78, 18%188 (1990)
lmmunohistochemical Localization of Corticotropin-Releasing Factor- and Arginine Vasotocin-like lmmunoreactivities in the Brain and Pituitary of the American Bullfrog (Rana catesbeiana) during Development and Metamorphosis JAMESA. CARR’ AND DAVID 0. NORRIS Laboratory of Comparative Endocrinology, Department of Environmental, Organismal and Population Biology, University of Colorado, Boulder, Colorado 80309-0334 Accepted June 22, 1989 Immunoperoxidase staining for corticotropin-releasing factor (CRF) in the median eminence was sparce or absent from premetamorphic tadpoles, but increased dramatically by late prometamorphosis. Quantitative photometry revealed that CRF-like immunostaining material in the median eminence was most dense in metamorphic climax tadpoles. Arginine vasotocin (AVT)-like immunostaining material was visualized in perikarya of the magnocellular nucleus, with extensive fiber staining seen in the medial basal and infundibular hypothalamus as well as in the median eminence and pars nervosa of the pituitary. AVT-like immunoreactive perikarya were virtually absent in premetamorphic tadpoles, but their number increased greatly by Taylor-Kollros stage XII and continued to increase after this stage. Quantitative photometry revealed that AVT-like immunoreactivity in the pars nervosa increased greatly at Taylor-Kollros stage XII and remained intense after this stage. AVT-like immunoreactivity did not appear in the median eminence until Taylor-Kollros stage XVI. Localization of AVT-like immunoperoxidase staining around portal vessels in the median eminence suggests an anatomical mechanism for delivery of AVT to anterior pituitary corticotropes. These results indicate that both CRF and vasotocinergic neuronal systems develop just before the activation of inferrenal steroidogenesis which occurs during the later stages of metamorphosis in this species. o WO Academic F-MS, IIK.
Steroidogenesis in the amphibian interrenal gland is generally believed to be under stimulatory control by adrenocorticotropin (ACTH) secreted from the pars distalis of the pituitary gland. Hypophysectomy results in atrophy and decreased steroidogenic capacity of the interrenal gland while replacement treatment with mammalian ACTH generally reverses these effects (Dongen et al., 1966; Kemenade et al., 1968, Buchmann et al., 1972; Liversage and Price, 1973; Laub et al., 1975; Mazzi et al., 1984; cf. Holmes and Ball, 1974). ACTHlike immunoreactivity has been demon’ To whom reprint requests should be addressed at University of New Mexico, Department of Anatomy, School of Medicine, Albuquerque, NM 87131.
strated in cells of the rostral pars distalis in all species examined (cf. Holmes and Ball, 1974), and chromatographic evidence suggests that amphibian ACTH is similar in structure to the mammalian peptide (Dot-es et al., 1989).
Little is known about the regulation of ACTH production and release in amphibians. In mammals, ACTH release is stimulated by a hypothalamic factor recently identified as the 41-residue peptide corticotropin-releasing factor (CRF) (cf. Antoni, 1986). Lesioning studies by Buchmann et al. (1972), Notenboom et al. (1976), and later Vellano et al. (1985) suggested that ACTH release in amphibians is similarly regulated by a hypothalamic factor. Using antisera raised against mammalian CRF, 180
001~6480&O $1.50 Copy&M Au tights
0 1990 by Academic Press, Inc. of reproduct& in any form reserved.
several investigators have successfully identified the immunoreactive distribution of this peptide in the hypothalamus of the amphibian (cf. Peter, 1986). CRF is a potent releaser of ACTH from the anuran pituitary in vitro (Tonon et al., 1986 in Rana ridibundu), while in vivo treatment with this peptide is effective in elevating plasma levels of the interrenal steroid corticosterone in the newt, Taricha granulosa (Moore and Miller, 1984). The neurohypophysial hormone arginine vasopressin (AVP) also regulates release of ACTH from the mammalian pituitary (cf. Antoni, 1986). There is indirect evidence that the homologous peptide in amphibians, arginine vasotocin (AVT), plays a similar role in stimulating ACTH release from the amphibian pituitary (Tonon et al., 1986). AVT-like immunoreactive perikarya and efferents are present in the preoptic hypothalamus and median eminence of the frog, R. ridibunda (Tonon et al., 1985), while in vitro treatment of frog pituitary corticotropes with neurohypophysial hormones stimulates ACTH release (Tonon et al., 1986). The later stages of amphibian metamorphosis are characterized by a dramatic increase in steroidogenic activity of the interrenal gland, as gauged by measurement of steroidogenic enzymes as well as plasma levels of interrenal hormones (Hsu et al., 1980; Jaffe, 1981; Kikuyama et al., 1986; Carr and Norris, 1988). The specific mechanisms regulating this burst in interrenal secretory activity remain unresolved. The purpose of the present study was to test the hypothesis that activation of interrenal steroidogenesis during spontaneous metamorphosis is accompanied by the differentiation of CRF- and AVT-like neuronal systems in the tadpole brain. METHODS Animals. Premetamorphic and prometamorphic bullfrog larvae were obtained from Charles Sullivan
Inc., Nashville, Tennessee, in April and May of 1988, respectively. Late prometamorphic tadpoles, obtained in June 1988, were allowed to undergo metamorphosis spontaneously in the laboratory. Larvae were maintained on a light regime of 12L:12D in 6 x 4-ft tiberglass tubs containing aged, aerated tap water. Boiled spinach and a prepared food (Frost, 1982) were available to larvae ad libitum. Tadpoles were staged according to the method of Taylor and Kollros (1946). Immunoperoxidase staining. Premetamorphic (stages V-K), prometamorphic (stages XII-XIX), and climax (stages XX-XXIII) tadpoles were weighed and sacrificed by rapid decapitation. Superficial cranial structures were removed from the skull and the brain fuced in situ in Bouin’s fixative. Brains fwed in Bouin’s fmative were processed for para& embedding and sectioned sagitally at 10 urn. Serial sections for a single brain were mounted sequentially on one to two series of six slides apiece, beginning on the first slide and ending on the sixth for each series. Mounted sections were air dried in an oven set at 35” for at least 48 hr before staining. Immunoreactive CRF-like and AVT-like materials were visualized using the avidin biotinylated-peroxidase method employing an ABC kit from Vector Laboratories. Briefly, tissue sections mounted on glass slides were deparaffinized, rehydrated, and then exposed to a trypsin solution (0.3 mg/ml) for 3 min followed by several rinses in 0.1 M phosphate-buffered saline (PBS, pH 7.5). Slides were then incubated for 30 min with normal goat serum (Vector Laboratories), drained, and incubated with primary rabbit antiserum (l/250-1/1000 in PBS) raised against CRF (purchased from Chemicon, with human CRF as the antigenic determinant) or AVT (Dr. R. E. Jones, Boulder, Co., and Dr. G. Ervin, UCLA Medical School) for 30-60 min at room temperature. Slides were then rinsed in PBS and incubated successively for 30 min at room temperature with biotinylated goat anti-rabbit IgG (l/ 1000 in PBS) followed by avidin biotinylated-peroxidase complex (Vector Laboratories). After rinsing in PBS, sections were treated with a solution containing 0.4% DAB (diaminobenzidine tetrachloride, Sigma Chemical Co.) in Tris buffer (pH 7.2) and 0.08% hydrogen peroxide. Sections were then dehydrated and coverslips mounted with Pet-mount. The CRF antiserum used in this study does not cross-react with Lys-vasopressin, Arg-vasopressin, Met- or Leu-enkephalin, ACTH l-39, or 8-endorphin. AVT antiserum showed no cross-reactivity with AVP and slight cross-reactivity with mesotocin as determined by RIA (personal communication, Dr. G. Ervin, UCLA Medical School). Blocked controls were performed by preincubating either antiserum with 1Opg human CRF (Chemicon) or 50 ug AVT for 18 hr. Quantification of immunoperoxidase staining. Immunoperoxidase staining in the median eminence and
pituitary was quantified with the aid of a standard darkroom photometer (Olympus EMM-7) attached to an Olympus microscope using a procedure modified after the method of Saland et al. (1988). One tissue section per animal, taken midway through the pituitary or median eminence region, was examined using a 40x objective. Light intensity from the median eminence or neural lobe of the pituitary was then recorded on the light meter. To control for differences in nonspecific staining, background light intensity from the anterior lobe was also measured for each section and the value was subtracted from the recorded value for the median eminence or neural lobe. The overall reading was then converted to a positive number since light meter readings were reduced with more intense staining and nonspecific staining of the anterior lobe was generally very low. Statistics. Variance around group means was tested using a one-way ANOVA procedure. Differences between group means were tested using Fisher’s test for least significance. An Q value of 0.05 was used for all tests.
photometry (Fig. 2) revealed that when pooled together, the median eminentia of tadpoles in metamorphic climax (XXXXII) and late prometamorphosis (XVIIXIX) showed significantly greater staining (F = 12.78, df = 17, P < 0.0005) than those of premetamorphic tadpoles. There was no statistical difference in the staining of median eminentia between tadpoles in prometamorphosis and metamorphic climax. AVT-like immunostaining during spontaneous metamorphosis. There was an AVT-
like immunoreactive signal present in the pars nervosa of tadpoles from all developmental stages that was absent from both the pars intermedia and pars distalis (Fig. 3). Intensity of the AVT signal in the pars nervosa increased during prometamorphosis and metamorphic climax (Fig. 4). Staining of the AVT antibody was blocked by prior RESULTS incubation with 50 p,g AVT but not by preCRF-like immunosiaining during spontu- incubation with 10 pg CRF. neous metamorphosis. During metamorAVT-like immunoreactivity was also phosis there was a gradual and generalized seen in the median eminentia of prometathickening of the median eminence. CRFmorphic and climax tadpoles (Figs. 3 and like material in the median eminence and 4). AVT-like immunostaining fibers and terinfundibular stalk was visualized as a dark minals in the median eminence were contadbrown precipitate (Fig. 1). No staining was sistently absent from premetamorphic seen in the pars intermedia, pars nervosa, poles but were present by stage XII and or other areas of the brain. CRF staining in were seen in all prometamorphic and clithe median eminence was abolished by max tadpoles (Fig. 4). AVT-like immunoprior incubation of the antiserum with 10 p,g peroxidase staining in the median eminence human CRF but not preincubation with 50 could be seen located around portal vespg AVT. sels, suggesting an anatomical mechanism Although CRF-like material was present for delivery of AVT to anterior pituitary in the median eminence as early as stage corticotropes . VI, CRF-like immunostaining in the median In the brain of prometamorphic and clieminence of premetamorphic tadpoles was max stage tadpoles, labeled fibers (Fig. 5) clearly less intense than that of prometawere observed projecting caudally toward morphic and climax tadpoles (Fig. 1). CRFthe infundibular hypothalamus while a roslike material was restricted to the external tral projection appeared to course through zone of the median eminence in all groups the ventral telencephalon. AVT-like immuand was found only in the most medial sec- nopositive fibers were seen in the brain of a tions of the median eminence in premetafew premetamorphic tadpoles, and they apmorphic tadpoles, while there was a grad- peared to increase in number as metamorual dispersal of CRF-like material to more phosis progressed. lateral areas of the median eminence as AVT-like immunoreactive perikarya metamorphosis progressed. Quantitative were localized in the dorsal (Mgd) and ven-
FIG. 1. Immunoperoxidase staining for CRF-like material in median eminence (arrows) of climax (stage XX, A) and absence of CRF-like immunoreactivity in premetamorphic (stage V, B) tadpoles. Note the lack of staining in any of the pituitary regions (AL, anterior lobe; IL, intermediate lobe; NL, neural lobe). Original magnification = 33X. (C, D) Higher magnification showing CRF-like material (arrows) in median eminence of prometamorphic (stage XVIII, C) and climax (stage XX, D) tadpoles. Note the greater intensity of staining in the climax condition. c, capillary. Original magnification = 132x.
tral (Mgv) regions of the magnocellular nucleus (Mg) in the caudal portion of the preoptic hypothalamus (Fig. 5). AVT-like immunoreactive perikarya in the Mgd formed
a thin band of cells delineating the ventral borders of vertically laminated cell groups in the ventromedial nucleus. An immunoreactive signal for AVT was seen in cells of
0.6 .zul 2 5E
0 ii f 0.2
FIG. 2. Graph demonstrating the results from photometer readings of CRF-like material in the median eminence during metamorphosis. Bars represent a single value or the mean of two values for each stage. When pooled together, tadpoles in climax (stages XXXXII) and late prometamorphosis (XVII-XIX) had signiticantly more intense immunoperoxidase staining in the median eminence than premetamorphic tadpoles.
the Mg nucleus in only a few premetamorphic tadpoles, although AVT-immunopositive perikarya were present in greater number by stage XII and increased in number after this stage (Fig. 6).
FIG. 3. Immunoperoxidase staining for AVT in median eminence (arrows) and neural lobe of climax (stage XX, A) and the absence of AVT-like immunoreactivity in premetamorphic (stage V, B) tadpoles. Note the lack of staining in other pituitary regions (AL, anterior lobe; IL, intermediate lobe; NL, neural lobe). Original magnification = 33x.
DISCUSSION This is the first study to examine the localization of CRF- and AVT-like material in the brain and pituitary of a larval anuran. The results presented here indicate that maturation of the CRF-like and AVT-like fiber systems occurs during metamorphosis. The time course of CRF-like immunostaining in the median eminence corresponds well with the pattern of interrenal steroidogenesis seen during metamorphosis (Hsu et al., 1980), providing indirect support for the argument that activation of interrenal steroidogenesis during metamorphosis is a result, at least in part, of maturation in the CRF fiber system. The
observation that the greatest intensity of CRF-like immunostaining in the median eminence was observed in tadpoles undergoing metamorphic climax, a period when endogenous circulating thyroid hormone levels are maximally elevated (Suzuki and Suzuki, 1981), lends support, albeit indirect, to the idea that thyroid hormones mediate maturation of the CRF fiber system during natural metamorphosis. Development of the AVT-like immunoreactive fiber system also corresponds well with the known effects of AVT on water balance in amphibians. Administration of AVT in vivo results in water uptake from
BRAIN CBF AND AVT IN TADPOLE
B 0.6 a .Z s f $ E 0 5 f
0.4 0.2 00 _._ >---x=-==xx-== >SZmx> >
Taylor-Kollros Stage FIG. 4. Graph demonstrating photometer results for AVT-like immunoperoxidase staining in pars nervosa (A) and median eminence (B) during metamorphosis. Bars represent a single value or the mean of two values for that stage. AVT staining was not detectable in the median eminence until stage XVI.
the skin and bladder of most amphibians (Bentley, 1971). Responsiveness of bullfrog tadpoles to AVT depends upon developmental stage: young larvae show little response to AVT (as judged by water uptake), while preclimax tadpoles show a much greater response (Bentley and Greenwald, 1970). Thus, the ability to retain water under the influence of AVT develops at a period when the animal begins to experience more desiccating environments. Localization of CRF-like immunostaining in the most external zone of the median eminence of bullfrog tadpoles stands in
FIG. 5. (A) AVT-like staining in fibers in medial basal hypothalamus of a late prometamorphic (stage XVIII) tadpole. Fibers were first observed by stage XII and were observed in all later stage animals. Original magnification = 132x. (B) AVT-inununostaining perikarya in a late prometamorphic tadpole. Labeled perikatya were most commonly seen in the ventral arm of the magnocellular nucleus and were first seen at stage XII @‘OR, preoptic recess). Original magnitication = 33 X .
6. Graph depicting the change in AVT-like immunostaining perikarya in the magnocellular (Mg) nucleus during metamorphosis. Labeled cells were counted at 70-p.m intervals through the Mg nucleus. Points represent a single value or an average of two values for each stage. FIG.
good agreement with the location of CRFlike material in the median eminence of adult bullfrogs (Gonzalez and Lederis, 1988) and adult R. ridibundu (Olivereau et al., 1987). Absence of CRF-like material from the pars intermedia and pars nervosa of bullfrog larvae is supported by a similar observation in adult bullfrogs (Gonzalez and Lederis, 1988). Localization of AVT-like material in the Mgd and Mgv nuclei correlates with the retrograde transport study of Pasquier et al. (1980) demonstrating that fiber tracts ending in the pars nervosa originate in the dorsal and ventral arms of the Mg nucleus. This observation also tends to support the contention of Neary and Northcutt (1983) that the Mgd not be included as part of the ventromedial nucleus, as AVT-like material was clearly absent from cell bodies in the ventromedial nucleus. The observation that AVT-immunostaining fiber tracts leading from the Mg could be traced en route to the
infundibular hypothalamus and median eminence further supports the Mg nucleus as a site of AVT production in the bullfrog. However, the fact that this antiserum cross-reacted slightly with mesotocin (as determined by RIA) suggests that some of the perikarya and fiber tracts showing an immunopositive reaction could be mesotonergic since mesotocin-like material is found in perikarya and fiber tracts of the hypothalamus in other amphibian species (Tonon et al., 1985). The time course of immunostaining in the Mgv and Mgd nuclei during metamorphosis corresponds well with the appearance of intenselystaining,AVT-likeimmunoreactivity in the median eminence and pars nervosa. AVT-like immunoreactive perikarya are virtually absent in the premetamorphic hypothalamus, corresponding with relatively light AVT-like immunostaining in the pars nervosa and the complete absence of AVTlike material in neurosecretory fibers of the median eminence. However, by early prometamorphosis there was a dramatic increase in the number of AVT-like immunoreactive perikarya in both arms of the Mg nucleus, consonant with the appearance of AVT-like immunostaining in the median eminence, and a consequent increase in the intensity of AVT-like immunostaining in the pars nervosa. ACKNOWLEDGMENTS Portions of this work were supported by a Grantin-Aid of Research from Sigma Xi and a Doctoral Fellowship from the University of Colorado. This work was submitted by J.A.C. as partial f&ilhnent for the Ph.D. degree at the University of Colorado. Thanks to T. Gleeson, J. Hanken, and R. E. Jones for critical review of this manuscript, to R. M. Dores for expert technical advice on immunocytochemistry, and to L. C. Saland for use of her Olympus microscope.
F. A. (1986). Hypothalamic control of adrenocorticotropin secretion: Advances since the dis-
covery of 41-residue corticotropin-releasing factor. Endoc. Rev. I, 351-318. Bentley, P. J. (1971). “Endocrines and Osmoregulation.” Springer-Verlag, New York. Bentley, P. J., and Greenwald, L. (1970). Neurohypophysial function in bullfrog (RUM catesbeiuna) tadpoles. Gen. Comp. Endocrinol. 14, 412-415. Buchmann, N. B. Spies, I., and Vijayakumac, S. (1972). Hypophysial corticotropic function and its hypothalamic control in Bufo byfo evaluated by the plasma concentration of corticosterone. Gen. Comp.
Can-, J. A., and Norris, D. 0. (1988). Interrenal activity during metamorphosis of the tiger salamander, Ambystoma 71, 63-69.
Dongen, W. I. van, Jorgenson, C. B., Larsen, L. O., Rosenkilde, P., Loft, B., and Oordt, P. G. W. J. van (1966). Function and cytology of the normal and autotransplanted pars distalis of the toad, Bufo bufo. Gen. Comp. Endocrinol. 6, 491-500. Dores, R. M., Meza, J. C., Schenk, L. M., Carr, J. A., and Norris, D. 0. (1989). Detection of adrenocorticotropin-related and a-melanocytestimulating hormone-related substances in the anterior pituitary of larval and adult Ambystoma tigrinum (class: Amphibia). Endocrinology 124, 1007-1016.
Frost, J. S. (1982). A time-efficient, low cost method for the laboratory rearing of frogs. Herp. Rev. 13, 73-77.
Gonzalez, G. C., and Lederis, K. (1988). Sauvaginelike and corticotropin-releasing-factor-like immunoreactivity in the brain of the bullfrog (Runu catesbeiana).
Tissue Res. 253, 29-37.
Holmes, R. L., and Ball, J. N. (1974). “The Pituitary Gland, a Comparative Account.” Cambridge Press, London. Hsu, C. Y., Yu, N. W., and Chen, S. J. (1980). Development of A5 -3 8-hydroxysteroid dehydrogenase activity in the interrenal gland of Runa cutesbeiuna. Gen. Comp. Endocrinol. 42, 167-170. Jaffe, R. C. (1981). Plasma concentration of corticosterone during Runu catesbeiunu tadpole metamorphosis. Gen. Comp. Endocrinol. 44, 314-318. Kemenade, J. A. M. van, Dongen, W. J. van, and Oordt, P. G. W. J. van (1968). Seasonal changes in the endocrine organs of the male common frog, Ranu temporuriu. II. The interrenal tissue. Z. Zellforsch.
Kikuyama, S., Suzuki, M. R., and Iwamuro, S. (1986). Elevation of plasma aldosterone levels of tadpoles at metamorphic climax. Gen. Comp. Endocrinol.
Kntg, E. C., Honn, K. U., Battista, J., and Nicoll, C. S. (1983). Corticosteroids in serum of Runu catesbeiuna during development and metamorphosis. Gen. Comp. Endocrinol. 52, 232-241. Laub, J. M., Callard, G. V., and Callard, I. P. (1975). The role of adrenal steroids in the negative feedback control of the amphibian interrenal gland. Gen.
Liversage, R. A., and Price, B. W. (1973). Adrenocorticosteroid levels in adult Diemictylus viridescens plasma following hypophysectomy and forelimb amputation. J Exp. Zool. 185, 259-264. Mazzi, V., Colucci, D., Andreoletti, G. E., and Vellano, C . (1984). ACTH dependency of aldosterone incretion in the newt, Triturus cristatus carnifex Laur. Accad. Sci. Torino 118, 339-343. Moore, F. L., and Miller, L. J. (1984). Stress-induced inhibition of sexual behavior: Corticosterone inhibits courtship behaviors of a male amphibian (Turicha
Neary, T. J., and Northcutt, R. G. (1983). Nuclear organization of the bullfrog diencephalon. J. Comp. Neurol.
Notenboom, C. D., Terlou, M., and Maten, M. L. (1976). Evidence for corticotropin-releasing factor synthesis in the preoptic nucleus of Xenopus laevis tadpoles. Cell Tissue Res. 169, 23-31. Olivereau, M., Vandesande, F., Boutique, E., Ollevier, F., and Olivereau, J. M. (1987). Immunocytochemical localization and spatial relation to the adenohypophysis of a somatostatin-like and a corticotropin-releasing factor-like peptide in the brain of four amphibian species. Cell Tissue Res. 241, 317-324. Pasquier, D. A., Cannata, M. G., and Tramezzani, J. H. (1980). Central connections of the toad neural lobe as shown by retrograde neuronal labelling: Classical and new connections. Brain Res. 195, 37-45.
Peter, R. E. (1986). Vertebrate neurohormonal systems. In “Vertebrate Endocrinology: Fundamentals and Biomedical Implications,” Vol. 1, “Morphological Considerations” (P. K. T. Pang and M. P. Schreibman, Eds.), pp. 57-104. Academic Press, New York. Saland, L. C., Gutierrez, L., Kraner, J., and Samora, A. (1988). Corticotropin-releasing factor (CRF) and neurotransmitters modulate melanotropic peptide release from rat neurointermediate pituitary in vitro. Neuropeptides 12, 59-66. Suzuki, S., and Suzuki, M. (1981). Changes in thyroidal and plasma iodine compounds during and after metamorphosis of the bullfrog, Rana cufesbeiuna. Gen.
CARR AND NORRIS
Taylor, A. C., and Kollros, J. J. (1946). Stages in the normal development of Rana pipiens larvae. Anat. Rec. 94, 7-23. Tonon, M. C., Burlet, A., Lauber, M., Cuet, P., JCgou, S., Gouteux, L., Ling, N., and Vaudry, H. (1985). Imrnunohistochemical localization and radioimmunoassay of corticotropin-releasing factor in the forebrain and hypophysis of Rana ridibunda. Neuroendocrinology 40, 109-l 19. Tonon, M. C., Cuet, P., Lamacz, M., Jegou, S., C&e, J., Gouteux, L., Ling, N., Pelletier, G., and
Vaudry, H. (1986). Comparative effects of corticotropin-releasing factor, arginine vasopressin, and related neuropeptides on the secretion of ACTH and a-MSH by frog anterior pituitary cells and neurointermediate lobes in vitro. Gen. Comp. Endocrinol. 61, 438-445. Vellano, C., Andreoletti, G. E., Mazzi, V., Collucie, D., and Peyrot, A. (1985). Effect of permanent deafferentiation of the anterior preoptic area on serum aldosterone levels in the crested newt. Gen. Comp. Endocrinol. 60, 104-108.