separately those undergoing yolk resorption and glandular organization (stages I, II, and III) and others which were completely sclerotic and easily recognized by their brown color (stage IV). The ovaries of pregnant animals (n = 8) contained only class 1 follicles and the uteri contained freshly ovulated eggs and/or fetuses whose sizes ranged from 1 to 12 cm in length. Uterine fluid was collected by puncturing the uterus and aspirating the fluid with a syringe. During this procedure, care was taken to avoid damage to the vascular supply, in order to minimize blood contamination of the uterine fluid samples. Hormone determination. Concentrations of estradiol-17B, progesterone, and androgens (testosterone + So-dihydrotestosterone) in plasma, methanohc tissue extracts, and uterine fluid were determined by radioimmunoassay (Pierantoni ef al., 1984, 1987). Ovarian follicles from single animals were completely deyolked before being counted, weighed, and extracted in 70% MeOH (1:4, w/v) overnight at 4°C. Aliquots of tissue extracts or plasma samples were pipetted into 16 x 150-mm glass tubes and extracted twice (2 x 7 ml) with diethyl ether (anesthesia grade). The aqueous layer was snap-frozen in an acetone/dry ice bath, and the organic layer was decanted into tubes maintained at 36°C under a stream of nitrogen. The evaporated extract was reconstituted in phosphate buffer (pH 7.0) with 0.1% gelatin. Samples (0.1 ml) were incubated overnight at 4°C with 0.1 ml of 3H-labeled steroid (containing 12,000,9500, and 7000 cpm for androgens, estradiol-17B, and progesterone, respectively) and 0.1 ml of diluted antisera. The final concentrations of the antisera were androgens, 1:96,000, estradiol-17B,

ET AL.

1:105,000, progesterone, 1:135,000. Antisera were provided by Dr. Bolelli (Servizio di Fisiopatologia della Riproduzione, CNR Bologna, Italy). The characteristics of the antisera have been reported previously (Pierantoni et al., 1984, 1987). Bound/free separation was carried out using a charcoal-dextran (0.5-0.05%) suspension in phosphate buffer. After mixing, the tubes were incubated for 5 min at 4°C and centrifuged at 2000s for 15 min at 4°C. Supematants were decanted into S-ml minivials containing 4 ml of Insta Gel (United Technologies Pakard) scintillation cocktail. Validation of the method has been reported previously (Pierantoni et al., 1984, 1986, 1987). Percentage of recoveries of assayed hormones ranged from 85 to 90%. Sensitivities were 3 pg/tube for progesterone and 1.5 pg/tube for both estradioLl7B and androgens. The intra-assay coefficients of variation were 6.5% for progesterone, 7.4% for estradiol-l7B, and 6.2% for androgens. The inter-assay coefficients of variation were 10.4% for progesterone, 9.3% for estradiol-17B, and 11.1% for androgens. Statistics. Significance of differences were evahrated by analysis of variance followed by Duncan’s test for multigroup comparisons or by Student’s t test, when appropriate, using a BIOSTAT computer program (FABER s.r.l., Napoli, Italy).

RESULTS Pregnant Animals

The ovaries of pregnant fish contained only class 1 follicles, and the hormone levels of these follicular tissues were similar to those of nonpregnant fish possessing only class 1 follicles (Fig. la). In contrast, plasma progesterone levels of pregnant animals were significantly higher (P < 0.02) than those of nonpregnant animals though ovaries contained only class 1 follicles. Progesterone (109.42 k 36.61 pg/ml) was measurable in the uterine fluid (not shown), but estradiol-17p and androgens were not detectable. Nonpregnant Animals

Fish with ovaries containing

only class 1

FIG. 1. Progesterone (P), estradiol-17B (E), and androgen (A) contents in de-yolked follicles and plasma of pregnant and nonpregnant Torpedo marmorata. Number of follicle assayed is shown in brackets. Data of hormonal follicular tissue content included in each panel were evaluated by one-way ANOVA and Duncan’s test. Bars indicate means * SD and the same letter superscript indicates not significant differences (P > 0.05). For comparison of plasma values (means ? SD) see Results.

STEROID

LEVELS

IN

329

ELASMOBRANCH

bl

WN-REGNANT c*sr 1 foIl*l. enwnt

cl

NON=MfiGNANT clru 1.2 kMlifh

0

2 B

prount

t 1411 I

a

l

InI

.

P mol bml . . l-l I

,185rl.n dl

NON-PREGNANT ChU 1.23 follicle pr.*d

.I

MN-REGNANT clrsr 1.2.3 pr*Wlt

d 700

120

i

foIlid*

330

FASANO

follicles had follicular tissue concentrations of progesterone that were lower than those of estradiol-17P and androgens (P < 0.01, Fig. lb). Conversely, in plasma, concentrations of estradiol-17P and androgens were below the sensitivity of the assay, while progesterone was detectable. In animals having class 1 and 2 follicles, follicular tissue progesterone titers remained lower than those of estradiol-17P and androgens (P < 0.02; Fig. lc), plasma levels of estradiol- 17p and progesterone did not differ significantly, and plasma androgens were not detected. Animals developing ripe ovaries were further divided in two groups on the basis of the follicular hormonal changes. In one group (Fig. Id), estradiol-17l3 and androgen concentrations were higher in class 2 follicles than in class 3 (P < 0.01). The other group (Fig. le) was characterized by follicular tissue hormone concentrations that were proportional to follicular size. In the latter group, plasma levels of estradiol17p were higher than in all other animals examined (P < 0.001). Atretic

Preovulatory

Follicles

Atretic follicles showing features of yolk resorption were found in animals during the early stages of pregnancy, and these follicles had progesterone concentrations that varied (widely) from 0.053 to 9.75 pg/mg of tissue. However, progesterone concentrations were always 2- to 20-fold higher than estradiol-17P and 2 to 7-fold higher than androgens (not shown). Follicles of stage III (Fig. 2a) showed the true morphological aspects of an epithelial gland being characterized by the lengthening and anastomosis of the villi and by the hypertrophy of the vascular net. Atretic follicles of stage IV showed pigmentary degeneration due to the sclerosis of the vascular net of the villi (Fig. lb) and all measured hormone concentrations were below the sensitivity of the assay. DISCUSSION

Our results show that in Torpedo marmo-

ET AL.

rata, plasma steroid hormone levels fluctuate according to the different reproductive stages. Although differences in plasma progesterone concentrations were not apparent in nonpregnant animals, estradiol- 17p levels were elevated in animals approaching the maturational stage before ovulation. In the viviparous skate, Raja erinacea, progesterone is involved in events occurring at ovulation (Koob et al., 1985; Koob and Callard, 1986). In Squalus acanthias, Scyliorhinus canicula, and Raja erinacea (Koob et al., 1985; Tsang and Callard, 1987a,b; Sumpter and Dodd, 1979), plasma estradioLl7P and testosterone concentrations are positively correlated with follicular diameter. In particular, in Squalus acanthias the large follicles synthesize estrogen most actively (Tsang and Callard, 1987b). Our results suggest that Torpedo marmorata is similar to other elasmobranchs examined with respect to plasma estradiol17p. Plasma androgens are undetectable in Torpedo marmorata, and this contrasts with the positive correlation found between plasma testosterone and follicular size in Squalus acanthias, and Raja erinacea

Scyliorhinus

canicula,

(all preceding references). Whether the lack of detectable plasma androgens in Torpedo marmorata reflects a particular property of this species requires further study, because only a limited number of elasmobranch species have been investigated appropriately to date. Anyhow, the possible existence of elasmobranch species lacking circulating androgens (T + DHT) is of interest since this would represent a unique finding in vertebrates (Chietfi and Pierantoni, 1987). In pregnant Torpedo marmorata, progesterone is the only steroid hormone that we detected in plasma, and its levels were significantly higher than in nonpregnant animals. The ovaries of pregnant fish contained only small (class 1) follicles, suggesting that follicular growth is inhibited during pregnancy in this species, as suggested by Chieffr (1961). Since progesterone may in-

STEROID

LEVELS

IN ELASMOBRANCH

FIG. 2. (a) Stage III atretic follicles showing a richly vascularized glandular structure (x250). (b) Stage IV atretic follicles showing the sclerosis of the glandular vascular net (x312).

331

332

FASANO

hibit estrogen-induced vitellogenin synthesis in Squalus acanthias (Callard et al., 1980), we suggest that the lack of follicular growth during pregnancy observed in Torpedo marmorata is at least in part the result of high progesterone levels exerting a negative effect on vitellogenesis. In the aplacental viviparous Squalus acanthias, progesterone is produced by corpora lutea arising from postovulatory follicles (Tsang and Callard, 1987a). It has been suggested that corpora lutea originate from preovulatory rather than postovulatory follicles in Torpedo marmorata. Indeed, postovulatory follicles are characterized by a scarce steroidogenic activity (Chiefft, 1961). Therefore, our data support the hypothesis that preovulatory follicles undergoing atresia provide a source of progesterone in pregnant Torpedo. Specifically, we found that steroid concentrations in healthy follicles of pregnant and nonpregnant animals were not different, and that stage I-III atretic follicles contained substantial progesterone concentrations. This suggests that the additional amount of progesterone in the plasma of pregnant fish may come from atretic follicles in which yolk resorption is occurring. In particular, the glandular morphological aspect of stage III further strengthens this hypothesis. Titers of progesterone in follicular tissue were lower than those of estradiol-17P and androgens in all groups of fish examined. The relatively high levels of plasma progesterone compared to the low levels observed in follicular tissue are difftcult to reconcile. In this regard, the possibility of rapid release into the blood or an extragonadal source of progesterone (adrenal?) should be taken into consideration. It is interesting to note that, although androgen concentrations in follicular tissue were comparable to those of estradiol-17l3, androgens were not detected in any of plasma samples. This suggests that androgens may be utilized within the follicular layers. Testosterone in vertebrate ovaries

ET AL.

generally is the substrate for aromatase activity (Chiefft and Pierantoni, 1987). It is worth noting that ripe Torpedo marmorata ovaries containing follicles of class 1, 2, and 3 showed two distinct follicular tissue hormonal patterns: in one group of fish, estradiol-17s and androgens concentrations were higher in class 2 than in class 3 follicles, while in the other group, hormone concentrations increased progressively from the smallest to the largest follicles. Whether this difference resulted from different maturational stages of class 3 follicles during a preovulatory hormonal peak is an intriguing possibility requiring further investigation. ACKNOWLEDGMENT We thank Professor A. Gorbman (Department of Zoology, University of Washington, Seattle, Washington) for commenting on the manuscript.

REFERENCES Callard, I. P., Ho, S. M., Gapp, D. A., Taylor, S., Danko, D., and Wulezyn, G. (1980). Estrogen and estrogenic action in fish, amphibians and reptiles. In “Estrogens in the Environment” (J. A. McLachlan, Ed.), Vol. 5, pp. 213-237. Elsevier/ North Holland, New York. Chief& G. (1961). La luteogenesi nei Selaci ovovivipari. Ricerche istologiche ed istochimiche in Torpedo oceilata e Torpedo marmorata. Pubbl. Staz. Zool. Napoli 32, 45-166. Chiefft, G., and Pierantoni, R. (1987). Regulation of ovarian steroidogenesis. In “Hormones and Reproduction in Fishes, Amphibians and Reptiles” (D. 0. Norris and R. E. Jones, Eds.), pp. 117137. Plenum, New York/London. Dodd, J. M. (1975). The hormones of sex and reproduction and their effects in fishes and lower chordates: Twenty years on. Am. Zool. 15, 137-171. Jenkins, N., and Dodd, J. M. (1990). Effects of synthetic mammalian gonadotropin-releasing hormone and dogfish hypothalamic extracts on levels of androgens and estradiol in the circulation of the dogfish (Scyliorhinus canicula L.) .I. Endocrinol. 86, 171-177. Koob, T. J., and Callard, I. P. (1986). Progesterone treatment causes early oviposition in Raja erinatea. Bull. Mont. Desert Islands Biol. Lab. 25, 13g-139. Koob, T. J., Tsang, P., and Callard, I. P. (1985).

STEROID

LEVELS

Plasma estradiol, testosterone and progesterone levels during the ovulatory cycle of the skate (Raja erinacea). Biol. Reprod. 35, 261-215. Pierantoni, R., Iela, L., Dehio, G., and Rastogi, R. K. (1984). Seasonal plasma sex steroid levels in the female Rana esculenta. Gen. Cornp. Endocrinol. 53, 126-134. Pierantoni, R., Varriale, B., Minucci, S., Di Matteo, L., Fasano, S., D’Antonio, M., and Chieffi, G. (1986). Regulation of androgen production by frog (Rana esculenta) testis: An in vitro study on the effects exerted by estradiol, Sa-dihydrotestosterone, testosterone, melatonin and serotonin. Gen. Comp. Endocrinol. 64, 405-410. Pierantoni, R., Varriale, B., Fasano, S., Minucci, S., Di Matteo, L., and Chieffi, G. (1987). Seasonal plasma and intraovarian sex steroid profiles, and influence of temperature on gonadotropin stimulation of in vitro estradiol-178 and progesterone

IN ELASMOBRANCH

333

production, in Rana esculenta (Amphibia:anura). Gen. Comp. Endocrinol. 67, 163-N. Sumpter, J. P. (1977). “Biochemical and Physiological Studies on the Reproductive Hormones of the Dogfish (Scyliorhinus canicula L.).” Ph.D. thesis, University of Wales. Sumpter, J. P., and Dodd, J. M. (1979). The annual reproductive cycle of the female lesser spotted dogfish, Scyliorhinus canicula L., and its endocrine control. J. Fish. Biol. 15, 687-695. Tsang, P., and Callard, I. P. (1987a). Luteal progesterone production and regulation in the viviparous dog&h, Squalus acanthias. J. Exp. 2001. 241, 377-382. Tsang, P., and Callard, I.P. (1987b). Morphological and endocrine correlates of the reproductive cycle of the aplacental viviparous dogfish, Squalus acanthias. Gen. Comp. Endocrinol. 66, 182-189.