Immunocytochemical and morphometric study of prolactin cells during amphibian morphogenesis

GENERAL

AND

COMPARATIVE

ENDOCRINOLOGY

69, 188-196 (1988)

lmmunocytochemical and Morphometric Study of Prolactin during Amphibian Morphogenesis’

Ceils

SOCORROGARCIA-NAVARRO, MARIA M. MALAGON, GREGORIO GARCIA-HERDUGO, AND FRANCISCO GRACIA-NAVARRO Department of Cell Biology, Faculty of Sciences, University of Cdrdoba, Cdrdoba, Spain Accepted September 4, 1987 The peroxidase-antiperoxidase (PAP) immunocytochemical technique and rabbit antihuman prolactin (PRL) antiserum were used to localize and identify PRL-producing cells in the pars distalis of Bufo calamita, Hyla meridionalis, Alytes cisternasii, Pelobates cultripes, and Rana perezi tadpoles at different stages of development as well as in l-year-old postmetamorphic animals. This cell type was located throughout the gland in P. cultripes and R. perezi and in the caudal two-thirds in B. calamita, H. meridionalis, and A. cisternasii in premetamorphic animals. These distribution patterns do not show changes throughout development. Morphometry was used to evaluate the changes observed in pars distalis volume and stereological parameters of PRL immunoreactive cells during development. Pars distalis volume increased during the larval growth period and decreased throughout the metamorphic climax. PRL volume density and cellular area showed different patterns in the different species, although the correlation between these parameters suggests a period of great proliferative rate followed by changes in cellular size. The changes observed in PRL total volume suggest the existence of two phases in amphibian development: (i) a period of PRL storage during pre- and prometamorphosis and (ii) a period of release at the metamorphic climax. 0 1988 Academic Press, Inc.

Immunocytochemical methods, in conjunction with more conventional experimental manipulations, have been used in the study of adenohypophyseal cells in adult anurans but in only a few cases have these techniques been applied to tadpoles (Doerr-Schott, 1980; Girod, 1983). In premetamorphic stages, cells immunoreactive to mammalian prolactin (PRL) antisera have been identified in Alytes obstetricans tadpoles in the anterior two-thirds of the adenohypophysis (RCmy and Dubois, 1973; Guyetant et al., 1977); in the rostra1 zone in Xenopus laevis (Moriceau-Hay et al., 1979), and in the whole pars distalis in Nectophrynoi’des occidentalis (ZuberVogeli and Doerr-Schott, 1984). The investigation in later stages demonstrates ’ This work has been subsidized by Grant 2184-83 conceded by the ComisGn Asesora (CAICYT) of Spain.

that the area occupied by this cell type increases rapidly and regularly during amphibian development. PRL has been well documented with respect to its growth-promoting and antimetamorphic effects in amphibian larvae (see review in Dodd and Dodd, 1976; White and Nicoll, 1981). PRL levels have been assumed to be high during pre- and prometamorphosis and then begin to decline once metamorphosis starts (Etkin, 1968). Nevertheless, Clemons and Nicoll (1977b) and Yamamoto and Kikuyama (1982a) have observed that bullfrog PRL plasma levels are relatively low during preand prometamorphosis and high at late climax. In a later study on the synthesis and storage of PRL in the pituitary gland during metamorphic climax, Yamamoto et al. (1986b) concluded that during the climax period, bullfrogs acquire the ability to synthesize, store, and release a considerable 188

0016-6480/88 $1.50 Copyright All rights

8 1988 by Academic Press, Inc. of reproduction in any form reserved.

PROLACTIN-PRODUCING

CELLS IN AMPHIBIAN

amount of PRL. The physiological importance of these data is yet to be made clear. Stereological methods could be an instrument that would permit the study of adenohypophyseal cell changes during amphibian morphogenesis in small-sized specimens where radioimmunoassay (RIA) and other physiological methods could be applied only with difficulty. Nevertheless, no morphometric studies concerning the evolution of adenohypophyseal cell types during amphibian morphogenesis have been reported. The aim of the present investigation is to carry out a comparative study of the occurrence, anatomical distribution, and morphometric changes of PRL immunoreactive cells in five species of amphibian anurans during developmental stages (pre-, pro-, and metamorphosis) and in postmetamorphic animals, to establish the nature of a common behavior of PRL cells during amphibian metamorphosis. MATERIALS

AND METHODS

Animals. Tadpales and juvenile animals of five species of amphibian anura @r&o calamitu, Hyla meridionalis, Pelohates crrltripes, Alytes cisternasii, and Rana peuezi) have been used. Eggs and tadpoles (in the case of A. cisteunasii) were captured in spring and bred in the laboratory at a constant temperature (I8 t 2”) and photoperiod (13 hr in light and 11 hr in darkness). Feeding and water renewal occurred every day. The animals were selected at different stages of development, as defined by Gosner (196(l), from the time of hatching to the end of metamorphosis: 12-month-old postmetamorphic animals were used as well. The animals were sacrificed under MS-222 (Sandoz) anesthesia. Their skulls were fixed in sublimate containing Bouin-Hollande fluid for 2 days, then dehydrated and embedded in paraffin. Serial sections (5 pm) were cut in sagittal planes and mounted in a sequential set on gelatinized slides. Irnm~~nocyfochernicaI technique. PRL immunoreactive cells were localized in histological sections by using the peroxidase-antiperoxidase (PAP) complex immunocytochemical technique (Sternberger et a/., 1970). The sections were dewaxed, hydrated gradually, and washed in Tris-buffered saline (TBS, 0.05 M, pH 7.6). Then the sections were treated sequentially

MORPHOGENESIS

189

with HZO, (3% in distilled water; for 30 min at room temperature); normal goat serum (30 min at room temperature) to eliminate background staining; rabbit antihuman PRL serum (l:lOOO, 48 hr at it” in moist chamber, NIADDK); goat anti-rabbit serum (l:lOO, 30 min at room temperature, Behring Institute); rabbit peroxidase-antiperoxidase soluble complex (PAP. l:lOO, 30 min at room temperature, Dako); and 3.3’-diaminobenzidine tetrahydrochloride (22 mg. Sigma) in Tris buffer (174 ml, 0.05 M, pH 7.6) with H,O, (0.03%) applied under agitation for 8 min at room temperature. Subsequently. sections were dehydrated and coverslipped. Specificity control of the antiserum was checked by treating adjacent sections with normal goat serum or with antiserum preadsorbed with homologous or heterologous hormones. Morphotnetric study. The stereologicai study was carried out using the hypophysis of tadpoles belonging to stage 28 (at which all the animals show immunoreactive PRL cells). stage 34 (the end of premetamorphosis), stage 40 (prometamorphosis). stage 42 (the beginning of metamorphic climax), stage 44 (midclimax), and stage 46 (the end of climax); 12-month postmetamorphic animals were included as well. The stereological study was performed with an Automatic Image Analyser IBAS (Kontron, FRG) using area analysis (Weibel, 1980). Four parameters were measured: (a) The pars distalis volume (PDV): The area occupied by all pars distalis sections of an animal (five per stage and species) were measured. To calculate the total volume of pars distalis in each animal, the following formula was applied:

t is the section thickness (always the same) and I;A is the sum of the areas. The mean and standard error of five pars distalis volumes per stage and species have been calculated. (b) Cellular section area (CA): To calculate cellular area, 30 PRL immunoreactive cells showing their nucleus were measured per animal (150 per stage and species). (c) Volume density (Vv) of PRL cells: This stereological parameter expresses the percentage of pars distalis voiume occupied by PRL cells. It was calculated using the formula Vv = AC/At, where AC is the profile area sections of PRL immunostained cells and At is the pars distalis area. Six sections of pars distalis from each of the five animals used per stage and species have been measured. The mean and standard error For a stage were calculated from the means and standard errors of all animals belonging to that stage. (d) Total volume occupied by PRL cells (PRLV): To calculate this parameter, the mean Vv and its standard error per animal (calculated from six sections) were

190

GARCIA-NAVARRO

multiplied by the PDV of the same animal to obtain the total volume occupied by PRL cells in its pars distalis. The mean PRLV and standard error for each stage were calculated from the means and standard errors of five animals of that stage. Data were analyzed statistically using the following programs written for the IBAS computer console: nested analysis of variance, the Student I test or, for nonhomogeneous data, the Kolmogorov-Smirnov U test.

RESULTS

Immunocytochemical

Controls

No reaction was observed when sections were incubated with normal goat serum instead of the primary antibody. The inhibition of human PRL (hPRL) antiserum with its homologous hormone abolished immunostaining, while the inhibition of hPRL antiserum with human TSH, ACTHi-,,, rat FSH, ovine LH, and human STH antisera neither suppressed nor diminished immunoreactivity. Cells Revealed

by hPRL Antiserum

Rabbit antiserum to human PRL stains numerous cells in the pars distalis which are polymorphous, average in size, and possess a nucleus that is generally spheroid and eccentric, The first signs of immunoreaction appear in stages 25-27 in the different species studied. In pars distalis midsagittal sections of premetamorphic animals, PRL immunoreactive cells are located in the posterior twothirds in Bufo calamita (Fig. la), Hyla meridionalis (Fig. 2a), and Alytes cisternasii (Fig. 3a). In Pelobates cultripes (Fig. 4a) and Rana perezi (Fig. 5a) a positive reaction is seen in cells distributed throughout the pars distalis. During development, PRL cells increase in number and at the end of metamorphosis they are present in the same distribution as in premetamorphic stages (Figs. lb, 2b, 3b, 4b, 5b).

ET

AL.

Pars Distalis

Volume

In all the species studied, the pars distalis volume increased until late prometamorphosis (Stage 40). Moreover, with the sole exception of R. perezi, growth continued, showing a maximum volume (three to six times larger than in premetamorphic animals) at the onset of metamorphic climax (stage 42). During climax, pars distalis size remained constant in H. meridionalis and decreased in the other species. The pars distalis was larger at postmetamorphosis than at late climax, except in P. cultripes in which we did not find significant differences (Fig. 6). Volume Density of PRL Immunoreactive Cells Although PRL Vv shows different patterns in the species studied, we can divide these patterns into two groups of behavior. In A. cisternasii, H. meridionalis, and B. calamita, Vv values of PRL cells increased during pre- and prometamorphosis, reaching a maximum value at stage 40. A decline in Vv was observed at the beginning or middle of climax. These values remained low until the end of metamorphosis. In these species, postmetamorphic animals showed a considerable increase (Fig. 7). After a decrease at the end of premetamorphosis (stage 34), PRL Vv values increased in R. perezi from stage 40 to the end of the metamorphic climax, and then decreased at postmetamorphosis; in P. cultripes Vv increased gradually from stage 28 to midmetamorphosis (stage 44), then decreased at stage 46 and remained low at postmetamorphosis (Fig. 8). Cellular Area of PRL Cells In A. cisternasii, H. meridionalis, and B. calamita, the cellular area of PRL immunoreactive cells showed a decrease from premetamorphosis to late prometamorphosis that continued during the metamorphic

PROLACTIN-PRODUCING

CELLS

IN

AMPHIBIAN

MORPHOGENESIS

FIGS. 1-5. Mediosagittal sections through the hypophysis showing the localization of PRL immu(Fig. 2), Alytes cisternusii (Fig. 3), Raru~ noreactive cells in Bufo calamita (Fig. I), Hyla meridionalis perezi (Fig. 4), and Pelobates cultripes (Fig. 5) tadpoles. Rostra1 to left. (a) Hypophysis at premetamorphosis (stage 28). (b) Hypophysis at metamorphic climax (stage 44). me, median eminence; pd, pars distalis; pi, pars intermedia. x 110.

*. L

192

GARCIA-NAVARRO

21.106

17.106

ET

AL.

l

B.

calamita

A H. meridionalis 0 A. cisternasii

0.35

. /

l

13.106

*fi -. “a s :

.

“s 2 c e

0.25

s l

0.15

9.106 l l

r;,L

l

S.lOE

28

.

34 -PRE-

-._t

l

l I

&

k,

, 28

*

34 -

PRE-

-

PRO

-

40 42 44 46 -CLIMAX -

STAGE

FIG. 6. Changes in pars distalis volume (PDV) during development of Bufo calamitu (O), Hylu meridionnlis

(A),

44

46

Postm.

I

0

6-

42

+-ClJ&fAX-+

STAGE l .

1.10

40 --PRO-

Alyres

cisternasii

(Cl), Ranu

perezi

(A),

and Pelobates c&tripes (B). Abscissa: developmental stage. Ordinate: pars distalis volume (km3). Standard errors, not represented in this case, were fess than 10% of the mean in each group. *Significantly different at P < 0.05 versus preceding stage by F and I tests (N = 5).

climax in B. calamita. In A. cistern&i and H. meridionalis, this parameter fluctuated during climax and increased in postmetamorphic animals (Fig. 9). In R. perezi and P. c&tripes, PRL cell size decreased in premetamorphosis and later increased during prometamorphosis. At the beginning of climax, this parameter showed a decrease that continued until late climax in P. cultripes, whereas it increased from mid- to late climax in R. perezi (Fig. IO). Volume Occupied

FIG. 7. Changes in volume density (Vv) of PRL immunoreactive cells in Bufo calamita (O), Hyin meridionalis (A), and Alytes cisternasii (Cl) throughout development. Abscissa: developmental stage. Ordinate: volume density (pm3/pm3). Standard errors, not represented in this case, were less than 10% of the mean in. each group. *Significantly different at P < 0.05 versus preceding stage by F and t tests (N = 5).

by PRL Cells

In all the species except B. calamita,

the total volume occupied by PRL cells showed a considerable increase until the beginning of the metamorphic climax (Fig. 11). In B. calamita, the volume occupied by PRL cells increased until late prometamorphosis. At the onset of climax, it showed a decrease that continued until the end of metamorphosis. In postmetamorphic animals, PRL cells occupied a volume similar to that observed at stage 40. In P. cultripes, a decrease was observed throughout the metamorphic climax which continued in postmetamorphic animals. PRL volumes of R. perezi and H. meridionalis decreased slightly at midclimax. At the end of metamorphosis they showed an increase that continued in postmetamorphic animals. The maximum PRL volume reached by A. cisternasii at stage 42 remained constant until midmetamorphosis, decreasing at the

PROLACTIN-PRODUCING

CELLS

IN

AMPHIBIAN

193

MORPHOGENESIS

/ 28

34 -PRE-

40 -

PFiO-

4’2 44 -CLIMAX

46 -

Postm

STAGE

,A / 28

34 -PRE-

40 42 44 - -CLIMAX

-PRO

46 -

Postm

STAGE

8. Changes in volume density of PRL immunoreactive ceils in Iiutru pewzi (LI) and Pelobures cnltripes (m) during development. Abscissa: developmental stage. Ordinate: volume density (pms/p.m’). Standard errors, not represented in this case. were less than 10% of the mean in each group. *Significantly different at P < 0.05 versus preceding stage by F and t tests (N = 5). FIG.

FIG. 10. Changes in PRL cellular area (CA) in Rnn~l perezi (A) and P&hates cc&ripes (M) during development. Abscissa: developmental stage. Ordinate: ceiMar section area (km’). Standard errors. not represented in this case, were less than 10% of the mean in each group. “Significantly different at P < 0.05 versus preceding stage by f and F tests (N = 150).

end of the same period. Postmetamorphic animals showed a considerable increase, reaching values greater than those observed at the beginning of climax. DISCUSSION

34 -Pm--

-

PRO -

40 42 44 -CLIMAX

46 -

Postm

STAGE

9. Changes in PRL cellular area (CA) in BrQo calamiru (O), Hyla meridionalis (A), and Alytes cisternasii (Ii’) during development. Abscissa: developmental stage. Ordinate: cellular section area (pm’). Standard errors, not represented in this case, were less than 10% of the mean in each group. *Significantly different at P < 0.05 versus preceding stage by r and F tests uv = 150). FIG.

The use of anti-mammalian PRL-antibodies to identify and locate amphibian PRL immunologically rests on the assumption that the hormones in the two vertebrate groups are chemically very closely related. Polyacrylamide gel electrophoresis shows that PRL of Rana cntesOeiurm (Nicoll and Nichols, 1971; Yamamoto and Kikuyama, 1981) and Bufo jnponicus (Yamamoto et al., 1986a) behave in the same way as mammalian PRL. On the other hand, the use of anti-mammalian PRL antiserum (Eddy and Lipner. 1975) produces the same metamorphic change acceleration as anti-frog PRL antiserum (@lemons and Nicoll, 1977a; Yamamoto and Kikuyama, 1982b). These properties indicate the close relationship between mammalian and amphibian PRL, which makes it reasonable to believe that the findings of the present study reveal the true identity of PRL and

194

_ il

GARCIA-NAVARRO

6.10

’

5.10

6

4.10

z

3.10

F

2.10

f

a 3 > 2

,

28

3’4 .-PRE-

40 W-PRO-

42

44

46

Postm.

-CLIMAX-

FIG. 11. Changes in the total volume occupied by cells during development of Bufo cahnita (O), Hyla meridionalis (A). Aiytes cisternasii (01, Rana perezi (A), Pelobates cultripes (m). Abscissa: developmental stage. Ordinate: volume occupied by PRL cells (pm3). Standard errors, not represented in this case, were less than 10% of the mean in each group. *Significantly different at P < 0.05 versus preceding stage by F and t tests (N = 5). PRL

the localization of its production sites in tadpoles. This is in accord with the extensive use of antisera against mammalian PRL as the primary antibody in immunocytochemical identification of PRL cells in amphibians (see review in Girod, 1983). There are no changes in the distribution pattern of PRL immunoreactive cells throughout development in any the species

ET

AL

studied. In R. perezi and P. &tripes, this cell type shows a localization similar to that described in Nectophrynofdes occidentalis (Zuber-Vogeli and Doerr-Schott, 1984), occupying the whole gland. In B. calamita, H. meridionalis, and A. cisternasii, this cell type is distributed in the caudal two-thirds of the pars distalis. In all species studied, pars distalis volume increased greatly throughout development until the end of prometamorphosis or the onset of the metamorphic climax, i.e., during the growth period of amphibian larvae. At climax a generalized decrease in gland size was observed. A similar nonquantitative correlation has been described previously (see Fig. 1 in Reyrel, 1970). The anatomical changes occurring at head level as well as the morphological changes suffered by the hypophysis during climax may imply an interior reorganization of the gland by diminishing intercellular spaces and/or by a decrease in cellular size/ number. Our data on diminishing PRL cellular area and the degranulation of other adenohypophyseal cell types observed by Reyrel (1970) and Van Oordt (1966) suggest that the decrease in pars distalis volume may be due to these phenomena, although the methods used do not permit us to reject other possibilities. The increase in PRL Vv observed during pre- and prometamorphosis in B. calamita, H. meridionalis, and A. cisternasii correlates with a decrease in cellular size. This correlation can be explained only by an increase in the number of PRL cells, probably as a result of an increase in the proliferative rate during these periods of development. Throughout the climax, PRL Vv showed a decrease that corresponded to increases and/or decreases in cellular area. This suggests that the decrease in Vv may be due to degranulation and/or degeneration of PRL cells instead of to an increase in the proliferative rate. In R. perezi and P. &tripes pre- and

PROLACTIN-PRODUCING

CELLS

prometamorphic tadpoles, the increase in volume density is due mainly to an increase in cellular size, although small increases in the proliferative rate cannot be rejected. In these species, PRL Vv continued to increase during metamorphic climax, showing behavior contrary to that observed in the other species. This increase correlates to a decrease in cellular size that is more marked in P. cultr’ipes than in R. perezi. These data suggest that the increase in PRL Vv is due mainly to an increase in cell number. The low blood PRL levels observed in R. cutesbeiana (Clemons and Nicoll, 1977b; Yamamoto and Kikuyama, 1982a), the high rate of PRL synthesis in the pars distalis of R. catesbeiana (Yamamoto et al., 1986b), and the great increase in total volume occupied by PRL cells during pre- and prometamorphosis all strongly suggest a high rate of PRL storage throughout the growing larval period of amphibian development . The authors mentioned above reported high PRL levels as well as a decrease in PRL synthesis in R. catesbeiana throughout the metamorphic climax. These data agree with the decrease observed in the volume occupied by PRL immunoreactive cells and lead us to conclude that PRL is released at different rates by each species during metamorphic climax. Aside from the cellular mechanisms that produce changes in PRL immunoreaction, the preceding data lead us to conclude that amphibian development is divided into two main periods: (i) a period of PRL storage (during pre- and prometamorphosis), characterized by low PRL levels in serum and increasing PRL volume; (ii) period of PRL release showing high blood PRL levels and decreasing PRL volume. All these findings contradict the theories on hormonal control of amphibian development enunciated by Etkin (1968) and Dodd and Dodd (1976) which suggest a high PRL

IN

AMPHIBIAN

MORPHOGENESIS

195

release rate during pre- and prometamorphosis and low PRL levels during climax. Further research is necessary to elucidate the actual role played by PRL during amphibian morphogenesis. The quantitative methodology (morphometry and stereology of immunostaine pituitaries) used in this paper can be considered a useful instrument permitting the study of amphibian development in those species in which physiological methods cannot be used because of their small size. ACKNOWLEDGMENTS The authors gratefully acknowledge the National Pituitary Program (NIADDK. Baltimore, MD) for the gift of the hormones and antiserum used in this study and appreciate the revision of the marxscript carried out by Mrs. M. Sullivan.

REFERENCES Clemens, G. K., and Nicoll, C. S. (1977a). Effects of antisera to bullfrog prolactin and growth hormone on metamorphosis of Rana caresheiuna. GPN. Comp. Endocrinol. 31, 495-497. Clemons, G. K.. and Nicoll. C. S. (1977b). Deveiopment and preliminary application of a homologous radioimmunoassay for bullfrog proiactin. Gen. Comp. Endocrinol. 32, 531-535. Dodd, M. H. I., and Dodd, J. M. (1976). The biology of metamorphosis. Pn “Physiology of Amphibians” (J. A. Moore. Ed.), Vol. 3. pp. 46?599. Academic Press, New York. Doerr-Schott, J. (1980). Immunohistochemistry of the adenohypophysis of non-mammalian vertebrates. Acta Histochem. SL&. 22s, 185-223. Eddy, L., and Lipner, H., (1975). Acceleration of thyroxine-induced metamorphosis by proiactin antiserum. Gen. Camp. Endocrfnol. 25, 462466. Etkin, W. (1986). Hormonal control of amphibian metamorphosis. In “Metamorphosis: A Problem in Developmental Biology” (W. Etkin and L. I. Gilbert. Eds.), pp. 313-348. Appleton. New York. Girod, C. (1983). The adenohypophysis of amphibians. In “Immunocytochemistry of the Vertebrate Adenohypophysis,” pp. 83-106. G. Fischer Verlag, Stuttgart. Gosner, K. L. (1960). A simplified table of staging anuran embryos and larvae with notes on identification. Herpetologicn 16, 183-190. Guyetant, R., Bugnon, C., Fellmann, D.: and Block, B. (1977). Etude cytoimmunologique comparative

196

GARCIA-NAVARRO

des cellules a prolactine et des cellules a somatohormone chez des tetards d’illytes obstetricans (L) eleves en groupe ou isoles. C.R. Acad. Sci. Paris 285, 255-262. Moriceau-Hay, D., Doerr-Schott, J., and Dubois, M. P. (1979). Mise en evidence par immunofluorescence des cellules a prolactine et des cellules somatotropes dans l’hypophyse du tetard de Xenopus, (Xenopus luevis D). Gen. Comp. Endocrinol. 39, 322-326. Nicoll, C. S., and Nichols, C. W., Jr. (1971). Evolutionary biology of prolactins and somatotropins. I. Electrophoretic comparison of tetrapod prolactins. Gen. Camp. Endocrinol. 17, 300-310. Remy, C., and Dubois, M. P. (1973). Les cellules somatotropes et les cellules a prolactine de l’hypophyse du tetard d’illytes obstetricans Laur. Identification par immunofluorescence. C.R. Acad. Sci. Paris 167, 1581-1584. Reyrel, F. (1970). Developpement embryonnaire et larvaire de l’hypophyse de Pelobates cu1tripe.s (Batracien, anoure). Ann. Embiyol. Morphogen. 3, 263-272. Sternberger, L. A., Hardy, P. H., Cuculis, J. J., and Meyer, H. G. (1970). The unlabeled antibody enzyme method of immunohistochemistry preparation and properties of soluble antigen-antibody complex (horseradish peroxidase-antihorseradish peroxidase) and its use in identification of spirochetes. J. Histochem. Cytochem. 18, 315-333. Van Oordt. P. G. W. J. (1966). Changes in the pitu

ET AL.

itary of the common toad B&o bufo during metamorphosis and the thyrotropic cells. Z. Zellforsch. Mikrosk. Anat. 75, 47-56. Weibel, E. R. (1980). “Stereological Methods.” Vol. 1. “Practical Methods for Biological Morphometry.” Academic Press, New York. White, B. A., and Nicoll, C. S. (1981). Hormonal control of amphibian metamorphosis. In “Metamorphosis: A Problem in Developmental Biology” (L. I. Gilbert and E. Frieden, Eds.), pp. 363-396. Plenum, New York. Yamamoto, K., and Kikuyama, S. (1981). Purification and properties of bullfrog prolactin. Endocrinol. Jupon. 28, 59-64. Yamamoto, K., and Kikuyama, S. (1982a). Radioimmunoassay of prolactin in plasma of bullfrog tadpoles. Endocrinol. Japan. 29, 159-167. Yamamoto, K.. and Kikuyama, S. (1982b). Effect of prolactin antiserum on growth and resorption of tadpole tail. Endocrinol. Japon. 29, 81-85. Yamamoto, K., Kobayashi, T., and Kikuyama, S. (1986a). Purification and characterization of toad prolactin. Gen. Comp. Endocrinol. 63, 104-109. Yamamoto, K., Niinuma, K., and Kikuyama, S. (1986b). Synthesis and storage of prolactin in the pituitary gland of bullfrog tadpoles during metamorphosis. Gen. Comp. Endocrinol. 62, 247-253. Zuber-Vogeli, M.. and Doerr-Schott, J. (1984). Localisation par immunofluorescence de differents principes hormonaux de l’hypophyse de Nectophrynoi’des occidentalis Angel, au cours de developpement embryonnaire. Gen. Comp. Endocrinol. 53, 264-271.