- Email: [email protected]
Ontogenetic Profile of FSH and LH in Rana esculenta
General and Comparative Endocrinology 116, 114–121 (1999) Article ID gcen.1999.7313, available online at http://www.idealibrary.com on
Ontogenetic Profile of FSH and LH in Rana esculenta M. Fiorentino,* K. Yamamoto,† S. Tanaka,‡ C. Pinelli,* B. D’Aniello,* S. Kikuyama,† and R. K. Rastogi*,1 *Dipartimento di Zoologia, Universita` di Napoli Federico II, 80134 Naples, Italy; †Department of Biology, School of Education, Waseda University, Nishiwaseda 1-6-1, Tokyo 169–50, Japan; and ‡Department of Biology, Institute of Science, Shizuoka University, Shizuoka 422-8529, Japan Accepted April 21, 1999
of pituitary gonadotropic activity in the adult amphibian (Licht et al., 1983; Itoh et al., 1990; Pavgi and Licht, 1989, 1993; Stamper and Licht, 1993, 1994; PolzonettiMagni et al., 1998; Kim et al., 1998). So far, the dynamics of pituitary gonadotropic activity remain poorly understood in amphibians during development. Few studies have examined the differentiation of gonadotropinexpressing cells in the pituitary, providing immunohistochemical evidence that FSH- and/or LH-secreting pituitary cells are differentiated during larval development in amphibians (Moriceau-Hay et al., 1982; Kar and Naik, 1986; Tanaka et al., 1991; Ogawa et al., 1995; Pinelli et al., 1996). GnRH is the major neuroendocrine regulator of synthesis and secretion of pituitary gonadotropins (Stamper and Licht, 1993; Rastogi and Iela, 1994; Muske, 1997; King and Millar, 1997, and references therein). The presence of GnRH has been described in the amphibian brain during larval development (King and Millar, 1981; Muske and Moore, 1990; Rastogi et al., 1990; Northcutt and Muske, 1994; Ogawa et al., 1995; D’Aniello et al., 1995; Di Fiore et al., 1996). This alone would suffice to argue that if the brain– pituitary anatomical connections, i.e., hypothalamo– hypophyseal portal system, were differentiated during larval development, the pituitary gonadotropes should receive and eventually respond to a stimulatory GnRH signal to regulate gonadal activity and related steps of reproduction. However, it is known that gonadal activity, in terms of the production of gametes and the adult plasma levels of sex steroids, sets in only when
Circulating levels and pituitary content of FSH and LH were determined by specific radioimmunoassays in Rana esculenta starting a few days after hatching until the completion of metamorphosis. Both gonadotropins were found in the pituitary as well as in the blood plasma at all stages of development examined here. The plasma concentrations of FSH and LH were more or less uniform during pre- and prometamorphosis, but increased significantly at the onset of metamorphic climax. The plasma levels of FSH and LH remained high at the completion of metamorphosis. The pituitary content of FSH and LH was low in early premetamorphosis. It increased slightly through prometamorphosis and metamorphic climax, following which a highly significant increase occurred. Whereas plasma concentrations of FSH and LH were essentially similar within a single stage of development, the pituitary FSH content was severalfold higher than pituitary LH. The significance of these results is discussed in relation to the functional maturation of the brain–pituitary–gonadal axis in the frog. r 1999 Academic Press The onset of the secretion of pituitary gonadotropins is a fundamental step in the functional maturation of the brain–pituitary–gonadal axis in vertebrates. There is a good body of literature on the reproductive cycle-related changes, influence of gonadotropinreleasing hormone (GnRH), and steroidal modulation 1
To whom correspondence should be addressed. Fax: (⫹39) 081 7903 336. E-mail: [email protected].
114
0016-6480/99 $30.00 Copyright r 1999 by Academic Press All rights of reproduction in any form reserved.
115
FSH and LH during Frog Development
sexual maturity is attained and the animal is ready for reproduction. The underlying message should be that a temporal cascade of developmental events occurs which progressively prepares anatomical substrates for the target organs to receive and respond to the regulatory molecule(s). In addition to the immunohistochemical analyses, using monoclonal antibodies against bullfrog FSH and LH -subunits, in Rana catesbeiana, Tanaka et al. (1991) also radioimmunoassayed the pituitary content of FSH and LH. Whereas some studies monitored the circulating levels of FSH and LH in adult anurans (Licht et al., 1983; Itoh et al., 1990; Itoh and Ishii, 1990; PolzonettiMagni et al., 1998; Kim et al., 1998), to our knowledge, no previous study reports the ontogenetic profile of circulating FSH and LH levels in any amphibian species. The objective of the current study, therefore, was to analyze, by radioimmunoassay, the developmental changes in plasma and pituitary levels of FSH and LH in Rana esculenta.
MATERIALS AND METHODS Animals Developmental stages of R. esculenta, starting a few days after hatching when external gills are no longer discernible under a dissecting microscope and hind limb buds are clearly seen, were collected from a pond in a field near Naples, as was done in previous studies on brain GnRH and pituitary gonadotropes (D’Aniello et al., 1995; Di Fiore et al., 1996; Pinelli et al., 1996). Developmental stages were classified according to the anuran staging scheme of Witschi (1956). In keeping with our earlier reports (D’Aniello et al., 1995; Pinelli et al., 1996; Di Fiore et al., 1996), Stages 26–28 and 29 represent premetamorphosis; Stage 30 represents prometamorphosis; Stages 31 and 32–33 represent early and late metamorphic climax, respectively; and Stage P represents newly metamorphosed (postmetamorphic, with the tail bud hardly visible) froglets. Under light MS222 (tricaine methanesulfonate, Sigma Chemical Co., St. Louis, MO) anesthesia, blood was collected directly from the heart with a fine heparinized glass capillary. Immediately after centrifugation,
at 4°, each plasma sample was lyophilized and stored until assayed. The number of individual blood plasma samples for Stages 26–28, 29, 30, 31, 32–33, and P was 9, 7, 8, 7, 6, and 3, respectively. For the determination of pituitary content of FSH and LH, pituitaries were pooled and lyophilized. Three such pools were used for each stage. A representative number (n ⫽ 3) of pituitaries for each stage of development was fixed in Bouin’s fluid, embedded in paraffin wax, and sectioned sagittally (5 µm). Two series of alternate sections were prepared and immunostained with monoclonal anti-bullfrog FSH and LH, respectively.
Radioimmunoassay of Plasma and Pituitary Gonadotropins Blood plasma and pituitary gonadotropins were measured by specific radioimmunoassays for bullfrog FSH (fFSH) and LH (fLH), employing fFSH and fLH as standard and radioligand, respectively. In these RIAs, antisera against the -subunits of fFSH and fLH were used instead of antisera against intact fFSH and fLH to avoid possible cross-reaction with the ␣-subunits of glycoprotein hormones (Tanaka et al., 1983). Radioiodination of fFSH and fLH with Na125I (career free, Radiochemical Centre, Amersham, England) was carried out at room temperature according to the modified lactoperoxidase method as described previously (Yamamoto et al., 1995). For details on radioimmunoassay of FSH and LH the reader is referred to PolzonettiMagni et al. (1998). Intact fLH and fFSH standard resulted in a long-dose inhibition of intact 125I-fLH and 125I-fFSH binding to the fLH and fFSH antiserum, respectively. The sensitivity, defined as the amounts of hormones that significantly decreased the counts by 2 SD from the 100% value, averaged 0.122 ⫾ 0.02 ng (mean ⫾ SEM) fLH and 0.122 ⫾ 0.01 ng (mean ⫾ SEM) fFSH per 100 µl of assay buffer for fLH and fFSH RIAs, respectively. The interassay coefficients of variation in RIAs for fLH and fFSH were 3.3 and 3.6%, respectively, when the estimated dose required for 50% inhibition in 10 assays was used. The intraassay coefficients of variation in the RIAs for fLH and fFSH were 4.3 and 3.8%, respectively, when determinations were repeated 10 times using 3.9 ng protein of standard hormone. The linear portion of the inhibition curve for the plasma of
Copyright r 1999 by Academic Press All rights of reproduction in any form reserved.
116
R. esculenta was parallel to the fLH and fFSH standard curves. Bullfrog ␣-subunit of gonadotropins and bullfrog GH showed no apparent displacement of the labeled intact standard in either RIA. The fLH and fFSH preparations showed a slight but negligible cross-reactivity in the fFSH RIA and fLH RIA, respectively. Bullfrog PRL also showed a slight crossreactivity in both RIAs. Cross-reactivity of fTSH in fFSH and fLH RIAs was not checked, since we did not have purified fTSH preparation. However, our unpublished immunohistochemical study revealed that neither antiserum against fFSH nor antiserum against fLH used in this study stained TSH cells recognizable with human TSH antiserum (NIDDK, NIH, Bethesda, MD).
Immunohistochemistry of Pituitary Gonadotropes and Cell Counting Monoclonal antibodies against bullfrog FSH (BF3B25; 1:8000–10,000; Tanaka et al., 1990) and bullfrog LH (BL4B11; 1:30,000–50,000; Park et al., 1987) were used. Antigen retrieval was performed as described earlier (Pinelli et al., 1996). Briefly, paraffin sections mounted on glass slides were rehydrated with PBS (pH 7.4; 0.1 M), covered with primary antiserum, and placed overnight, at 4°, in a black moist chamber. After a rinse in PBS, sections were stained with DAB, using ImmunoPure ABC staining kit (Pierce Chemical Co., CA), or with goat anti-mouse fluorescein isothiocyanate-conjugated secondary antibody (1:200; Pierce). From each pituitary, at least six 25-µm distant sections were selected for cell counting. To obtain the percentage of gonadotropes, FSH- and LH-immunopositive cells were counted together with immunonegative cells.
Fiorentino et al.
FIG. 1. RIA analysis of plasma FSH levels during development of R. esculenta. Sample sizes are given in parentheses. Each point is the mean ⫾ 1 SEM. Different superscripts indicate significant differences (a vs b ⫽ P ⬍ 0.001; b vs c ⫽ 0.005 ⬍ P ⬍0.001; a vs c ⫽ 0.05 ⬍ P ⬍ 0.01).
RESULTS Ontogenetic Profile of Plasma FSH and LH (Figs. 1, 2) Both gonadotropins, FSH and LH, were present in the plasma of early tadpoles (Stage 26–28). The concentration of plasma FSH and LH varied over a wide range within each stage, being 2.78–5.65 ng/ml FSH and 2.03–4.44 ng/ml LH in Stage 26–28, 3.17–5.79 ng/ml FSH and 3.21–4.6 ng/ml LH in Stage 29,
Statistics Results are given as the mean ⫾ 1 SEM. Data were evaluated for statistically significant differences by Kruskal–Wallis analysis of variance, and a one-way analysis of variance (ANOVA) followed by Duncan’s multiple range test for pairwise comparisons. Probability of 0.05 or less was considered to be statistically significant.
Copyright r 1999 by Academic Press All rights of reproduction in any form reserved.
FIG. 2. RIA analysis of plasma LH levels during development of R. esculenta. Sample sizes are given in parentheses. Each point is the mean ⫾ 1 SEM. Different superscripts indicate significant differences (a vs b ⫽ P ⬍ 0.001).
117
FSH and LH during Frog Development
3.35–6.3 ng/ml FSH and 3.23–5.38 ng/ml LH in Stage 30, 5.12–11.15 ng/ml FSH and 4.4–8.19 ng/ml LH in Stage 31, 5.45–6.98 ng/ml FSH and 4.51–7.56 ng/ml LH in Stage 32–33, and 4.77–6.07 ng/ml FSH and 4.54–7.23 ng/ml LH in postmetamorphic froglets (Stage P). The geometric means and ⫾ the SEM of plasma FSH and LH concentrations at each developmental stage are presented in Figs. 1 and 2, respectively. Within each stage of development, except in postmetamorphic froglets, the geometric mean of the plasma FSH level was comparatively higher (not significant statistically) than that of plasma LH. Developmental stage-related fluctuation in plasma FSH and LH was highly significant by the Kruskal– Wallis test (FSH: KW ⫽ 22.135, P ⫽ 0.0001; LH: KW ⫽ 23.124, P ⫽ 0.0001) as well as by ANOVA followed by Duncan’s multiple range test (FSH: F between group ⫽ 9.009, df ⫽ 5, N ⫽ 40, 0.05 ⬍ P ⬍ 0.01; LH: F between group ⫽ 8.419, df ⫽ 5, N ⫽ 40, 0.05 ⬍ P ⬍ 0.01) Figure 1 shows that of all stages examined, plasma concentration of FSH was lowest in Stage 26–28 tadpoles. It increased progressively through pre- and prometamorphosis (Stages 29 and 30, respectively) and showed a highly significant peak at the beginning of the metamorphic climax (Stage 31 vs 26–28, 29, or 30: P ⬍ 0.01). This peak followed a statistically significant decline at Stage 32–33 (P ⬍ 0.05) and in postmetamorphic froglets (P ⬍ 0.01). However, FSH values at Stages 32–33 and P did not differ and were significantly higher than those at Stage 26–28 (0.05 ⬍ P ⬍ 0.01). The pattern of changes in plasma LH levels was essentially similar to that observed for plasma FSH (Fig. 2). In fact, plasma LH levels increased significantly at Stage 31 (P ⬍ 0.01) compared to those at Stages 26–28, 29, and 30. Furthermore, in contrast to plasma FSH values which declined in late metamorphic climax, plasma LH values remained high until after metamorphosis. Indeed, plasma LH values at Stages 31, 32–33, and P represented a plateau and were all significantly higher than those at Stages 26–28 and 29 (0.05 ⬍ P ⬍ 0.01). It is thus plausible that, during development, peak plasma levels of both gonadotropins are reached during the metamorphic climax. The overall increase in circulating levels of FSH and LH during the metamorphic climax and soon after was hardly twofold over that observed in Stage 26–28 tadpoles.
FIG. 3. Pituitary FSH content determined by RIA during development of R. esculenta. Sample size is three for each stage. Each point is the mean ⫾ 1 SEM. Different superscripts indicate significant differences (a vs b ⫽ 0.05 ⬍ P ⬍ 0.01).
Pituitary Content of FSH and LH during Development (Figs. 3, 4) Radioimmunoassayable amounts of both FSH and LH were present in the pituitary of the earliest tadpole stage examined (Stage 26–28). Developmental changes in the pituitary content of both gonadotropins were significant by the Kruskal–Wallis test (FSH: KW ⫽ 15.409, P ⫽ 0.009; LH: KW ⫽ 13.491, P ⫽ 0.02). ANOVA followed by Duncan’s multiple range test
FIG. 4. Pituitary LH content during development of R. esculenta. Sample size is three for each stage. Each point is the mean ⫾ 1 SEM. Different superscripts indicate significant differences (a vs b ⫽ 0.05 ⬍ P ⬍ 0.01).
Copyright r 1999 by Academic Press All rights of reproduction in any form reserved.
118
showed that, although progressively increasing during development, the pituitary content of FSH and LH did not change significantly until the completion (total tail resorption) of metamorphosis, when a sharp and highly significant increase occurred (FSH: F between group ⫽ 5.621, F within group ⫽ 1.052, df ⫽ 5, N ⫽ 18; LH: F between group ⫽ 5.209, F within group ⫽ 1.641, df ⫽ 5, N ⫽ 18). In fact, pituitary FSH and LH content in postmetamorphic froglets was significantly higher than that at any earlier stage of development (0.05 ⬍ P ⬍ 0.01). In addition, the pituitary content of FSH within any stage of development was of several magnitudes higher than LH, the FSH:LH ratios being 3:1–10:1 at Stage 26–28, 6:1–23:1 at Stage 29, 10:1–13:1
Fiorentino et al.
at Stage 30, 11:1–18:1 at Stage 31, 7:1–13:1 at Stage 32–33, and 9:1–24:1 at Stage P. The total FSH content of the pituitary from newly metamorphosed froglets (Stage P) was nearly 15 times higher than that from the earliest tadpole stage (Stage 26–28) examined (Fig. 3). In comparison, LH content of the pituitary of froglets was barely sixfold that of Stage 26–28 tadpoles (Fig. 4).
Immunohistochemistry of Pituitary Gonadotropes (Figs. 5a–5d) Using monoclonal FSH and LH, only FSHimmunoreactive cells were seen in the developing pituitary of tadpoles from Stages 26–28, 29, and 30
FIG. 5. (a) FSH-immunoreactive cells in the pars distalis of the pituitary of a Stage 27 tadpole. Scale, ⫻80. (b) FSH-immunoreactive cells in the pituitary of a Stage 29 tadpole. Scale, ⫻80. (c and d) FSH- and LH-immunoreactive cells in consecutive sections of the pituitary of a postmetamorphic froglet. Scale, ⫻110.
Copyright r 1999 by Academic Press All rights of reproduction in any form reserved.
FSH and LH during Frog Development
(Figs. 5a, 5b). A few weakly immunoreactive LH cells were revealed for the first time in tadpoles at the beginning of metamorphic climax (Stage 31). In early premetamorphic tadpoles (Stage 26–28) the relative frequency of FSH cells was rather low, being 0.9–1.6% of pituitary cells. Their number increased progressively through stages 29 and 30 of development (1.8– 2.6 and 3.8–4.9%, respectively). During the metamorphic climax (Stages 31 and 32–33) and in postmetamorphic froglets the percentage of gonadotropes increased remarkably (6.1–9.9, Fig. 5c) and LH gonadotropes were clearly discernible. The latter, very few in number during early climax (0.6–0.9%), were almost as numerous in postmetamorphic froglets as FSH cells (8.3–9.3%, Fig. 5d). As previously demonstrated in our laboratories (Pinelli et al., 1996; Tanaka et al., 1990), FSH and LH-immunoreactivity was colocalized in many pituitary gonadotropes (Figs. 5c, 5d).
DISCUSSION The present study, using specific radioimmunoassay, is the first to quantify circulating levels of FSH and LH in an amphibian during larval development until the completion of metamorphosis. A major finding of this study is that plasma levels of FSH and LH increased significantly at the onset of metamorphic climax. The pituitary content of both gonadotropins increased significantly at the completion of metamorphosis. However, it is not feasible to ascertain from the foregoing results if the ontogenetic changes in plasma FSH and LH levels are associated with changes in synthesis and/or release. The presence of radioimmunoassayable FSH in the pituitary and plasma of early tadpoles is consistent with previous immunohistochemical analysis of developing pituitary in which a few FSH-immunoreactive cells were detected (Pinelli et al., 1996). Similarly, both studies, viz the present study and that by Pinelli et al. (1996), are consistent in that LH-immunoreactive cells become detectable by immunohistochemistry starting at the metamorphic climax. However, LH was quantified, using the same monoclonal antibody, by radioimmunoassay in the plasma as well as in the pituitary of early tadpoles. It may be argued that either low LH content of pituitary cells escapes detection by
119
immunohistochemistry because it is below the limit of the immunohistochemistry technique or radioimmunoassay is more sensitive and can detect low amounts of this gonadotropin. In a previous analysis of brain GnRH during development of R. esculenta, a similar situation was encountered, and radioimmunoassay revealed GnRHs in the brain of tadpoles much earlier than did immunohistochemistry (D’Aniello et al., 1995; Di Fiore et al., 1996). As is the case in other vertebrates, there is substantial in vivo and in vitro evidence in amphibians to indicate GnRH regulation of pituitary gonadotropic function and involvement of sex steroids in GnRHstimulated pituitary FSH and LH secretion (see Pavgi and Licht, 1993; Stamper and Licht, 1993, 1994; and references therein). The question is how the developmental pattern of pituitary gonadotropin synthesis and release relates to changes in brain GnRH production and, ultimately, to gonadal activity (gametogenesis and sex steroid production). Coincident with a significant rise in brain GnRH content and plasma levels of FSH and LH only after prometamorphosis, we are inclined to believe that the brain–pituitary– gonadal axis becomes a functional unit as the metamorphic climax sets in. This is sustained by the fact that mGnRH axons in this frog reach the median eminence at around the time when tadpoles begin to metamorphose (see D’Aniello et al., 1995). Support for this concept is also provided by our unpublished observations that two intraperitoneal injections of mGnRH, given on 2 consecutive days, do not induce FSH/LH secretion in the prometamorphic tadpoles, whereas they do during the climax. This would lead us to infer that the frog pituitary becomes responsive to GnRH only when the tadpoles begin to metamorphose and not before. Furthermore, during development, brain GnRH content remains low until around prometamorphosis (Di Fiore et al., 1996), at which time it begins to increase, showing a substantial elevation during the metamorphic climax and again at the completion of metamorphosis (Di Fiore et al., 1996). As far as the plasma levels of sex steroids are concerned, we have preliminary data to show that premetamorphic tadpoles contain 44 ⫾ 18 pg/ml of androgens and 54 ⫾ 16 pg/ml of estradiol-17, whereas at the term of metamorphic climax the circulating levels of sex steroids are significantly higher at 166 ⫾ 32 pg/ml of andro-
Copyright r 1999 by Academic Press All rights of reproduction in any form reserved.
120
gens and 250 ⫾ 81 pg/ml of estradiol-17 (Di Fiore, Pinelli, D’Aniello and Rastogi, unpublished data). Thus, in Stage 26–28 tadpoles, FSH and LH are detectable in the pituitary and plasma as early as the GnRH is detectable in the brain and the sex steroids are detectable in the plasma. Our present results and those published earlier (D’Aniello et al., 1995; Di Fiore et al., 1996) support that GnRH and FSH/LH are most likely produced only after hatching has occurred. It has also been established that tadpole (pre- and prometamorphic) gonads possess the capacity to synthesize and secrete the sex hormones, an activity independent of the pituitary until metamorphosis (see Hsu et al., 1985). This would allow us to assume that during pre- and prometamorphic development each endocrine component of the hypothalamus–pituitary–gonadal axis grows, differentiates, and functions independently. Indeed, it was proposed several decades ago that during embyronic development of vertebrates, the main steps of the endocrine axes are initially autonomous and that only at a later time during development are endocrine feedback systems established (Willier, 1939, 1955). In the chick, it was conclusively determined that the endocrine components of the brain– pituitary–gonadal axis function independently of each other during early development (see Woods and Thommes, 1998). Analogously, although hormones typical of the brain–pituitary–gonadal axis, viz. GnRH, pituitary gonadotropins, and sex steroids, were measured in great tit nestlings until 9 days of age, the hypothalamus– pituitary axis becomes a functional unit at a later time (Silverin and Sharp, 1996). Among vertebrate species, the developmental dynamics of the brain–pituitary–gonadal axis of the frog seem to differ remarkably from that of mammals, e.g., the rat. In fact, in the latter, brain GnRH is detectable as early as day 12 of gestation, at which time GnRH binding sites can be found in the developing anterior pituitary, while pituitary LH becomes detectable around fetal day 17 and FSH around fetal day 19 (see Ojeda and Urbanski, 1994). In this mammal, it has been suggested that brain GnRH plays a trophic function essential for the differentiation of pituitary gonadotropes. In birds and amphibians, this does not seem to be the case or is, at best, simply a matter of speculation. In the frog, even though it is clear that, during larval development, the brain contains radioimmunoas-
Copyright r 1999 by Academic Press All rights of reproduction in any form reserved.
Fiorentino et al.
sayble GnRH, we have no evidence to suggest an active involvement of GnRH in the control of the tadpole pituitary. So far, there is no evidence of the presence of GnRH receptors in the tadpole pituitary. On the other hand, within fish it is suggested that the brain–pituitary–gonadal axis is intact at the time of hatch in the rainbow trout, but conclusive experimental evidence is still necessary (see Feist and Schreck, 1996). In conclusion, it becomes obvious that, among vertebrates, there are group-specific differences in the temporal appearance of hormone-producing units of the brain–pituitary–gonadal axis and in the establishment of endocrine feedback mechanisms.
ACKNOWLEDGMENTS Financial aid from MURST and Universita` degli Studi di Napoli Federico II is appreciated.
REFERENCES D’Aniello, B., Pinelli, C., Di Fiore, M. M., Iela, L., King, J. A., and Rastogi, R. K. (1995). Development and distribution of gonadotropin-releasing hormone neuronal systems in the frog (Rana esculenta) brain: Immunohistochemical analysis. Dev. Brain Res. 89, 281–288. Di Fiore, M. M., King, J. A., D’Aniello, B., and Rastogi, R. K. (1996). Immunoreactive mammalian and chicken-II GnRHs in Rana esculenta brain during development. Regul. Pept. 62, 119–124. Feist, G., and Schreck, C. B. (1996). Brain–pituitary–gonadal axis during early development and sexual differentiation in the rainbow trout, Oncorhynchus mykiss. Gen. Comp. Endocrinol. 102, 394–409. Hsu, C.-Y., Chang, L.-T., Ku, H.-H., and Lu, M.-H. (1985). In vitro estradiol synthesis and secretion by tadpole ovaries of different developmental stages. Gen. Comp. Endocrinol. 57, 393–396. Itoh, M., Inoue, M., and Ishii, S. (1990). Annual cycle of pituitary and plasma gonadotropins and plasma sex steroids in a wild population of the toad, Bufo japonicus. Gen. Comp. Endocrinol. 78, 242–253. Itoh, M., and Ishii, S. (1990). Changes in plasma levels of gonadotropins and sex steroids in the toad, Bufo japonicus, in association with behavior during the breeding season. Gen. Comp. Endocrinol. 80, 451–464. Kar, S., and Naik, D. R. (1986). Cytodifferentiation and immunocharacteristics of adenohypophsial cells in the toad, Bufo melanostictus. Anat. Embryol. 175, 137–146.
FSH and LH during Frog Development Kim, J. W., Im, W. B., Choi, H. H., Ishii, S., and Kwon, H. B. (1998). Seasonal fluctuations in pituitary gland and plasma levels of gonadotropic hormones in Rana. Gen. Comp. Endocrinol. 109, 13–23. King, J. A., and Millar, R. P. (1981). TRH, GH-RH, and LH-RH in metamorphosing Xenopus laevis. Gen. Comp. Endocrinol 44, 20–27. King, J. A., and Millar, R. P. (1997). Coordinated evolution of GnRHs and their receptors. In ‘‘GnRH Neurons: Gene to Behavior’’ (I. S. Parhar and Y. Sakuma, Eds.), pp. 51–77. Brain Shuppan, Tokyo. Licht, P., McCreery, B. R., Barnes, R., and Pang, R. (1983). Seasonal and stress related changes in plasma gonadotropins, sex steroids, and corticosterone in the bullfrog, Rana catesbeiana. Gen. Comp. Endocrinol. 50, 124–145. Moriceau-Hay, D., Doerr-Schott, J., and Bubois, M. P. (1982). Immunohistochemical demonstration of TSH-, LH-, and ACTH-cells in the hypophysis of tadpoles of Xenopus laevis. Cell Tissue Res. 225, 57–64. Muske, L. E. (1997). Ontogeny, phylogeny and neuroanatomical organization of multiple molecular forms of GnRH. In ‘‘GnRH Neurons: Gene to Behavior’’ (I. S. Parhar and Y. Sakuma, Eds.), pp. 145–180. Brain Shuppan, Tokyo. Muske, L. E., and Moore, F. L. (1990). Ontogeny of immunoreactive gonadotropin-releasing hormone neuronal system in amphibians. Brain Res. 534, 177–187. Northcutt, R. G., and Muske, L. E. (1994). Multiple embryonic origins of gonadotropin-releasing hormone (GnRH) immunoreactive neurons. Dev. Brain Res. 78, 279–290. Ogawa, K., Suzuki, E., and Taniguchi, K. (1995). Immunohistochemical studies on the development of the hypothalamo–hypophysial system in Xenopus laevis. Anat. Rec. 241, 244–254. Ojeda, S. R., and Urbanski, H. F. (1994). Puberty in the rat. In ‘‘The Physiology of Reproduction’’ (E. Knobil and J. D. Neill, Eds.), 2nd ed., pp. 363–409. Raven Press, New York. Park, M. K., Tanaka, S., Hayashi, H., Hanaoka, Y., Wakabayashi, K., and Kurosumi, K. (1987). Production and characterization of a monoclonal antibody against -subunit of bullfrog lutropin. Gen. Comp. Endocrinol. 68, 82–90. Pavgi, S., and Licht, P. (1989). Effects of gonadectomy and steroids on pituitary gonadotropin secretion in a frog, Rana pipiens. Biol. Reprod. 40, 40–48. Pavgi, S., and Licht, P. (1993). Inhibition of in vitro pituitary gonadotropin secretion by 17-estradiol in the frog Rana pipiens. Gen. Comp. Endocrinol. 89, 132–137. Pinelli, C., Fiorentino, M., D’Aniello, B., Tanaka, S., and Rastogi, R. K. (1996). Immunohistochemical demonstration of FSH and LH in the pituitary of the developing frog, Rana esculenta. Gen. Comp. Endocrinol. 104, 189–196. Polzonetti-Magni, A. M., Mosconi, G., Carnevali, O., Yamamoto, K.,
121 Hanaoka, Y., and Kikuyama, S. (1998). Gonadotropins and reproductive function in the anuran amphibian, Rana esculenta. Biol. Reprod. 58, 88–93. Rastogi, R. K., and Iela, L. (1994). Gonadotropin-releasing hormone: Present concepts, future directions. Zool. Sci. 11, 363–373. Rastogi, R. K., di Meglio, M., and Iela, L. (1990). Immunoreactive luteinizing hormone-releasing hormone in the frog (Rana esculenta) brain: Distribution pattern in the adult, seasonal changes, castration effects, and developmental aspects. Gen. Comp. Endocrinol. 78, 444–458. Silverin, B., and Sharp, P. (1996). The development of the hypothalamic–pituitary–gonadal axis in juvenile great tits. Gen. Comp. Endocrinol. 103, 150–166. Stamper, D. L., and Licht, P. (1993). Further studies on the influence of GnRH on the biosynthesis of gonadotropins in female frogs (Rana pipiens). Gen. Comp. Endocrinol. 92, 104–112. Stamper, D. L., and Licht, P. (1994). Influence of androgen on the GnRH-stimulated secretion and biosynthesis of gonadotropins in the pituitary of juvenile female bullfrogs, Rana catesbeiana. Gen. Comp. Endocrinol. 93, 93–102. Tanaka, S., Hanaoka, Y., and Wakabayashi, K. (1983). A homologous radioimmunoassay for basic gonadotropin. Endocrinol. Jpn. 30, 71–78. Tanaka, S., Park, M. K., Hayashi, H., Hanaoka, Y., Wakabayashi, K., and Kurosumi, K. (1990). Immunocytochemical localization of the subunits of glycoprotein hormones (LH, FSH, and TSH) in the bullfrog pituitary gland using monoclonal antibodies and polyclonal antiserum. Gen. Comp. Endocrinol. 77, 88–97. Tanaka, S., Sakai, M., Park, M. K., and Kurosumi, K. (1991). Differential appearance of the subunits of glycoprotein hormones (LH, FSH, and TSH) in the pituitary of bullfrog (Rana catesbeiana) larvae during metamorphosis. Gen. Comp. Endocrinol. 84, 318–327. Willier, B. H. (1939). The embryonic development of sex. In ‘‘Sex and Internal Secretions’’ (E. Allen, C. H. Danforth, and E. A. Doisy, Eds.), 2nd ed. pp. 64–144. Williams & Wilkins, Baltimore. Willier, B. H. (1955). Ontogeny of endocrine correlation. In ‘‘Analysis of Development’’ (B. H. Willier, P. A. Weiss, and V. Hamburger, Eds.), pp. 574–619. Saunders, Philadelphia. Witschi, E. (1956). ‘‘Development of Vertebrates.’’ Saunders, Philadelphia. Woods, J. E., and Thommes, R. C. (1998). Avian reproductive system, developmental endocrinology. In ‘‘Encyclopedia of Reproduction’’ (E. Knobil and J. D. Neill, Eds.), Vol. 1, pp. 337–347. Academic Press, San Diego. Yamamoto, K., Yamashita, K., Hayakawa, Y., Hanaoka, Y., and Kikuyama, S. (1995). Development and application of a homologous radioimmunoassay for Xenopus prolactin. Gen. Comp. Endocrinol. 99, 28–34.
Copyright r 1999 by Academic Press All rights of reproduction in any form reserved.
Ontogenetic Profile of FSH and LH in Rana esculenta M. Fiorentino,* K. Yamamoto,† S. Tanaka,‡ C. Pinelli,* B. D’Aniello,* S. Kikuyama,† and R. K. Rastogi*,1 *Dipartimento di Zoologia, Universita` di Napoli Federico II, 80134 Naples, Italy; †Department of Biology, School of Education, Waseda University, Nishiwaseda 1-6-1, Tokyo 169–50, Japan; and ‡Department of Biology, Institute of Science, Shizuoka University, Shizuoka 422-8529, Japan Accepted April 21, 1999
of pituitary gonadotropic activity in the adult amphibian (Licht et al., 1983; Itoh et al., 1990; Pavgi and Licht, 1989, 1993; Stamper and Licht, 1993, 1994; PolzonettiMagni et al., 1998; Kim et al., 1998). So far, the dynamics of pituitary gonadotropic activity remain poorly understood in amphibians during development. Few studies have examined the differentiation of gonadotropinexpressing cells in the pituitary, providing immunohistochemical evidence that FSH- and/or LH-secreting pituitary cells are differentiated during larval development in amphibians (Moriceau-Hay et al., 1982; Kar and Naik, 1986; Tanaka et al., 1991; Ogawa et al., 1995; Pinelli et al., 1996). GnRH is the major neuroendocrine regulator of synthesis and secretion of pituitary gonadotropins (Stamper and Licht, 1993; Rastogi and Iela, 1994; Muske, 1997; King and Millar, 1997, and references therein). The presence of GnRH has been described in the amphibian brain during larval development (King and Millar, 1981; Muske and Moore, 1990; Rastogi et al., 1990; Northcutt and Muske, 1994; Ogawa et al., 1995; D’Aniello et al., 1995; Di Fiore et al., 1996). This alone would suffice to argue that if the brain– pituitary anatomical connections, i.e., hypothalamo– hypophyseal portal system, were differentiated during larval development, the pituitary gonadotropes should receive and eventually respond to a stimulatory GnRH signal to regulate gonadal activity and related steps of reproduction. However, it is known that gonadal activity, in terms of the production of gametes and the adult plasma levels of sex steroids, sets in only when
Circulating levels and pituitary content of FSH and LH were determined by specific radioimmunoassays in Rana esculenta starting a few days after hatching until the completion of metamorphosis. Both gonadotropins were found in the pituitary as well as in the blood plasma at all stages of development examined here. The plasma concentrations of FSH and LH were more or less uniform during pre- and prometamorphosis, but increased significantly at the onset of metamorphic climax. The plasma levels of FSH and LH remained high at the completion of metamorphosis. The pituitary content of FSH and LH was low in early premetamorphosis. It increased slightly through prometamorphosis and metamorphic climax, following which a highly significant increase occurred. Whereas plasma concentrations of FSH and LH were essentially similar within a single stage of development, the pituitary FSH content was severalfold higher than pituitary LH. The significance of these results is discussed in relation to the functional maturation of the brain–pituitary–gonadal axis in the frog. r 1999 Academic Press The onset of the secretion of pituitary gonadotropins is a fundamental step in the functional maturation of the brain–pituitary–gonadal axis in vertebrates. There is a good body of literature on the reproductive cycle-related changes, influence of gonadotropinreleasing hormone (GnRH), and steroidal modulation 1
To whom correspondence should be addressed. Fax: (⫹39) 081 7903 336. E-mail: [email protected].
114
0016-6480/99 $30.00 Copyright r 1999 by Academic Press All rights of reproduction in any form reserved.
115
FSH and LH during Frog Development
sexual maturity is attained and the animal is ready for reproduction. The underlying message should be that a temporal cascade of developmental events occurs which progressively prepares anatomical substrates for the target organs to receive and respond to the regulatory molecule(s). In addition to the immunohistochemical analyses, using monoclonal antibodies against bullfrog FSH and LH -subunits, in Rana catesbeiana, Tanaka et al. (1991) also radioimmunoassayed the pituitary content of FSH and LH. Whereas some studies monitored the circulating levels of FSH and LH in adult anurans (Licht et al., 1983; Itoh et al., 1990; Itoh and Ishii, 1990; PolzonettiMagni et al., 1998; Kim et al., 1998), to our knowledge, no previous study reports the ontogenetic profile of circulating FSH and LH levels in any amphibian species. The objective of the current study, therefore, was to analyze, by radioimmunoassay, the developmental changes in plasma and pituitary levels of FSH and LH in Rana esculenta.
MATERIALS AND METHODS Animals Developmental stages of R. esculenta, starting a few days after hatching when external gills are no longer discernible under a dissecting microscope and hind limb buds are clearly seen, were collected from a pond in a field near Naples, as was done in previous studies on brain GnRH and pituitary gonadotropes (D’Aniello et al., 1995; Di Fiore et al., 1996; Pinelli et al., 1996). Developmental stages were classified according to the anuran staging scheme of Witschi (1956). In keeping with our earlier reports (D’Aniello et al., 1995; Pinelli et al., 1996; Di Fiore et al., 1996), Stages 26–28 and 29 represent premetamorphosis; Stage 30 represents prometamorphosis; Stages 31 and 32–33 represent early and late metamorphic climax, respectively; and Stage P represents newly metamorphosed (postmetamorphic, with the tail bud hardly visible) froglets. Under light MS222 (tricaine methanesulfonate, Sigma Chemical Co., St. Louis, MO) anesthesia, blood was collected directly from the heart with a fine heparinized glass capillary. Immediately after centrifugation,
at 4°, each plasma sample was lyophilized and stored until assayed. The number of individual blood plasma samples for Stages 26–28, 29, 30, 31, 32–33, and P was 9, 7, 8, 7, 6, and 3, respectively. For the determination of pituitary content of FSH and LH, pituitaries were pooled and lyophilized. Three such pools were used for each stage. A representative number (n ⫽ 3) of pituitaries for each stage of development was fixed in Bouin’s fluid, embedded in paraffin wax, and sectioned sagittally (5 µm). Two series of alternate sections were prepared and immunostained with monoclonal anti-bullfrog FSH and LH, respectively.
Radioimmunoassay of Plasma and Pituitary Gonadotropins Blood plasma and pituitary gonadotropins were measured by specific radioimmunoassays for bullfrog FSH (fFSH) and LH (fLH), employing fFSH and fLH as standard and radioligand, respectively. In these RIAs, antisera against the -subunits of fFSH and fLH were used instead of antisera against intact fFSH and fLH to avoid possible cross-reaction with the ␣-subunits of glycoprotein hormones (Tanaka et al., 1983). Radioiodination of fFSH and fLH with Na125I (career free, Radiochemical Centre, Amersham, England) was carried out at room temperature according to the modified lactoperoxidase method as described previously (Yamamoto et al., 1995). For details on radioimmunoassay of FSH and LH the reader is referred to PolzonettiMagni et al. (1998). Intact fLH and fFSH standard resulted in a long-dose inhibition of intact 125I-fLH and 125I-fFSH binding to the fLH and fFSH antiserum, respectively. The sensitivity, defined as the amounts of hormones that significantly decreased the counts by 2 SD from the 100% value, averaged 0.122 ⫾ 0.02 ng (mean ⫾ SEM) fLH and 0.122 ⫾ 0.01 ng (mean ⫾ SEM) fFSH per 100 µl of assay buffer for fLH and fFSH RIAs, respectively. The interassay coefficients of variation in RIAs for fLH and fFSH were 3.3 and 3.6%, respectively, when the estimated dose required for 50% inhibition in 10 assays was used. The intraassay coefficients of variation in the RIAs for fLH and fFSH were 4.3 and 3.8%, respectively, when determinations were repeated 10 times using 3.9 ng protein of standard hormone. The linear portion of the inhibition curve for the plasma of
Copyright r 1999 by Academic Press All rights of reproduction in any form reserved.
116
R. esculenta was parallel to the fLH and fFSH standard curves. Bullfrog ␣-subunit of gonadotropins and bullfrog GH showed no apparent displacement of the labeled intact standard in either RIA. The fLH and fFSH preparations showed a slight but negligible cross-reactivity in the fFSH RIA and fLH RIA, respectively. Bullfrog PRL also showed a slight crossreactivity in both RIAs. Cross-reactivity of fTSH in fFSH and fLH RIAs was not checked, since we did not have purified fTSH preparation. However, our unpublished immunohistochemical study revealed that neither antiserum against fFSH nor antiserum against fLH used in this study stained TSH cells recognizable with human TSH antiserum (NIDDK, NIH, Bethesda, MD).
Immunohistochemistry of Pituitary Gonadotropes and Cell Counting Monoclonal antibodies against bullfrog FSH (BF3B25; 1:8000–10,000; Tanaka et al., 1990) and bullfrog LH (BL4B11; 1:30,000–50,000; Park et al., 1987) were used. Antigen retrieval was performed as described earlier (Pinelli et al., 1996). Briefly, paraffin sections mounted on glass slides were rehydrated with PBS (pH 7.4; 0.1 M), covered with primary antiserum, and placed overnight, at 4°, in a black moist chamber. After a rinse in PBS, sections were stained with DAB, using ImmunoPure ABC staining kit (Pierce Chemical Co., CA), or with goat anti-mouse fluorescein isothiocyanate-conjugated secondary antibody (1:200; Pierce). From each pituitary, at least six 25-µm distant sections were selected for cell counting. To obtain the percentage of gonadotropes, FSH- and LH-immunopositive cells were counted together with immunonegative cells.
Fiorentino et al.
FIG. 1. RIA analysis of plasma FSH levels during development of R. esculenta. Sample sizes are given in parentheses. Each point is the mean ⫾ 1 SEM. Different superscripts indicate significant differences (a vs b ⫽ P ⬍ 0.001; b vs c ⫽ 0.005 ⬍ P ⬍0.001; a vs c ⫽ 0.05 ⬍ P ⬍ 0.01).
RESULTS Ontogenetic Profile of Plasma FSH and LH (Figs. 1, 2) Both gonadotropins, FSH and LH, were present in the plasma of early tadpoles (Stage 26–28). The concentration of plasma FSH and LH varied over a wide range within each stage, being 2.78–5.65 ng/ml FSH and 2.03–4.44 ng/ml LH in Stage 26–28, 3.17–5.79 ng/ml FSH and 3.21–4.6 ng/ml LH in Stage 29,
Statistics Results are given as the mean ⫾ 1 SEM. Data were evaluated for statistically significant differences by Kruskal–Wallis analysis of variance, and a one-way analysis of variance (ANOVA) followed by Duncan’s multiple range test for pairwise comparisons. Probability of 0.05 or less was considered to be statistically significant.
Copyright r 1999 by Academic Press All rights of reproduction in any form reserved.
FIG. 2. RIA analysis of plasma LH levels during development of R. esculenta. Sample sizes are given in parentheses. Each point is the mean ⫾ 1 SEM. Different superscripts indicate significant differences (a vs b ⫽ P ⬍ 0.001).
117
FSH and LH during Frog Development
3.35–6.3 ng/ml FSH and 3.23–5.38 ng/ml LH in Stage 30, 5.12–11.15 ng/ml FSH and 4.4–8.19 ng/ml LH in Stage 31, 5.45–6.98 ng/ml FSH and 4.51–7.56 ng/ml LH in Stage 32–33, and 4.77–6.07 ng/ml FSH and 4.54–7.23 ng/ml LH in postmetamorphic froglets (Stage P). The geometric means and ⫾ the SEM of plasma FSH and LH concentrations at each developmental stage are presented in Figs. 1 and 2, respectively. Within each stage of development, except in postmetamorphic froglets, the geometric mean of the plasma FSH level was comparatively higher (not significant statistically) than that of plasma LH. Developmental stage-related fluctuation in plasma FSH and LH was highly significant by the Kruskal– Wallis test (FSH: KW ⫽ 22.135, P ⫽ 0.0001; LH: KW ⫽ 23.124, P ⫽ 0.0001) as well as by ANOVA followed by Duncan’s multiple range test (FSH: F between group ⫽ 9.009, df ⫽ 5, N ⫽ 40, 0.05 ⬍ P ⬍ 0.01; LH: F between group ⫽ 8.419, df ⫽ 5, N ⫽ 40, 0.05 ⬍ P ⬍ 0.01) Figure 1 shows that of all stages examined, plasma concentration of FSH was lowest in Stage 26–28 tadpoles. It increased progressively through pre- and prometamorphosis (Stages 29 and 30, respectively) and showed a highly significant peak at the beginning of the metamorphic climax (Stage 31 vs 26–28, 29, or 30: P ⬍ 0.01). This peak followed a statistically significant decline at Stage 32–33 (P ⬍ 0.05) and in postmetamorphic froglets (P ⬍ 0.01). However, FSH values at Stages 32–33 and P did not differ and were significantly higher than those at Stage 26–28 (0.05 ⬍ P ⬍ 0.01). The pattern of changes in plasma LH levels was essentially similar to that observed for plasma FSH (Fig. 2). In fact, plasma LH levels increased significantly at Stage 31 (P ⬍ 0.01) compared to those at Stages 26–28, 29, and 30. Furthermore, in contrast to plasma FSH values which declined in late metamorphic climax, plasma LH values remained high until after metamorphosis. Indeed, plasma LH values at Stages 31, 32–33, and P represented a plateau and were all significantly higher than those at Stages 26–28 and 29 (0.05 ⬍ P ⬍ 0.01). It is thus plausible that, during development, peak plasma levels of both gonadotropins are reached during the metamorphic climax. The overall increase in circulating levels of FSH and LH during the metamorphic climax and soon after was hardly twofold over that observed in Stage 26–28 tadpoles.
FIG. 3. Pituitary FSH content determined by RIA during development of R. esculenta. Sample size is three for each stage. Each point is the mean ⫾ 1 SEM. Different superscripts indicate significant differences (a vs b ⫽ 0.05 ⬍ P ⬍ 0.01).
Pituitary Content of FSH and LH during Development (Figs. 3, 4) Radioimmunoassayable amounts of both FSH and LH were present in the pituitary of the earliest tadpole stage examined (Stage 26–28). Developmental changes in the pituitary content of both gonadotropins were significant by the Kruskal–Wallis test (FSH: KW ⫽ 15.409, P ⫽ 0.009; LH: KW ⫽ 13.491, P ⫽ 0.02). ANOVA followed by Duncan’s multiple range test
FIG. 4. Pituitary LH content during development of R. esculenta. Sample size is three for each stage. Each point is the mean ⫾ 1 SEM. Different superscripts indicate significant differences (a vs b ⫽ 0.05 ⬍ P ⬍ 0.01).
Copyright r 1999 by Academic Press All rights of reproduction in any form reserved.
118
showed that, although progressively increasing during development, the pituitary content of FSH and LH did not change significantly until the completion (total tail resorption) of metamorphosis, when a sharp and highly significant increase occurred (FSH: F between group ⫽ 5.621, F within group ⫽ 1.052, df ⫽ 5, N ⫽ 18; LH: F between group ⫽ 5.209, F within group ⫽ 1.641, df ⫽ 5, N ⫽ 18). In fact, pituitary FSH and LH content in postmetamorphic froglets was significantly higher than that at any earlier stage of development (0.05 ⬍ P ⬍ 0.01). In addition, the pituitary content of FSH within any stage of development was of several magnitudes higher than LH, the FSH:LH ratios being 3:1–10:1 at Stage 26–28, 6:1–23:1 at Stage 29, 10:1–13:1
Fiorentino et al.
at Stage 30, 11:1–18:1 at Stage 31, 7:1–13:1 at Stage 32–33, and 9:1–24:1 at Stage P. The total FSH content of the pituitary from newly metamorphosed froglets (Stage P) was nearly 15 times higher than that from the earliest tadpole stage (Stage 26–28) examined (Fig. 3). In comparison, LH content of the pituitary of froglets was barely sixfold that of Stage 26–28 tadpoles (Fig. 4).
Immunohistochemistry of Pituitary Gonadotropes (Figs. 5a–5d) Using monoclonal FSH and LH, only FSHimmunoreactive cells were seen in the developing pituitary of tadpoles from Stages 26–28, 29, and 30
FIG. 5. (a) FSH-immunoreactive cells in the pars distalis of the pituitary of a Stage 27 tadpole. Scale, ⫻80. (b) FSH-immunoreactive cells in the pituitary of a Stage 29 tadpole. Scale, ⫻80. (c and d) FSH- and LH-immunoreactive cells in consecutive sections of the pituitary of a postmetamorphic froglet. Scale, ⫻110.
Copyright r 1999 by Academic Press All rights of reproduction in any form reserved.
FSH and LH during Frog Development
(Figs. 5a, 5b). A few weakly immunoreactive LH cells were revealed for the first time in tadpoles at the beginning of metamorphic climax (Stage 31). In early premetamorphic tadpoles (Stage 26–28) the relative frequency of FSH cells was rather low, being 0.9–1.6% of pituitary cells. Their number increased progressively through stages 29 and 30 of development (1.8– 2.6 and 3.8–4.9%, respectively). During the metamorphic climax (Stages 31 and 32–33) and in postmetamorphic froglets the percentage of gonadotropes increased remarkably (6.1–9.9, Fig. 5c) and LH gonadotropes were clearly discernible. The latter, very few in number during early climax (0.6–0.9%), were almost as numerous in postmetamorphic froglets as FSH cells (8.3–9.3%, Fig. 5d). As previously demonstrated in our laboratories (Pinelli et al., 1996; Tanaka et al., 1990), FSH and LH-immunoreactivity was colocalized in many pituitary gonadotropes (Figs. 5c, 5d).
DISCUSSION The present study, using specific radioimmunoassay, is the first to quantify circulating levels of FSH and LH in an amphibian during larval development until the completion of metamorphosis. A major finding of this study is that plasma levels of FSH and LH increased significantly at the onset of metamorphic climax. The pituitary content of both gonadotropins increased significantly at the completion of metamorphosis. However, it is not feasible to ascertain from the foregoing results if the ontogenetic changes in plasma FSH and LH levels are associated with changes in synthesis and/or release. The presence of radioimmunoassayable FSH in the pituitary and plasma of early tadpoles is consistent with previous immunohistochemical analysis of developing pituitary in which a few FSH-immunoreactive cells were detected (Pinelli et al., 1996). Similarly, both studies, viz the present study and that by Pinelli et al. (1996), are consistent in that LH-immunoreactive cells become detectable by immunohistochemistry starting at the metamorphic climax. However, LH was quantified, using the same monoclonal antibody, by radioimmunoassay in the plasma as well as in the pituitary of early tadpoles. It may be argued that either low LH content of pituitary cells escapes detection by
119
immunohistochemistry because it is below the limit of the immunohistochemistry technique or radioimmunoassay is more sensitive and can detect low amounts of this gonadotropin. In a previous analysis of brain GnRH during development of R. esculenta, a similar situation was encountered, and radioimmunoassay revealed GnRHs in the brain of tadpoles much earlier than did immunohistochemistry (D’Aniello et al., 1995; Di Fiore et al., 1996). As is the case in other vertebrates, there is substantial in vivo and in vitro evidence in amphibians to indicate GnRH regulation of pituitary gonadotropic function and involvement of sex steroids in GnRHstimulated pituitary FSH and LH secretion (see Pavgi and Licht, 1993; Stamper and Licht, 1993, 1994; and references therein). The question is how the developmental pattern of pituitary gonadotropin synthesis and release relates to changes in brain GnRH production and, ultimately, to gonadal activity (gametogenesis and sex steroid production). Coincident with a significant rise in brain GnRH content and plasma levels of FSH and LH only after prometamorphosis, we are inclined to believe that the brain–pituitary– gonadal axis becomes a functional unit as the metamorphic climax sets in. This is sustained by the fact that mGnRH axons in this frog reach the median eminence at around the time when tadpoles begin to metamorphose (see D’Aniello et al., 1995). Support for this concept is also provided by our unpublished observations that two intraperitoneal injections of mGnRH, given on 2 consecutive days, do not induce FSH/LH secretion in the prometamorphic tadpoles, whereas they do during the climax. This would lead us to infer that the frog pituitary becomes responsive to GnRH only when the tadpoles begin to metamorphose and not before. Furthermore, during development, brain GnRH content remains low until around prometamorphosis (Di Fiore et al., 1996), at which time it begins to increase, showing a substantial elevation during the metamorphic climax and again at the completion of metamorphosis (Di Fiore et al., 1996). As far as the plasma levels of sex steroids are concerned, we have preliminary data to show that premetamorphic tadpoles contain 44 ⫾ 18 pg/ml of androgens and 54 ⫾ 16 pg/ml of estradiol-17, whereas at the term of metamorphic climax the circulating levels of sex steroids are significantly higher at 166 ⫾ 32 pg/ml of andro-
Copyright r 1999 by Academic Press All rights of reproduction in any form reserved.
120
gens and 250 ⫾ 81 pg/ml of estradiol-17 (Di Fiore, Pinelli, D’Aniello and Rastogi, unpublished data). Thus, in Stage 26–28 tadpoles, FSH and LH are detectable in the pituitary and plasma as early as the GnRH is detectable in the brain and the sex steroids are detectable in the plasma. Our present results and those published earlier (D’Aniello et al., 1995; Di Fiore et al., 1996) support that GnRH and FSH/LH are most likely produced only after hatching has occurred. It has also been established that tadpole (pre- and prometamorphic) gonads possess the capacity to synthesize and secrete the sex hormones, an activity independent of the pituitary until metamorphosis (see Hsu et al., 1985). This would allow us to assume that during pre- and prometamorphic development each endocrine component of the hypothalamus–pituitary–gonadal axis grows, differentiates, and functions independently. Indeed, it was proposed several decades ago that during embyronic development of vertebrates, the main steps of the endocrine axes are initially autonomous and that only at a later time during development are endocrine feedback systems established (Willier, 1939, 1955). In the chick, it was conclusively determined that the endocrine components of the brain– pituitary–gonadal axis function independently of each other during early development (see Woods and Thommes, 1998). Analogously, although hormones typical of the brain–pituitary–gonadal axis, viz. GnRH, pituitary gonadotropins, and sex steroids, were measured in great tit nestlings until 9 days of age, the hypothalamus– pituitary axis becomes a functional unit at a later time (Silverin and Sharp, 1996). Among vertebrate species, the developmental dynamics of the brain–pituitary–gonadal axis of the frog seem to differ remarkably from that of mammals, e.g., the rat. In fact, in the latter, brain GnRH is detectable as early as day 12 of gestation, at which time GnRH binding sites can be found in the developing anterior pituitary, while pituitary LH becomes detectable around fetal day 17 and FSH around fetal day 19 (see Ojeda and Urbanski, 1994). In this mammal, it has been suggested that brain GnRH plays a trophic function essential for the differentiation of pituitary gonadotropes. In birds and amphibians, this does not seem to be the case or is, at best, simply a matter of speculation. In the frog, even though it is clear that, during larval development, the brain contains radioimmunoas-
Copyright r 1999 by Academic Press All rights of reproduction in any form reserved.
Fiorentino et al.
sayble GnRH, we have no evidence to suggest an active involvement of GnRH in the control of the tadpole pituitary. So far, there is no evidence of the presence of GnRH receptors in the tadpole pituitary. On the other hand, within fish it is suggested that the brain–pituitary–gonadal axis is intact at the time of hatch in the rainbow trout, but conclusive experimental evidence is still necessary (see Feist and Schreck, 1996). In conclusion, it becomes obvious that, among vertebrates, there are group-specific differences in the temporal appearance of hormone-producing units of the brain–pituitary–gonadal axis and in the establishment of endocrine feedback mechanisms.
ACKNOWLEDGMENTS Financial aid from MURST and Universita` degli Studi di Napoli Federico II is appreciated.
REFERENCES D’Aniello, B., Pinelli, C., Di Fiore, M. M., Iela, L., King, J. A., and Rastogi, R. K. (1995). Development and distribution of gonadotropin-releasing hormone neuronal systems in the frog (Rana esculenta) brain: Immunohistochemical analysis. Dev. Brain Res. 89, 281–288. Di Fiore, M. M., King, J. A., D’Aniello, B., and Rastogi, R. K. (1996). Immunoreactive mammalian and chicken-II GnRHs in Rana esculenta brain during development. Regul. Pept. 62, 119–124. Feist, G., and Schreck, C. B. (1996). Brain–pituitary–gonadal axis during early development and sexual differentiation in the rainbow trout, Oncorhynchus mykiss. Gen. Comp. Endocrinol. 102, 394–409. Hsu, C.-Y., Chang, L.-T., Ku, H.-H., and Lu, M.-H. (1985). In vitro estradiol synthesis and secretion by tadpole ovaries of different developmental stages. Gen. Comp. Endocrinol. 57, 393–396. Itoh, M., Inoue, M., and Ishii, S. (1990). Annual cycle of pituitary and plasma gonadotropins and plasma sex steroids in a wild population of the toad, Bufo japonicus. Gen. Comp. Endocrinol. 78, 242–253. Itoh, M., and Ishii, S. (1990). Changes in plasma levels of gonadotropins and sex steroids in the toad, Bufo japonicus, in association with behavior during the breeding season. Gen. Comp. Endocrinol. 80, 451–464. Kar, S., and Naik, D. R. (1986). Cytodifferentiation and immunocharacteristics of adenohypophsial cells in the toad, Bufo melanostictus. Anat. Embryol. 175, 137–146.
FSH and LH during Frog Development Kim, J. W., Im, W. B., Choi, H. H., Ishii, S., and Kwon, H. B. (1998). Seasonal fluctuations in pituitary gland and plasma levels of gonadotropic hormones in Rana. Gen. Comp. Endocrinol. 109, 13–23. King, J. A., and Millar, R. P. (1981). TRH, GH-RH, and LH-RH in metamorphosing Xenopus laevis. Gen. Comp. Endocrinol 44, 20–27. King, J. A., and Millar, R. P. (1997). Coordinated evolution of GnRHs and their receptors. In ‘‘GnRH Neurons: Gene to Behavior’’ (I. S. Parhar and Y. Sakuma, Eds.), pp. 51–77. Brain Shuppan, Tokyo. Licht, P., McCreery, B. R., Barnes, R., and Pang, R. (1983). Seasonal and stress related changes in plasma gonadotropins, sex steroids, and corticosterone in the bullfrog, Rana catesbeiana. Gen. Comp. Endocrinol. 50, 124–145. Moriceau-Hay, D., Doerr-Schott, J., and Bubois, M. P. (1982). Immunohistochemical demonstration of TSH-, LH-, and ACTH-cells in the hypophysis of tadpoles of Xenopus laevis. Cell Tissue Res. 225, 57–64. Muske, L. E. (1997). Ontogeny, phylogeny and neuroanatomical organization of multiple molecular forms of GnRH. In ‘‘GnRH Neurons: Gene to Behavior’’ (I. S. Parhar and Y. Sakuma, Eds.), pp. 145–180. Brain Shuppan, Tokyo. Muske, L. E., and Moore, F. L. (1990). Ontogeny of immunoreactive gonadotropin-releasing hormone neuronal system in amphibians. Brain Res. 534, 177–187. Northcutt, R. G., and Muske, L. E. (1994). Multiple embryonic origins of gonadotropin-releasing hormone (GnRH) immunoreactive neurons. Dev. Brain Res. 78, 279–290. Ogawa, K., Suzuki, E., and Taniguchi, K. (1995). Immunohistochemical studies on the development of the hypothalamo–hypophysial system in Xenopus laevis. Anat. Rec. 241, 244–254. Ojeda, S. R., and Urbanski, H. F. (1994). Puberty in the rat. In ‘‘The Physiology of Reproduction’’ (E. Knobil and J. D. Neill, Eds.), 2nd ed., pp. 363–409. Raven Press, New York. Park, M. K., Tanaka, S., Hayashi, H., Hanaoka, Y., Wakabayashi, K., and Kurosumi, K. (1987). Production and characterization of a monoclonal antibody against -subunit of bullfrog lutropin. Gen. Comp. Endocrinol. 68, 82–90. Pavgi, S., and Licht, P. (1989). Effects of gonadectomy and steroids on pituitary gonadotropin secretion in a frog, Rana pipiens. Biol. Reprod. 40, 40–48. Pavgi, S., and Licht, P. (1993). Inhibition of in vitro pituitary gonadotropin secretion by 17-estradiol in the frog Rana pipiens. Gen. Comp. Endocrinol. 89, 132–137. Pinelli, C., Fiorentino, M., D’Aniello, B., Tanaka, S., and Rastogi, R. K. (1996). Immunohistochemical demonstration of FSH and LH in the pituitary of the developing frog, Rana esculenta. Gen. Comp. Endocrinol. 104, 189–196. Polzonetti-Magni, A. M., Mosconi, G., Carnevali, O., Yamamoto, K.,
121 Hanaoka, Y., and Kikuyama, S. (1998). Gonadotropins and reproductive function in the anuran amphibian, Rana esculenta. Biol. Reprod. 58, 88–93. Rastogi, R. K., and Iela, L. (1994). Gonadotropin-releasing hormone: Present concepts, future directions. Zool. Sci. 11, 363–373. Rastogi, R. K., di Meglio, M., and Iela, L. (1990). Immunoreactive luteinizing hormone-releasing hormone in the frog (Rana esculenta) brain: Distribution pattern in the adult, seasonal changes, castration effects, and developmental aspects. Gen. Comp. Endocrinol. 78, 444–458. Silverin, B., and Sharp, P. (1996). The development of the hypothalamic–pituitary–gonadal axis in juvenile great tits. Gen. Comp. Endocrinol. 103, 150–166. Stamper, D. L., and Licht, P. (1993). Further studies on the influence of GnRH on the biosynthesis of gonadotropins in female frogs (Rana pipiens). Gen. Comp. Endocrinol. 92, 104–112. Stamper, D. L., and Licht, P. (1994). Influence of androgen on the GnRH-stimulated secretion and biosynthesis of gonadotropins in the pituitary of juvenile female bullfrogs, Rana catesbeiana. Gen. Comp. Endocrinol. 93, 93–102. Tanaka, S., Hanaoka, Y., and Wakabayashi, K. (1983). A homologous radioimmunoassay for basic gonadotropin. Endocrinol. Jpn. 30, 71–78. Tanaka, S., Park, M. K., Hayashi, H., Hanaoka, Y., Wakabayashi, K., and Kurosumi, K. (1990). Immunocytochemical localization of the subunits of glycoprotein hormones (LH, FSH, and TSH) in the bullfrog pituitary gland using monoclonal antibodies and polyclonal antiserum. Gen. Comp. Endocrinol. 77, 88–97. Tanaka, S., Sakai, M., Park, M. K., and Kurosumi, K. (1991). Differential appearance of the subunits of glycoprotein hormones (LH, FSH, and TSH) in the pituitary of bullfrog (Rana catesbeiana) larvae during metamorphosis. Gen. Comp. Endocrinol. 84, 318–327. Willier, B. H. (1939). The embryonic development of sex. In ‘‘Sex and Internal Secretions’’ (E. Allen, C. H. Danforth, and E. A. Doisy, Eds.), 2nd ed. pp. 64–144. Williams & Wilkins, Baltimore. Willier, B. H. (1955). Ontogeny of endocrine correlation. In ‘‘Analysis of Development’’ (B. H. Willier, P. A. Weiss, and V. Hamburger, Eds.), pp. 574–619. Saunders, Philadelphia. Witschi, E. (1956). ‘‘Development of Vertebrates.’’ Saunders, Philadelphia. Woods, J. E., and Thommes, R. C. (1998). Avian reproductive system, developmental endocrinology. In ‘‘Encyclopedia of Reproduction’’ (E. Knobil and J. D. Neill, Eds.), Vol. 1, pp. 337–347. Academic Press, San Diego. Yamamoto, K., Yamashita, K., Hayakawa, Y., Hanaoka, Y., and Kikuyama, S. (1995). Development and application of a homologous radioimmunoassay for Xenopus prolactin. Gen. Comp. Endocrinol. 99, 28–34.
Copyright r 1999 by Academic Press All rights of reproduction in any form reserved.